Man, Technology and Risk 
A Study on Intuitive Risk Assessment 
and Attitudes towards Nuclear Energy 
by 
O. Renn 
This translation was undertaken by the 
Tec:hnik-Sprachendienst G. Beyer GmbH 
Aachener StraBe 340 
D-5000 Koln 41 
and they are responsible for the accuracy af the text 
- I -
Abstract 
Using the instruments of empirical social sciences, a 
crass-section study was conducted comprising experLments 
on qualitative risk characteristics, in-depth interviews 
on mechanisms of risk perception and representative 
surveys of the public on technical risk sources, in par-
ticular with regard to nuclear energy. The results of 
these studies show that person-related expectations in 
respect of risk consequences, the possibility of personal 
influencing control, the severeness of risk consequences 
and one's own risk propensity playa significant role in 
the evaluation of risks. 
- II -
Introduction 
Risk - a multifarious concept: A gambler hoping to win, 
an actuary computing 11fe expectations, an industrialist 
estimating his marketing opportunities, a family evaluat-
ing the future usefulness of a consumer product, a pa-
tient visualizing the success or failure of surgery, a 
technician investigating the probabilities of skiing 
accidents or malfunctions - all of them start out from 
a eommOn concept : risk. But do they always mean the 
same thing? 
What is risk? The scientist has a clear-cut and fast answer 
,. 
to this question: Risk is the sum total of all probabilities, 
P, multiplied by the valuations of possible consequen-
ces,c. It 1s on the basis of this formula that risk anal-
yses are computed for technological systems, nuclear pow-
er stations, road traffic, dams and bridges, but also 
for the side effects of phar.maceutlcals, for environmen-
tal pollution or for noxious substances. 
But can we use such a formula when it involves personal 
risk considerations when buying consumer goods, problems 
of the individual or collective acceptance of technologi-
cal risk sources, usefulness or absurdity of leisure oc-
cupation, or the psyche of compulsive gamblers? Certain-
ly notl But to this date psychology, the science respon-
sible for this field, has rather neglected the topic of 
risk. Many works on decision-making theory, as stated 
by the social psychologists Janis and Mann, do not 
even mention the risk concept. This has probably been 
due to the fact that this word is used to define so many 
different ideas that its definition has become vague. 
- III -
The present study represents an attempt to bridge 
this gap in psychological research and to furnish a con-
tribution toward a psychology of risk. Such an under-
taking of course requires the selection of an exemplary 
risk source, unless the investigator accepts being bogged 
down in a multitude of examples and their peculiarities. 
Nuclear energy was selected as this exemplary risk source. 
Why? There is hardly any technological facility which has ever 
stirred the consciences of the western World as nuclear 
power stations and their associated plants. Hardly any 
facility has become so associated with risk, and 
hardly any technology has so shaped the perception 
of our industrial culture. The selection of this example 
is less a tribute to the topicality of this conflict than 
a calculus in the researcher's ter.ms of reference, 
using the past discussion on the nuclear energy risk and 
its rational and emotional assimilation by the populace 
as a data base for an empirical study. For in this exam-
ple, attitudes already exist, and object perception struc-
tures are well developed. 
These advantages of object selection can be gained only 
by accepting an important disadvantage. In a phase of po-
litical confrontation between opponents and supporters, 
studies in the social sciences often have only a rational-
izing character. They are used either to substantiate 
one's own judgement of this issue via a social science 
analysis, or to provide one's reference group with data 
and arguments for better armament 1n its conflict with the 
opponent. Neither was our intention. In order to avoid 
the danger of allowing our own attitudes to become part 
of the interpretation, this study includes only analyses 
on perception processes and attitude structures of the 
survey respondents, without any attempt at evaluation or 
criticism of the responses. Of course, this approach does 
not preclude that concepts such as risk, risk estimation and 
risk perception are subjected to conceptual clarifica-
tion. 
- IV 
In spite of the exemplary approach the discussion 1s not 
confined to the nuclear energy example. The theoretical 
analysis and its empirical translation include other risk 
sources from pharmaceuticals to skiing as comparison cri-
teria, 1n order to cover higher-level patterns of risk 
perception and risk assimilation. Comparative analysis 
of different ~lsk sources broadens the spectrum of the 
perception fields analyzed and allows the discussion of 
results covering several objects. This book describes, 
1n concise and easily understandable form,the most impor-
tant results and conclusions of more than two years of 
research in this fleld . For those readers particular-
ly interested 1n this subject, the final report on this 
research project has been published in six volumes in 
the form of JU1-Reports by the JUlich Nuclear Research 
Centre. Most of the Figures have been adopted from this 
comprehensive version. 
The book is organized into five major chapters: Part I 
presents a critical discussion and summary of the exist-
ing literature on risk perception problems, intuitive 
technology evaluation and nuclear energy attitudes. 
Part II provides a review of our own theoretical approach 
and the basic concept used for the empirical analysis. 
In spite of all our attempts to achieve easy understand-
ing in terms of phraseology and terminology, the two 
chapters are characterized by social science concepts 
and, therefore, may be somewhat difficult to follow for 
readers not experienced in this field of research. There-
fore, all the important methodological stipulations are 
repeated once again in the presentation of the empirical 
resul ts in Parts III - V, so that readers who are inter-
ested only in the results can pass over the theoretical 
chapters . I ask the reader's indulgence for the result-
ing, unavoidable redundance in the presentation. 
- v -
The third chapter is concerned with two social psychology 
experiments intended to contribute toward clarification 
of the intuitive evaluation of risky situations. Part IV 
serves the same purpose. It describes and interprets the 
results of two comprehensive surveys on the evaluation of 
different risk sources and their determinant factors. 
The last part 1s exclusively concerned with nuclear energy 
perception and an analysis of attitude components with 
respect to nuclear energy facilities and their deter.mlnlnq 
quantities. The results presented in this Part are based 
on a representative survey conducted in five towns 
in the State of North Rhine-Westphalia, a state within the 
Federal Republic of Germany. 
Finally, the book includes a brief synopsis of the entire 
study and an index listing and defining the most important 
concepts . 
Publication of this book would have been impossible without 
the cooperation of Prof. Dr. G. Wiswede, Prof. Or. E. K. 
Scheuch and Dr. E. MUnch. I am obliged to Professors 
Wiswede and Scheuch, who supported part of this study as 
a doctoral thesis, for their critical reviews and many 
suggestions and improvement proposals. My special appreci-
ation is owed to Dr. E. MUnch for his support in the 
realization of the research project described in this book 
and for his vigorous support of my work. I am equally 
obliged to the late Mr. Beyer for the excellent job of 
translating the German original version into English. 
Finally, my special gratitude is extended to Mrs. Tanz, who at 
considerable effort produced this book from my handwritten 
manuscript. 
JUlich, January 1981 
Dr. O. Renn 
Table of Contents 
Abstract 
Introduction 
Table of Contents 
List of Figures 
Foreword 
Swmnary 
Part I 
- VI -
Page 
I 
II 
VI 
IX 
XII 
XV 
Risk Perception and Acceptance : 1 
A Bibliographical Review 
Systematic Summary of Different Research concepts 2 
Decision Analysis and Risk Theory Concepts 6 
Normative Decision-Making 6 
Descriptive Decision-Making 9 
conclusions on Risk Theory and DecisionAnalysis 19 
Economic Concepts 20 
Individual Psychology Concepts 22 
Social Psychology Concepts 26 
Sociological Concepts 41 
Open Questions and PhenomenaRequiri"ng Explanation 60 
Part II 
Risk and Risk Perception -
Basic Theoretical Concept 
Defini tions 
Risk Concept 
Risk Perception Concept 
Risk Acceptance Concept 
Probabilistic Risk Analysis for Technologies 
The Graduated Rationality Model 
63 
P,4 
64 
67 
70 
80 
88 
Atti tudes and Acceptance 94 
Description of Different Research Projects for 102 
Empirical Analysis 
Survey Instruments and Study Cases 104 
No. 
- VII -
Part III 
Socio-Psychological Experiments on Specific 
Risk Perception Problems 
- The Medical Capsule Experiment -
Experimental Setup 
- Results of the Medical Capsule Experiment 
- The Poisoned Water Experiment -
Experimental Setup 
- Results of the Poisoned Water Experiment 
Part IV 
Individual Risk Perception -
Result of Comprehensive Studies 
- Effect of Risk Levels (Expected Values) on the 
Evaluation of Risk Sources 
- Homogeneity of Risk-Benefit Estimates Made by 
Different Population Groups 
109 
110 
113 
116 
119 
125 
126 
142 
- Effect of Ideas on the Characteristics of the 149 
Individual Risk Sources (Belief System) 
- Effect of Qualitative Risk Characteristics 167 
- Effect of Risk Propensity and Benefit Orientation 185 
- Effect of Demographic and Social Characteristics 190 
- Overall Analysis of the Factors Influencing the 199 
Perception and Acceptance of Risks 
Part V 
Nuclear Energy Attitudes and their Determinants - 217 
Result of Sampling Surveys 
- Post-War Nuclear Energy Attitudes of the Population 218 
of the Federal Republic of Germany 
- Effect of the Three Mile Island Accident on the ~8 
population 
- Nuclear Energy Compared to Other Energy Sources 234 
- Attitudes Toward Nuclear Energy and their Internal 249 
Structure 
- Significance of Personal Value Preferences and ~4 
Socio-Political Beliefs for the Formation of 
Nuclear Energy Attitudes 
- Effect of Social and Demographic Characteristics 
on Nuclear Energy Attitudes 
- Nuclear Energy and Sex 
- Nuclear Energy Attitudes and Age 
275 
276 
278 
- VIII -
- Nuclear Energy Attitudes and Stratification 
- Nuclear Energy Attitudes and Political Party 
Preference 
- Nuclear Energy Attitudes and Other Social 
Categories 
278 
289 
- Confidence 1n Institutions and Other Reference 289 
Groups 
- Attitude of the Population to Specific Questions ~7 
of Nuclear Energy Use 
- Economic Growth and Energy Consumption 
- Attitude Toward Citizens' Initiatives 
297 
299 
- Perception of the Waste Disposal Problem 301 
- Summary of all Factors Influencing Nuclear Energy 304 
Attitudes 
- General Conclusion 
- 1. Information Reception 
- 2. Development of Beliefs 
- 3. Attribution of Probabilities 
- 4. Evaluation of Beliefs 
- 5. Development of a Consistent Attitude 
- 6. Acceptance Decision 
- 7. Rationalization 
- 8. Risk Perception and Social Research 
JO~ 
311 
312 
313 
314 
317 
318 
318 
319 
Index of Technical Terminology with Definitions 32~ 
References 335 
- IX -
List of Figures Page No. 
Fig. 1: Risk Acceptance Model in a Two-Factor 
Psychological Space (cf . R6glin) 
Fig. 2: Risk Acceptance as a Function of Risk 
Characteristics (cf. Fischhoff et al.) 
Fig. 3: The Perception of Various Risk Dimensions 
(Risk Evaluation of Nuclear Waste) 
Fig . 4: Attitude Dimensions According to Fishbein 
Model (IAEA Risk Assessment Group) 
Fig. 5: Types of Risk-Benefit Distribution 
Fig. 6: Outcome-Exposure Pathways 
Fig. 7: Methods to Determine Optimum Risk Minimization 
Fig . 8: The Model of Graduated Risk Perception 
Fig . 9: Model of the Risk Acceptance Weighing Process 
Fig. 10: Basic Attitude Concept 
Fig. 11: Results of the Medical Capsule Experiment 
27 
33 
35 
39 
73 
83 
85 
91 
95 
99 
111 
Fig . 12: Discrepancy Between True and Perceived Fatalities 129 
(U.S. Data acc. to Fischhoff) 
Fig . 13a: Discrepancy Between True and Pereeived Fatalities 
for an Average Year (German Data, Logarithmic Scale) 133 
Fig. 13b: Discrepancy Between True and Perceived Fatalities 
for a Catastrophic Year (German Data, Logarithmic 115 
Scale) 
Fig. 13c: Discrepancy Between True and Perceived Fatalities 
for an Average Year (German Data, Mixed Scale) 137 
Fig. 14: Mean ValUes of Three Independently Measured 
Risk-Benefit Estimations 
Fig. 15: The Estimation of Benefit and Risk - A Comparison 
Between German and American Data 
Fig. 16. Factor-Score Patters of the Belief Scale: 
Percentage of Declared Variance 
143 
153 
- x -
Fig. 17: Aggregate Belief Structure Factors (Summation 
of Factorial Loads) 
a) Alcohol, Smoking, Heroin 157 
b) X-Rays, Electrical Appliances, Skiing 159 
0) Automation, Motorcars, Air Traffic 161 
d) Pesticides, Nuclear Energy, Coal 163 
Fig. 18: Factor-Score Patterns of the Qualitative Risk 
Characteristics: Multiple Regression Coefficients 
FiS!. 19 I Comparison of German and American Values for 
Qualitative Risk Characteristics 
FiS!. 20: The Qualitative Risk Characteristics as a Function 
of Risk-Benefit Estimation and the Risk Evaluation 
FiS!. 21a: Aggregate Qualitative Risk Factors (Voluntariness 
Versus Equal Risk-Benefit Distribution) 
FiS!. 21b: Aggregate Qualitative Risk Factors (Voluntariness 
Versus Severeness of Consequences) 
FiS!.21c: Aggregate Qualitative Risk Factors (Equal Risk-
Benefit Distribution Versus Severeness of 
Consequences) 
FiS!. 22: Weighting of Belief Factors as a Function of 
Class Membership (Pretest Results) 
FiS!. 23: Weighting of Aggregate Belief Factors as a 
Function of Class Membership (Main Study) 
FiS!s.24a 
to 24e: Graphic Representation of the Multlple Regression 
Values of all Independent Variables for the" 
Semantic Differential of Nine Risk Sources 
Flga 25: Estimation of Nuclear Energy as a Personal 
169 
173 
175 
177 
179 
181 
195 
197 
201 
al3 
205" 
207 
209 
Preference for the Future, as a Function of the 239 
Belief System and the Semantic Differential for 
Nuclear Energy 
FiS!. 26: Discrepancy Between Desired and Expected Future 
of Nuclear Energy 241 
- XI -
Fig . 27: Attributive Profiles for Coal, Solar and 
Nuclear Energy (Semantic Differentials) 
Fig. 28: Relationship Between Risk-Benefit Estimation 
and Behavioural Intention 
Fig. 29: Path Analysis Model of Nuclear Energy Attitudes 
and Positions in a Referendum 
Fill· 30: Specific Pattern of General and Personal Values 
in a TWo-Factor Psychological Space 
Fill· 31: Effect of Socio-Political Attitudes in the 
Risk-Benefit Estimation of Nuclear Energy 
Fill· 32 : Attitude-Forming Variables as a Function of 
Stratum and Position 
247 
259 
261 
267 
273 
231 
- XII -
Foreword 
This book is primarily concerned with the perception of 
technological risk and, in particular, with attitudes to-
ward nuclear energy. In the case of this controversial 
topic there 1s the danger of sciencQ being forced into a mere-
ly palliative role of post-factum justification and ra-
tionalization of positions already taken. Therefore, I 
should like to assure the reader from the outset that the 
author of this book has exhibited a well-balanced and 
aloof attitude, and has succeeded well in avoiding this 
trap and communicating a great variety of differentiated 
information. 
The following problem areas are involved: 
(1) risk perception and intuitive risk assessment 1 
(2) development and structure of attitudes toward 
nuclear energy; 
(3) psychic and social influencing factors which 
are the causes of these attitudes. 
The author initially discusses perception and acceptance 
for different risk sources on a quite general basis. 
Risk acceptance 1s defined as a cognitive process of 
weighing between subjectively weighted benefits and de-
trimental consequences. Previous studies on risk accept-
ance are critically discussed and found to be excessively 
limited . The author's own approach leads to a phased mod-
el consisting of the following steps : Collection of rel-
evant cognitions on risks, attribution of subjective pro-
babilities to these cognitions, weighting of these proba-
bilities and comparison of these weighted cognitions. 
- XIII -
In order to determine the ranking of (naive) perception 
and acceptance of risks the author proposes an additional 
comparison to a type of objective risk measures, using 
the expert evaluations as a basis. But such a comparison 
is highly problematic since everyone knows how often the 
experts have been wrong in this branch of the natural 
sciences and how often they hold different views. More-
over, everybody knows that many imponderables enter into 
the quasi-objective risk evaluation of an expert. Apart 
from the problematic basic questions (Who is an expert? 
To what degree is he an expert? To which group does he 
belong?), and apart from the risk evaluations which are 
difficult to grasp and often can be weighted only arbi-
trarily, the problem in the final analysis involves a 
finite selection of possibilities and occurrences which 
can be included in the calculation and, finally, finite 
propagation paths with respect to detrimental consequen-
ces: it is impossible to calculate probabilities of the 
occurrence of events of which nobody has ever thought before. 
The author is quite aware of these reservations with re-
spect to "objectivity", but would not forgo this field 
of reference of the quasi-objective evaluation of the ex-
perts, because otherwise the naive perceptions of laymen 
and the calculations of experts would be compared on a 
level of equality. Rather, Renn attempts to develop the 
gradual differences in reality content by means of a mo-
del of graduated rationality in the perception of risks, 
which can certainly be considered a pace-setting achieve-
ment. 
The empirical analysis itself is concerned with percep-
tion processes of risk sources and with the development 
and structure of attitudes, specifically with respect to 
nuclear energy, as well as the internal and external fac-
tors which influence these attitudes. The implementation 
of the empirical study - ranging from its operationaliza-
tion, which was sometimes extremely difficult and complex, 
all the way to the well-balanced interpretation of the 
- XIV -
results - reveals mastery of the methodological instru-
ments of the social sciences. ~Some of the results con-
firm previous studies, while others go far beyond the 
present state of the art and reveal a great variety of 
interesting data. For instance, one important informa-
tion item 1s that nuclear energy opinions are highly af-
fected by emotions but are still ambivalent to the large 
majority. This result also appears to reflect interna-
tional aspects since comparable data from the United 
States and Japan also demonstate this ambivalence and 
rigidity of the opinion body. In contrast to the pic-
ture broadcast by the media it 1s also characteristic 
that most of the nuclear energy opponents feel isolated 
and that - in a resigned attitude? - they do not believe 
in the necessity of nuclear energy but are nevertheless 
convinced that it will come about in the Federal Republic 
of Germany, regardless of whether the citizens want it or 
not . Insofar it appears that these results represent an 
important correction to the widespread opinion that the 
opposition to nuclear energy is straight-lined and un-
compromising. 
The author has presented an empirical study which ranks 
at the very forefront of the presently topical scientific 
discussion and is probably the best-founded study of 
this subject at present. On the other hand, the author in 
his mastery of the sophisticated methods of empirical 
social research and, at the same time, of socia-psycho-
logical and sociological phenomena, proves to be a good 
advocate of interdisciplinary, problem-oriented research. 
Cologne, January 1981 
Prof. Dr. G. Wlswede 
Cologne University 
- xv -
Sununary 
Previous studies on risk perception by the population have 
been based predominantly on three methodological concepts: 
- the descriptive decision analysis model 
- themodel of risk-specific perception influence 
factors such as voluntariness or personal control, and 
- the attitude model. 
The main emphasis of studies conducted to date has been in 
the field of perception-specific risk approaches which, on 
the one hand, were developed quasi indirectly 1n the 
"Revealed Risk Analysis" from the history of the acceptance 
of risk sources (approach by Starr) or, on the other hand, 
measured by direct surveys with a view to their significance 
(approach by Fischhoff et al.). Studies on the attitude 
concept, which is largely pursued by the IAEA Risk Assessment 
Group 1n Vienna, relate more strongly to the cognitive 
structure of risk perception. Finally, decision analysis 
studies are presently being conducted at the University of 
Southern California (von Winterfeldt,Edwards) , dealing with 
the value conceptions of relevant groups in the nuclear 
controversy, where the perception of risk aspects is being 
understood as an attribution process of prior value com-
mitments to concrete risk objects. 
On the basis of these studies, our aim was to develop an 
integrative approach to the measurement of perception 
patterns for a variety of risk sources. The relevant 
theoretical starting pOint was a mental risk decision model 
intended to observe and empirically COver the process 
ranging from information uptake to the behaviour intention 
in respect of the object in question. The individual 
stations of risk perception were characterized as follows: 
- information uptake, 
- the forming of general beliefs, 
-~-
_ the attribution of probabilities according to the 
degree of one's own involvement and typical patterns 
of generalization, 
_ the evaluation of beliefs by means of attributive 
biases, qualitative risk characteristics and 
dispositive as well as sltuatlve factors, 
- the internal coordination of beliefs (consistence 
finding) 
- the acceptance decision, 
- the forming of an attitude and of behavioural 
intentions resulting therefrom, 
_ the immunization of the attitude against inward and 
outward doubts (rationalizing). 
For the empirical transformation of this concept, two 
different comprehensive interviews to elucidate belief 
patterns about various risks and their weights, two 80c10-
psychological experiments on the significance of qualitative 
risk characteristics and one representative survey on 
attitude formation and its determinants were conducted in 
the Federal Republic of Germany. Part of the questions 
were selected. so as to per.m1t direct comparisons with 
parallel American and French studies. 
The results of this very extensive study, comprising twa 
years of research, may be summarized as follows: 
- Most of those interviewed possess relatively homogeneous 
evaluation criteria for an intuitive risk assessment and 
arrive at similar results in the evaluation of risk-benefit 
relations for different risk sources. The classification 
of risk sources according to the level of their overall 
risk practically does not reveal any regional, temporary 
of stratum-specific differences. Even a comparison between 
the populations in the United States and the Federal 
Republic of Germany results in similar patterns of risk 
estimation. 
- XVII -
- Although most of those interviewed can reflect relatively 
well the expected values of losses (fatalities, injured 
persons, damage to property per year), these estimations 
are taken as evaluation criteria for the risk-benefit 
estimation to a minor extent only. The evaluation of 
risks is governed essentially by the following factors: 
- perceived consequences of a risk source for oneself 
or close associates, 
- perceived consequences relative generally to health, 
safety and future scope of freedom, 
- personal posslbl11ts of influencing the risk, 
- the possibility of far-reaching and fatal losses, 
- the personal readiness to take risks. 
The first two aspects were defined as cognitive, the 
following two aspects as qualitative risk characteristics, 
and the fifth as dispositive risk propensity. With the 
aid of these five variables, between 44 and 72% of the 
variance of the risk-benefit estimation could be explained, 
depending on the risk source. 
- Demographic and social characteristics play a subordinate 
role in the evaluation of risks. Only the sex of those 
interviewed has an influence on perception, however not 
directly, but via the devious path of 'special sensitivity 
to risk consequences relative to health and genetics" and 
"personal risk propensity". Cognate attitude patterns 
and general value orientation are of significance only 1f 
the associations with the risk source are concerned 
primarily with risk consequences for the economic,society 
and future life style. This 1s the case, for instance, 
in connection with nuclear energy and plant-protective 
agents, but not in connection with coal energy, computers 
(despite the problems of data protection) or x-ray 
equipment. 
- XVIII -
_ The nuclear energy risk has a particularly ambivalent 
perception structure due to attributive biases, on 
account of the negative, perceived qualitative risk 
characteristics (non-voluntary, high hazard potential) 
and because of the fear of long-term damage possibilities 
which cannot be recognized today. Although the majority 
of those interviewed was convinced of the fact that 
nuclear energy 1s an economic necessity and will play 
a significant role for the future supply of energy, they 
argue, however, that it involves completely new risk 
dimensions and cannot be calculated in a scientific 
analysis. 
- This ambivalence has left the basic proponents of nuclear 
power with a dissonant attitude structure, i.e. the 
beliefs of those interviewed are very inhomogeneous and, 
in part, contradictory. Their behaviour intentions are 
accordingly limited, and their doubts about themselves 
are particularly pronounced. 
- Determined opponents of nuclear energy, on the other hand, 
are neither convinced of the economic necessity nor of 
the harmlessness of the risk. They possess a very homo-
geneous attitude structure and, accordingly, a stronger 
behaviour motivation. They exhibit, however, a pro-
nounced resiqnative belief that "it is too late for the 
nuclear case now· and that, in the long run, they can no 
longer do anything to prevent the expansion of the 
nuclear energy sector. This pessimistiC view of the 
future was defined by us as reality stress . It often has 
a socially destabilizing effect, expressing itself, for 
instance, 1n a disintegration from society (withdrawal 
from society) or, On the other hand, by rigorous activism. 
- XIX -
- The possibilities of a meaningful dialogue between pro-
ponents and opponents of nuclear energy are by no means 
exhausted. Both the opponents and proponents of nuclear 
energy are still rather inclined to trust in the state-
ments of expert scientists as long as these cannot be 
clearly identified as representing the interests of 
industry. Moreover, 1n the case of the proponents, a 
special dependence on pro-nuclear reference groups 1s to 
be observed which, no doubt, is based on the fact that 
they, in particular, seek the support of others due to 
their ambivalent attitude structure. The opponents have 
less confidence 1n institutional reference groups and feel 
more strongly attracted to information groups (e.g. local 
environmentalists). 
As a recommendation for continuing the dialogue between 
proponents and opponents of nuclear energy, we consider 
it necessary that the topics of such conversation should 
rather be oriented to the perceived consequences of 
nuclear energy utilization than to the discussion of risk 
expected values, which are in any case little contested. 
- 1 -
Part I 
R1sk Perception and Acceptance: 
A Bibliographical Review 
- 2 -
Systematic Summary of Different Research Concepts 
The problems involved in the risk acceptance of nuclear 
facilities in almost all western industrial nations has 
resulted in an avalanche of psychological and social 
science literature devoted to the questions of the 
development, origins and consequences of the 
present risk acceptance crisis. In spite of this broad 
spectrum of available scientific analyses most of these 
studies have been limited to the application or transfer 
of sub-theories to the present-day situation and attempt, 
on the basis of a particular theoretical perspective, to 
describe the nuclear energy conflict and reveal its 
causes. On the other hand, generally oriented and inter-
disciplinary studies on a broad empirical basis are 
rarely encountered. 
A brief selection of some important research concepts 
and theories 
flicts will 
on risk acceptance and nuclear 
be presented in this chapter. 
energy con-
It goes 
without saying that a selection of this type can never 
be comprehensive or representative: rather, it 1s the 
objective of this description to provide a systematic 
review of the broad spectrum of scientific literature 
avai~able and comparative explanations of the intentions 
and results of studies accomplished to date. This re-
view does not include publications which also provide a 
summary of existing concepts but do not represent original 
theories of their own (cf., for instance, von der Che, 
179,288; Paschen, '79,307). [9uotattons are ordered 1n the 
follOWing manner: author, year of publication, number 1n 
alphabetic IndexJ In accordance with SCientific conventions 
and the personal categorization of the authors,the individual 
studies will be classified into the following scientific fields: 
- economics-oriented concepts: 
deCision analysis concepts: 
- risk theory concepts! 
- indiVidual psychology concepts; 
SOCial psychology concepts! 
- SOCiological and political SCience concepts. 
- 3 -
Table 1 contains a systematic classification of concepts and 
their most important proponents. This list,ef course,is incom-
plete, and the classification is often problematic. 
However, it appears suitable and sufficiently accurate 
for an initial orientation on the spectrum of possibilities. 
Table 1: 
Systematic Classification of ExplanatopY Concepts 
SCientific field 
of classification 
Economics 
Decision analysis 
Theoretical 
framework 
Marginal 
utility 
theory 
Economic 
theory of 
politics 
Normative 
risk assess-
ment 
Normative 
decision-
making 
theory 
Descriptive 
decision-
making theory 
Summary de-
scription of 
the concept 
Proponents 
(selection) 
Determination 
of marginal 
cost and util-
ity, comparison 
of individuals, 
groups and 
society 
Felix 
Renn 
Demand maximl- Downs, Frey, 
zatlon via Tltz 
resource 
mobilization, 
pol! tical in-
fluence 
Optimum proce-
dures for cost 
and benefit 
estimation, 
risk minimiza-
tion and selec-
tion of alter-
natives 
Optimization of 
procedures to 
guide decision-
makers to the 
most rational 
decision pos-
sible 
Definl tion of 
the determi-
nants of the 
factual deci-
sion-making 
process 
Rowe, Lowrance, 
Kates, 
Fischhoff, 
Sagan 
Coombs, Okrent, 
Raiffa,Edwards 
Janis/Mann, 
Pollatsek, 
Tversky, Vlek, 
Stallen,Kogan, 
von Winterfeldt 
Jungermann 
(contd.) 
,ntific field 
:lassification 
< theory 
:hology 
lal psychology 
- 4 -
Theoretical 
framework 
Revealed pre-
ference 
analysis 
Referred or 
expressed 
preference 
analysis 
Psycho-
analysis 
Psychological 
reduction 
theory 
Risk per-
ception 
Risk social-
ization 
Attitude 
concept 
Communica-
tion concept 
Symbolic re-
duction 
theory 
Summary de-
scription of 
the concept 
Past risks as 
indicators for 
the evaluation 
of new risks 
Empirical deter-
mination of risk 
elements through 
surveys 
Transfer of 
psychoanalytical 
concepts and ar-
chaic types to 
the underlying 
causes of nuclear 
energy attitudes 
Proponents 
(selection) 
Starr, Cohen 
Fischhoff, 
Slavic, 
Lichtenstein 
WHO Report, 
Schild, 
v . Erichsen, 
WUnschrnann, 
Tublana 
Transfer of Pahner, 
psychic mechanisms Pelicier 
to the perception 
of nuclear energy 
Perception 
effects in risk 
assessment (at-
tributve biases) 
Learning of 
interpretation 
patterns for risk 
evaluation 
Attitude toward 
the subject re-
quires risk 
acceptance 
Maynard, 
Tversky, 
Fischhoff, 
Slovlc,Vlek, 
Kogan, Bler-
brauer, 
Frantzen, 
Schmid-Jllrg 
(Battelle) 
Gutmann 
(Battelle) 
Cohen/Hansel 
Otway,Niehaus 
Davis, 
(v . Buiren) 
Fishbein 
The conflict is Goerke 
the result of (Eisenhart, 
misguided and Crebsbach) 
distorted communi-
catlon 
The nuclear energy 
conflict is rep- Rllglin, 
resentatlve for Dumenil, 
symbolic societal HofstKtter 
or psychological 
issues 
(contd.) 
Scientific field 
of classification 
Sociology 
- 5 -
Theoretical 
framework 
Sociology of 
the sciences 
Empirical-
functional-
istic concept 
System ana-
lysis concept 
Conflict 
theory con-
cept 
Summary de-
scription of 
the concept 
Conflicts are 
the result and 
expression of 
scientific 
orientation pro-
cesses 
Conflicts are 
functional/dys-
functional for 
future value 
orientation 
Crisis in the 
interchange 
relationship 
between nuclear 
system and polit-
ical periphery 
Dispute repre-
sents conflict 
of interests and 
participation 
Proponents 
(selection) 
Nowotny, 
Tschledel, 
Nelkln, 
Weingast 
LUbbe,Schoeck 
Douglin, 
Tognaco!, 
Battelle, 
BMFT II Study 
Gripp 
Normative- Conflicts lead 
Scharloth, 
Paschen, 
Andritzky 
Nelkin, 
Kitscheet 
Schumacher, 
Mo6mann, 
Gladitz 
democratic to more demo-
(participa- cracy 
tory research) 
Neo-Marxism 
Orthodox 
Marxism 
Conflicts are Offe, 
caused by clasp Metz 
contradictions 
and capitalistic 
crises 
Conflicts are Lob 
caused by the 
realization of 
monopolistic 
capital exploita-
ti"on interests 
The overall matrix seen from this Table exhibits a very 
broad spectrum of concepts whose internal logic frequently 
predetermines the approach and sometimes even the results. 
It would be beyond the scope of this study to discuss in 
detail each of these concepts and to subject them to critical 
- 6 -
examination. However, it seems important to sununarlze briefly the 
basic concepts underlying the existing studies and to 
highlight the spectrum of questions which have remained 
unresolved. 
Decision Analysis and Risk Theory Concepts 
~Q~~S!Y~_Q~g!~!Qn=~~~!ng 
Decision-making and risk theory studies are motivated 
either by the problem as to an optimum procedure for 
selecting the best option from a number of alternatives 
(normative theory), or . by the problem of describing the 
actual selection process, 1.e. describing how individuals, 
groups or institutions select an option (descriptive 
theory). In addition, reference 1s made to a so-called 
explicative theory which has been used to develop the 
mathematical and logical framework for rational decision-
making and sometimes also to a typological theory which, 
in a manner of speaking, involves a systematic identifica-
tion and classification of all types and possibilities 
of decision-making. 
The distinction between normative and descriptive or 
explicative and tyPological deCision analysis 
should not be understood as implying strict compart-
mentalization. Even the development of proposals for 
rational decision-making cannot forgo the subjective 
evaluat.ion of alternatives for action.. Also, the descrip-
tion of actual behaviour must be based on certain logical 
assumptions of behaviour because otherwise the 
typification or generalization of behaviour patterns 
would be impossible. Rather, this distinction applies 
to the purpose of a study: either to give decls1on-
making aids or to define existing forms of decision-making. 
- 7 -
But how do these theories relate to the problem as to 
the 
ing 
risk acceptance of nuclear energy? 
a definition of risk acceptance at 
Without present-
this early point 
it appears to be intuitively eVident that any con-
sideration as to the acceptance of risk represents a 
decision-making problem. Is the probability of negative 
effects worth the benefit associated with the source of 
risk? In order to respond to this question the likely 
procedure is to collect the ris k aspects, to weight them 
with the probability of their occurrence and to compare 
them to the benefit aspects . The result of this balancing 
process 1s a risk acceptance decision . 
Let us first consider the proposals presented in the 
literature to justify normative risk acceptance thresholds, 
i.e. the determination of a point where risk accep-
tance appears justified . The great variety of exist-
ing solution models can be reduced to s even basic variants. 
These include the following: 
- The arbitrary establishment of a risk acceptance 
threshold on the basis of plausible considerations 
(such as the use of production processes optimized for 
safety engineering), natural standards <such as the 
natural radiation level) or the dispersion spectrum of 
a potential type of detriment which can be found 1n 
nature (such as the difference b e tween the radiation 
levels in the Black Forest and in the North German 
Plains as a measure for the maximum acceptable risk) 
(Rowe, 177, 349, p. 74 ff.) ; 
The quantification of benefit and risk. Risks are con-
sidered acceptable when the value of their benefit covers 
the potential detriment. However, the value of different 
benefit and/or detriment dimensions must be defined sub-
jectively (Fischhoff et al., '79, 114, ~. 18 ff.) to 
allow their comparability and intercomparabillty. 
- 8 -
_ The marginal cost analysis of risks (cost-effectiveness) . 
Under this concept the expected cost of the risk is 
compared to the cost required to reduce the risk . The 
risk acceptance threshold can be defined as that point 
where the marginal cost is identical for both cost 
types. Since this concept, too, requires a subjective 
evaluation of different cost dimensions in order to 
obtain a yardstick for comparison, Black et a1. proposed 
to use only health detriment as the unit of measure 
and to select that polnt on the cost curve where the 
expected losses caused by the risk source are equal 
to the expected losses caused by the 
of safety measures (Black et al . , 
p. 5 ff.). While this method allows 
adoption 
'79,481, 
rational decision-
making as to the limits up to which safety devices are 
still justified, it fails to provide any indication 
as to whether the risk source as such is acceptable; 
- The historical definition of a risk acceptance threshold 
(revealed preference analysis), using as a yardstick 
the manner in which society has accepted risks in the 
past. Risks which are not greater than risks accepted 
in the past may be conSidered worthy of acceptance 
(Starr, '69, 405, p. 1232 ff.); 
- Empirical surveys of decision-makers or of relevant groups 
among the public to identify preference structures 1n re-
lation to risk. In this theory, potential risks are 
quantified and individually presented to decision-makers 
for voting and evaluation so that subjective estimations 
are always taken into consideration (Social Judgement 
Theory; Hammond et al., '75, 494). At the same time, 
parallel sensibility analyses are conducted to deter-
mine the spectrum of possibilities and alternative 
structures of risk effects through surveys; 
- 9 -
- Weighting of risk consequences through empirically 
determined, qualitative risk criteria. This so-called 
"referred preference analysis" is used to determine, through 
general surveys, the basic types and criteria of risk 
evaluations, to identify the perception and evaluation 
of different risks and to pass on the results to 
decision-makers as a basis for their risk assessment 
(Fischhoff et al . '78, 112, p. 127 ff.). Although the 
empirical studies have been conceived as a means for 
the optimization of decision-making processes, they 
also provide a good insight into the process itself 
and the determinants of risk perception. 
- Elicitation of underlying values from groups involved 
in the decision-making process and breaking down of the 
values into dimensions, sub-dimensions and detailed 
specifications until a level of measurable attributes 
is reached, on which alternatives can be evaluated by 
measurements, judgements or both (Value Tree Analysis). 
This technique is part of multiattributive utility 
measurement and is aimed at finding an overall utility 
function for different conflicting groups. This very 
new method has nat yet been worked out in all its de-
tails, but appears to be very promising for combined 
normative and descriptive decision analysis (v. Winter-
feldt et al., '80, 533, p. 3 ff.). 
The expressed risk analysiS and the value tree technique 
already complete the transition to descriptive risk ana-
lysis. In the extreme case the decision-maker is expected 
to base his political judgement on the empirically deter-
mined risk preferences of the population. But what are 
the preferences of the population and how are they struc-
tured? Twa different approaches are available to respond 
to this question: 
10 
- The mathematical-axiomatic representation of actual 
risk acceptance behaviour with reference to formalized 
risk and lottery games using the two basic variables, 
the probability of winning or losing in a lottery and 
the distribution of expected values; 
- The systematic, empirical identification and descrip-
tion of the decision-making process in "genuine", 
true-to-life risk problems involving an incalculable 
number of effective variables. 
Again, a brief description' of the different concepts 
and their explanatory value appears necessary . The 
mathematical-axiomatic models will be discussed first: 
11_ 
- The perceived risk is equivalent to the probabil-
ity of winning or of losing a certain amount 
of money (Expected Value Theory). This classical 
risk perception theory 1s now considered dis proven by 
empirical evidence. 
- The perceived risk is equivalent to the product of the 
probability of loss and its benefit evaluation. This theory 
was put forward by the economists Damar and Masgrave 
(44, 487, p. 388 ff.). It, too, is considered disproven 
today (cf. Vlek, Stallen, '79, 529, p.3). 
- The perceived risk is equivalent to the product of the subjective 
probabilities of benefit and loss and their individual 
weighting (Subjectively Expected Utility, SEU). To this 
time, SEU theory is used as a basic model of risk per-
ception, but its empirical relevance 1s highly disputed 
since the subjective weighting has the result that any 
experimental result can be considered as confirmation for 
the SEU theory. It fails to meet the falsifiability 
postulate (cf. Kogan, Wallach, 172, 504, p. 137; Svenson, 
'75, 525, p.187 ff . ). 
- The perceived risk depends on the variance of probability 
distribution and is independent of the expected value 
of the lottery (Coombs, Pruitt, '60, 484, p.265 ff.). 
The model is no longer propounded with this rigidity today 
(Coombs, '72, 60, p.5 ff.). Even modified versions 
(greater weighting of the loss variance or semi~variance 
models) were not convincing under empirical aspects. 
- The perceived risk is equivalent to a linear combination 
of expected values and variance (cf. Pollatsek, Tversky, 
'70,516, p.547). The subjective weight of the two 
influencing variables is considered to be a free para-
meter. Existing empirical studies have not, or only 
partially confirmed this theory (cf. Coombs, Bowen, '71, 
485, p.27 f.; Payne, '75,515, p. 86 ff.). 
- 12 -
- The perceived risk cannot be mathematically determined be-
cause there are certain concepts on equivalence relation-
ships between alternatives but no clearly transitive 
ranking order. Individuals will evaluate risks on the basis 
of a single-peaked preferenc e fUnction and a subjective risk 
ideal, and two decision-making theorems are stipulated: 
, . Out of ~wo lotteries having the same expected value I 
that lottery is selected which has the smaller variance. 
2. Out of two lotteries having the same variance, that 
lottery is selected which has the greatest expectance 
value (Portfolio Theory). This theory has been tested 
empirically on several occasions and has been proved to 
be basically valid (cf. Coombs, '72, 60; Schafer, ' 78, 
380, p.40 f.). However, in spite of the mathematical 
elegance of the axiomatic derivation of this portfolio 
theory, the question remains as to whether it does not 
expend great efforts to prove the trivial discovery 
that individuals subjectively judge risky actions according to 
internal preferences and optimize the remaining factor. 
- The perceived risk results from the combination of an 
evaluation function of opportunities and a preference 
function of risks (Krelle, '68, 507, p. 140 ff.). The 
salient characteristic of this theory is that a distinc-
tion is made between subjective benefit evaluations 
and risk attitudes (Sch~ter, '78, 380, p. 32). Quite 
apart from some theoretical problems, this concept has 
not been empirically validated to date (Schafer, '78, 
380, p. 33 and p. 41). 
Still other models have been devised, but most of these 
represent only intermediate concepts "between the variants 
described above. Therefore, we feel that this list will 
suffice for a brief review of the field. The primary 
purpose of presenting the characteristics of these con-
cepts was to point out that the process of theory develop-
ment in the mathematical estimation of empirical risk 
-13 
assessment 1s not yet complete and, at the same time, 
to demonstrate the limited information content and 
artificial nature of these concepts. The transfer of 
laboratory experiments to real risk situations, the 
highly hypothetical experimental arrangements which are 
especially susceptible to systematic errors (response 
sets), the absence of a realistic emotional or motiva-
tional decision-making pressure and the occurrence of 
ambiguous results in the testing of these theories are in 
themselves sufficient to restrict the application of these 
conce pts for the development of social sclence~tbeorles to 
very few cases. The principal argument is that the limited 
scope of the experimental situation (hypothetical game 
of chance) and the limitation to the calculatory quantities 
of probability, winnings and losses do not allow any 
generalization of results, not even to include "genuine" 
games of chance (cf. Vlek, Stallen, '79, 529, p. 4 f.; 
Schafer, '78, 380, p. 38 ff.: Sji6berg, '77, 522, p. 8 f.: 
Kogan, Wallach, '72, 504, p. 137 f.). And it is even 
more open to doubt whether the results of these experi-
ments could represent something in the nature of universal 
risk evaluation criteria . Both the studies in the field 
of insurance risk (cf. Slovic, '78, 369, p. 58 ff.) and 
the results of risk research in the consumer field (cf. 
Cunningham, '67, 486, p. 84 ff.l Bettmann, '73, 480, 
p. 184 ff.) make it a natural assumption that the in-
fluencing factors of risk perception vary between different 
sUbjects and situations. 
AS a counterbalance to the mathematically oriented de-
cision-making theories a number of process-oriented models 
have been devised in the meantime where risk perception 
and acceptance are considered to be functions of informa- . 
tion processing and evaluation and of the resulting con-
flicts. Again, some basic types can be identified among 
the great variety of models on the basis of their under-
lying concepts and/or fields of application. 
- 14 -
_ Models of cognitive stress and common sense reasoning; 
_ Models to describe the decision-making process; 
_ Models .for subjective weighting of consequences; 
- Integrative risk perception models. 
All four types of model will be described in detail, below: 
- The concept of common sense reasoning and its biases 
is based on the quest10n as to how a multitude of complex 
information items can be reduced to major notions 
(salient beliefs). The empirical approach consists 
predominantly in assigning problems to test subjects 
which can be objectively resolved (on a logical or 
statistical basis), and explaining the subjective devia-
tions from the "correct" solution as a consequence of 
"erroneous conclusions" (biases). A great variety of 
distorting mechanisms in information processing have 
been identified experimentally: individuals will struc-
ture randomly distributed events on the basis of a content 
pattern (1n games of chance, for Instance~ based on the 
idea of poetic justice); in developing their conclusions 
they are excessively self-confident and fail to recognize 
the laws of statistical regression; they tend to over-
estimate the dependability of the occurrence of events 
when the events are still fresh in their memory (availa-
bility) and when they have correctly predicted them on 
an earlier occasion (learning effect); finally, they 
tend to prefer alternatives which are subject to a lesser 
degree of ambiguity in their expected value, although 
this means that, object1vely, they accept a higher risk 
(in this case, reference should be made only to the 
summarizing references: Ross, '77; 346, p.280 ff.; 
Sj~berg, '77, 522, p. 9 ff.; Slovic, F1schhoff, '79, 
370, p. 6 ff.; Cohen, Hansel, '61,483, p. 17 ff.). 
_ 15 _ 
- The concept of describing the decision-making 
process attempts to clarify the individual decision-
making phases, the balancing of consequences and the 
strategy of information processing. The starting poInt 
for most theoretical models of this type 1s the time 
dimension . Since decisions have to be made within a 
certain period of time, individuals in a group are 
forced to resolve, through certain strategles,the con-
flict existing between the time-consuming acquisition 
of information which would provide more dependability 
and the urgency of the decision (in terms of economics: 
the opportunity cost of waiting) (cf. Miller, Starr, '67, 
514, p. 60 ff.). The following procedures have 
been conceptualized as decision-making strategies: 
• Threshold value analysis (Satisficing Strategy). This 
means that individuals will check each consequence 
as to whether a threshold value considered necessary 
is exceeded (cf. Simon, 176, 521, p. 83 ff.). But 
this model must fail with a great number of alternatives 
and consequences because a rule for the selection of 
the optimum variant does not exist . 
• Satisfying K.O. conditions (Elimination by Aspects). 
The variants requiring a decision are screened against 
a graduated K.O. system of conditions until that variant 
remains which satisfies most of the conditions (cf. 
Tversky, '72, 527, p. 281 ff.). Although this concept 
may be qUite meaningful and close to reality in the 
case of some product decision-making (such as purchasing 
a motor car), it cannot escape the danger of the ambi-
guity of results or the constraints of a strict ranking 
order of K.O. conditions which is hardly feasible in 
reality. 
- 16 -
e Muddling Through. A noncommittal sub-decision (such 
as asking someone to dance) results 1n a new social 
situation which 1n turn causes a reaction of the en-
vironment (such as an attachment, a request for another 
meet1ng etc.) which 1n turn requires further decls1on-
making (whether or not to meet again) . Following a 
number of additional ad-hoc decisions,the social situa-
tion narrows to the poInt where a fundamental deci-
sion, not originally intended, becomes necessary (such 
a6 marriage). This form of "stumbling" into decisions 
is considered predominantly a characteristic of bureau-
cracies in democratic societies since the reaction of 
the societal institutions continuously modifies the 
sub-decision-making process (cf. Lindblom, '59, 511, 
p. 79 ff.). 
el!Ursuit of sub-objects by rational decision-making while showing 
consideration for external influenc.e groups (Mixed Scanning) • 
This concept is intended to indicaee an intermediate approach 
between the optimization of action alternatives and a 
continuous, reactive transformation of these objectives 
and means on the basis of social influences, where the 
rational selection of alternatives must always be measured 
against the democratic realization potential of these 
strategies (cf. Etzioni, '67,491, p.387 f.). 
e The balance model of conflict theory. In this mOdel 
the variables of "time constraints", "utilitarian gains 
and losses for oneself and others·, "perception and 
evaluation of alternatives", "approval or disapproval 
from significant other" and "congruity with self-percep-
tion" are varied systematically in order to analyse the 
expected consequences and reactions of individuals for 
all the possible combinations (cf. Janis, Mann, '77, 499, 
p. 70; 86, p. 137 ff.). In particular, the mechanisms 
of stress reduction are problemized through the dis-
placement of regrets, post-decisional justification 
_ I7 _ 
(rationalization), selective information 
input and cognitive striving for consonance 
(bolstering) (cf. p.91 ff., p.144 ft.). The two authors 
define the overall model not only as a description of 
the actual decision-making process but even more so as 
a theoretical reference frame for the entire spectrum 
of decision-making situations. 
- The concept of subjective weighting of consequences 
represents an attempt to investigate the 
types of consequences involved in decision-making and 
their weighting by the subjects. Unlike the cognitive 
stress models (Type 1), this does not concern the prob-
lems of distorted or erroneous perception but personal 
values, preferences and aversions used by individuals to 
evaluate alternative actions. It is not by chance 
that this theoretical approach is predominant in market 
research because the purchasing behaviour of individuals 
is decisively determined by a subjective pre-evaluation 
of the consequences of making a purchase. AgAin, some 
basic models can be summarized: 
• The two-component model by Cunningham (cf . '67, 486, 
p.82 ff.). This concept means that the purchase decision 
1s determined by the product of the perceived hazards 
of the purchasing consequences and the subjectively 
perceived degree of uncertainty inherent in these con-
sequences . 
• The two-component model by Betmann (cf. '73,480, p.18S ff.). 
It means that the decision can be explained by the 
strength of the belief in positive or negative consequences 
and by the personal significance of the different con-
sequences. 
- 18 -
• The multi-component model by Kaplan et al. (cf. '74, 
500, p.287 ff.). This approach typifies the con-
sequences and classifies them in accordance with the criteria 
()f Ihstrumental utility, financial losses, social re-
actions, psyaholoqioal effects and physical safety risk . 
• The multi-component factorial model by Schweiger (cf. 
'76, 526, p.26). This model further differentiates 
between the criteria of the model presented by Kaplan 
et al. on the basis of the additional components of 
"subjective uncertainty· and "importance". 
All four of these· models have so far had little effect 
on the scientific discussion of risk acceptance. More-
over, some of them are highly controversial from the 
empirical point of view (cf. von Rosenstiel, Ewald, '79, 
518, p.95 ff.). 
- The concept of integrative risk perception 
is based on the impossibility of constructing 
risk perception components. In reality, it is stated, 
man evaluates risks exclusively on the basis of an 
inner, one-dimensional evaluation scale (judged seriousness 
of risks). For instance, Wilde holds that each individual 
has a constant risk threshold with reference to risky 
activities (cf. Wilde, '78, 530, p. 134 ff.). Starr's 
Revealed Preference Concept (refer to Page ) also matches 
this concept on a higher level. One general objection 
to this model type is that it creates an artificial "black 
box· and, consequently, is interested merely in the result 
of weighing risks but not its causes and influencing 
factors. At the same time it lacks an empirically validated 
basis for the stipulation of an individual or societal 
risk acceptance threshold . However, this concept 1s 
interesting in that it allows the stUdy of dependencies 
- 19 -
on third variables such as social or demographic data 
with the aid of a risk evaluation scale. Kogan and 
Wallach conducted an experiment to study the effect of per-
ceived self'-value (ego-defensiveness) and. the degree of anxiety on 
risk acceptance and concluded that both variables are 
significant factors (cf. Kogan, Wallach, '64, 503, p.190 f). 
However, other experiments have so far failed to reveal 
additional evidence to the effect that personality 
characteristics have an effect on risk evaluation (cf. 
Kozielecki, '74, 50S, p.3 ff.). 
£9!}£!!!2!2!!!a __ o_n_Ji_i_s_k __ T_h_e_o_rj!._a_n_<t'p_e_c_i_s_i_o_n_llR.aJ...v_i_s_ 
A review of the great variety of risk and decision-making 
theory concepts will reveal a dicrepancy between the 
mathematical models which are highly formalized and com-
plex, but far from reality, and the partial, often poorly 
structurized but empirically more relevant procedural 
models. The following conclusions for the theoretical 
and empirical concept have been drawn from the existing 
studies on risk perception and decision analysis: 
- Risk perception is based only partially on the numerical 
quantities of "subjective probability estimation" and 
"expected value·s" and is not determined by SEU theory. 
- Decision-making on risks 1s strongly affected by cog-
nitive stress factors such as complex information pro-
cessing, decision-making under pressure of time, un-
certainty as to potential consequences etc. 
- Risk perception and evaluation are not based on a 
standard risk-welghing pattern but are specific for 
classes of risk sources and dependent on subjects and 
situations. 
- 20 -
_ The question as to whether some generally applicable 
risk evaluation strategies do exist and the 
degree to which personal characteristics correlate 
with these specific strategies, has not been clarified 
to date. 
_ Other problems which appear to require clarification 
include a typology of risk consequences, the question 
as to its internal weighting or its effect on the per-
ception of the overall risk, the problems involved in 
psychic processing of the uncertainty factor and the 
effects of additional risk-specific variables. 
These insights are used for both the basic theoretical 
concept and the interpretation of empirical results for 
the purposes of this study. However, the risk theory 
concept an the whole proves to be too narrow for use as 
the sole basis of further analysis. Therefore, a review 
of the economiC, psychological and sociological concepts 
is also required in order to develop additional information. 
EconomiC Concepts 
The explanation of risk acceptance by the economic models 
follows closely the reasoning of deCision-making theory. 
Two concepts exist so far which purport to describe and 
explain the risk acceptance of technological innovation 
on the basis of economic theories: 
- the marginal utility concept, 
- the "New Economic Theory· concept. 
Using the analogy of product purchasing where the con-
sumer relates the potential utility of the last-consumed 
unit to the resulting cost and makes his purchase deCision 
on that basis, the marginal utility concept compares the 
additional utility of the risk source for nearby residents 
- 21 -
to the subjectively perceived marginal risk of the source, 
and strikes the balance of these two quantities. Al-
though numerical quantification of these two components 
is hardly possible, this concept has the advantage of 
creating a universal starting concept which facilitates 
systematic structurizatlon of the empirical results and 
their interpretation. If one subdivides the elements 
involved in each balance between utility and risk into 
direct elements (such as financial losses or advantages), 
indirect elements (such as detrimental effects on health 
or improved infrastructure), symbolic elements (such as 
loss of status or greater social prestige) and dis-
tributive elements (such as relative income losses or 
advantages to certain groups) it is seen immediately that 
the different agents of a society such as individuals, 
groups, institutions etc . also attribute different weights 
to these elements when they determine their marginal 
utility. For instance, in the case of major technologies 
the individuals will start out predominantly from the in-
direct advantages and disadvantages while groups tend 
towards placing more emphasis on distributive elements 
and decision-makers tend to see primarily symbolic and 
direct elements. This results in a rationality gap be-
tween the different levels; for, although each of them 
takes the most rational decision on the basis of his own 
utility maximization, the result does not tend in the same 
direction (Renn, '79, 333, p.161 ff.). 
The economic theory of politics model is based on a 
similar concept: each individual and each group attempts 
to maximize advantages and minimize disadvantages in the 
most rational manner possible. However, different ways 
of life, interests and economic situations result in 
different risk notions in each instance. Downa attempted 
- n -
to prove that it is especially the ecology and environ-
mentalism topic which, due to the existing structure of 
society, will lead to polarization of the utility maximiza-
tion to be expected . With improving organization of the 
opponents , this polarization will reach a climax in 
social conflict until new types of conflict management 
have been established (Downs, '72 , 84, p.38 ff.). Pro-
gressing beyond the theoretical description of social 
conflicts the New Economic Theory, using the analogy of 
the risk model applicable for market events, considers 
the political decision-making process an optimization 
process where the groups of society struggle for relative 
advantages over other social forces, at the same time 
evaluating rationally the uncertainty and the risk of 
failure (Frey,'78, 492, p.214 ff; Himmelmann, '77, 496, 
p.193 f.). In its basic features this model corresponds 
to the classical concept of optimization of the expected 
value and, therefore, remains inferior to the presently 
developed theories in the field of descriptive risk per-
ception. In particular, the information theory assumptions 
on which the risk perception of the New Economic Theory 
1s based, are not yet convincing in the light of com-
munications research (cf. the criticism of Abplanalp/ 
Hettlage, '79, 478, p.259 ff.). 
Individual Psychology Concepts 
Since the start of industrialization, man and technology 
have been the subject of philosophical, anthropological 
and psychological research . The central aspect of these 
considerations is the problem of the autonomy of techno-
logical developments, the resulting material constraints 
and the feedback effects of these constraints on the 
psychic assimilation of environment and external stimuli. 
-23 
It would exceed the scope of this study to review the 
voluminous literature on this subject. Therefore, only 
a few studies shall be mentioned here which were con-
cerned specifically with the problems involved in the 
acceptance of nuclear facilities. 
During the very early phases of oppposltlon to nuclear 
energy in the 'fifties - this protest exhibits all the 
characteristics of problems in adaptation to modernization 
trends and aversions based on experience with nuclear 
bombs - several attempts were made to explain the op-
position to nuclear energy and major technology with 
psychoanalytical methods. Mythological anxieties 
(Prometheus), transfer of food ingestion and excretion 
to food contamination and radioactive waste, and suppres-
sion mechanisms 1n coming to terms with nuclear weapons 
were cited as problems (von Erichsen, 162, 441, p.161 ff.; 
WHO Report No. 13, '60, 450, p.14 ff.). In recent times, 
WUnschmann has retrieved these studies and transferred 
Jung's archetypal concept to the present-day conflict on 
solar and nuclear energy (WUnschmann, 'BO, 460). WUnsch-
mann holds that, in addition to the rational and factually 
oriented conflicts on the advantages and disadvantages of 
nuclear energy, the psychically subconscious mechanisms 
of environmental assimilation are the main causes of the 
acceptance crisis of nuclear energy. For this purpose 
he uses the following hypothetical effective variables 
(WUnschmann, '80, 460, p.25): 
Individual Awareness: 
- Paragon of civilization which enslaves man; 
- Isolation and alienation from nature; 
- Cultural and metaphysical rootlessness; 
- Suppression of emotional values and intellectual loneliness; 
- Lobbying, capitalistic exploitation; 
- Dissatisfaction and frustration on the job; 
- Fear of a world exclusively determined by reason and 
will-power. 
_ Uneasiness and suspicion towards an enigmatic bureaucracy. 
Collective subconscious: 
- Archetypal punishment myth; 
- Paradise archetype: 
- Solar archetypeJ 
- Shadow archetype . 
In applying these effective parameters to the nuclear 
energy conflict, WUnschmann arrives at the conclusion 
that the association between bomb and nuclear power 
station, the perception of a monolithic block of a 
nuclear power lobby , the hope for a riskless solar future 
and the rejection of anonymous, centralized institutions 
have caused a subconscious opposition of man towards 
nuclear energy. 
Similar arguments are used by Tubiana ('78, 428) who 
transferred categories of Freudian psychoanalysis to the 
nuclear energy conflict. Tubiana postulates that human 
behaviour is predominantly determined by dogma, fear, 
tradition and myths. He identifies as particular sup-
pression mechanisms the latent technology fear, nature 
cult, cognizance myth and fear of indeterminable environ-
mental structures . These four mechanisms, he postulates, 
cause a back-up of emotions whose energy is then dis-
charged against a symbolic object . Nuclear power stations 
as symbols for technology, human achievements and un-
accustomed types of risk, he feels, are natural targets 
for irrational and subconscious criticism of everyday 
problems (Tubiana, '78,428, p.4 ff.). 
Pahner ('76, 305) and Pelicier ('77, 312, p.198 ff.) 
also attempted to determine some psychic variables which 
could have an effect on attitudes toward nuclear energy. 
In this study, the opposition is related to the transfer 
of personal conflicts to external scapegoats and to the 
25-
compensation of a lack of orientation in a value-
pluralistic society which the individual can no longer 
handle. We are indebted to Schrenk for pointing out the psy-
chological characteristics of nuclear energy advocates 
(Schrenk, '76, 398, p.87 ff.). In particular, he 
characterized the personal identification with techno-
logical facilities as an ego-stabilizing orientation 
toward fixed regulatory structures. 
Psychological analyses of the nuclear energy conflict 
have frequently met the criticism that the emphasis on 
subconscious motives and suppressed or compensated 
anxieties insinuates that the groups involved in develop-
ing the arguments are irrational and that opposition to 
nuclear energy 1s either deviant psychological behaviour or 
a meaningful protest against the wrong object (cf. Paschen, 
'79, 307, p.17 ff.; Kitschelt, '80, 532, p.175 f). This 
criticism certainly appears justified where the psycho-
logical approach is used as the exclusive psychological 
explanation pattern, since both nuclear energy protest 
and nuclear energy support can be justified with rational 
considerations, as demonstrated by the marginal utility 
concept, for instance. On the other hand, it cannot be 
denied that unconscious or subconscious asaimllatlon 
mechanisms of risk information predetermine the direction 
of the developing attitude toward the object in question. 
However, it is very difficult to determine the proportion 
of unconscious factors with sufficient exactitude by 
empirical means. As a rule, these can only be postulated 
theoretically but are very difficult to demonstrate in 
empirical studies. Therefore, the present stud~whose 
primary emphasis is on empirical research/has neglected 
an explicit development of depth psychology arguments 
without questioning the justification of a transfer of 
depth psychology theories to the individual technology 
acceptance process. 
- 26 
Social Psychology Concepts 
A fluid transition zone exists between psychology and 
social psychology so that the classification 'of the 
different studies was not self-evident but had to be 
accomplished on the basis of self-assessment (in the 
case of R6g1in, for instance) or on the basis of sCien-
tific conventions. Moreover, they are 1n close proximity 
to the descriptive decision-making models. 
That most risk acceptance studies have originated in the 
field of social psychology may also be due to the fact 
that the individual component of risk perception and the 
society-oriented component of the social protest move-
ment require a dual approach. With respect to the nuclear 
conflict, social psychology research has concentrated on 
two explanatory concepts: 
- Acceptance models (symbolic reduction), 
- Attitude-formation models. 
One very simple, more intuitive model was developed by H. 
Ch. RBglln (RBglln, '77, 338, p.S8 ff.). RBglln's basis 
is that the nuclear energy conflict is primarily due to 
a projection of the anXiety-envy syndrome typical 
of our society to nuclear energy facilities (RBglin, '77, 
337, p.21). This syndrome induces individuals to accept 
only those things which exhibit both problem solution 
potential (projection) and potential identification with 
the object (identification). In the case of nuclear 
energy, there is a large deficit of identification pos-
sibilities, i.e. of emotional warmth, so that it is per-
ceived as an alien element and, at best, a necessary 
evil (cf. Figure 1). 


- 29 -
Although the plausibility and validity of 
this model cannot be dented, its theoretical and empirical 
justification and the spread of the spectrum which can 
be explained by this model remain obscure; in particular, 
there 1s a lack of interpretation patterns to explain 
the causes of deviations from the midline and their 
effects. 
A similar explanation model was developed by Hofstatter 
('79, 172). Based on the observation that individuals 
attempt a semantic simplification of the characteristics 
of objects, he transferred the ideas associated with the 
antonyms of male and female to a dual field of words by 
allocating highly correlating attributes to each of these 
two terms. Thus, the attributes active, dominating, 
strict, serious-minded, but also cruel and destructive 
are classified under the male generic and, stmilarly, 
the attributes passive, emotional, warm-hearted, soft 
but also untidy and vague under the female generic. 
Technology and, in particular, technolog1cal progress, 
are predominantly attributed to the male polarity, i.e. 
they are associated with the attributes clear, well-
ordered and sober, but also with destructive and cruel. 
Hofstatter emphasizes that this is not due to an inherent 
ambivalence of technology (which is not impossible) but 
to intuitive simplification mechanisms. Consequently, 
the ambivalent estimation of technology 18 not only deter-
mined by the object itself but also by people's concepts 
and prejudices. However, Hofstatter's theory does 
not allow any statement whether some technologies are 
forced to justify their existence while this does not 
apply to other technologies, or only to a minor extent. 
The explanatory value of this theory consists in the 
description of an initial semantic situation which is 
used to filter information. The degree to which this 
bias becomes equally effective for all technologies cannot 
be derived from the model. 
The risk acceptance studies by Dumenil et al. start out 
from the social role of technology in order to extract 
psychic factors. Man's attitudes toward a multitude of 
politically relevant issues with symbolic values (ranging 
from premarital love all the way to the foreign 
legion) are measured, correlated and the correlation 
values, representative of the degree of similarity of 
the individual issues, transferred to a coordinate 
system. The authors use the identified dot aggregations 
in this coordinate system to develop patterns of related is-
sues to which they assign appropriate labels. Empirical 
application of this model revealed a positive correlation 
between the attitude towards nuclear energy on the one 
hand and a conservative attitude toward life and preference 
for a -strong- government on the other (Dumen!l, 177, 85, 
p.97 ff.; Agraphiotis, Pages, '77, 3, p.139 ff.). For 
Dumenil and his collaborators the attitude toward nuclear 
energy is merely symbolic of biased socio-political per-
ception patterns used to evaluate new phenomena on the 
baSis of an inherent code. 
This interpretation is based on the concept that the 
inherent characteristics of an object and their individual 
perception have hardly any effect on individual attitudes. 
The independence of attitude development on the character-
istics of an object which is postulated by these authors 
cannot be maintained on the basis of existing theoretical 
analyses and empirical studies. At least this concept 
fails to clarify why - of all things - nuclear energy 
has become the symbol of an entire popular movement, and 
not supersonic aircraft, industrial robots or oil re-
fineries. 
, 
• 
- 31 -
The attributive risk perception theory which has been 
developed in close proximity to risk theory studies is 
also based on object-independent structures for risk 
evaluation and perception. One of the first acceptance 
models developed in this field of research is the Dread-
voluntary Model developed by Fischhoff et al. which pro-
vides that risk acceptance - regardless of the type of 
risk source involved - decreases with increasing disasterpoten-
tial, involuntar.iness and increasing certainty that the detrimental 
effects will be lethal (cf. Figure 2). In the meantime, 
these models have been refined, and an entire spectrum 
of influencing factors has been developed (cf. Fischhoff 
et al., '78, 112, p.146). 
Another approach to determine the structure of risk 
perception and risk evaluation by individuals was adopted 
in the study by Maynard et al. on the disposal of radio-
active waste (Maynard et al., '76, 244). The risk 1s sub-
divided into several sub-dimensions (short-term, long-term,cost, 
accident detection and recovery)and subsequently three 
independent measuring procedures are used to determine 
the structure of the dimensions considered optimum by 
the respondents. Although this model was developed ex-
clusively for the subject in question, the authors postulate 
the possibility of its transfer to other risk sources. 
They feel that it is an advantage of this subdivision that 
the dimensions for the development of attitudes towards 
a risk source are demonstrated. And, in fact, it was 
shown that different combinations of dimensions are selected, 
for instance between environmentalists and nuclear energy 
experts with respect to nuclear waste disposal. While 
the enVironmentalists stated that their acceptance decision 
was based predominantly on the long-term risk aspects, 
the nuclear engineers emphasized the short-ter.m risks 
(cf. Figure 3). 
- 32 
The structure of this risk subdivision must be considered 
a shortcoming of this study. While the respondents 
stated post factum that the decisions made appeared 
logical and reasonable to them, this subdivision was 
disproved in a preliminary test preceding our empirIcal 
study. It appears, therefore, that there 1s no inter-
subjectively valid form of risk dimensions. Moreover, 
it remains an open question whether a scale listing the argu-
ments for and against nuclear energy could not have pro-
duced the same results via a much simpler method. 
Factor I 
Technological RI.k 
Food 
Colouring 
Food 
Pre.ervatlve. 
Spray 
Can. 
Pre.crlptlon Medical 
Anti-Biotic. X-Ray. 
Vaccinations 
Pe.tlclde. 
Surgery 
Railroad. 
Elactrlc 
Power 
Home 
Appliances 
Bicycle. Smoking 
SkIIng 
Swimming 
Nuclear 
Powar 
Commercial 
Aviation 
Factor II 
Dread Risk 
Motor Vehicle 
Oeneral 
Aviation 
Mountain 
Climbing 
Location of RI.k Item. In a Two-Factor 
Psychological Space (ace. to FI.chhott. et al.) 
- 33 -
Faktor I 
Technlsch Induzierte Rislken 
Gespritzte 
Lebensmlttel 
Konservierungs-
stoffe 
Antl-Blotlkal 
Riintgenstrahlen 
Impfungen 
Spray-
dosen 
+ Kernenergle 
Pflanzenschutzmlttel 
Kommerzleller 
Operative Eingriffe Flugverkehr 
Faktor II 
TiSdllches Rlslko 
Schienenverkehr 
Elektrlscher Pkw 
Strom 
Haushalts-
gerite 
Fahrrad 
Sklfahren 
Rauchen 
Bergstelgen 
Schwimmen 
-
Ole speziflsche Anordnung von Rislkoquellen 
1m Zwei-Faktoren Raum nach Fischhoff u. a. 
Fig. 2: Risk Acceptance as a FUnction of Risk Characteristics 
(cf. Fischhoff et al.) 
Average Rankings for Six Respondent Clusters 
1 1 1 1 1 1 
1.ALTI 
..... n ".lILTI 
2 2 2 1." ADa. 2 2 1.11LTS 2 ".nLTI 2.1. IT' 2M LT' z.oa aTl 2.031'1 2.1'1'1, 
ADAK 
a.JlADM l.aIADa. 
2.qADI" 2.MAD&1I 
t.e'LT. 2.1'1T' 
3 3 3 3 3 3 
3.12 Coet 
, ... eo.t u.c:.t 
3.to Coet .... CooI a.nCoet 
4 4 4 4 4 4 
~ t ~.~ 1, ~~ ~ ~ ~ ~ .-s, ~ "6~ ~ .. 
•• 
'S II'~ . .. ~ to ~ C'" .. to'S ~ ~ ~. ;. 'iS1b ~ ~i! ~ 
" .. 
.C' ,..0- ~ .,~, ~ 
~ \ ~ ~ .", ~ ~O! 
-.. 
Ck ~O! o't.. ~ \, '0." . " ~ .,  ~ 
-.. 
~ 
•• 
-LTS - Long Term Safety 
STS - Small Term Safety 
AD. - Adjustment/Reduction of RI,k 

- 37 -
It seems much more rewarding to use the value 
tree technique already described in connection 
with the decision analysis concepts in order 
to subdivide the dimensions of risk perceptions. 
The tree analysis permits the investigator to 
start out with the actual dimensional scope of 
each respondent group and to avoid a pre-structured 
risk concept which 1s superimposed on the subjects 
(cf. von Winterfeldt, 'SO, 333, p. 47 ff.). 
The risk studies carried out by the German Battelle 
Institute also included a risk perception study (Frantzen, 
Schmid-JBrg, '16, 120). Identified perception factors 
included reinforcement through learning processes, dis-
sonance reduction, conformity and nonconformity, types 
of cognitive reinforcement (rewards, coercive, leqitima-
tive, referent, expert power) and conflict management 
on the basis of game theory. With respect to nuclear 
energy, three influence levels are considered to be of 
central significance: overt anxiety (concrete and diffuse) , 
environmental awareness and attitudes towards potential 
alternatives. 
The second major field of psychological research is con-
cerned with attitude i ~ ' formation. The most interesting 
approach from the point of view of methodology is that 
developed by otway and Fishbein under the auspices of 
the IAEA/IIASA Risk Assessment Group (IAEA = International 
Atomic Energy Agency, a UN agency based in Vienna which 
also maintains a risk perception research group; IIASA Q 
International Institute for Applied Systems Analysis in 
Laxenburg, Austria), since it starts out directly from 
the descriptive risk concept (Otway, Fishbein, '76, 302; 
Bowmann, Fishbein et a1., '78, 41; Thomas et al., '79, 420; 
Thomas et aI., '79, 421; Niehaus, '77, 276, p.454 ff . ). 
- 38-
The object-independent evaluation matrices used to weight 
statements on a risk source are measured independently 
via an evaluation scale (Items such as "Affluence is good" 
are evaluated on the basis of a good - poor continuum) , 
and sUbsequently the cognitive content inherent in this 
evaluation is referred to the risk object and measured 
by means of a belief scale (nuclear energy creates affluence). 
The probability of correctness stated by each individual 
is measured separately for each statement. Once the 
evaluations and beliefs have been determined, the pro-
duct of both components will yield the weighted cognitions on 
a certain risk object (Otway, Fishbein, '76, 302, pp.2 - 8). 
Factor analysis can be used to reduce these cognitions to 
a few effective parameters (salient beliefs) which deter-
mine the attitude: when this measurement technique was 
applied to attitudes toward nuclear energy, four different 
risk factors developed. environmental risks, psycho-
logical risks, socIa-political risks and economlc-
technical benefits. As in any factor analysis, 
this labelling system is, of course, arbitrary. The 
selection of generic terms is not very convincing, 
especially in this field, because general risk statements 
are classified under psychological aspects while the con-
crete dangers derived from them are classified under socio-
political or environmental aspects (cf. also the criticism 
Of von der Ohe, '79, 288, p.12). Opponents and advocates 
of nuclear energy differ especially in that the advocates 
primarily emphasize the economic-technical benefits 
while the opponents place maximum emphasis on psycho-
logical and socio-political risks (cf. Figure 4) (Thomas 
et al., '79, 421, p.e ff.). 
In spite of some methodological reservations on Fishbein's 
model for attitude measurement, these studies have been 
used as prototypes for our own empirical investigations. 
Beliefs about Five Energy Sources Held by Those 
Pro and Con the Use of Nuclear Energy 
Pro Nucleer Group 
Nucl.ar Solar Hydro 
Con Nuclear Group 
IndIrect RIsk 
EconomIc aeneflt 
EnvIronmental RIsk 
PsychologIcal and Physical RIsk 
Technology Development 
Coal 011 

- 41 
In addition to the IAEA studies, a number of other 
attitude studies have been carried out where the eXistence 
of attitudes was stipulated and correlations with external 
variables were sought. Examples include the studies by 
the American Battelle Institute (Melber et al. '77, 250), 
the studies of the Institute for Behavioural Research 
in Canada (Davies, '76, 68; Greer-Wootten, Hitson, '76, 
141), two studies by Goerke (Goerke, 175,331; Goerke, 
'78, 132) and an earlier study by this author (Renn, '77, 
331) • 
A nwnber of other studies deserve to be mentioned because 
they do not fit into the two classification character-
istics of acceptance models and attitudes. These include 
the study by C. Krebsbach and G. Eisenhart (Battelle 400/3, 
178, 99) on communication processes between politicians, 
experts and the public, a study by Gutmann on risk socializa-
tion (Gutmann', Battelle 401, '77 , . 147) and, finally, .. 
methodological study by Zebroski · on balance sheets in risk-
benefit perceptlops (Zebroski, '75, 467). 
These three studies shall not be discussed in detail 
because they are not particularly related to the present 
study. 
SOCiological Concepts 
Studies determined predominantly by sociological 
questions and thought processes are distinguished 
primarily by their scientific theory approach and their 
concept of SOCiety . The closer the proximity to the 
macro level of society, the more important the personal 
positions of the analysts become, so that the results 
are hardly comparable. Therefore, it is merely possible 
to summarize a few general, basic lines on the different 
argument levels. 
- 42 -
One study which starts out from the microstructure of 
expert attitudes and from that basis attempts to analyze 
the conflict as a crisis in science perception, is the 
summary published by Helga Nowotny on her activities as 
scientific advisor to the Nuclear Energy Referendum in 
Austria (Nowotny, '79,282). Helga Nowotny explicitly 
questioned the experts selected for the dialogue with 
the public and, at the same time, andy zed the public 
discussions and hearings conducted with these experts. 
This publication is too voluminous to allow a brief 
summary of its results. Of special interest is the sub-
division of experts into types which differ not only by 
the manner 1n which they present their arguments but 
also by their understanding of science and by their 
identification with external points of reference (such 
as institutions, social values, self-perception in their 
role as scientists). 
Thus, scientific attitudes toward nuclear energy are 
determined not only by factual variables but also by 
personality and situation variables. 
This finding is further supported by the studies of 
D. Nelkin on the role of scientific expertise in techno-
logical decisions (Nelkin, '71, 274 and '78, 275). In 
case studies, Nelkin analysed the structure of the arguments 
advanced by both proponents and opponents of nuclear 
energy and revealed latent trans-scientific values and 
issues which to a large extent determine the pos1tions of 
the scientists involved. 
- 43 
The highly differentiated evaluation by the different 
expert types identified by H. Nowotny or D. Nelkin, which 
is further extended into biographical and behavioural 
characteristics, has found a counterpart, although a 
simple one, in a study by Zetterberg in Sweden who 
classified the conduct of arguments by experts as pre-
dominantly seeking instrumental rationality in the case 
of advocates and value rationality in the case of the 
opponents, to use Max Weber's terminology (Zetterberg, 
'78,470, p. 23 f.). In addition, the role of the 
sCientist and the functioning of the overall science 
system of which nuclear energy and nuclear research are 
a part have been the subject of considerations and ob-
servations reported by Tschledel, HUlsmann and Prua which 
- 44 -
are more concerned with the barriers and resistance to 
a critical understanding of science induced by the system. 
These studies include few approaches of significance to 
the present publication (Tschiedel, '77, 424, p.96 ff.: 
HUlsmann, '77,176, p.112 ft.: HUlsmann, '78,177, p.147 ft.: 
PrUB, '74, 320). 
The macro-sociological discussion on the ranking of 
nuclear energy in SOCiety represents a contest between 
functionalistic, systems analysis, conflict theory and 
Marxist-neo-Marxist approaches. All of these studies 
start out from the basic question as to which social and 
SOCiety-oriented conditions have caused the nuclear con-
flict, and the consequences which result from this situation. 
With respect to content, the sociological studies can be 
classified under seven theoretical aspects, 
1. Social determinants of the nuclear energy protest, 
2. Diffusion and innovation research, 
3. Sociological aspects of technology, 
4. Conflict theory, 
5. Theory of the environmentalist movement, 
6. Participation theories, 
7. Moral and ethical aspects of nuclear energy. 
1. The discussion of the social determinants of nuclear 
energy protest 1s concerned with the question as to the 
SOCiety-oriented conditions of the protest movements and 
the description of the SOCially active partiCipants in 
this conflict. 
A review limited to the most important studies yields 
the following claSSification of SOCiety-determined 
initiators, 
- 45 -
- Inability to meet the requirements of society, and 
transfer of this stress to 
(Tubiana, '79, 428, p.9); 
functional, dead objects 
-
- Increasing awareness of external effects of technology 
(Scharioth, '77, 382, p.340); 
- Myst1fication of nature as a model for the regulatory 
processes of society (D6d~leln, '78, 80); 
- Environmentalism as a pseudo-religious movement (Garrison, 
'77, 126, p.156; Maxey, '76, 240, p.656 ff.); 
- The changed role, functions and perception of science 
among the public (Ravetz, '79, 328, p.8); 
- Negative marginal utility of major technology as a 
cause of soclal tension (Renn, '79, 333, p.165); 
- Mistrust of bureaucracy and state planning authorities 
(Lakoff, '77, 508, p.367 ff.); 
- Increasing protection of minorities against government 
planning power, and transfer of generally accepted 
humanitarian principles to technolo~icalissues (Scheuch, 
'74, 389, p.38); 
- Increased unemployment due to rationalization measures, 
creating an awareness of the ambivalence of technological 
progress (Scharioth, '77, 383, p.342); 
- Desire for more participation (Battelle II, 177, 25, 
p.95 ft.); 
- Changing values in developed industrial societies 
(Paschen, '79, 307, p . 28) • 
The above list does not claim to be complete, but it 
provides an idea of the broad spectrum of explanation 
phenomena and demonstrates that any abstract search for 
causes in society is practically doomed to failure since 
fishing for really determinant factors in a sea of potential 
variables becomes a lottery game when no theoretical pre-
structurlzatlon has taken place. 
- 46 -
More promising is the second approach where positions 
in the nuclear energy debate are referred to specific 
characteristics of the individuals occupying these posi-
tions. The danger of this approach is that the causes 
of the exhibited characteristics are related to 
attitudes. 
American studies (such as Oouglin, '76, 81) characterize 
active nuclear energy adversaries as follows: 
_ They originate from the upper middle class or from 
the lower classes (i.e. they have either above-average 
education or very little education), 
- They emphasize the necessity of environmentalism; 
- They are active in civil rights movements; 
- They are also committed to more social justice and 
participation. 
The personality pattern of German nuclear power ad-
versaries is less clearly developed. While this author 
in 1977 was also able to identify a coalition of less 
informed, highly emotionally arguing individuals from 
the lower classes with exceptionally well informed, 
highly rationally arguing members of the upper classes 
among nuclear energy adversaries, other studies have 
revealed a general overrepresentatlon of the lower soclal 
strata (Renn, '77, 331, p.ll0 ff.; Battelle II, '77, 25, 
p.A67). More recent polls (Infratest '77, Allensbach '79) 
have shown that,as a rule, income correlates to positive 
nuclear energy attitudes while average education and 
low popUlation density (rural areas) exhibit negative 
correlation to that attitude. 
_ 47 
At the present state of the art, typification of positions 
at best appears to have succeeded to a rudimentary degree 
only, especially since many of the traditional classifica-
tion criteria (political orientation, political party 
preference, social structure criteria) can explain only 
a minimum of the differentiation in the positions taken. 
Moreover, it is in any event questionable whether a de-
scription of opponents and adherents would better reveal 
the motives and behavioural causes. On the contrary, it 
appears more important to us to classify attitudes to-
wards nuclear energy under an individual or group-specific 
matrix of related attitudes in order to determine the 
ranking of this topic in the perception structure. 
Re 2. A first step towards a higher-level approach is to 
incorporate the specific problem of nuclear energy 
, 
acceptance into the problem field of technological innova-
tion acceptance. In particular, this approach was taken 
for comparative studies which attempted to develop common 
factors of society's perception and acceptance of differ-
ent innovations (cf . for instance Mazur, '75, 246, p.58 ff.) 
Innovation and diffusion research is based on a long 
tradition. It would exceed the scope of the present 
study to review the theoretical concepts involved which 
are particularly concerned with the developing nations 
(cf. Flora, '78, 108, p.19 ff. and Rogers, '63, 334, 
p.38 ff.). However, some theoretically postulated and 
empirically demonstrated circumstances shall be briefly 
discussed here in order to obtain possible explanation 
patterns of technological risk acceptance. Primary 
emphasis in diffusion research is on two explanation levels: 
- The question as to which social and cultural factors 
affect the acceptance of innovations within a social 
system, and 
_ The question of 
istics determine 
-48 
which social and personal character-
positions in reacting to innovation 
(from acceptance to refusal). 
According to Rogers ('63, 334, p.124 ff.) there are 
primarily five variables which play a determinant role 
in the acceptance of innovations: the relative advantages 
which an innovation would yield to the recipients, the 
compatib1lity with prevailing cultural "alues and social ideas, 
the complexity of the consequences for the social 
system, the possibilities for reversal (divisibility) 
and the communicability of an innovation . 
Although these variables were determined primarily in 
the SOCial context of innovation research involving 
American farmers, and although the problems involved in 
the ambivalent effects of the innovation whicb is to be 
diffused were not raised at all (Katz et al., '63, 199, 
p.249 ff.), these five innovation-promoting parameters 
can still be transferred - after some re-interpretation 
- to the present-day situation of the acceptance of large-
scale technology and nuclear energy. For this purpose, 
the following relationships can be used aa hypotheses. 
- The more uniform the distribution of benefits and 
risks of a risk source between individuals or groups, 
the sooner will relative advantages be revealed and 
the sooner can acceptance be expected (distributive 
equivalence) • 
- Tbe better the innovations match the values of a social 
group or institution and its concepts of the future, 
the sooner can acceptance be expected within this 
social field (and, probably, the leas can it be 
expected outside this field when the group draws 
SOCial demarcation lines) (intentional value congruence) • 
_ 49 _ 
-The better the response of innovations to the primary 
human environment and the more directly they meet 
human needs, i.e. the fewer detours there are between 
stating a need and the manner of satisfying it, the 
sooner acceptance can be expected (functional deter-
minacy) . 
-The fewer the constraints to freedom of action caused 
by the realization of an innovation and the fewer the 
number of perceived factual constraints, the sooner 
acceptance can be expected (safeguarding the freedom 
of action). 
T The more consistently positive or negative information 
is passed by different social agents to the public, 
the sooner positive or negative acceptance can be ex-
pected (communicative consistency). 
These five hypotheses cannot be derived directly from 
Rogers' statements; rather, his classification was used 
as a matrix to find an adequate explanation for today's 
innovation protest. We shall come back to these hypotheses 
in the interpretation of the empirical results. 
Less relevant than determination of social factors 
which inhibit or promote innovation is the study of the 
second sub-area of diffusion research, i.e. the innovation 
acceptance process, and the description of individual positions 
as a function of certain social or psychic characteristics. 
However, one trend which is contrary to the present-day 
situation should be mentioned: the traditional character-
istics of innovation-opposed individuals such as con-
servatism, advanced age, retention of given social condi--
tions (Katz et al., '63, 199, p.240 ff.: Rogers, '63, 334, 
p.172 ff.; Barnett, '62, 24, p.73 ff.l are practically 

51 -
On the whole, the s1milarity of the structure for "risk 
perception", "ignorance" and "alienation" appears to be 
interesting_ In earlier studies of innovation acceptance, 
opponents frequently exhibited a low degree of information, 
little reality in their ideas of the degree of hazard and 
a high degree of alienation. In the case of nuclear 
energy this applied until as late as the mid-sixties 
(Erskine, '63, 102, p.180 ff., Douvan/Withey, '54, 82,p.2). 
Therefore, while information on the functional principles 
of an innovation and a preference for rational declsion-
making used to be sufficient for a positive attitude, 
today I B innovation protest 1s based on a notion of per-
ceived ambivalence, not in the sense of prejudice-burdened 
semantic reduction (Hofstatter) but in the sense of a 
clear perception of negative aspects. This perception 
of ambivalence has been revealed in surveys on techno-
logical progress conducted in both the United States 
(Taviss, '72, 418, p.620) and in the Federal Republic of 
Germany (Battelle II, '77, 25, p.A105 ff.). 
The new types of encounters between man and innovation 
have been typified in a stUdy by Kates and phrased as 
counterparts to older typifications (such as Barnett's). 
Kates makes a distinction between three different innova-
tion perception types: 
- The Worry Bead School: the perceived risks frequently 
appear greater and more hazardous than they really are. 
Risk assessment is a very limited tool for providing 
adequate information on the hazardous possibilities, it 
directs attention towards very improbable catastrophic 
results. Technological innovations should be implemented 
by institutional procedures and a decision should be 
made on the basis of experience as to the degree of their 
retention in the future. Existing institutional super-
vision is quite adequate to ensure the safety of the 
population. 
52_ 
- The Tip of the Iceberg School: the actual risk is 
much greater than perceived by the population and by 
the experts. Therefore, prior to their implementation, 
innovations should be studied and analyzed carefully. 
The preferred tools are scenarios on the most pessimistic 
possibilities which could be involved in the innovation. 
- The Count the Bodies School: the critical boundary has been 
reached where new technological innovations will produce 
more detrimental than beneficial effects on man's quality 
of life. The hazards are well known and managed, but the 
values and expectations towards taking risks have changed. 
Every man must have the right to protect himself against 
the effects of high-risk technology. Therefore, reduction 
of the technological apparatus or, as a minimum, rigid de-
centralization are necessary (Kates, '76, 197, p. 4 ff.). 
The empirical relevance of these types has been postulated by 
Kates, but it has not yet been proven by the data he used. 
However, even as a theory this typification is useful in 
arriving at a new differentiated notion of innovation 
perception • 
. 3. Studies on the relationships between technology and 
SOCiety are based more on cultural philosophy and are 
not as easy to translate into empirical terms. Following 
the German tradition of contrasting culture and civiliza-
tion, more conservative soc1al researchers such as Gehlen, 
Freyer and Schelsky have denounced the constraints of a 
technological society and its attendant devaluation of 
ideals and the uprooting of man, but at the same time 
have emphasized the rationality of the technological pro-
cess· and the organizations developed from it (Gehlen, '57, 
127, p.8 ff ·.; Freyer, 70, 121; Schelsky, '61, 384, p.440 ff.). 
Now that H •. Marcuse (Marcuse, '67, 237) and E. Bloch 
(Bloch, '59, 37, p.783 ff.) have introduced the Subje~t 
53 
of "materialized rule through technology", the nec-Marxist 
literature typifies the role of technology as a camou-
flage instrument of the ruling class (Bahr, '70, 21, 
p.67 ff.; Ullrich, '77, 434, p.165 ff.). This position 
holds that certain technological developments tend to 
preserve privileges and consolidate the existing structure 
of society, especially through different forms of mate-
rializing the rule. To that degree, opposition to certain 
innovations is part of a revolutionary restructurizatlon 
process. 
Although these theses contribute little to the subject 
of the causes for the social conflict on nuclear energy 
(unless information on exploitation through technology 
were a determining action principle of the opponents), 
this interpretation of the ·nuclear energy conflict pre-
cisely fits the observation described above that a re-
structurizatlon of value orientations from promoting 
to inhibiting information is taking place. 
te 4. One significant component of this new value orienta-
tion is the awareness of the environment. Unlike the 
diffusion researchers, the theoreticians of the environ-
mentalist movement hold that this protest movement is 
motivated less in terms of innovation or technology than 
in terms of the effects of this technology (or of any 
other measure) on the environment. Deeply concerned 
with environmental pollution, aroused by the limits of 
growth and full of self-doubts as to whether the consumer-
orientation of our society is meaningful, new groups have 
formed and existing groups have changed their objectives 
to use their influence actively in favour of our natural 
environment and to protect it (McEvoy, '72, 247, p.217 ff.). 
The different phases of the environmentalist movement, 
- 54 -
from a more voluntary self-control all the way to major 
demands for participation, have been frequently described 
and analyzed (Means, '72, 249, p.209 ff.; Tognacci, '72, 
422; Mayer-Tasch, '78, 242, p.8 ff.; Rammstedt, '77, 324, 
p.447 ff.). 
One observation with respect to the environmentalist 
movement 1s of special interest for the question as to 
the acceptance of technological innovations: this is 
the fact that those who suffer most from environmental 
pollution are least involved in action or protest against 
it (McEvoy, '72,247, p.226 ff.). Three theoretical 
concepts are advanced to explain this phenomenon: 
_ The inadequate reciprocation 1n social exchange 
(origi!lally advanced by Blau, 1964; Turner, '69, 429, 
p. 819); 
_ Relative deprivation (originally advanced by Merton, 
Stouffer, '50: Turner, '69, 429, p.819), 
- The hierarchic needs structure (originally advanced by 
Maslow, '54; Mauss, '75,239, p.578). 
The degree to which these three concepts can explain the 
fact that, of all people, individuals privileged in terms 
of enVironmental quality have the greatest perception of 
environmental pollution, is difficult to verify by empirical 
means. However, all three explanation concepts are plausible: 
resistance in the lower classes cannot develop due to lack of 
social and pol! tical resources; compared to other problems, 
environmental conservation 1s of secondary importance 
for these classes (while it is of primary importance to 
the middle class); and in this SOCial situation the desire 
to mee~ primary needs supersedes any concern 
for the environment. It is likely that all three explana-
tions represent part of the true motives. The political 
conflict on environmental conservation also includes the 
55 -
thesis that, in promoting environmental conservation, 
the middle classes wish to erect a barrier to inhibit 
the social climbers of the lower classes by strangling 
economic growth (Gilliam, '72, 129, p.41). However, 
empirical proof for this thesis is still outstanding. 
In the meantime, some authors have developed normative 
social designs for the future embodiment of ·technology 
and society from theories on the environmentalist move-
menta A rich literature has developed under the headings 
of "soft technology- and the ideal of a "new life-style" 
which, recognizing a deficiency of major technology -
its centralized organizational structure, its hlqhinstltutio~ 
cost and its social vulnerability and resilience - describes 
new development lines towards a more decentralized supply 
structure, standardization of the roles of consumer and 
producer and greater political participation at the local 
level (Traube, '78, 425; Dickson, '78, 74: Huber, '78, 175: 
Amery, '78, 111 Lovins, 177, 127). There is no doubt 
that these theories have motivated part of the anti-nuclear 
energy movement as utopian designs (Kitschelt, 180, 532, 
p.309). 
Re 5. Stimulated by the environmentalist movement and the 
fact that many young people have started out toward new 
Utopias, the conflict theoreticians have developed models 
of their own. In these models they make a distinction 
between an identity crisis of industrial man induced by 
the ecological movement on the individual level and a 
politically induced legitimacy crisis on the control level 
of ·society (cf. Habermas, '79, 149, p.SO ff., Rammstedt, 
'77, 323, p.40 ff.). While the identity crisis involves 
a rethinking process on the part of man to turn fr.om 
consumption and growth orientation to an awareness of the 
environment, the legitimacy crisis is ,concerned with the 
inability of the political system, perceived by ·the . popu-
latlon, to meet adequately a threat to _ our .worl~: . en~ir9n­
mental pollution and waste of natural resources (Kielmanseg9, 
- 55-
'79, 205, p.31 ff.: C. Offe, '74, 286, p.129: H. Gripp, 
'78, 142, p.248 ff.). It is postulated that 
competitive democracy would never take harsh and incisive 
measures to avoid disappointing the voters and 
that the four or H ,ve' year cycle in which the voters 
approve their government can no longer ensure a long-
term political action perspective. Authors of neo-
Marxist or Marxist orientation feel that the environmental 
problems are an expression of the capitalistic 
production crisis which continues to aggravate due to 
the insurmountable conflict between profit management and 
quality of life (Pickshaus, '75, 315, p.71 ff.: Bernt, 
'74, 30, p . 40 f.). 
More closely integrated with the technological innova-
tion problem is LUbbe's thesis that the protest against 
technologies compensates for the oppressive experience 
of diminishing cultural confidence (LUbbe, '79, 230, p.3). 
He feels that an increasing marginal utility experience 
with respect to technology has led to the erroneous 
interpretation that the crisis is a value reflection 
crisis while, in reality, it is a management crisis since 
it is not the society targets generally associated with 
technology which are under dispute but the value-means 
relations and the alternative approaches which can be 
used to achieve these targets (LUbbe, '79, 230, pp. 10, 15). 
Similar arguments are used by Scheuch who feels that the 
protests against technology can also be understood a8 a 
reduction of technological complexity via moralization, 
a means used espeCially by the cultural elite which is 
favoured as a reference group, in order to be able to 
discuss anything and, at the same time, become immune to 
actual attacks (Scheuch, '74,389, p.38). It seems to us 
that ' the crisis theories, although often excessively bur-
dened with ideologies, represent quite meaningful approaches 
towards a macro-sociological theory of the protest movement. 
- 57-
Re 6. The participation concepts remain to be discussed. 
In this field it would be presumptious to present a 
summary review of the entire literature. With respect 
to the nuclear energy conflict, participation research 
centres on two major questions: 
- How can the democratic, institutionalized and represen-
tative political structures be reconciled with the new 
demands and campaigns for more political pazticipation, 
and what types of conflict resolution models can be 
identified? (ZilleBen, '78, 472; Nelkin, '78, 275; 
Guggenberger, '78, 146, p. 172 ff.). 
- How can the various interests of society which are neglected 
by the social and political exchange system be so pre-
dicted and organized that approximation to a normative-
democratic structure can be achieved? (Paschen, '78, 
307, p.34 ff.; Battelle II, '77, 25, p.B29 ff.; WUsten-
hagen, '76, 462, p.13 ff . ) 
Both questions reveal the basic attitude in the evalua-
tion of the nuclear conflict. the risks of nuclear energy 
cause disadvantages - in reality or in the perception .of 
the affected individuals - to certain groups which defend 
themselves and, 1n the process, incorporate potentially 
affected individuals and other disadvantaged groups 
(Turner, '69,492, p.826). The political system is 
forced to react since it haa to operate on a broader 
base due to party coalitions, Up to this point the analyses 
of most authors are identical. Subsequently, however, 
opinions differ: conservative theoreticians fear that 
adoption of the participation demands would result in 
the adoption of particularist interests and an institu-
tionalization of parochial politics (von Unruh, '74, 435, 
p. 74) I systems analysis theoretiCians either doubt the 
- 58 -
capability of the political system to perceive the 
adopted demands as they relate to a commitment to re-
duce complexity (Luhmann, '72, 231, p.156 f . ), or they 
feel that it is only the adoption of these participation 
demands which can guarantee an acceptable reduction 
(Battelle II, Part B, '77, 25, p . B31, Gripp, ' 78, 142, 
p.283 f.). Advocates of a normative theory of democracy 
or a pragmatic political approach, are in favour 
of the introduction of new types of participa-
tion and have designed the appropriate models (Dienel, 
'77, 75; Zillessen, '78, 474, p.122 ff.). In addition 
to the theoretical debate on participation, objectives 
and structure of citizens' action groups, especially in 
the German-speaking countries, are studied empirically, 
and their relevance to present-day politics is investigated 
(Andritzky, Wahl-Terlinden, '78, 13; Kodolitsch, ' 75, 
210, p.264 ff., Kempf, '78,201, p.358 ff . ). 
Although participation and nuclear energy are closely 
related, the problems and topics arising in this field 
cannot be generally tranferred to risk acceptance and 
techno'logical innovation. However, acceptance problems 
can initiate the desire for participation and may lend 
one of its components, such as nuclear energy, a symbolic 
character for the entire movement . And yet, participation 
research has hardly ever clarified the step from a diffuse 
innovation protest to an organized movement of symboliC 
character. At best there is the useful note that it is 
not the distorted perceptions or symbolic reductions (refer 
to the criticism of the psychological approach) which 
initiate the conflict, but genuine disadvantages and 
shifts of interest. 
- 59 -
Re 7. We should like to forgo a more detailed discussion 
of the ethical value analyses of the nuclear energy con-
flict which originate either from a normative attitude 
towards democracy or from a transfer of institution-
specific value matrices to the conflict (by the churches, 
for instance). These studies are in most cases not relevant 
for a description of the scientific analysis of the 
nuclear energy protest. 
It is hardly possible to write a transparent critical 
analysis of the soc10logical literature on nuclear energy, 
especially not in the concise form required here. Some 
significant items of criticism have already been mentioned 
in the description of the concept. Review of the different 
sociological research concepts appears to allow the fol-
lowing conclusions: 
- Modern diffusion research,which incorporates determinant 
research on nuclear energy attitudes and research in-
to influencing social factors, is potentially able 
to lead to better clarification of attitude formation 
and of the acceptance proceSSJ 
- The analysis of the environmentalist movement can de-
scribe and analyze the origins and development of move-
ments, the value adjustments on which they are based 
and the specific problems of involvement representationl 
- Conflict theories afford a suitable framework of explana-
tions for macro-sociological stimuli which have short-
term effects on SOCiety. The multiple-level application 
of the crisis concept to individuals, groups and society 
as a whole also makes this theory useful a8 an inte-
grative explanation matrix, although the restrictions 
of the theoretical environment (crises presuppose abrupt 
disturbances) allow only partial application. 
- ro -
_ Participation theories shift the basic question of 
acceptance and perception towards the level of decision-
making power on the imminent realization of innovations. 
Combined with this is the idea that conflicting interests 
are the decisive causes of innovation protest which, 
as a rule, can be bridged and resolved only through 
normative democratic mechanisms or restructuring of 
society. 
Open Questions and Phenomena Requiring Explanation 
The difficulties of studying nuclear energy problems 
with the tools available to the social sciences are 
the topicality and the dynamics of the conflict 
which make Kimple, static decisions quite 
useless, and the fact that ideologies already 
dominate the positions. The demand for a theoreti-
cally well-founde« and empirically justifiable research 
concept results in the necessity of clearly developing 
the phenomena requiring explanation and identifying that 
area which we intend to analyze thoroughly. If we review 
once again the results and theories of acceptance research 
presented in this chapter we must arrive at the conclu8ion 
that the following problema have not yet been satisfactorily 
resolved: 
- The types and structures of the decision-making pro-
cess in the face of risk-burdened circumstances, and 
the types of risk perception, 
- The weighting and evaluation of perceived ri.k components 
and their combination into an acceptance deCision, 
- The SOCial and personal factors which have positive or 
negative effects on such an acceptance deCision, 
- The formation of attitudes towards nuclear energy and 
its connections to other external objects a8 well as to 
psychological and social determinants, 
- 61 -
- The stress situation in the populace caused by intuitive 
risk perception and anxiety on the one hand, and 
loyalty to reference groups and confidence 1n the in-
stitutions of society on the other hand; 
- The opinion-fanning process on risk problems by relevant social 
groups, 1n particular the discrepancy existing be-
tween the attitude of the leadership and that of the 
group members which can already be observed (for instance 
in the political parties); 
- The effect of organized protest groups and non-organized pro-
testers on public opinion and the formation of 
political beliefs; 
- The increasing loss of confidence in political planning 
authorities and 1n the experts who support them, and 
the loss of legitimacy in the acceptance of political 
decisions; 
- Political and social developments as well as institu-
tional barriers for a reorientation of the value 
system of society toward alternative objectives in life. 
It would be presumptuous to attempt discussion of all of 
these problem aspects and corresponding theoretical con-
cepts within the scope of just one stUdy without taking 
recourse to trivial or superficial explanation patterns. 
Therefore, a selection from the great variety of problem 
items 1s required; at least under the present state of 
the art a comprehensive review, as desirable as it may 
be, is still impossible. In order to achieve progress 
toward this objective over time, it appears to me a 
priority task to discuss in particular the basic struc-
tures of the present-day conflict, since it is only the 
knowledge of this structure which allows an approach to-
ward the developments and endeavours of society which 
are based on it. While the latter certainly follow their 
- 62 -
own laws, they are easily misinterpreted in the absence 
of information of the basic motives which initiate them. 
In our view, the following three problem areas are of 
basic significance for the present-day nuclear energy 
conflict. 
- Risk perception and its intuitive evaluation; 
- Formation and structure of attitudes towards nuclear 
energy: 
- Psychic and social influence patterns which explain 
these attitudes. 
These aspects, which still appear abstract here, will 
be filled with more life and content in the next section 
devoted to the development of the conceptual tools. 
63 -
Part II 
Risk and R1sk Perception 
Basic Theoretical Concept 
- 64 
Definitions 
The risk concept is one of the major topics in the 
analysis of the conflicts involving modern technologies. 
This does not merely concern the problem of the factual 
quantifiability of risk .consequences but also the per-
ception and weighting of these consequences by the public. 
A more detailed analysis of these two complexes requires 
clarification of the terms risk, risk perception and 
risk acceptance. For this purpose, some definitions and 
the associated explanations shall be given. 
Risk shall be defined as the probability of negative or 
positive consequences which result from the realiza-
tion of an event or course of action. 
Risk perception shall be defined as the mental repre-
sentation and memorization of the negative or positive 
consequences of a risk source and the probability of 
their occurrence by individuals, groups of institutions. 
Risk acceptance shall be defined as the result of a de-
cision-making process on risk sources which is significantly 
affected by a combination of subjectively weighted and 
evaluated consequences and their subjective probabilities. 
Risk Concept 
Risk has been defined as the probability (1) of negative 
or positive consequences (2) which 'result from the 
realization of an event or a cours~ of action(3). These 
three components require some explanations 
(1) The term probability is difficult to define since it 
may have different meanings 1n different contexts. 
A major distinction can be made betw'een three defini-
tions of ' probability. 
65 _ 
- The logical concept which holds that probability 
represents the degree of certainty by which an 
event can be classified as being true; 
- The distributional concept which holds that 
probability represents the relative frequency of 
an event over time; 
- The personalistic concept which states that prcbability 
represents the strength of a conviction that a 
certain event will 1n fact occur (Lee, '71, 509) 
Sj~berg, '77, 522, p . 6 ff.). 
The third, personalistic definition is irrelevant for 
risk definition as an objective quantity. However, the 
logical and distributional definitions must be com-
bined since it 1s necessary, on the one hand, to specify 
the probability for the accuracy of a statistical con-
clusion and, on the other hand, to state the expected 
value (including the statistical confidence interval) 
for the event under consideration. For instance, a cor-
rect statement taking into consideration both of these 
criteria would read approximately as follows: As a con-
sequence of risk source X, 30 to 35 persons per annum 
will die with a certainty of 95'. 
Unlike other risk definitions, the present study does 
not merely relate risk to the probable consequences but 
also to the certain consequences. While this 1s con-
trary to the intuitive risk concept (Rowe, '77,349, p.25), 
it is significant for further analysis because the certain 
detriments and probable detriments are not independent 
of each other. 

- 67-
(3) The causes for potentially neqative or positive consequences 
arise from the realization of events or actions. Con-
sequently, it is siqnificant that consequences can be 
traced back to concrete objects or Actions, or can be 
anticipated: diffuse anxieties or concerns should be 
explicitly excluded. The connection to an initiating 
event or action must be clearly established in order to 
satisfy this part of the definition. 
However, for the existence of a risk it 1s not necessary 
that the protagonist knows this connection or has 
deliberately caused the hazardous conditions. 
Risk Perception Concept 
Risk perception has been defined as the mental represen-
tation and memorization (1) of negative or positive con-
sequences (2) and of the probability of their occurrence 
(3) by individuals, groups or institutions (4). 
(1) In this definition, perception is analytically 
separated from the weighting of the content items, 
although there is no doubt that these have a feed-
back effect on the selection of certain conditions 
and the attention devoted to them. Many types of 
subjective perception exist. An attempt at a rough 
classification will yield the following in summary 
form: 
- Perception of risk dimensions and calculation of 
their probability based on the individual's own 
experience (observational perception); 
- Perception of risk dimensions and calculation of 
their probability on the basis of publically docu-
mented experience (documentary perception); 
-68-
_ Perception of risk dimensions and calculation of 
their probabllity on the bu1a of exchange of ex-
perience (communicative perception), 
_ Perception of risk dimensions and calculation of 
their probability via association chains (associa-
"tive perception), 
_ Calculation of probability and uncertainty dimensions 
via subjective assimilation mechanism8 (intuitive 
perception) • 
consequently, perception by laymen and by scientiste 
differs merely in the degree of application of the 
different perception modes. Thus, scientific risk 
analysis can only utilize that personal experience which 
is reproducible, or it can CIlly employ that calculation 
which is acceptable within the framework of formal 
logic and scientific conventions. 
In addition to the perception of circumatances, 
memorization of the perceived circumatances is a 
second constituent criterion of risk perception. 
(2) In the perception of a risk, positive and negative 
consequences are again included .. r1ak conaequence., 
but again this includes only the perception of these 
consequences, not their evaluation. At first glance 
it appears audacious that this definition includes 
only two variables of riak perception. consequences 
and their probability. Considering the great variety 
of"" factors which have bean established as coristituent 
perception criteria in decision-making theOry 8X-
per1mante (distribution "variance, d1atribution semi-
variance, internalized risk threshold values, qualita-
tive risk criteria etc.), "thia limitation to the 
clasaical componente of consequence. and probability 
does not appear justified. On the other hand, however, 
_ 69 
these third-level and fourth-level factors as a rule 
determine the evaluation of a risk, not its percep-
tion, and, moreover, they can be interpreted as 
subjectively perceived consequences of a risk source. 
Thus, voluntary risk acceptance,which 1s defined as 
a qualitative characteristic, can be considered to 
be a conclUsion drawn from the structure of certain 
risky circumstances, stated approximately as follows: 
if the detrimental effects do occur, I will not have 
any influence on the dimensions of these effects and 
will be dependent on the assistance of other people. 
(3) Unlike the case of risk definition, probability in 
the context of perception is understood to be a 
personalistic parameter, meaning that, similar to 
the risk consequences, the calculation 1s based on 
the subjective yardstick used in each case. It is 
not the strength of the belief that a certain event 
will occur which is considered to be the operational 
definition but the subjective notion of relative 
frequencies . While this limitation of the probability 
concept simplifies the complex semantic scope of 
subjective probability, it represents the only meaning-
ful possibility for measuring the same dimension of 
probability each time when considering different 
individuals. 
(4) Individuals, groups and institutions are named as 
risk perception agents. Experiments on the "Risky 
Shift" phenomenon have demonstrated that group-
specific types of risk perception exist. Individuals 
in a group state a greater risk acceptance than they 
do in individual decisions (Stoner, '61, 524). How-
ever, the exact opPOSite, a so-called "Cautious Shift", 
can also occur (Stallen, '77,523, p.6). Both shift 
types can be considered to be elements of a qroup-dynamic 
process tendinq toward polarization • . Institutions, 
- 70 
too ,will arrive at other forms of risk perception than 
the individuals working in them, based on their ob-
jectives and value system. Therefore, an analytical 
distinction between individuals, groups and institu-
tions appears justified. 
Risk Acceptance Concept 
At the outset of this chapter, risk acceptance was de-
fined as follows. 
Risk acceptance shall be defined as the result of a 
decision-making process (1) which is significantly (2) 
affected by a combination of subjectively weighted andevalu-
ated consequences and their objective probabil1ties (3). 
Again, a brief explanation of the definition components 
shall be presented. 
(1) Risk acceptance is not defined as a behavioural 
pattern but as a result of a reasoning process of 
weighing between subjectively weighted benefit and 
risk consequences. Precisely speak1ng, acceptance 
includes five steps. 
- Collection of risk information, 
- Allocation of subjective probabilities to these 
information items, 
- Weighting of theBe cognitive elements, 
- Comparison of the weighted elements, 
- Comparison of the results to potential alternatives 
and to non-use of the risk source. 
Consequently, it is assumed that indiViduals in some 
manner process the multi-dimensional positive and 
negative elements to · obtain a total weighting which 
allows comparison to other risk sources. And this 
appears intuitively plausible since every individual 
is forced in everyday life to select from several 
71 -
hazardous alternatives, one alternat·lve which 1s optimum 
for that individual. It is immaterial whether the 
best option in the sense of normative optimization 
is actually selected. The only important aspect is 
that, in the final analysis, comparison between risk 
sources can only take place when the individual has 
found a yardstick for comparison which is valid for 
risk sources. 
However, the reduction of perceptual elements to an 
acceptance decision must not be equated with the existence 
of a standard yardstick for all types of risks. 
Risk yardsticks are referred to the subjectively 
perceived common features of the risks to be compared, 
and all the available empirical research has shown 
that a universal risk evaluation pattern (such as a 
product of losses and probabilities) is not likely 
to exist. In spite of this limitation, it is meaning-
ful to retain the artificial concept of risk acceptance 
as a generic term. In the case of different classes of 
hazardous actions or events, a selection will certainly 
not always take place according to the same criteria; how-
ever, s~larltle8 in the strateqles of the approach, 
in the pattern used for allocating information and in 
the specific weighting mechanisms might also exist in 
the case of heterogeneous risk decisions. 
(2) It is not intended to claim that acceptance · is ex-
cluaively determined by the weighted cognitions of 
the risk source. However, it appears meaningful to 
postulate that the greatest part of the variance of 
an acceptance decision can be explained by the latter. 
This assumption is in keeping with the notion that 
man decides predominantly on an object-rational baSiS, 
- 72 -
i.e. after weighing subjectively recognized advantages 
and disadvantages, and that effects alien to the ob-
ject (such as perceived or subconsciou8 anxieties) 
play a role in this process but do not determine the 
result of the decision-making process. This view of 
the decision-making process which is primarily ad-
vocated by Janis and Mann (Janis, Mann, '77, 499, 
p.95 ff.) is frequently contradicted by studies based 
on depth psychology (such as Wnnschmann,'SO, 460). 
The empirical results rather tend to be in favou~ of 
the thesis of an object-related deCision-making pro-
cess since it has been poasible, as a rule, to ex-
plain the major portion of the decision's variance 
with the aid of statement scales and weighting fac-
tors (Fischhoff et al., '78, 116, Thomas et al., '79, 
421) • 
(3) The acceptance process requires a combination of the 
perceived elements before a decision can be made at 
all. The manner of calculating this combination, 
whether by eliminating insignificant elements, sum-
mation of all elements, or other types of summarizing, 
will be disregarded for the generally applicable 
definition. Views on this subject have differed 
greatly until today and it is also doubtful that all 
risk sources are summarized by means of identical 
patterns. 
The most important step between risk perception and 
acceptance is the process of weighting and evaluating 
the cognitive elements on the basia of personal value 
estimation factors. In thi. proces., the perceived 
consequence., the subjective probabilities and their 
combination can be attributed weights of their own. 
Again, the type . of combination will be di.regarded 
here. A review of the evaluation level. postulated in 
the literature will yield a di.tinction between five 
different claases of factors, 
Congruent Ca.e: Equal Dlatrlbutlon of Rlak 
and Banaflt 
Altrulatlc Caae: All Shara tha Banaflt, 
Faw Taka tha Rlak 
Ollgocentrlc C .. a: Faw Shara tha Banaflt, 
All Taka tha Rlak 
Inaequlvalant Caaa: Faw Share Rlak and 
Banaflt, Moat Have Onl)' 
tha Rlak or the Benefit 
Disjoint C .. e: The Ones Take the Risk, 
the Others Have the 
Benefits 

- 75 
a. Distribution of benefits and detriments between the in-
dividual, the individual's reference group and society 
as a whole. According to the graph below (cf. Figure 5), 
a distinction Can be made between five basic distribution 
types: 
- The congruent case: 
Equal distribution of risk and benefit for a person or 
a group. 
- The altruistic CAse: 
All have benefits, but only a few have detriments. 
- The oligocentric case: 
Few have benefits, but all have detriments. 
- The inequIvalent case: 
A few share the benefits and detriments, while others 
have only benefits and/or only detriments. 
- The disjoint case: 
One group bears all the detriments while the other 
group has all the benefits. 
The question as to the distribution of benefits and detri-
ments not only arises anew for each risk sour,ce, but also 
for each type of risk element~' (direct, indirect, intangible 
or symbolic consequences) . From the great variety of 
possible combinations, four special cases shall be developed 
here which have special weight in view of the present-
day conflict: 
- Relatively uniform distribution of direct benefit. and 
detrtments, combined with a low degree of external 
effects. This case is least problematic and is usually 
accepted (for instance, in the case of consumer products) . 
76 _ 
_ Relatively uniform distribution of direct benefits and 
direct detriments, but greater disparity of the indirect 
consequences, or broad distribution of indirect benefits 
but not of the detrimental effects. In this case, a con-
flict between the rationality of the planning authorities 
as an element of society and the individual rationality 
of the persons accepting the risk will result (for 
instance, in the case of major technological facilities 
producing a public or semi-public good such as 
electric power) • 
- Relatively uniform distribution of direct benefits and 
direct detriments, one group has the indirect benefits 
while the indirect detriments are distributed. This 
situation will result in conflict between the groups 
of a society (for instance, when industries are estab-
lished in a location where great benefits are granted 
through tax advantages and creation of new jobs, while 
the detriments such as a waste air cloud have the 
greatest effect in neighbouring towns) • 
- Imbalance of direct detriment. and direct benefits. 
This case requires a socially or ethically motivated 
·acceptance on the part of the disadvantaged, i.e. a 
transfer legitimation, because otherwise open conflict 
Can be expected (example: open prison) • 
Typification of risk sources on the basis of their dis-
tribution structure is neither exhaustive nor complete. 
The literature hardly yields any approach towards a 
theoretical penetration of risk distribution/ empirical 
studies are entirely non-eXistent (Rowe, '77, 349, p.129 ff./ 
. Lowrance , '76, 128, p.94 ff./ Wagner, '78, 443, p.3 ff.). 
71 
b. Inherent characteristics of risk source and risk con-
sequences (qualitative risk ' characteristics). 
Unlike distribution effects, the so-called qualitative 
risk characteristics have been more frequently the sub-
ject of studies. The Table below (cf. Table 3) provides 
a review of most characteristics which have been identified 
so far. 
Table 3: Qualitative Risk Characteristics 
1. Factors Concerning the TyPe of Risk Conseguence: 
- Voluntary exposure 
- Risk avoidance possibility 
- Discounting in time 
Spatial distribution 
- IdentIfiable or anonymous risk taker 
- Personal control over the outcome 
- Supervision by society 
- Perceived degree of control 
- Degree of familiarity 
- Degree of scientific and technological maturity 
- Habituation 
2. Factors Concerning ' the Order of Magnitude of the Risk 
Consequence, 
- Fatal versus limited damage 
- Delayed versus immediate effects of damage 
- Catastrophic versus chronic damage 
- Extreme cases of the ratio of the magnitude of 
consequences and their probabilities 
(continued) 
n -
Table 3 (continued) 
3. Factors Concerning the Nature of the Risk Source: 
- Human, social and artificial risk sources 
- Individual avoidability of potential hazards 
- Possible perception with sensory organs 
- Reversibility or irreversibility of the potential 
damage 
- Existence of alternatives 
- Distance from the risk source 
4. Factors Specifically Referred to Benefits: 
- Exclusivity of benefit experience 
- Public versus private good 
- Uniform distribution of benefits 
(Rowe, '77, 349, p.42, Otway, '77, 297, p.4, Otway et al., 
'75, 296, p.3 ff.; D~erlein, '78, 80, p.17 ff., Starr, '77, 
409, p.446 ff . , Slovic et al., '79, 872, p.ll ff., Swaton 
et al., '76, 379; Maderthaner et al., '76, 236, p.5 ff., 
Vlek, Stallen, '79, 529, p.19 f.). 
Empirical studies have confirmed that voluntariness, 
familiarity and the severity of damage have a special 
effect on risk acceptance (Fischhoff et al., '78, 112, 
p. 144 ff.). In addition, personal monitoring possibili-
ties, habituation effects and avoidability have been 
identified as significant effective parameters (Swaton 
et al., '76, 379, p. 7 ff., Maderthaner et al., '76, 
236, p. 6 ff., Peters, '79,314, p. 96 ff.). However, 
none of these studies reveal the significance of the 
qualitative risk characteristics for risk acceptance 
or their general applicability to different types of 
risk. 
79 
c . Attributive, possibly universal erroneous interpretation 
patterns of individuals (attributive biases). 
These sources of perception and weighting were already 
identified in our description of the decision-making 
theory and risk analysis studies. The overestimation 
of risk sources with dramatic effects, the underestima-
tion of the probabilities in a catastrophic risk con-
sequence, the perception of one's own experience as a reli-
able yardstick and the negative relationship existing 
between the complexity of a risk source and its intuitive 
acceptance are only a few of the mechanisms used to &8-
s~11ate information on risk sources under the ·common 
sense- concept . The prtmary question 1s to determine 
how individuals transfer risk to their own lives 
and develop notions on damage possibilities and probabilities, 
the other question involves the generalization process 
where an awareness of exposure, once developed, 1s expanded 
to include general circumstances. The first area has 
been represented as "availability" in the mostly English-
language literature, and the second a8 -representativeness-
or "intuitive regression" (Stallen, '77, 523, p.7 ff., 
Slovic et al., '79, 371, p.15 ff., Tversky!Kahneman, '75, 
432) • 
The topiC can be summarized by stating that memory, the 
capability to imagine consequences, perceived publicity 
and dramatic effects of the potential damage have an 
effect on the strength of the potentialexpoaurevariable, 
and that the extrapolation from small numbers to an entity, 
the neglect of statistical dispersion and the overestima-
tion of redundant information determine the type of in-
tuitive generalization (Roaa, '77, p.l') ff., Blerbrauer, 
'77, 34, p.71 ff . ). 

















































All Accld.nla 
Botull'm P,egn.ncr 
Flood 
Torn.do 
lIoto, V.h. Acc. 
HOIllIcId. 
Tb 
A.thm. 
Electrocullon 
Sm.llpox V.ccln.tlon 
Actu.1 Num"', 01 O •• th. pe' Y •• , 
All 
C.ncer 
All 01 ..... 
H •• rtOI .. _ 
Stroke 
Stom.ch C.nc., 
O ..... ta. 

131_ 
Elpecially problematic was the inclusion of culturally-
induced risks such as leisure occupations, sports and 
artistic activities. The pretests had revealed heavy 
resistance on the part of the respondants even to per-
ceive these forms of risk as being risky at all and 
to fill in a belief scale on these relationships. The 
only risk sourceswhich yielded a response pattern which 
could be interpreted in any meaningful way at all re-
main motor car racing, gliding and skiing. 
Therefore, in order to avoid overlapping with motor car 
driving and air traffic in general, only the skiing risk 
was included in the main study. However, this risk 
lource haa nothing in common with nuclear energy. There-
fore, it is merely used as the control variable in the 
analyses to follow. 
The fir.t ta.k assigned to the interview subject was to 
estimate the looses (deaths, injuries, property damage) 
for each of the risk sources listed. The figures for 
the field of private cars were provided to the 
respondento so that they would have a starting point for 
their e.timationo. The result. of thele interviews 
have been compiled in Table 10, and several indices have 
been listed in this Table to summarize the data. 
The abstract eotimateo are of secondary significance. 
Rather, the difference between the estimate made by the 
population and the data contained in the official sta-
tilticl are of interest. Figures 13 a-c show a compari-
Son of the median values for the two cla.ses of vari-
ables. In Figure 13 a, logarithmic ocale was used to 
establish a relation between real losles and eltimated 
1088es during an average year, in Figure 13 b the same 
relation has been plotted for the los" estimates in a 
132 -
hic Year and in pigure 13 c the data are aqain catastrop d 
plotted for an average year but this time usinq a mixe 
1 ithmic for the official data and linear Bcale: ogar 
for the estimated values. -- What -are the conclusions to 
be drawn from the Table and from these fiqures? 
The span for the estimated deaths (normal year) 
ranges from 130 to approx. 17,000, while the real, 
statistically determined values ranqe from 22 to 
99,000. 
Low 10S8 rates are easily overestimeted, while hiqh 
1088 rates are more likely to be underestimated. 
The multiplier for 108s estimates between normal 
years and a catastrophic year i. on the averaqe 
approximately 2.04 (and, therefore, sliqhtly qreater 
than in the American studies). Some -sensational or 
often-discussed risks and risks cauo1nq conceivably 
great 1088es are extremely overe-sUmated with re-
spect to catastrophic perception capabilities. 
For pesticides and X-rays, 10 times the number of 
deaths was estimated, for coal-fired power station. 
17 times the number of deaths, for skiinq accident. 
32 times and for nuclear facilities -2,360 times 
the real number of deaths. The differences between 
average year-and -catastrophic year are sl1qhtly re-
duced when an index is computed from deaths; inju-
ries and property damage -instead of merely usinq the 
fatal1 ties as the reference '. parameter 1 however, the 
basic relations -remain -\Jnchanged. 
The estimate. of -Ioues for an average year approxi-
mate the true data frOm the statistics fairly well. 
The correlation between -the -respective medians is 0.78. 



estimated 
Fatalities 
Real Fatalities 
H.roln 
Home Appliances 
Aircraft 
Nuclear 
Energy Pesticide. 
SkIIng X-Ray 
Accident •• t Work 
Smoking 
Motor Vehlcl. 
Alcohol 
Aeeldenl •• 1 Home 
Coal Power 
.. 
.. 
'" • 18000 
;;; geschiitzte 
n Verluste • Arbellsunlalle 
.. 16200 
- '" " 
" 0 .. ~ 11 ~ 14400 
a PI " 
PI " " ::J > 'g 
" < " 12600 
PI " 0 
,. " '< • Alkohol 
PI PI 
• '" to 
" " 10800 • ~ 0< ~ He~oln Haushaltsunlalle 
>< (I) (f) I--' 
ro~CtI w 
p., t1 ::s 9 000 -...,J 
'" .. 
e. ~ 7200 
" - PI it 
t'Ij 540°1 Elektr. Haushaltsgeriite. 
" . 
" o 
" 3600 .. 
~ l~ 0. KKW .Pflanzenschulz hj 1 800 • • Kohlekraftwerk ~ statistlsch _ _ _ .Rontgen e. errechnete Tote 0 
::;: r-0,87 r2-0,77 0 0,5 1 1,5 2 2,5 3 3,5 4 45 5 
.. ' 
~ log X 

_ 140-
If the different spreads are incorporated and a 
mixed logarithmic-linear correlation i. used, the 
relation between the two median classes will rise 
to a value of .0.81. 
Comparison between the computed statistical data 
for a catastrophic year and the corresponding sub-
jective estimates reveals a greater difference be-
tween the two classes of variables. Apparently, 
moat ri8ks are given a relatively 8imilar estimate, 
since ten of the thirteen risk sources are located 
within a narrow range between values 6 and 8, while 
the real value8 show a dispersion between 2 and 10 
(logarithmic scale). Therefore, in a normal year 
. it appears that ·common sense- can produce a rela-
tively good estimate of 108.es. However, where 
intuitive yardsticks do not exist, which applies to 
the claasification of cataatrophic developments, 
relatively constant 10.8 rates are stated without 
more ,detailed differentiation between risk type., 
or extreme orders of magnitude are selected. As 
a result, the correlation coefficient between real 
and 'perceived values i. only 0.39 in thi8 easel the 
relation i8 not 8ignificant (p • 0.10). 
~f we atart out from the result that the risk level and 
ita,J',rob8bil1 ty i,; perceived relatively correctly by the 
population ~r an average year ,tlte question immediately 
ariaes whether th~8e perceived parameters also determine 
th ' ' 
. e . subjective evaluation of the rial< source. And here 
a Surprising r~8ult ,is obtained. Almost all correlation 
analY8eS ~de betwee~ ~ialt estimates and information on 
future, procedure with respect' -t~ "risk on the one hand 
and 'all the possibl 1 
" , ' , e O. 0.. indices , on the, other hand 
practically fail to reveal any relationship. Regardless 
of whether we start out from the perceived number of 
deaths, injured or property damaged, regardless of 
whether we construct several indices of these three 
parameters, whether we divide the risk level by the 
number of potentially affected persons, whether we use 
the date for normal or catastrophic years, whether we 
average the data for normal or catastrophic years under 
several different procedures (using ratios of 1.50, 
1.100 or 1.250) - it is impossible to find a clear-cut 
relationship to risk-benefit evaluation. 
The same result has been found in the American 
studies. Similar to the present study, theBe 
investigators found a very low agreement 
between estimAted risk level and risk evaluation. 
However, the correlation improved significantly when the 
data for estimAting the catastrophic year were included 
(Slovic et al, '79, 372, p. 13 ff.). For the German 
semple even this relationship is not significant. 
Several studies on the acceptance of medical risks have 
arrived at the same conclusion. The acceptance problem 
is only slightly influenced by the estimate of the risk 
level but predominantly by the evaluation of risk con-
8equences and the motivations for risk acceptance 
(Pochin, '75, 317, p. 189). 
All of these studies confirm the result that the perceived 
risk levels represent only one, possibly low, influencing 
factor in the estimation of risk sources. At normal 
averages the risk level estimates of experts end laymen 
are relatively homogeneous. However, the experts con-
sider these estimates to be a normative baais for the 
acceptance decision (this has been confirmed empirically 
through a survey of experts conducted by Slovic et aI, .'79, 
372, p. 12), while laymen consider it to be only one 
Weighting factor among many. 
142 -
Homogeneity of Risk-Benefit Estimates Made by Different 
Population Groups 
Prior to discu.sing ·the na~alquestion a. to the causes 
of the deviation of the intuiUve ri.k-benefit e.timates 
from the level of the perceived 10 •• rate, a brief ex-
planation of the structure of risk-benefit estimates i. 
of special significance. In all three surveys the inter-
view subjects were requested to state a value on a acale 
from -3 to +3 which in their opinion beat represented 
the ratio of benefit to detriment for the rlak source 
in question. This variable ia referred to a. risk-
benefit estimate below. In addition, the interview sub-
jects were requested in questionnaire II to grade bene-
fits and detriments separat.ly on a scal. from 0 to 10. 
Another index ~r the risk-benefit estimate was obtained 
fram these two variables by summation. Consequently, the 
following variables were available for the comparative 
study of independent and dependent measured values on 
risk~benefit estimation, 
_. ·estimate 11 risk-benefit esUmate for 100 interview ' 
Subjects using questionnaire 1/ 
estimate 2, risk-benefit estimate for 100 interview 
subjects using questionnaire 2/ 
estimate 3. medians of the d1fferencebetween the 
. sep~rately determined benefit and detri-
~nt,.e.timat" (questionnaire 2) / 
elt1miot.: 4. ' rtak'-benefit estimats for 507 interview 
".ubject. ' Using ·questionnair. 3. 
The ·distdbut.ion of-th,,"medians for ' .ach of th ... four 
8stimat •• ··h • "b" . 
. . as , aen: plotted in Ficjtir. 15; Good ·agreem.nt 
between" the'mealilns la°seeneat -first irlahce. .This i. 
R'.k .. aenem: 
£.f'maUon 
Survey 1 
Survey 1 V8 Survey 2 
Survey 2 va Survey 3 
Alcohol 
Smoklnll 
Heroin 
Correlatlone 
Home Appllanc •• 
Home Appllanc •• 
X-Rey 
Coel X-Rey 
Motor V.hlcle Alrcre't 
Alrcr.1t 
Motor Vehicle 
Automation at Work Skllnll 
Automation at Work 
Pe.tleld •• Nucl.ar Energy 
Nuclear Energy 
P •• tlcld •• 
RI.k·Benetit 
eatimaUon 
Survey3 
Rlek- Benefit Eetimatlon Survey2 

145 _ 
a striking result, considering that it involved three 
completely different samples. 
The same result is also obtained after differentiating 
by survey locations. With the exception of the nuclear 
energy variable, there are no significant differences 
in risk estimation between interview subjects living in 
JUlich and in Kerpen. This applies equally to the first 
and the second questionnaire. 
The dispersion spread of the individual categories is 
also relatively low. This homogeneity of the response 
behaviour is retained as a basic structure even in the 
case of highly different criteria differentiation. 
Apparently the evaluation pattern. of individuals are 
similar - at least for the 13 risk sources included in 
these studies. This fact allows two conclusions. 
Risk-benefit estimates of different risk sources 
follow a relatively rigid judgement pattern in the 
case of all individuals. 
The results of this judgement process agree to a 
large degree ' so that it is probable that the deter-
minants of this process have similar interpersonal 
structures. 
With respect to the question regarding the future procedure to 
be adopted for risk-handling,the uniformity of response-behaviour is 
less pronounced between questionnaires 1 and 2. The 
correlation coefficients for the 13 risk sources vary 
from 0.13 (motor car) to 0.77 (heroIn). This may be 
due, on the one hand, to the fact that the individual 
questions as well as the response categories were phrased 
differently. On the other hand, it is also possible that 
. risk level and benefit receive the same interpersonal 
- 146 
estimate, but that this does not apply to the desired 
consequences. 
If we use the separate estimates of benefits and detri-
ments as a basis for analysis, we encounter a reciprocal 
relationship' the greater the benefit estimate, the 
lower will be the estimate of the detrimental effects. 
Both quantities correlate at an intensity of -0.96. 
This result is contrary to the values obtained in the 
Fischhoff study ('78, 114, p. 132 ff.) and an IAEA study 
by Otway ('77, 297, p. 15) where benefit and detriment 
were given different weights for overall judgement. 
However, different methodologies were used for these 
two studies. 
In spite of this discrepancy, a glance at Fiqure 15 
reveals that, considering the point aggregation, the 
American values reported by Fischhoff have a structure 
which is highly similar to that of the German values. 
On the Whole there i. merely a parallel shift toward 
the left-hand bottom corner. This may be due to the 
linear transpOSition of the American data into the 
existing coordinate system or to the use of a different 
measuring technique. For FisChhoff et al divided their 
test subjects into two groups, one of which evaluated 
only the benefits and the other only the risks. ' 
Yet there is great agreement. both the relative spac-
ing of the jIOints and their distance from the respective 
regression , line are .almost identical. This observation 
once again supports the theSis of,. a relatively universal 
estimation of "risk iUld ,benefit levels of known risk 
sources. 
Eatlmatlon 
of Benetlt 
X-Ro, 
Hom. AppU.nc •• 
MotorV.hlcle 
Cool 
Automation at Work 
Motor Vehlcl. 
Alrcroft Nuel •• , Energy 
SkIIng P •• tlcld •• 
X-Ro, 
Alrcroft 
H_ Appllonco. 
P •• tlcld •• 
Nuel •• r Energy 
Alcohol Alcohol 
Smoking Smoking 
Heroin 
Estimation of Risk 
German Data American Data 

- 14~ -
Effect of Ideas on the Chllracterhtics 
of the Individual Risk Sources (Belief System) 
Two 1mportant results of the risk perception analysis 
must be retained so far: 
Risk estimation is only slightly dependent on the 
perceived loss rate per unit of tima. 
The evaluation of risk sources on the basis of their 
intuitive benefit-detriment distribution or risk-
benefit estimate is relatively universal for all the 
interview subjects. 
The present chapter is concerned with describing a class 
of variables which might provide an explanation of the 
level and uniformity of risk-benefit estimates of differ-
ent risk sources. the subjective ideas and arguments 
given for each risk source (the Bo-called belief system). 
For this purpose, a special scale including 46 statements 
on economic, health, personal, society-related and poli-
tical aspects has been designed, and the interview sub-
jects were requested to state the probability for the 
occurrence of each of these consequences.. Contrary to 
the otherwise customary Bcales, positive and negative 
consequences were measured separately to acquire the 
ranking of each statement with respect to both content 
(right or wrong) .and direction (encountered more frequent-
ly AS a negative consequence or encountered more frequent-
·ly as a positive consequence). Fina~ly, similar to 
Fischbein's attitude measurement model (where the scale 
values tor the beliet system are multiplied by the gener-
al subjective evaloation of these statements), the 
responses to each statement were weighted with the values ' 
of the so-called evaluation Bcale (the personal signifi-
Cance ot each argument to the interview subject). 
However, due 
belief scale 
150 -
to reasons of methodology, the non-weighted 
was used predominantly .for evaluation. 
Complex statistical techniques had to be used for eval-
uation of thes~ scales in order to ensure a meaningful 
and convincing reduction of the total of 1,196 variables 
to major parameters and factors . For this purpose, 
three statistical test procedures were used consecutively . 
- multiple regression of beliefs, using each risk-
benefit est1mate and each evaluation index as depen-
dent variables, in order to exclude irrelevant state-
ments, 
factorial analysis of the remaining statements to 
reveal the basic structure and acquire the major 
factors I 
- -variance and reqre88ion analyses of thea. factor., 
;using the risk perception variables which must be 
explained. 
The schematic in .Figure ·16 showa the results of this 
evaluation • .. This graph demonstrates the aignificance 
of the ·individual.belief factors tor the affective' evaluation 
of each risk source. TIle evaluation VUiable for each risk source 
must be seen separately 'from the variable of risk-
benefit estimation. _,· It · was . determined as . the sum ot 
a so-called ' semantic 'differential, a technique which . 
has been tried 'and tested .in the .aocial sciences .as a 
valid instrument ' for the "acquisition of . aUecUve represent-
ation of objects :. or>ogroupS .of penons In Qne's perceptive !.mages. 
The numerical.data on the ordinate represent the value of the 
declared variance .(R~) ' !i for the ,dependent variable "risk. 
source evaluat1on~ . ! 
- 151 -
A number of interpretation possibilities exist for a 
comparison of the columns for the 12 risk types. The 
following relationships appear to have special interest: 
A common characteristic of the positive leaders 
in risk-benefit evaluation is that subject-
related values and advantages play a dominant role 
(electrical appliances, X-rays, motor car, coal). 
The risk sources given a relatively positive evalua-
tion (intermediate position) are characterized by 
strong emphasis on benefits to society and to the 
economy which have only minor effects on the individ-
ual (aircraft, automation). 
Risk sources given a highly ambivalent evaluation 
(nuclear power stations, pesticides) are associated 
predominantly with socio-political and society-
related disadvantages, while health aspects and per-
sonal economic advantages are approximately balanced. 
In the case of risks given a clearly negative evalu-
ation (smoking, alcohol, heroin) the aspect of the 
individual's own health detriment becomes largely 
dominant. This means that subject-referred values 
have a strong effect on risk-benefit estimation in 
both the positive and the negative range. The accep-
tance of habit-forming products,in particular, appears 
to be dependent on the ratio of the perceived enjoy-
ment factor to the perceived potential health detri~ 
ment. While the enjoyment factor of alcohol ranges 
one percentage point above the health detr~nt 
factor, the evaluation is inverted in the case of 
nicotine, and for heroin the ratio is 1.5. The riak-
benefit estimates are given in the same negative 
sequence. This might be a useful fact"" for instance 
- 152-
for public information campaigns. Health detriments 
are perceived for all habit-forming products, but 
their estimation and acceptance is siqnificantly 
determined by the ratio of perceived potential enjoy-
ment and beUeved deqree of potential health detriment , 
regardless of the absolute level of the ccrresponding 
values . 
The skiing risk fall. slightly out of the general 
pattern: positive and personal aspects by far pre-
dominate, but the risk-benefit 
a more intermediate position. 
thi. reflects symbolic beliefs 
estimation occupies 
It i. possible that 
not taken into consid-
eration, such as "skiing is only for the affluent". 
Last but not least, the behavioural role component 
(where the interView subjects were skiers themselves) 
is likely to have a significant effect on the per-
ception of skiing risk. 
- Another interesting area is the comparison between 
the alternatives of coal and nuclear energy. Both 
have the same purpose, the generation of electric 
power . However , the perception structure for these 
two ~ystema differs greatiy. Apparently coal is 
directly, asso,ciated with electric P>wer supply and 
hence with ci,uality of life, pleasure and enjoyment, 
whi,l" in" ~e case of nuclear energy the society-
related, and, social constraints appear to be the 
mest,', important, linka in the association chain. The 
per~ived pers~nai advantages of coal energy over-
shadow the also perceived disadvantages to the environ-
men t and to !leu t1i. 
In the case of nuclear 
energy" health aspects and 
accident risks are a~ao found ,a. import~t negative 
:tnerefore, the predoounance of' negative items. 
Muftt .... It' 
.-... 
A_ 
ROokS-
t p.,.on.' H •• lth 
2 PI •• lur. end Comrort 
:I Socl.1 Burden 
4 Oener.1 H •• lth 
-
.... er ..... 
5 SocI.1 Aetr09r .... on 
• P.r.onaID.pend.nc. 
7 Pe~.IAd.a~ 
• Acckient RtaII; 
, .. tickle, "-
...... u."u. 
• ".tenanceor 
"""It Condttlon 
to o-eral H •• lth 
t t Per.onll DI,.dvanta, •• 
t 2 EI'IvIrOftm.ntlH •• lth 
HtIde., 
.-.. 
Coal Motor Vehk:" Aln:ta" 
AutOM.t;lon 
.t; Work 
13 Economic Adv.nt •••• 
14 Sociopolitical 
Dian .. nt .... 
1 5 Socia' Progr ••• 
1. P.rlOtt.1 .nd 
Economic Adv.nt •••• 
, 7 Futur. Quality of Llf. 
t 8 Rlak or Llf. and limb 
1 9 Environmental Hazarda 
20 Accld.nt Rlak 
21 Pr ••• nt~Day and 
Economic Technology 

- 155 -
perception contents may be partly offset by pro-
jected expectations in the future and advantages 
to society and to the economy. 
In addition to the individual interview data the medians 
for each risk source were also analysed as basic data 
for an aggregate evaluation in order to ensure still 
better comparability of the risk sources. This procedure was 
probably used in the study by Fischhoff et al. ('79, 112) but 
using geometric means, and approximately simulated in the IAEA 
studies by using three-dimensional factor analysis. 
However, the following must be kept in mind for this 
type of aggregate evaluation: 
the greater part of the variance is arbitrarily 
excluded; 
excessively high coefficients occur in correlation 
analyses/ 
factor analys~s will necessarily yield an identi-
cal set of factors for each risk source. 
Of course, the last-named item describes a desirable 
result, that, unlike the graph presentation in Figure 
1', the information contained in the different scales 
for each risk source can be combined into absolutely 
identical factors and that the values in each instance 
are comparable. However, this is an arbitrary homoge-
nization procedure whose results can be interpreted 
only with reservations . 
This type of factor analysis of the medians yielded 
the following basic dimensions: 
-156 -
consequences for the individual 
(pleasure, happiness, health, leisure, 
los,es, ~Qnyenlenc •• education), 
effects on welfare and society 
financial 
(~ality of life, supply, general health, 
protection of life, standard of living, security, 
aocial justice, conservation of the environment); 
social and technological progress or retrogression 
(modernization, progress of society, advantages to 
future generations)r 
socio-political values (freedom, progress, democratic 
rights, citizen's participation, social justice), 
direct personal advantages and disadvantages (plea-
sure, personal advancement, happiness, convenience). 
Figures 17a to l1d represent the content structure of 
the aggregate data. A presentation similar to that of 
the IAEA study (cf. Page 34 ) was selected. 
The five factors are plotted as block diagrams and, 
therefore, allow direct viaual comparison for each risk 
source. Again, the three habit-forming products exhibit 
the sequence of alcohol-smoking-heroin in almost all 
evaluation asPects; The greateat personal advantagea 
are "i,sociate,!" wi th alcohol, followed by Cigarette smok-
ing and heroi.n; The same appliea approximately to the 
perception of SOCiety-related consequences. While the 
social progress factor has no interpretative signifi-
cance (as alreadY'"descr1bed in the case" of the indivi-
dual analysis), alcohol and" smoking are " reversed for 
the socio-political effects. Although the difference 
Alcohol Smoking 
81 - Pe_nol Direct ondlndlrect Adyonteg .. ond D_yonteg .. 
82 - Eflecte for lho Public end the Society 
·83 - 8ocIoiend Technologlcol Prog_ (Retrog,...lon) 
84 - 8oclopolltlcol Volue. 
al - Peraon •• Convenience .net Pleeeure 
Heroin 

"l 
..... 
lQ 
~ 
-..J 
tr tr 
X 0 :t' 
I ..... lQ 
:0 lQ 
'" 
"l '1 
'< 
'" 
ro 
cn Cl lQ 
rt 
'" 0 rt t'l '1 ro 
I-' ..... 
ro 
'" 
tIl 
Cl I-' ro 
rt I-' 
'1 t'" ..... 
I-'- 0 ro 
Cl 
'" 
..... 
'" 
0-
I-' cn Ul 
rt 
:t' '1 
'0 C 
'0 Cl 
I-' rt 
..... C 
'" 
'1 
::l ro 
Cl 
ro "l 
cn 
'" Cl rt 
Ul 0 ;..- '1 
..... cn 
..... 
::l 
lQ Ul 
~ 
'" rt I-'-
0 
::s 
• 51 
. 52 
0 53 
.54 
. 55 
Röntgen elektr. Haushaltsgeräte 
persönliche direkte und indirekte Vor- und Nachteile 
qesellschaftli('he und soziale Vor- und Nachteile 
sozialer und technologischer Fortschritt (Rückschritt) 
sozia lpolitische Argumente 
direkte persönliche Vor- und Nachteile 
Skifahren 
..... 
(J1 
<D 
X-Ray Home Appiliances 
S 1 P ..... n.1 Dlr.ct .nd Indirect Ad •• ntall" .nd DI •• d •• ntall •• 
S 2 Ef1ecta lor the Public .nd the Society 
S 3 - Socl.1 .nd Technolotlc.1 Prog_ (R.trog,...lon) 
S 4 - Soclopolltlc.1 V.lu .. 
8 5 - Penonal Conyenlence and PI •• au ... 
SkIIng 
Automation at Work Motor Vehicle 
81 - Pereonal Direct and Indirect AcI ... ntagaa and Dlaaclwanl8gaa 
82 - Effacta for the Public and the SocIety 
83 - 80clel and Tachnolog ..... Prog .... (Retrog .......... ' 
84 - 8oclopolltlcal Valu .. 
85 - Penonal Conwen_ end ..... au .. 
Aircraft 


PMtIcIdee Nuclear Power 
a 1 - P ..... n.1 Direct .nd Indirect Ad •• nleg ••• nd DIM.h.nleg •• 
a 2 - Elfecle for the Public and the aoclety 
a 3 - Socl.1 .nd T.chnologlc.1 progr ••• (R.trog .... lon) 
S 4 - Soclopolltlc.1 V.lu .. 
S 5 - P.reon •• Cony.nl.nee .nd Pteaeure 
Coal 
Ib5 _ 
is minor and, considering the small number of cases, 
cannot claim any significance, this inversion appears 
plausible since the consequences of alcohol to health 
and social life are more easily perceived than the 
consequences of smoking. In the case of heroin, the 
most negative beliefs, compared to alcohol and nicotine, 
are found on all factorial levels . 
An especially ambivalent structure of belief factors 
is found in the case of pesticides and nuclear energy. 
The balance values for pesticides are clearly distrib-
uted about zero, expressing the perceived ambivalence 
of the advantages and disadvantages of pesticides. The 
nuclear energy values, on the other hand, exhibit a 
well-developed high in. the question of projected ad-
vantages and a well-developed low in the question of 
consequences to society and politics. On the whole, the 
subject- related consequences are given a slightly posi-
tive evaluation, but not the society-related and econom-
ic aspects. These relationships support the theory that, 
today, the perceived negative consequences of nuclear 
energy can at best be compensated by its projected ad- · 
vantages. 
USing coal energy as a comparison yardstick again, the 
great difference in the evaluation of personal and 
society-related aspects is seen at first glance. The 
median for the subject-related beliefs is 26 points 
higher for coal than it is for nuclear energyl in the 
case of the society-related and economic value., the 
difference i. even 68 points. Progress alone, as 
the third dimension,is attributed more to nuclear anergy 
than it is to coal energy. 
166 -
Comparison between the two meana of transport, motor 
car and aircraft, reveals some interesting relation-
ships. On the whole, the aircraft haa been able to 
achieve a positive evaluation ahead of the motor car, 
and air travel was attributed both more subject-related 
and more society-related advantages. The motor car can 
acquire a better rating only with respect to "political 
effects", this is probably due to the fact that the 
motor car allows greater expandon of consumer patterns 
to include all the c1as8es, while the aircraft still 
suffers a little from the stigma of exclusivity. The 
differences are significant but yet fairly low. 
Another remarkable fact is that the society-related and 
economic consequences of motor vehicle traffic are 
perceived to be negative on the average. As perceived 
by the population, the advantages of motor car traffic 
such as high standard of living, flexibility and qual-
ity of life can no longer offset the high accident rates, 
conge_ted streets, pollution and over-motorization. 
However, the percentage of personal advantages is 80 
great that an acceptance cri_is in motor car traffic 
must not be expected. Yet the protest against further 
development of the highway network and the demands that 
motor car traffic be reduced which have been ~ut forward 
recently in Germany appear to haVe! had their effect on general 
perception by the population. 
Electrical' home appliances and skiing are given a posi-
tive evaluation in all categories where, due to their 
nature, the 'factors of progress and socio-political 
aspects are not very pronounced, especially in the case 
of skiing ' a"" a 'risk oiource. The relatively poor rating 
of X-ray diagnostiCS as a risk source deviates from the 
analysi_ of the individual data. But it is especially 
here that the impo_ed factorial structure is likely to 
16'7 -
have been least justified since, for instance, the 
factor "direct advantages" is composed primarily of 
the elements of pleasure, happiness and convenience 
which have little significance in X-ray diagnostics. 
Finally, the surprisingly positive evaluation of the 
variable "automation on the job" should be pointed out. 
Positive quantities were obtained for all five factors. 
Positive beliefs on modern machinery on the job appear 
to prevail not only with respect to future generations 
and society as a whole but also for the personal wel-
fare of the individual. Automation in terms of humaniz-
ing the job i8 more deeply rooted in the awareness of 
the population than the potentially negative effects of 
unemployment and conditioning by mechanized work. 
Effect of Qualitative Risk Characteristics 
The psychological literature in the field of risk per-
ception is primarily concerned with the effect of so-
called qualitative risk characteristics, i.e. specific 
characteristics of risk consequences in terms of 
voluntarines8 or verification possibil1ties (cf. Fischhoff 
et aI, '78, 112, Vlek and Stallen, '79, 5291 Slovic et 
aI, '78, 368). These inherent characteristics are often 
considered to be the decisive variables of risk per-
ception. In order to test thi. theory, a Bcale of 12 
POlarity pairs (senlantic differential method) was designed which 
inclUded the following notion.: 
-
-
voluntary/involuntary, 
/ f 1 1 to some groups of society, useful to all use u on y 
_ hazard is directly perceivable/hazard is not per-
ceivable/ 
168 -
risk is known to science/risk is not known to 
science, 
immediate detriments/delayed detriments; 
risk can be personally influenced/risk cannot be 
personally controlled; 
individual has benefits himself/individual has no 
personal bensfits/ 
i th t of failure/catastroph-minor consequences n e even 
ic consequences in the event of failure; 
known, calculable risk/unknown, incalculable risk; 
every-day risk/entirly unaccustomed risk/ 
safety supervised by the public/safety not supervised 
by the public; 
same benefits can be achieved through alternat1ve 
facilities of lesser risk/no alternatives exist to 
achieve the same benefits at lesser risk. 
These scalar values were again evaluated statistically 
and, using multiple regression, compared to tne evalua-
tion of each risk source. A graphic review of the 
results can be seen in Figure 18. The 12 risk sources 
have been plotted on the abscissa and the value of the 
relationship on the ordinate (multiple correlation co-
effiCient). Unlike the belief Bcale, the simple correla-
tion values, not the declared variance, were selected 
(squared correlation values). For, on the average, the 
summated proportion of the declared variance is only 
about 18' while the six factors of the belief scale 
reached ,apprcximately 30 to 38 percentage points. Con-
sidering the explanation .tructure revealed here, the 
assumption·,.that. the qualitative character1etics have a 
dOminant ,position ' in risk ,perception becomes questionable. 
-~ 
Smoking H ..... x...., N?ICf •• '~ Cool 
Rlsk-QbJect 
1 Se.,er. Consequenclee 5 HUIIfdti W.U-t<:nown 9 S.f.ty under 
2 Volon,ary e Av ...... Altetnllti" •• Public Control 
RlakTaklng wtth Equal Benefit to Inperceptlble RJak 
3 ,.rlonll CofttroI 7 E .. ryd.y RI.k 11 No Penonal Control 
.. P.non.' .... fItIR .. k 8 EquoI_1I1 12 ex'rllOfdlnlry Hlzard 
DIotrIbutlon 
Aut ........ 
.t Wortt p--
13 I""ofontl" Rlak 
,. '",mHl,t. Co .... -
qUlndea 
15 Unknown Huard 
18 UhtorCon .. quendel 
-
Motor'lletMct.. A ..... '....... p  IICft
17 No A • .u.tMe Altern.· 
tl., •• with Equal Benefit 
1. Del.,.., (L"",-
T ., .. , ConNquencIH 
It •• ,oty_ 
Public Control 
20 T.chnk:elly Ad.-need 

_ 171 
This assumption becomes all the more questionable 
because it is the benefit aspects not included by 
Fiachhoff et al which play the most important role 
in the regression equation for six of the nine risk 
sources. The risk-related characteristics are closely 
related only to the extremes "nuclear energy and 
pesticides" on the one hand and "home appliances" on 
the other. It is likely that benefit and detriment 
aspects perceived in risk source evaluation are primary 
factors in judgement-forming, while qualitative risk 
characteristics have an effect only where certain risk 
sources are associated with positive or negative ex-
tremes of the qualitative perception. They are used 
for additional orientation when perception thresholds 
are exceeded. Similar to the Fischhoff studies, two 
determining risk areas were developed in this study: 
thevoluntarines80f risk-taking and the possibility 
of personal verification; 
the degree and type of potential detriments. 
However, the ranking of the8e two factors must be 
assessed conSiderably lower than suggested in the 
American studies. In order to demonstrate the differ-
ences between the two studiea, ' the comparable results 
from the interviews by Fischhoff et al were transformed 
in linear fashion to match the German scale, and the 
different pairs of values were plotted in a coordinate 
system (cf. Figure 20). 
In the four most important criteria, "voluntary, person-
al verification, scientificaily known and im-
mediate effects", over 60' of all values- ranged within 
an interval of ±1 about the median line (uniform dis-
tribution). Consequently, the difference in the amount 
172 -
of the explanation value for the qualitative charac-
teristics can hardly be attributed to the different 
data structure. Rather, it is a product of the aggre-
gate analysis used in the American version which, as 
described above, inevitably raises ths correlation 
values. 
This will become evident immediately when the German 
data are also aggregated. Comparison of the medians 
between the qualitative variables and the risk evalua- , 
tion in each case yields a linear correlation of 0.68 
and a curvilinear correlation (exponential curve profile) 
of 0.74. F1gure 20 shows this in graph form. 
USing an approach similar to the American study, the 
aggregate qualitative scale values were sub-
j ected to factor analysis. This analys'is resulted in 
-three factors: one factor for voluntarineSB, personal 
control, public controllability and scientific familiar-
ity, a lecond factor for the severeness of consequences, 
everyday nature and short-term effects. 
These two factors wele alao found in the American study 
(Fi.chho!! et al, '78, 112, p. 145). A comparison of 
the individual factor load.l.ngs is shown in Table 11. The pre.-
ent stUdy ail.ao included the benefit diatribution as an 
additional factor which was not. part of the compar-
able , Amerigan stUdy. The absance of this aspect must 
be interpreted as a deficiency 8ince this dimension 
covers , approximately one third of the explained variance 
within the scale used here. 
The .,value! ~~; ,th. factor loadings w .. re plotted in a coor-
d1nate system (Piqures .. 2.,a. to 21c) for better interpretation of 
tint abstract nUlllerical .valuea. In the campat.isan of factors 1 and 1 
..... ,0< 
illtGorm.n 
Doll 
H.A. 
N. E. 
II. V. 
... v. Smoking 
Smoking 
H. E. 
... v. 
H.A . P •• "e,d •• 
H. E. X-R.y 
X-R.y H.E . 
... v. P •• "eld •• 
COil eo., 
P •• Ueld •• H. A. 
X-Ra, eo.1 
H.E. 
... V. H. A. 
Smoking 
Smoking 
Volontary Rlek-Taklng 
Immediate Effect 
Pereonal Control 
P •• Ueld •• 
X-R., 
H.A. 
Se.'. for .he U.S. Da.a 
Known to Science (In Germen: Technically Advanced) 
Home Appllancee 
Nuclear Energy 
Motor Vehicle 




OF 1 
Volontary Risk-Taking 
German Data 
American Oat. 
N ucl •• , En.'OJ 
Motor V.hlcle 
Motor V.hlcl. 
P .. tlcid .. 
Xofta, 
Nuel •• , Enefgy 
P •• tlcld •• 
SmokinG 
Smoking 
Home Appliance. 
Horn. Applianc •• 
Automation at Work 
Co.1 
Xofta, 
OF 3 
Severity of Consequencies 

QF2 
Equal Benefit 
Distribution 
H.roln 
Motor Vehlcl. 
X-Roy 
Automation at Work 
P.IUcld •• 
Nucl •• , Energy 
Home Appliance. 
eool 
Smoking 
OF 3 
Severity of Consequencies (Immediate, Severe, Dread) 

- 183 
the American values were plotted in addition. Similar 
to the graph presentation of the medians, a relatively 
81m11ar basic pattern is found in both studies, but 
the American values are subject to greater dispersion 
and more frequently occupy extreme positions. This 
might be due to the relatively homogeneous sample of 
the interview subjects in the American studies. 
The information content of these three figures can be 
summarized 1n three observational 
- Pesticides and nuclear energy are given negative 
evaluation in the case of all three qualitative fac-
tors, i.e. on the average there is a negative per-
ception of benefit distribution, personal control-
lability and the degree of consequences. In view of 
the fact that these negative evaluations include a 
relatively high proportion of the explained variance 
for the risk evaluation 1n question, the conclusion 
i. justified that .extreme ~alues in the quali-
tative risk characteristics have a marked effect on 
risk acceptance, while risk sources in the inter-
mediate value range of the scale of qualitative char-
acteristics are directly affected only by the benefit 
distribution component. 
Apart from the extremes discussed above, the central 
position of the benefit component (overall distribution and 
the individual's perceived self-utility) is clearly 
documented. The positive evaluation given to x-ray 
diagnostics, coal-fired power stations and motor cars 
runs parallel to the allocation of the benefit factor 
in each case (correlation values 0.71, 0.84, 0.76), 
while the more negative attributes of these three 
risk sources have only a minor effect on the overall 
evaluation in the case of the two risk-related factors 
(correlation value. between 0.21 and 0.48). 
)~4 -
Table 11, Comparison of Factor Loads to Qualitative 
Risk Characteri~tic8 in the German and 
American Studies (Fischhoff et a!. • 79. 112. p.14S) 
German American 
values values 
Factor 1 voluntary 0.93 0.89 
known to 0.83 0.88 
sCience 
personal control 0.87 - 0.83 
familiar 0.90 - 0.87 
imnediate 0.45 0 . 70 
Explained variance 52.8' 58.9' 
Factor 2 severity of con- 0.89 0.91 
sequences 
everyday risks- 0.82 0 . 60 
immediate riSks 0 . 47 - 0.45 
Explained variance 10.2' 21.1\ 
Factor 3 total benefits 0.93 
benefit to the 
individual 0 . 93 not included 
safety monitored 0.92 in the scale 
iumediate risks 0.72 
Explained variance 37.0' 
_. 
in American survey, common risk 
lAS -
_ In the oPPOsite direction, the same applies to the 
evaluation of habit-forming products. Here we have 
a high negative correlation between benefit distribu-
tion and risk evaluation (smoking: r. -0.79; 
heroin: r· -0.91); however, the extremely positive 
value. on the voluntariness scale have a compensating 
effect on the acceptance of the habit-forming product. 
Thus, in the case of the question as to future han-
dling of the risk it is less a prohibition than a 
voluntary abstention which is demanded (at lea.t with 
respect to alcohol and smoking). Consequently, the 
possible individualization of the risk influences 
the risk-benefit evaluation les8 than the type of 
acceptance decision. The tolerance of the smoking 
risk, therefore, does not ariee from a lack of per-
ception of risk consequences or their denial, but 
grows from the belief that the voluntariness of risk-
taking requires only voluntary instruments of risk 
management. 
On the whole, the qualitative characteristics appear to 
exhibit a preponderance of benefit distribution as the 
weighting factor for all risk sources. The two risk-
related factors have a direct influence on risk evalua-
tion only when certain threshold values which cannot be 
quantified here are exceeded. 
!ffect of Risk Propensity and Benefit Orientation 
In empirical studies, characteristics such as high risk 
propensity or risk aversion can be measured only under 
great methodological difficulties . In the l1tera~ure 
On decision-making theory these two characteristic. 
play an important role for the determtnation of differ-
ent types of deCision-making processes Creckle.. or 
le5 -
cautious). In the present study, risk propensity was 
measured by determininq how consistently all 12 risk 
sources were overestimated or underestimated in the 
risk-benefit estimation. The scale values represent-
inq extremely hiqh and extremely low estimates were 
selected from the responses for the benefit and detri-
ment variables and from the values for the risk-benefit 
estimate as a combined variable. In addition, the 
frequency of the zero cateqory indicated was introduced 
as a new variable . The newly formed variables were 
labelled as follows : 
benefit orientation (extreme emphasis on the benefits 
of risk sources) I 
detriment orientation (extreme emphasis on the detri-
ments of risk sources), 
hiqh risk propensity (extremely positive risk-benefit 
estimate) I 
risk neutrality (predOminantly neutral risk-benefit 
estimate) I 
risk aversion (predOminantly neqative risk-benefit 
estimate) • 
Consequently, five new characteristics were computed 
internally for each interview subject, the numerical value of 
Which represents the deqree of consistency of the risk 
propensity. For correlation analyses of the thsse five 
variables with one of the 12 risk-benefit estimates or 
with the risk evaluations, the associated benefit or 
detrime.n,~ value. were, of course, excluded from the 
conSistency index so that, in the final analysis, each 
of the 12 risk sources had 
. '. . 
DWn .. (the .values for the 11 
• L _. • • . , ' ",' ': , 
a correlation partner of ita 
other risk sources). This 
ensured _ that .. ~he .variables used for correlation were 
independent from each other wi th r~apect to their com-
poaltion. 
187 -
The correlation values between the five consistency 
variables and the variable for risk evaluation and 
ri8k-benefit estimation have been compiled in Table 
12. 
The variables on risk propensity are highly significant 
for both dependent variables (risk-benefit estimation 
and semantic differential). Persons of high risk 
propensity or interview subjects with a preference for 
high benefit estimations gave a highly positive risk 
balance estimation for motor cars, aircraft and Auto-
mation, but also nuclear energy and electrical appli-
ances. The opposite relation, i.e. the estimates given 
by rilk-aversive or detriment-oriented persons revealed 
the same trend, but the individual correlation values 
are only half as great. Persons who frequently marked 
the zero category were especially critical toward air 
traffic and had a slightly negative attitude toward 
motor cars and nuclear energy_ These relations suggest 
the conclusion that a positive risk balance in the 
evaluation of the risk sources of air traffic, motor 
cara and nuclear energy requires an overall positive 
attitude toward risk-taking, while a neutral risk-taking 
attitude already leads to negative evaluations • . Another 
striking feature of this table is that the risk evalu-
ation for alcohol and smoking exhibits only slight dependence 
on the risk propenBi ty characteristic. Only the risk-
aversive persons transfer their negative risk-taking 
attitude to the habit-forming products. This results 
in an inverted relationship. while a positive risk 
evaluation of technological and industrial risk sources 
is associated with a positive risk propensity, this 
attitude i8 not requi~ed for the habit-forming p.oducts. 
thia is indicative of a trend where habit-forming pro-
ducts which can be individualized are not classified a8 
Table 12: Dependence of the Risk-Benef~t Est~tion and the Semantic Differential 
on External Variables 
Correlations High rJ.sk lU..k Ri.k Benefit Detriment 
External variables propenaity aversion neutrality orientation orientation 
Smokinq Risk 0.13 ' -0.44 
-0.16 0.12 -0.37 
SD , 0707 -0:'29 -0.13 0.05 -O:2i 
Electrical ' Risk 0.22 -0.12 -0.21 O.SJ -0.09 
appliances · SD. ~ 0.03 -0.00 o:1i -0.09 , 
Nuclear po- Riak ~ -0.4B -0.24 0.3B -0.60 
" :0:30 0.46 :o:4J ver stations SD 0.46 -0. 40 
• 
. ;Ri.k 0.63 -0. 29 -0.47 0.57 , -0.49 Aircraft SD o.sJ -O:T7 -o.n 0.55 ~ 
- 0.20 Alccbol Risk -0.37 -0.02 0.15 -0.39 so 0:17 -0:39 -0.03 0:02 -6:33 
----
Coal Risk 0.21 -0.14 -0.22 0.10 -0.17 so o:n -0.16 -0.04 0 . 14 -6:n 
Motor car. Risk 0.60 -0.31 -0.29 0.40 -0.41 so Q."ff -0.15 0.01 0.26 -o.IT 
---
i Beroin Ri.k 0.10 -O.OB 0.02 0.07 0.05 so 0.01 . -0.13 0.05 0.13 -0.15 
---
Autc:aaticn Risk ~ -0 . 36 -0.27 0.41 -0.37 SD 0.40 -0:30 -o.i2 Q.3i -0.20 
----
Sk1.inq Risk 0.34 -0.29 -0.09 0.38 -0.23 SD 0.30 -'O:2ir -O.OS ii:'i2 -0""726 
SkUn .. lll.lc. 0.29 -0.23 -0.24 0.23 -0 . 26 SD 0.:.22 ~ -0:09 0:33 -0:12 
\-- .... ....,.~- "':L_"" : '.'1, ="!:}} -0. 1.:' 0. 2. ~;» -. -co , .. '" ._ .. ____ ¥~~1f . 
.- ,.--.- -
-
Ilisk - riak-
benefit est.1.!lation 
SD - risk evalua-
tion based on the 
8U11D8d values of 
the semantic diL-
Ler.nUal 
p - 0 . 99 
P - 0.95 
... 
&l 
TAb~e 13~ Dependence o~ the R1.k-aene~~t E.t~mat~on and the Bemant~o D1~rerent1a1 
on Externa1 Var~ab1e8 
Correla.tions Sex 
EXternal variable. 
&laking Risk 0.04 SD 0 . 01 
Blectrical Risk 0.13 
appliances SO 0.08 
Nuclear po- Risk -0.13 
war BtationS SO , -0.17 
Aircraft Risk -0.21 SO -0:15 
---
Alcohol Risk -0.17 so -0:19 
Coal Ri.k -0 .13 SO -0. 05 
Motor cars Risk 0.06 SO 0.03 
Beroin Risk -0.01 SO -0.04 
Auto.at1on Risk -0 .01 SO -0.05 
Sk1ing Riolt 0.06 
SO 0 .. 02 
-
Skiing Rillk -0 .. 24 SO -0.11 
Peeticides Riolt -0.46 
' : : SO -0.39 
"". 
Clas. 
-0.21 -0.17 
-O:T8 -0:16 
0.07 -0.09 
0.05 -0.12 
+0.19 +0.14 
.0:15 <0:14 
-0.10 0 . 14 
-0.19 o.I6 
0.08 -0.21 
-0.15 -o-:I9 
0.10 0.11 
0.16 0.06 
0.08 0.07 
0 . 04 0.01 
-0 . 22 +0.05 
-0:16 0.04 
+0. 13 0.15 
+0.17 0:16 
--- ---
0 .. 10 -0.08 
0.1l -o.lS 
0 . 25 0.13 
Q.19 0.08 
0 . 22 -0.10 
5:j6 -0:1; 
Pol~t~c&1 party prefer.nee 
COO gPD P'DP 
-0.04 0.01 -0.08 
-0.02 0.02 -0. 07 
0.13 -0.08 -0.12 
0.16 -0.07 -0.12 
0.18 -0.17 -0.09 
0:22 -0:22 -0.10 
0.06 -0.09 -0.03 
0 . 05 -0 . 09 -0.09 
-0.02 o.Os -0.06 
-0.08 0.04 -0.04 
-0.02 0 . 08 -0.18 
0 .. 01 0 . 02 -0:19 
-0.02 -0.11 -0.06 
0.02 -0.15 -0.12 
-0.10 0 .. 10 -0.04 
-0.11 0.14 -0.08 
-0.13 0.03 -0.03 
-0.12 0 . 00 -0 .. 02 
-0.05 0 .. 01 -0.13 
-0 .. 07 0.02 
-0.13 
0.02 0.03 0 .. 01 
0 .. 01 0 . 08 0 .. 03 
0 .. 01 -0.07 0.06 
0 .. 07 -0.08 0.03 
. -
-- --
Rialt - risk-
benefit estimation 
so - risk evalua-
tion based on the 
.~ values of 
the semantic dif-
ferential 
p - 0.99 
p - 0.95 
CJXJ - Christian 
Democratic Party 
SPO - Social 
DeDocratic Party 
PDP - Liberal Party 
... 
ex> 
'" 
less acceptable until a deliberately negative relation-
ship to risk on the whole exists, technological or 
industrial risks, on the other hand, must already prove 
their justification when no positive attitude toward 
risk-taking exists. However, this inverted relation 
does not apply to heroin which is given such a high 
negative evaluation by all the interview subjects that 
significant correlations cannot exist at all. 
However, the limits of interpretation capabilities must 
be pointed out once again: high correlation on the 
risk propensity scale merely means that interview sub-
jects who gave very high or very low evaluations to 
11 risk sources also did this consistently in the case 
of the twelfth source. Whether this trend toward con-
sistency actually reflects the dispositive characteris-
tic of risk propensity, the degree to which this is 
merely verbal, but not actual behaviOur, whether some 
persons do not have different matrices and evaluation 
patterns for numerical values, cannot be clarified by 
the present study. 
Effect of Demographic and Social Characteristics 
Some demographic and social characteristics were also 
investigat~ as part of the risk perception study. 
Again, a few simple correlation analyses to determine 
the correlations with risk evaluation and risk-benefit 
- estimation can reveal the effect of this variable. 
The coefficients have been compiled in Table 13. 
First, it is clearly seen that political party preference 
does not play any role, with the exception of nuclear 
energy risk estimation. That nuclear energy would be 
191 -
more likely to receive a positive evaluation by conservative 
voters (Christian Democratic Party) and a negative evaluation 
by socialist party voters (Social Democratic Party) 
could have been expected on the basis of the most recent 
opinion polls, a detailed discussion will follow in 
Part V. But the variables of sex, age and clas8 are 
of greater interest. This applies especially to pes-
ticides, where women, younger persona and members of 
the upper classes are more likely to give negative 
evaluations. With the exception of class membership, 
the same applies to nuclear energy. In either case 
it i. likely that the "artificial nature" of the ri.k 
source, the involuntary risk-taking which cannot be 
controlled by the individual and the possibility of 
cataatrophic detriments have a greater repulsive effect 
on women and younger people. Even at this early time 
the prediction appears justified that pesticides and -
through them - all chemical interventions in the food 
chain, from the point of view of the overall ri.k per-
ception structure will encounter the same acceptance 
problema from which nuclear energy suffers today, al-
though the topic may not be as controversial because 
the specific risk-benefit distribution for nuclear en-
ergy is still further removed from the congruent model 
than it 18 in the case of pesticides. 
The sex variable also affects the risk estimation for 
aircraft, alcohol (but not smoking) and X-ray diagno.-
tics. Again, only speculative explanation patterns 
can be given. It ia possible that a fear of radiation 
which Cannot be perceived by the sensory organs and a 
apecial sensibility with respect to health hazards are 
deCisive for the negative evaluation of X-ray diagnos-
tic. by women; it is likely that women .uffer from 
alcohol abuse by their mates more frequently than vice 
192 -
versa, and it is a known fact from the psychology of 
air traffic that women suffer greater anxieties prior 
to flying than men (which, however, is not yet an 
explanation of this phenomenon). The degree to which 
socialization-related factors (learning to express 
feelings), role-specific patterns (women are responsible 
for family, health and food) or even endogenous phenom-
ena (hormone changes, pregnancy) are potential influenc-
ing parameters, would be an interesting problem for 
future studies. 
Age- or class-specific relationships are less clearly 
developed. Nicotine and alcohol are given a more 
negative evaluation by the members of the upper clasle., 
although they do not demand more stringent measures 
against them. Automation on the job exhibits a slightly 
positive correlation with class and a slightly negativs 
correlation with age. contrary to the intuitive assump-
tion that it is especially the older employees who feel 
insecure because of the introduction of modern machinery 
on the job, it is especially the younger employees with 
low occupational prestige who perceive the automation 
risk to be negative. Another surprising feature is the 
greater risk estimation for X-ray diagnostics given by 
young people: it is possibla that this discrepancy is 
due to better information on potential radiation hazard. 
on the part of the younger generation. 
Finally, it shOUld be mentioned that younger persons and 
members of higher classes are more likely to vote again.t 
a prohibition of heroin (and instead would prefer volun-
tary abstinence), and that both types of persons would 
also prefer more stringent measures against nuclear 
energy. 
193 -
The comprehensive preparation for the present study 
in 1978 by accident yielded an interesting result. 
In the factor analysis of the evaluation scale, i.e. 
the EiRbbein weighting factors (64 items) for the 
belief scale, three factors were revealed as signifi-
cant dimensions which could be labelled as follows . 
- direct, subject-related advantages and disadvantages 
(more benefits, greater income etc.): 
- indirect, subject- and group-related advantages and 
disadvantages (health, education etc.), 
- society-related and socio-political advantages and 
disadvantages (more democratic rights, economic 
advantages etc . ). 
The subsequent correlation of these factors with class 
membership revealed a clear-cut relationship which haa 
been plotted in Figure 22. This shows that the lower 
class attributes major weight to direct advantages, the 
lower middle class to indirect advantages and the upper 
middle class to the society aspects, while the upper 
class aSSigns the same weight to all three factors. 
This result,which reveals class-specific weighting 
patterns for argumentation levels,was already presented 
aBa thesis at the Miami Energy Forum in 1978 (Engelmann, 
Rann, '80, 96, p. 364). 
In the survey itself, using 100 interview subjects each 
and a stricter belief scale , more differentiated claas-
Specific differences in weighting behaviour were re-
vealed; Similar to the factorial analysis of the aggre-
gate belief scale, the analysis of the Fishbein weights 
(evaluation scale) revealed five factors . 
I~ -
direct and indirect effects on health, life, 
happiness and security; 
economy- and society-related values such as quality 
of life, supply, social justice, environmental 
pollution; 
projective and political advantages and disadvantages 
such as modernization, advantages or disadvantages 
to future generations, progress of society, democrat-
ic rights; 
socio-political values such as freedom, participation 
by citizens, education, social justice; 
direct personal advantages and disadvantages such as 
pleasure, financial advantages, convenience, happi-
ness. 
Pigure 2l shows the relative significance of these five 
factors for each class category. Similar to the pre-
liminary studies, the significance of subject-related 
aspects declines with class level, but without leading 
to any differentiation between direct or indirect con-
sequences. At the same time, the importance of society-
and economy-related values rises with the class index 
on the average. Projective advantages and disadvantages 
are given a higher evaluation from the lower middle 
class up, but the differences are relatively small here. 
Also, political aspects such as freedom and participation 
by the citizens are evaluated a8 important only from 
this class level, and the proportion of this factor 
rises continuously until the upper class is reached. 
The lower cla8s and ~ to a slightly lesser degree - the 
upper cla88 attribute more weight to direct, personal 
advantage8 such as pleasure and convenience than the ~ 
middle classes. 
" 
Low Lower 
Stratum 
High Lower 
Stratum 
direct benefit/coat 
Indirect benefit/coat 
eltruletlc benefit/coat 
Low Middle 
Stratum 
High Middle 
Stratum High Stratum 

("I 
Low Lower 
Stratum 
High Lower 
Stratum 
Pereonal Adventages/Pleasure (Factor 5) 
Sociopolitical Values (Factor 4) 
Low Mlddla 
Stratum 
High Middle 
Stratum High Stratum 
Projected Consequencles/Polltlcal Values (Factor 3) 
Economic and Social Values (Factor 2) 
Personal Direct and Indirect Advantages (Factor 1) 

A detailed interpretation of these class-specific evaluation 
patterns would go beyond the bounds of this study. 
Thus, it could be discussed in detail whether the 
middle class,which is especially receptive to 
career stimuli, deliberately reduces the importance 
of aspects reflecting a positive attitude toward life, 
such as pleasure and convenience (in the form of 
sublimation), or whether it has internalized the 
ascetic way of life in the sense of Max 
Weber's theories on the relationships between capital-
ist. advancement and anti-consumption values. Also in 
reference to participation research, the question con-
cerning the degree to which demands for democratic 
participation rights, citizen's participation and 
political freedom are not typical interests of an 
upper class which is privileged in any event, would 
have to be clarified. In the present study these 
questions can only be mentioned briefly. 
Overall Analysis of the Factors Influencing the 
Perception and Acceptance of Risks 
Up to this point, the analysiS of the results on risk 
perception was limited to a gradual incorporation of 
the relevant claaaes of variables: first, the specific 
beliefs on each risk source, subsequently the qualita-
tive characteristics and finally the external factors 
were discussed. In addition, evaluation of the seman-
tic differential whose summated value was used aa tha 
indicator for risk source evaluation revealed several 
additional influencing factors which were labelled as 
follows: 
- security, 
- aCientificality, 
- analogue classification (auch aa sympathetic, ethical), 
- POWer. 
200 
In order to determine the ranking of each of these 
factors for the evaluation of the risk source. a 
multiple regression was made by introducing all the 
relevant factors of the different classes of variables 
and the demographic and social characteristics as 
independent parameters and determining their relative 
weights on the basis of the values for the explained 
variance of the two dependent variables (risk-benefit 
estimation and risk source evaluation). However. this 
type of overall statistical evaluation must be inter-
preted with special caution since common interaction 
effects between the variables are taken into considera-
tion but their internal. causal structure is concealed. 
In particular. it should be pointed out that social 
structure data such as age and class have a quite dif-
ferent relationship to the dependent variable than. 
for instance. the different qualitative characteristics 
which in turn can depend on the aocial data. 
The six belief factors. the five most important quali-
tative characteristics (not summarized) and five of the 
six factors of the semantic differential (but only for 
the regression of the risk-benefit estimation). exclud-
ing the "general evaluation" factor. and the external 
variables of "dispositive risk propensity" and social 
structure data were selected as independent variables 
for each risk source. The total number of variables 
was 32 since dummy variables were uaed for the data 
which were not subjected to interval-scaling (sex. pol-
itical party preference). As in the preceding chapter •• 
the risk-benefit estimation and the sum indices for the 
semantic differential served as i!ependent variables. 
The individual classes of variables are colour-coded in 
Figures 24& to 24e (green • social and demographic 
characteristics 1 bl~.e • beUef scale factors, red .. 









u_, 
Con"QulneIN 
,--NIIk .. u. ... ~ 
VA: tIf, fttIIk 
-. ""',; ........ _ ..... 
........ 
Attanwtt,. 
AlQIMr, ..... 
u .... ~'1.,.... 
Rlek-Benefit &tlmetlon 
Dlapoeltlve and External Factor. 
F actore of the Belief - Syetem 
Qualitative Risk - and Beneflt-Properties 
F actore of the Semantic Differential 
Heroin 
A".Of"~t 
Soc:la' Risk, 
u_, 
"-'"'" 
Volont.ry RIIIk 
hneflt Oriented 
Per.onal D~nd.I't(:Y 
Willing to Take RIIk. 
All Oth.r Factor, 
UM.~ V.-nc. 
Semantic Differential 
All Other Facto", Accounting lor Le .. Than Two Percent of the Explained Variance 
- 211 -
qualitative characteristics, and yellow. semantic 
differential factors) so that the lignificance of 
these classes becomes evident at first glance. The 
first striking feature is the low explained value of 
the semantic differential factors, where the basic 
factor, general evaluation, had of course not been 
included. Factor 2 of the semantic differential, 
lecurity and risk, plays a role only in the caae of 
coal, nuclear energy and electrical appliances, while it doe. 
not appear at all in the case of the habit-formin9 
products . But, belief scale facto... of 
similar content are represented in the case of smoking, 
heroin, pesticides, X-rays and motor car., 80 that it 
is a nat"ral assumption that then factors have better 
correlation with the dependent variables, thUB, they con-
ceal the information content of the semantic differ-
ential factors which have a similar con-
tent, so that they no lonqer yield any additional 
explained values. 
In the case of the qualitative riak-benefit character-
istics, the previously obaerved trend ia confirmed that 
risk-benefit estimation or risk Bource evaluation are 
affected only when certain threshold values have been 
exceeded. Thus, it 1a Been in the CAse of the hablt-
fOrming products that a clear emphasi8 exist. on the 
voluntary nature of risk-takin9, and in the case of 
nuclear energy and pesticides that there is a strong 
wei9htinq in terms of the type and severeness of ri.k 
consequences. Also, the individual controllability 
component in the case of the motor car i8 more clearly 
expresed in this final analysis than it was in the indi-
vidual evaluations. On the other hand, the benefit com-
ponent which dominated in the analyais of qualitative 
ifi ce probably due characteristics decline. in si,.n can , 
already covered to the fact that distribution aspects are 
by the statements on the belief acale. 
- 212 -
The belief 8cale factor8 and dispo8itive characteri8tics 
or social data (blue, green) account for the greatest 
proportion of the explained value. And the ri8k evalua-
tion values can claim special validity because the in-
ternally acquired variables of risk propen. i 1., and benefit 
detriment orientation are genuinely independent . 
What are the conclusions which can be drawn for the 
individual risk sources? 
In the case of smoking and heroin the belief factor8 
of health, risk to life and plealure are predominant. 
The effect of general benefit-or-detriment orientation 
18 le88 developed. In the caae of nuclear energy and 
pesticides, risk propensity appears to be a necessary 
prerequisite for positive evaluation. In both case8, 
this variable plays a primary role. Socio-political 
and security-specific problema as well a8 the qualitative 
risk characteristics of -delayed and catastrophic con-
sequences· take second place. The positive aspects of 
benefits, supply, economic and projective advantage I are 
characterized by low variance components only. On the 
other hand, belief in possible personal advantages and 
consequences to health dominate in the cale ·of coal and 
X-rays. Theae 
influence an 
subject-related variable. have a decisive 
attitude. toward these risk 80urces. 
Ri8k propensity is a180 an important explanation factor 
in the case of motor cara and automation. Persona who 
exhibit a high verbal risk propenai.ty apparently tran8-
fer thi. attitude especially to motor car. and automation 
and, to a Ilightly le88er degree, to pesticides and 
nuclear energy, while persons of greater risk-8cepticism 
qive these risk sources an excessively negative evalua-
tion. 
- 213 -
The positive and negative belief factors are approxi-
mately balanced in the case of automation and motor 
cars. Economic and personal advantages are in favour 
of technology on the job and , at the lame time, the 
perception of aocial di8advantagel and the health 
hazard play a role a8 well. The motor car is aS80ci-
ated with the fulfillment of personal desirea and 
values, but alao with pollution and accident risk aa 
negative effects . 
Benefit-orientation i& of primary significance for the 
evaluation of electrical appliance., meaning that per-
aona who, on the whole, consider all riak sources from the 
point of view of their advantages, tend especially to 
tranafer this attitude to electrical home appliances. 
In addition, the factors of voluntarines8, economic 
advantages and progress to society have an effect. 
The presentation of the different component. of the 
explanation level in the form of circular legmenta 
demonstrates that the belief structure, the qualitative 
risk-benefit characteristics, some dispositive charac-
teristic. and - to a lesser degree - semantic differen-
tial factors can be used to explain approximately 50 
to 60' of the variance encountered in the risk-benefit 
estimations and in the .um index of the semantic dif-
ferential. 
Figure. 24& to 248 al.o demonstrate the weaknesses of 
aggregate analyses. new, hardly typlflable explanation 
pattern. result for each risk source which in themselVes 
are quite plausible but can certainly not be reduced to 
general structures which can be qeneralized tor risk 
acceptance. It we introduce aggregate data into the 
analyais, we create not only an aztit1cially homogenized 
- 214 -
framework 80 that the different risk sources can be viewed 
through standardized spectacles, we also obtain rela-
tive values over 0.90 which are not confirmed by the 
individual analys1s, or only to a much lesser degree. 
Impressive proof for this data manipulation can be 
found through a regression of the aggregate influencing 
factors for all 12 risk sources. Compared to the cor-
relation values to which we have become accustomed in 
the soclal sciences, Table 14 reveals a truely sensAtion-
al reBult: 
Table 14: Aggregate Regression Analysis for the Sum 
Index of the Semantic Differential 
Summated Declared Additional Simple 
Independent 
variable 
correlation variance explanation correlation 
value 
S 6 (summated 
beliefs) 
S 5 (personal 
advantages 
and dis-
advantages) 
0 2 (risk dis-
tributiotl1 
0 1 (voluntari-
nes8) 
0 12 (no alter-
natives) 
o 8 (minor con-
sequences) 
0.959 
0.974 
0.986 
0.992 
0.999 
1.00 
0.92 
0.95 
0.974 
0.98 
0.998 
1.00 
, 0.92 
0.03 
0.03 
0.01 
0.01 
0.001 
0.96 
0.86 
0.96 
-0.30 
-0.38 
0.47 
With the aid of six variables, the median of the indepen-
dent v~rlable, risk evaluation, can be predicted to an accu-
Tacy of two decimal places. This type ,of determiniatic 
relationshlp 1s ~arely f~und even in the natural sclences. 
Aa much as we might be pleased by this clear-cut and 
satisfying result, we must conSider it A product of the 
-
215 _ 
statistical procedure used where the greater part of 
the non-clarified variance has been excluded by form-
ing the mean-values, sinoe the individual data were 
aggregated into nine cases (risk Boureea representing 
the individual case) . If we neglect, for the moment, 
the quantitative value. and study only the influence 
chain as Buch, the results of these regressions can be 
interpreted in a qulte meaningful way. 
The most important influencing factor is the sum index 
of the belief scale, followed by the perceived subjec-
tive advantages and disadvantages (which 1n part are 
already contained in the sum index). The third place 
i. occupied by risk distribution, again an indication 
of the siqnificance of this factor to risk perception, 
followed by the qualitative risk characteristics of 
·voluntarlne88, no alternatives possible and minor con-
8equences~. If we use only the aggregated value., we 
can draw the conclusion that a risk Bource which 1_ 
characterized by an overall positive belief structure, 
by perceived subjective advantages, by voluntary riak-
taking, by an equal distribution of benefits and risk., 
by the absence of catastrophiC effects and by the ab-
sence of any alternative of equivalent benefits, 1s 
accepted without any problem. 
This rule provides little information on the riak sources 
whose acceptance 1s subject to conflict: 1n the case 
of nuclear energy and pesticides as well as the habit-
forming products, specifiC ri,k perception pattern, 
occur, auch as the socia-political aspects, each of 
Which have an effect on the risk-taking decision, but 
Without contributing to the explanation model on the 
aggregate level. At best, this six-variable model can 
be used as a starting point for risk evaluation, pro-
Vided that the specific characteristics of the ri'k 
-m-
source. 1n question are al.o analysed 1n detail. To 
that extent. a possibility develops in principle 
whereby a global model can first be used for a rough 
acceptance determination. to be followed by individual 
analyses to clarify the degree to which special factors 
in the different factor cla •• es influence the empirical 
acceptance by the population. 
- 217 _ 
Part V 
Nuclear Energy Attitudes 
and their Determinants -
Results of Sampling Surveys 
- 218 -
Post-War Nuclear Energy Attitudes of the Population 
of the Federal Republic of Germany 
Since the early fifties, nuclear energy and nuclear 
power stations have been popular topics for commercial 
and academic opinion polling. While, in the fifties 
and sixties, public interest was centered on nuclear 
weapons applications, interest shifted to the peaceful 
uses of nuclear energy in power-generating reactors 
toward the mid-sixties. A brief historical description 
of the different phases of nuclear energy attitude 
formation is useful for following the trend of public 
opinion-forming over time (cf. Table 15). As far as 
the fifties are concerned, American studies must be used 
predominantly (Erskine, '63, 102, p. 180 ff.,; Douvan and 
Withey' 54, 82, p. 1 ff-; Levine and Modell' 65, 221, p. 275 ff.; 
Fischer et al, '51,113, p. 86ff;Back and Gergen '63, 20, 
p. 428 ff.) since German polling institutions did not 
conduct major nuclear energy surveys until 1969. 
Phase I: Until apprOXimately 1950/51, there was an 
ambivalent attitude toward nuclear energy: 
on the one hand, there was the fear of the 
horrors of nuclear Warfare, and on the other 
hand the acknowledgement that nuclear weapons 
were significant to ensure the military superi-
ority of one's own country. During this pe-
riod, the peaceful Use of nuclear energy plays 
no role at all and is either not perceived, or only 
vaguely so (Erskine, '63, 102, p. 180; Paschen, 
'78, 307, p. 3; Bieber, '77, 33, p. 3 ff;; 
Renn, '77, 331, p. 7 ff.). 
Phase II: During the fifties, there was a clear teversal 
of public opinion against nuclear weapons 
which found an expression in the Easter marches. 
- 219 -
A nuclear weapons stop was demanded world-
wide, and nuclear fall-out was considered a 
threat to one's own health, especially in the 
United States. The Atoms for Peace movement, 
initiated by President Eisenhower, created a 
counter-movement to the military threat and 
hopes for a reversal of man's destructive 
intellectual potential into a harnessing of 
natural resources for humane purposes. This 
metaphysical embellishment of nuclear energy, 
promoted for political reasons, prepared the 
first step toward a symbolic role of nuclear 
energy as a paragon of technology and innovation 
(Douvan and Withey, '54, 82, p. 2; Rosi, '65, 
345, p. 290 ff.;Levine, Modell, '65, 221, p. 275/ 
Hlifele, '75, 154, p. 44 ff.;Renn, '77, 331, 
p. 11 ff.) 
Phase III: When the test stop agreement was signed between 
the United States and the USSR, the spearhead 
of protest against nuclear weapons was broken; 
nuclear weapons took second place after the 
problems of substitute wars (such as Vietnam). 
The theory of the arms balance also gained ground 
among the population and, in addition, the re-
nunciation of nuclear bomb tests in the atmos-
phere reduced the general anxiety threshold. 
The 'question of peaceful uses still had little 
relevance. While there were still protests 
against the construction of research reactors 
in the fifties, where the typical characteris-
tics of traditional technological adaptation 
problems existed, the first commercial nuclear 
power stations could be commissioned in the 
late sixties without any appreciable protest or 
resistance (Bieber, '77, 33, p. 82 ff.; Schuster, 
'71, 403, p. 113 ff.; Erskine, 102, p. 162 ff./ 
Renn, '77, 331, p. 9 ff.) 
- 220 -
Table 15: Popular Opinions on Nuclear Energy Over Time 
(1945 - 80) 
Phases Years Attitude to- Attitude to- Knowledqe Polari-
ward nuclear ward peace- level zation 
weapons ful uses 
I 40-50 slightly neq- not yet low hardly 
ative, but differen- any 
acknowledqe- tiated 
ment of the 
nuclear bal-
ance of 
forces 
II 51-62 very neqative sl1qhtly low greater 
positive, for weap-
but still ons appl1-
not very cations 
relevant 
III 63-69 fairly neu- fairly better, none 
traI, less poSitive, but still 
relevant hardly neg- low 
at1ve, hiqh 
degree of 
indifference 
IV 69-7.5 no longer as qreat major- fairly greater 
relevant as ity poSitive, adequate before, ex- 20-301 neqa-
cept for t1ve, 10-151 
. prOliferation indifferent 
fears 
V 76-78 no longer 301 POSitive, relatively qreat 
relevant 25' negative, high 
45' ambiva-
lent 
VI 79-80 reqains pub- dualism be- relatively great, but lic influence ween per- hiqh weakening 
ceived inse- in confron-
curity and 
belief in tadon 
the necessity 
of nuclear 
energy 
- 221 -
Phase IV: In the late sixties and early seventies, the 
criticism of peaceful nuclear energy uses 
which had been aroused in the United States 
was seized upon in the Federal Republic of 
Germany. The problems were first discussed 
among scientists and only then presented to 
the public by professional critics. At the 
same time, local resistance to the construction 
of nuclear facilities developed. 
None of this would have had the result that 
nuclear energy grew into a major political 
problem if the symbolic nature of nuclear en-
ergy which had developed in the fifties had 
not been seized again and reinterpreted. The 
developing awareness of the limits of growth, 
the greater interest in post-material values 
on the part of younger people with university 
education, the greater sensitization of the 
population for environmentalism and nature, and 
the disappointed hopes aeto the effects of techno-
logical and scientific progress (such as space 
flight or cancer research) had the result that, 
in the minds of many individuals and groups, 
the symbolic attributes of nuclear energy, such 
a8 "progressive, clean, centralized and complex", 
turned into a negative connotation. 
In spite of the first opposition fronts against 
nuclear energy, there was initially a solidarity 
with pro-nuclear energy groups in the Federal 
Republic of Germany, most of which were close 
to the established institutions in politics and 
society (cf. Table 16). Opinion polls confirmed 
(cf. Allensbach, Table 33 and Bieber, '77, 33, 
p. 87) that the number of strong and moderate 
- 222 -
advocates of nuclear energy rose to approxi-
mately 70t, while less than one tenth of the 
population considered themselves opponents of 
the new technology. This surprising fact can 
be partly explained by the energy crisis in 
1973 where nuclear energy was perceived as an 
escape from the ,011 dilemma e Moreover I the 
persons who were opponents of nuclear energy in 
the fifties and sixties (characteristics: 
conservative, advanced Age, low specific know-
how, fear of losing social status) failed to 
find a social equivalent in the new protest 
group against nuclear facilities (persons of 
younger age with university education and a 
more leftist political trend) and, therefore, 
fe·lt insecure for political reasons. On the 
other hand, the protest movement was not yet 
strong enough to convince the potential adher-
ents of the new innovation protest (Paschen, 
'78, 307, p. 5 ff.; Bieber, '77, 33, p. 85 ff.; 
Renn, '78, 332, p. 10 ft.). 
Phase V: The characteristic feature of Phase V which 
started approximately with the protest activities in 
the Ge~an town of Wyhl (19.75) was the increasing consolida-
tion of nuclear energy opponents. Support by persons 
engaged in science and technology, the popular 
base in the local environment, the organization 
of citizens' initiatives and the reactions on 
the part of official politics which were clumsy, 
to say the . least, had the result that more and 
more citizens questioned the motives of nuclear 
poliey and adopted sceptical evaluations. Also, 
doubts as to the relevance of t he nuclear 
programme were increasingly reported in the 
media. 
- 223 -
Table 16: Nuclear Energy Attitudes in the Federal 
Republic of Germany (Summary of Various Polls) 
1) 2) 3) 4) 5) 6) 7) 
Posi- '75 Jan. '76 Dec.'76 Mid '77 End of May '78 Nov.' 78 
tiona '77 
Adher- 60 39 57 59 51 39 32 
ents 
Oppon- 16 20 41 40 27 18 36 
ents 
1ndif- 24 30 3 1 22 43 32 
ferent 
8) 9) 10) 11) 12) 13) 14) 
Dec. '78 Apr.' 79 July '79 July' 79 Aug.' 79 Jan.' 80 June 'SO 
Hesse Hesse 
Adher- 40 50 61 52 37 56 
ents 
Oppcn- 39 33 22 30 48 42 
ents 
1ndif- 21 12 17 18 15 2 
ferent 
1) Quoted from D. Goerke (Goerke, '75, 131, p. 1112) 
2) Infas study, FRG, No. 1315, 1976 
3) Sample Institut, Hamburg, source: letter to "Deutsches 
Atomforum", dated 4.2.77 
67 
32 
2 
4} Intermarket poll, source: Kernzeitung, New Information 
on Energy and Energy Policy, published by the 1nformations-
kreis Kernenergle, only edition, Bonn 1977 
5) Spiegel poll: Do We Need Nuclear Power? (Brauchen wir 
Atomkaft?), No.8, Hamburg '77, p. 163 
6) DlGOE Marketing Service, Vechta, May '78 (Goerke, '78 •.. , 
p. 133) 
7) Infra-Test, FRG, Nov. '78 
8) Stern, 17 .07.79 
9) F¥ankfurter Rundschau, 16.08.79, Infas-Hessen 
10) Same a. footnote 9 
11) Same as footnote 8 
12) Frankfurter Allgemeine Zeitung, 8.08.79, Forschungsgruppe 
Wahlen, Mannheim 
13) Der Spiegel, No. 19/1980, p. 44 
14) Emnid 1980 
- 224 • 
The discussion of the hazards of nuclear energy 
and some apocalyptic scenarios published by 
sceptics created a high potential of active 
resistance in the local environment of planned 
nuclear energy facilities which was discharged in 
ma&sivedemonstrations and sometimes even mili-
tant activities (Brokdorf, Stade, Kalkar). On 
the interregional level, the symbolic notion 
of nuclear energy for the environmentalist 
movement and for the decentralization movement 
- politically enriched by the Nixon scandal in 
the United States - developed further, and a 
llvely exchange of activists in the new anti-
nuclear power movement was initiated. Among the 
public this resulted in an increased perception 
of nuclear energy opponents who were ~egarded 
as being a majority. 
~uring this phase, a structurized attitude 
formed in most individuals; approximately 
25' were clear-cut opponents, approximately 
30' adherents, and the remaining 45' were 
ambivalent between the two extremes depending 
on the issues of the day and the political 
mood. Only a small proportion of not more 
than 10' were really indifferent (cf. Table 
16; Scharioth, Krebsbach, '77, 382, p. 3; 
Pas~ent '78, 307, p. 6 ff.; Renn, '78, 332, 
p. 108). 
- 225 -
Phase VI: While the years from 1975 to 1978 were char-
acterized by confrontation between opponents 
and adherents and the resulting polarization 
of views, the beginning of the Iran crisis and the 
Three Mile Island accident marked a trend toward 
less denunciation in the conflict and more 
attempts to find solutions that would allow 
political compromise. However, this should 
not be misunderstood as a levelling of positions. 
The fronts had neither softened, nor had they 
approached each other, but the conflicts had 
lost in virulence (partly due to lack of an oppar-
tunity),and both camps were undergOing an inter-
nal consolidation process to develop new argu-
ments and new ideas for future conflicts. 
In the general publiC, the opinion structures which had 
developed in the mid-seventies were more or less re-
tained. Following a brief opinion slump after Three 
Mile Island, an increasingly positive basic attitude 
in the question as to the necessity of nuclear energy 
developed again among the population. However, a number 
of contradictory results were obtained especially at the 
end of 1979 (for instance, between Infas and Studien-
gruppe Wahlen of Mannheim). An interesting feature in 
this conjunction was the increasingly critical evalua-
tion of the safety of nuclear facilities and their 
engineering maturity (reactions to the Three Mile Island 
accident) and the increasingly positive evaluation 
of their economic necessity. 
With this background on the social development of the 
nuclear energy conflict, the dat~ acquired in our studies 
can be better classLfied and understood. This empirical 
study to determine nuclear energy attitudes was carried 
out on the basis of population samples of five towns 
during the period from March to September 1979. These 
- 226 -
towns were selected using the following criteria: 
_ regional comparability (all towns are located in the 
same federal state); 
_ experience with nuclear facilities or confrontation 
with the planned construction of such a facility 
(however, the fifth town has no relationship to nuclear 
energy and was included as the control). 
Table 17, below, lists the data for the popUla-
tion's nuclear energy attitude. The indicator for the 
interview subject's attitude was his declared intention 
to vote pro- or contra- at a referendum on further utiliza-
tion of nuclear energy. 
With the exception of Hamm, the majority in all five 
survey units was in favour of continuing the construction 
, 
of nuclear facilities; about one third was in favour of 
a construction stop, and one fifth would either not go 
to vote or was still undecided. Significant differences 
are found between the values for Hamm (many opponents) 
and the other towns, and between JUlich (particuldrly few 
opponents) and the other towns. 
Little change in these basic relations is noted when the 
interview subjects are asked for their attitude toward 
nuclear faCilities, or the planned construction of such 
faCilities, in the vicinity of their homes. The 
correlation coefficient between voting behaviour in a 
referendum and a specific vote on a nuclear facility 
located close to the subjects' home is 0.68 (gamma 
coefficient). The high level of agreement in the fre-
quency distribution between these two variables may be 
considered evidence that a mentality which would approve 
nuclear power stations as long as they are not built in 
one's own vicinity, is not typical for the repondents 
under stUdy. 
- 227 -
Table 17: Voting Behaviour in a Nuclear Energy Referendum 
Hunic!pal:ttleB Kerpen Jlllich Bever-
"""en 
Samm Kalkar n 
Characteristics All survey Control Nuclear Nuclear Nuclear Fast 
areas town research power power ata- breeder 
facility station tion under under con-
construction etruction 
continued con-
struction of 46 51 58 44 31 47 
nuclear facili-
ties 
construction 
stop for 33 37 21 35 39 30 
nuclear 
facilities 
Undecided 
Would not 
to vote 
n 
16 9 15 18 25 16 
. 
go 5 3 6 2 5 7 
503 120 100 99 100 . 84 
Th.ls correspondence between gener,:ll attitude and 
local preference may be due to the fact that nuclear 
facilities already exist in four of these five survey 
areas, or that construction is in progress there, so that 
the existing reality has established a link between 
nuclear power stations in general and facilities located 
in the vicinity of the interview subjects' 
But the same parallelism of these two 
home. 
variables 
found in the present study was just as great for the 
control town of Kerpen (Spearman correlation index a 0.57) 
so that additional justification exists for speaking of 
a parallel trend in the evaluation of nuclear power 
stations far removed and close to one's home. This in-
terpretation is further supported by the fact that, in 
a previous study, this author found a comparably high 
correlation between these two variables (Renn, '77, 331, 
p.SO) . The perception of the necessity of nuclear power 
facilities and the appreciation that this might affect 
one's own way of life, would,therefore,not appear to be 
contradictory. 
232 
166 
80 
25 
503 
- 228 -
Effect of the Three Mile Island Reactor Accident 
on the population 
The most serious accident to date in the history of the 
peaceful use of nuclear energy, at least in the Western 
World, occurred en 28 March 1979 at the Three Mile Island 
Nuclear Power Station near Harrisburg, Pennsylvania 
(Kemeny Report, German edition, '80, 415, p. 18). This 
accident was initiated by mechanical malfunctions in the 
reactor and severely aggravated by a series of human 
errors made in attempts to correct these malfunctions. 
For days there were conflicting opinions as to whether 
an explosive oxygen-hydrogen mixture had developed and 
whether there was imminent danger to the population from 
explosions or from the emission of radioactive gases. 
Pregnant women were evacuated as a precautionary measure; 
thousands of citizens departed voluntarily from the area. 
Hardly any technological accident has ever triggered as 
broad a mass effect in the media as the Harrisburg 
accident. It is not the purpose of this study to describe 
or analyze the causes and consequences of this accident; 
rather, its object is to describe the reactions of the 
population to the Harrisburg accident. For this 
purpose, a description of the international reactions 
appears indicated first (cf. Table 18, below). 

- 230 -
In the United States and Sweden (but also in The Nether-
lands and Spain), there was a clearly negative shift in 
public opinion, while the drop in France and in the 
Federal Republic of Germany (but also in the United Kingdom 
and Switzerland) was limited and was recovered again in 
the course of the subsequent months. These different 
types of reaction to the Three Mile Island accident have not 
yet been given a theoretical explanation framework. 
The present study might yield sorne indications as to why the 
Three Mile Island accident caused such different reactions 
in different parts of the World. Due to 
the time phasing of the interviews conducted in JUlich 
the effect of this accident could be determined through 
comparison of the numerical data acquired prior to and 
after 28 March and, at the same time, several questions 
on the effects of the Three Mile Island accident were included 
in questionnaire No. III. The direct comparison shall 
be discussed first: 
Table 19: Three Mile Island Effect : Comparison of the ValUes 
for Nuclear Energy Attitudes Prior to and 
After the Accident (JUlich only) 
Attitude 
Pro nuclear 
energy 
Slightly 
positive 
Neutral 
Slightly 
negative 
Negative 
n 
Sig. - 0.151 
Lambda = 0.13 
Percentages Percentages 
prior to the accident after the accident 
6.0 4.1 
34.0 20.4 
28.0 38.8 
20.0 22.4 
12 . 0 14.3 
50 49 
- 231 -
Following the general trend in the Federal Republic of 
Germany, the purely quantitative effect of the Three Mile Island 
accident was relatively low in JUlich. Considering the 
small number of cases, the relation is not even signifi-
cant. An interesting feature of this distribution is the 
relatively great rigidity of the extreme attitudes and 
the greater migration from the moderately positive to the 
neutral attitude. This impression becomes even stronger 
when the values for the direct question as to the effect 
of the ~eeMile Island accident are consulted. 
Table 20. Reactions to the Thr.ee Mile Island Accident (in percent) 
Effects of the 
accident all towns Kerpen JdUch Beverungen Bamm ICalkar 
Positive 
attitude 1 3 0 1 1 2 
con£lrme4 
Opinion not 
influenced 22 19 21 26 18 27 
Merely stimu-
lated interest 33 36 39 30 34 21 
Slightly more 
negative 27 27 31 21 28 25 
attitude 
Much more 
negative 8 8 3 9 6 12 
attitude 
Previously 
negative 
attitude 10 6 6 12 12 12 
confirmed 
Median 3. 3 3.3 3.3 3.3 3.4 3.4 
In the case of more than SS percent of the interview subj~cts 
the Three Mile Island accident did not have any attitude-modifying 
effect; almost every third interview subject stated a 
slightly more negative attitude, and only one out of twelve 
subjects felt that this accident had caused a major attitude 
change. Again, marginal effects between neighboring 
- 232 -
forms of an attitude are found more frequently than a com-
plete deviation from a previously held opinion. 
In addition, subdivision of the reactions between nuclear 
energy adherents and opponents allows an estimate of the 
quantitative effects of the opinion change. Approximately 
10 , of the opponents stated that the accident had induced 
in them a much more negative attitude toward nuclear energy. 
The data for the national poll conducted by the Institut 
fUr Demoskopie in Allensbach are in the same order of 
magnitude. A more detailed study of our survey data 
revealed, moreover, that interview subjects claiming a 
large shift of attitude after the reactor accident gave 
especially inconsistent replies, had not developed a very 
strong position and exhibited below-average values on the 
knowledge scale. This leads to the natural assumption 
that persons who were concerned only marginally with the 
nuclear energy topiC prior to the Three Mile Island accident 
developed especially critical attitudes toward nuclear 
energy as a result of the extensive reporting on the 
accident without, however, having developed the strength 
of argument to support this opinion. The subsequent 
polls showed that, over time, some of the negative shift 
in the attitude of these persons was eliminated again. 
Interview subjects who had already developed a positive 
or negative attitude prior to the accident were given 
only slight cause to doubt their basiC attitudes. 
However, this evaluation of the Three Mile Island accident 
applies only to the general attitude on nuclear energy, 
not to the evaluation of the nuclear energy risk, where 
the available data indicate a greater shift towards 
negative evaluations. 
On the whole, the results of data analysis can be summarized 
to the effect that the variable consolidation of attitudes 
was responsible for the intensity at the Three Mile Island effect 
- 233 -
and that it failed to influence firmly established decisions 
on nuclear energy (invaccination effect), but that it did 
influence elements of these attitudes (such as the risk apprecia-
tion). Ih:.response to the Three Idj 1e Is land aceident the nuclear energy 
community frequently argues that the special safety pre-
cautions taken in German nuclear power stations are per-
ceived even by the public to be so exemplary that accidents of 
the Three Mile Island type cou~d not occur in this 
country. This hypothesis, too, was tested in the survey. 
In WUrgassen, Hamm and Ka1kar the nuclear power stations 
existing or 1n the process of construction were used 
for comparison, and in JU1ich and Kerpen the German nuclear 
power stations on the whole. 
The results of these interviews showed that most interview 
subjects evaluated the safety of German nuclear power 
stations to be neither poorer nor better than the safety 
of the American reactor on Three Mile Island. Almost 
half the interview subjects felt that an accident of the Three 
Mile Island type would be highly improbable in this 
country, but still possible. About 40' even believed 
that such an accident could occur in Germany at any time. 
On the whole, there was nO evidence of a greater confidence 
in German nuclear power stations. More than three quarters 
of the interview subjects were convinced that accidents of the 
Three Mile Island type could also occur in the Federal 
Republic of Germany. Consequently, a perception· of 
better safety in German facilities could not have been 
the cause for the minor shift of opinion after the 
Harrisburg accident. This lends further weight to the 
hypotheSiS of a high degree of consolidation of attitudes 
which cannot be upset by stressing events. 
- 234 -
Nuclear Energy Compared to Other Energy Sources 
The estimation of nuclear energy as a risk source can best 
be measured by the evaluation given to this energy source 
in comparison to alternative sources. The literature con-
tains a number of contradictory results on this topic: 
Maderthaner et al. in 1975 conducted a survey of 148 per-
sons in Austria, including persons residing 1n the Vicinity 
of reactor facilities, where the subjects were asked which 
of the following technological, industrial or social 
facilities involved the highest risk rate: a gasworks, 
a district heating system, an oil refinery, a psychiatric 
hospital, a nuclear reactor, a prison or an airport. 
The readiness of the interview subjects to move into the 
vicinity of one of these facilities or to accept its 
establishment in their neighborhood was evaluated as an 
indicator of intllitive risk assessment. Nuclear energy obtained 
the poorest rating among all seven risk sources. Compared 
to the other facilities it was perceived and evaluated 
as the (relatively) moat dangerous and least desirable 
risk source. Factor analysis revealed two risk types: 
hazard source due to technological equipment and due to 1magin-
ability of hazardous consequences (Maderthaner et al., '76, 236). 
- Unlike this study, a sampling survey conducted in 1976 
in the Federal Republic of Germany revealed precisely the 
opposite ranking order in the evaluation of risk sources: 
54 , of the interview subjects would prefer to move to 
the vicinity of a nuclear power station, 24 , to the 
vicinity of a coal-fired power station and 22 , to the 
vicinity of an oil-fired power station. Other risk 
sources were not included (Goerke, '76, 131, Part II, p.7). 
- A survey conducted by the Battelle Institute in different 
locations in the Federal Republic of Germany where a 
power station was in eXistence, under construction or 1n 
the planning phase, revealed a still more differentiated 
- 235 -
pattern. At the outset of this study in 1975 the pre-
vailing opinion was that, in the future, nuclear power 
stations should receive support with preference over 
all other energy sources. Throughout 1976 the vote in 
favour of nuclear energy declined significantly. Com-
pared to other energy sources such as 011, coal, gas 
and solar energy, an especially high risk was attributed 
to nuclear energy. Oil and nuclear energy were given 
approximately equal evaluations only in one city (Mannheim). 
In all other locations, nuclear energy always ranked 
last (Battelle, '76, 25, p. A74ff) . 
- In a survey conducted by this author in 1977, nuclear 
energy held the next-to-last rank when once again 
investigatLng neighborhood preferences. Popularity 
ranking: machinery-making plant (27 '), motorway (21 '), 
coal-fired power station (20 t), nuclear power station 
(18 ') and chemical plant (11 .). In the same study, 
nuclear energy was evaluated as being especially important 
for the future, compared to the other energy sources, 
i.e. 011, gas, solar energy and coal, but at the same 
time it was associated with an extremely high dislike 
(Renn, '77, 331, p.56ff). 
- In another survey conducted in 1978, a coal-fired power 
station was also more popular as a nelghboarthan a 
nuclear power station (42 t in favour of coal, 34 • in 
favour of nuclear energy). However, this survey also 
revealed that residents with long years of habituation 
to nuclear energy facilities preferred them on 
average, while the majority of citizens living in towns 
with coal-fired power stations would prefer to have 
nuclear energy facilities in their home towns (Goerke, 
'78, 132, p.93f). 
- 236 -
The present study confirmed the trend toward an increasingly 
negative perception of nuclear energy, compared to its 
alternatives. Coal- and oil-fired power stations were 
preferred by most interview subjects. Also, nuclear energy 
exhibited a high polarizing effect: one out of three 
subjectsselected it first, and about 40 'last. A similar 
effect can be shown for oil-fired power stations. 
However, the question as to the undesirable neighborhood 
of energy-producing systems provides very little information 
on general acceptance. This variable correlates with the 
aggregate belief scale on nuclear energy only with a factor 
of 0.17 (sig. ~ 0.02) and with the general nuclear energy 
evaluation variable only with a factor of 0.21 (sig. = 0.00). 
And with the potential voting behaviour in a nuclear energy 
referendum this variable has a correlation of only 0.11 
(sig. - 0.04). Consequently, a preference for coal and 
oil does not yet mean that nuclear energy is rejected. 
Rather, this preference merely means than most interview 
subjects when given a choice as to whether they would 
prefer coal-fired, oil-fired or nuclear power stations 
within the limits of their community, gave nuclear energy 
the lowest rating. 
Better than from the mere evaluation of energy sources, 
preferences for energy systems can be determined from the 
energy option selected by the interview subjects from 
several given possibilities for a longer forecasting 
period (in the present case, the year 2000). The interview 
subjects were requested to state a personal preference 
sequence of four different energy strategies for the 
future of energy supply. This resulted in the following 
values: 
- 237 -
Table 21: Priorities of the Desired Options (in percent) 
Options First Second Third Fourth 
Solar energy / 
alternative 35 28 23 14 
energy sources 
Coal energy 19 29 28 23 
Energy 
conservation 21 27 27 26 
Nuclear energy 25 16 22 37 
Compared to the three other energy options for the future, 
nuclear energy ranks last, as it did in the question of 
neighborhood preferences. The advocates of nuclear 
energy predominantly assign it the first or second place 
while almost all the opponents place it last . 
The correlation coefficient between voting preference in 
a nuclear energy referendum and the priority vote for pos-
sible energy options is 0.67 (gamma) . The correlation 
with the belief scale (eta ~ 0.62) and with the evaluation 
of nuclear energy on the basis of a semantic differential 
is on the sarne level. This relation is illustrated in 
Fig . 25 .• 
This close correlation between attitude-forming parameters 
such as belief system and general evaluation of 
nuclear energy with the desired future energy 
supply strategy could have been expected . But it is a 
surprising fact that energy system preferences undergo a 
major shift when the interview subjects are not questioned as to 
their personal preferences but as to the real develop-
ments. Table 22 clearly indicates this shift in favour 
of nuclear energy_ 
This inversion of Qrder between personal preferences and op-tions 
considered realistic indicates a gulf between the beliefs 
- 238 -
held by individuals as to what their future should be like, 
and their expectation as to what it will actually be like. 
MOre than half the interview subjects were convinced that 
nuclear energy will have become the most important energy 
source by the year 2000, but only 19.5 , would welcome 
this development. This contrast between preference and 
perceived reality is demonstrated in Fig. 26. In keeping 
with this estimate of real developments there is also a 
majority belief that most of the electric power will be 
generated from nuclear energy by the year 2000. A total 
of 65 , of the interview subjects attribute first place 
in this energy sector to nuclear energy, 15% believe 
in the priority of coal, and only 12 , believe that 
solar energy and other alternative energy sources will hold 
first place. At the same time, most of the interview 
subjects were quite aware that only a very small portion 
of our present-day electric power is generated from nuclear 
energy so that, obviously, nuclear energy must be expanded 
in order to assume its predicted dominant position in 
the future energy supply. 
Table 22: Priorities of the Expected Options (in percent) 
Options First Second Third Fourth 
Nuclear energy 52 20 18 9 
Coal energy 32 39 19 10 
Energy 
conservation 12 25 35 28 
Solar energy I 
alternative 5 16 28 53 
energy sources 
The results obtained in the present study clearly show 
that perceived future prospects and individual preferences as 
to the shape of the future are certainly not identical 
magnitudes but may even be diametrically opposed. This 
in percent 
Belief-Scale 
Semantic Differential 
Nuclear Energy 
Last Place Third Place Second Place First Place 
Rank Order of Nuclear Energy Compared to the Preference for Coal, Solar Energy 
and Energy Saving 



- 243 -
contradiction is not without relevance for the political 
acceptance of nuclear energy. The greater the conviction 
• 
on the part of the populace that their future way of life 
is beinq shaped aqainst their will and without any 
possibility of exertinq personal influence, the sooner 
we can expect political apathy and resiqnation 
as well as a retreat into sub-cultures or violent 
revolt. Anyone who would believe that the climax of the 
nuclear energy conflict has been passed already (R6thlein, 
'79, 339) misjudqes the explosive force of a development 
where the qulf between desired future livinq conditions 
and perceived reality widens continuously. 
Another questicnwhich concerned the allocation for research 
qrants for energy systems in a fictitious budqet 
of the Federal Ministry of Research and Technoloqy must 
also be seen under this aspect. The interview subjects 
were requested to assume the Research Minister's role 
and allocate the funds for the support of different enerqy 
options in accordance with their own preferences. 
Table 23&: Percentaqe Allocation in a Fictitious Enerqy 
Research Budget 
Energy Percentage allocations 
research 
grants 0 1-10 11-20 21-30 31-50 51-70 
Nuclear 
energy 13 11 9 31 16 8 
Fusion 
energy 38 15 32 9 1 4 
Solar 
energy 0 16 39 34 5 1 
Wind ener-
gy etc. 4 7 51 36 2 0 
Energy 
conserv. 11 22 46 14 2 2 
Coal liqu. 
and gasi-
fication 0 6 27 54 10 3 
71-100 
12 
1 
5 
0 
3 
0 
- 244 -
Table 23b: Mean Values of a Fictitious Energy Research 
Budget (in percent of the whole budget) 
Enerqy research All Advocates Opponents grants respondents 
Nuclear energy 27.4 30.0 21.4 
Fusion enerqy 15.8 15.3 13.3 
Solar energy 20.1 17.8 25.0 
Wind enerqy etc. 20.9 17.3 23.5 
Energy conservation 18.5 17 .2 21.3 
Coal liqu. and 
gasification 
21.4 20. 2 22.0 
- 245 -
The response pattern in this imaginary budget planning of 
research projects lies in the middle between the personal 
preference for a future energy option and the development 
considered to be realistic. The interview sub-
jects had to abstract themselves from their 
own person and perform the public function of a 
federal minister. The phrasing of these questions, however, 
emphasized the minister's policy-making authority to 
allocate the budget according to his own judgement. It 
is not surprising, therefore, that the values range 
between the extremes of personal preference and forecast future. 
With the exception of fusion energy, which meant little 
to the interview subjects, the percentages of the remaining 
energy sources range in close proximity. The values for 
coal liquefaction and gasification, solar energy and wind 
energy do not differ significantly. This applies especially 
to nuclear energy opponents who allocated 20 , of the 
research budget to each variant, with the exception of 
solar energy. 
The pOisoned water experiment had already revealed that 
personal risk-taking decisions are influenced by a subject's 
role as a public servant. And so it is possible that 
nuclear energy opponents, forced into the research minister's 
role, abandon personal preferences 1n favour of an overall 
approach and allocate support to nuclear energy, for 
instance I for economic reasons. However, the gulf 
which is revealed between personal preference and perceived 
development is more likely to be the result of a compensatory 
attitude in that nuclear energy opponents feel that it is already 
too late to do anything aqainst the nuclear programme. So they WQul§ 
allocate financial support to nuclear energy in order 
to optimize through research the safety aspects of a 
development which can no longer be stopped. It is likely 
that, in this case, the two explanation patterns overlap. 
- 246 -
In addition, subdivision of the percentage values into 
clasles reveals an interesting pattern of association with ex-
trl!me values (cf. Table 23a). Apart fran fusion energy which w •• 
included only because many well-informed interview 
subjects in the preteats had criticised ita absence, 
nuclear energy exhibited a greater fringe distribution 
than all other energy variant.. The high degree of 
polarization in the nuclear energy debate has induced 
a number of persons to asaume extreme positions which 
exhibit almost identical numbera at either end of the 
opinion scale. 
However, total rejection of nuclear energy does not imply 
total advocacy of aolar or wind energyJ nor does extreme 
adherence to nuclear energy imply total rejection of 
alternative energy sources. Coal liquefaction and gaSi-
fication and energy conservation do not even differ 
significantly in the contemplated research budgets 
Buggested by opponents and adherenta of nuclear energy 
(sig • • 0.34 and 0.11, respectively). 
The different distribution value. for nuclear energy in 
comparison to the remaining variants of energy production 
can be Itill further differentiated by including the 
lemantic differential for the different alternative. in 
the analYlis. 
Such a differential has been plotted in Pig. 27 for coal, 
nuclear energy and solar energy. The three curves 
demonstrate that .olar energy i. given highly positive 
evaluations in all categoriea and that coal is 
also given a positive evaluation under general and ana-
logous terms, with negative judgements only under the 
pollution and resourc t e was e aapects. However, nuclear 
energy i. regarded POsitively only in the categories of 
scientific and modern hil 
,w e it i. negatively perceived 
in all other categories. 



- 250 -
_ A- poa1tive bias scale compria1ng 3 items of the knOWledge leale: 
_ A negative bias scale compria1ng 3 items of the knOWledge leale: 
- Interest in the nuclear energy topiCS. 
In the free alsociations on nuclear energy, negative terms 
such a8 accidents, radiation risk, harmful effects on 
health etc. were dominant for both opponents and 
adherents. The number of these negative judgements is 
relatively independent of the attitude toward nuclear 
anergy or its risk estimation. The low correlation 
coefficient for both variables indicates that, initially, 
all the interview subjects associated nuclear energy 
with all the risks involved in it, regardless of whether 
these risks are considered to be very high or very low 
(gamma· 0.17). A detailed list of these free associa-
tions has been compiled in Table 24. Thla was an open-end 
question, and the interviewers had to clalsify the 
responses in accordance with a prescribed list of 
18 categories or, when the relponse did not appear to 
fit .any of these categories, report the response seperately. 
With respect to the estimation of the interview subjects' 
own level of information, more than two thirds of the 
subjects were convinced \hat they were fairly well or wellinfor:med 
on nuclear energy. In a 1'78 poll, an average of only 
40 , of the population stated they were adequately well 
informed (Goerke, '78, 132, p.122). However, the "adequate" 
category was the highest possible self-evaluation in that 
poll. In the survey implemented by this 
author in 1977, the number of peraona who felt fairly 
well or well infOrmed was alao much lower (Renn, '77, 331, 
p.66). However, the two surveys agree in one reapect. 
most individuals are sceptical with respect to the infor-
mation level of other citizens. ~n the 1977 survey 
77.7 , and in the present 1979 survey 82 , of the i~terview 
subjects believed that moat other citizens were poorly 
- 251 -
Table 24. Free Association. on Nuclear Energy 
(up to five entries per interview subject) 
Associations Absolute Percentage 
frequency Percentage of all case 
Accidents/incidents 229 10.5 45.8 
Three 1'l11e Island accident 227 
Reproces.ing plants 
(Gorleben) 173 
Citizens' initiatives 129 
Economical energy 
production 127 
Modern energy 
production 126 
Radioactivity damage 124 
Demonstrations 112 
Unsafeneas 91 
Reactor types 96 
Low risk 61 
Environmental degradation 59 
Dependability of supply 58 
Environmental pollution 56 
Efficient energy 
production 54 
Harmful effects on health 51 
Environmentally clean 35 
War/terrori... 31 
Miscellaneous 10 
10.5 
7.9 
5.9 
5.8 
5.8 
5.7 
5. 1 
4 . 4 
4.4 
2.8 
2.7 
2.7 
2.6 
2.5 
2.3 
1.6 
1.4 
0.5 
43.4 
34.6 
25.8 
25.4 
25.2 
24.8 
22.4 
21.8 
19 . 2 
12.2 
11.8 
11.6 
11.2 
10.8 
10.2 
7 . 0 
6.2 
2.0 
• zsz • 
1 informed on nuclear energy topics. or -fairly poor Y 
ubj t • self-evaluation of their information While the s ec s 
id a of the cognitive self-confidence level provides an e 
of the subjects, the next item of interest is a com-
If-evaluation and the objective scale parison between Be 
th kn ledne iroex. Some numer ical data shall value on e ow ~ 
illustrate this: 
Table 25. 
Variable 
Xnow-
ledge 
~ow~and Self-Evaluation of Individuals' 
Information Level (correlations) 
Se1£- Evaluation Positive Negattve 
evaluation of informa- bias bias 
of informa- tion level 
tion level of other 
persona 
Sign1fi-
cance 0.00 0.00 0 . 07 0.02 
Correla-
tion 0.23 - 0.17 - 0.07 - 0.14 (eta) 
It is seen clearly from this table that the subjective 
self-evaluation matches the objective know~ level to 
a ceresin degree. However. the low correlation coefficient 
of 0.23 appears to indicate that many individuals o~er­
estimate or underestimate their information level. 
The better the interview subjects can give correct answers 
on nuclear energy, the more sceptical they are with respect 
to the Information level of other citizens. This relation 
is also fairly weak (eta - - 0.17) but significant. A8 
expected, positive and negative bias exhibit a negative 
correlation with knowledge but the relatively low correla-
tion coeffiCient indicates that individual. with high 
negative or positive bias reject the whole of the opposite 
bias and probably give the correct answers to pure 
know~ questions on nuclear energy. High bias, therefore, 
- 253 -
does not necessarily imply les8 information on nuclear 
energy, but does imply unilaterally weighted knowledge. 
An interesting difference compared to the 1977 survey in 
Kerpen de&erves to be mentioned: in that town the nuclear 
energy advocates clearly exhibited more positive bias 
than the opponents exhibited negative bias. This relation-
ship has become approximately inverted in the present 
study. The median for positive bias is 1.75 in the case 
of the adherents, and the median for negative bias on the 
part of the opponents is 2.21. This relationship applies 
to all the towns surveyed. 
In addition to the positive and negative associations and 
the knowledge scale with the bias indices derived from it, 
a scale of 13 statements on nuclear energy was used to 
achieve a broad differentiation of nuclear energy beliefs. 
The numerical response categories were simply combined 
in a summated index. This scale is interpreted as a belief 
scale. Moreover, another, weighted belief scale was de-
signed by means of unrotated factor analysis. These two 
indices have a correlation coefficient of 0.90 so that, as 
a rule, they are interchangeable. 
The belief scale exhibits relatively high correlation 
with all cognitive variables so that, in the further 
analysis, it was used as being representative for the 
cognitive component. The emotional intensity within the 
attitude, used for this nuclear energy survey, was not 
measured by weighting the statements in accordance with 
Fishbein's model of the evaluation 8cale, as in the ri8k 
perception study, but by u8ing a semantiC differential. 
Based on unrotated factor analysi8, the value. were selected 
which ranked high in the general evaluation via the 
8emantic differential, two other variables 
- 254 -
were computed which are expected to prOvide information 
on the strength and consistency of the overall con-
viction. The strength index was formed from the sum 
of extreme response categories for discr~natively 
unique attitude variables (auch as belief, nuclear energy 
as the most-favoured option etc.). The consistency 
index describes the linearity of the response behaviour. 
The numerical value provides information as to the degree 
to which discr~natory questions yielded responses which 
matched the positions and attitudes. 
Table 26. Medians for the Consistency Index and 
Strength Index 
Arithmetic medians (Range 1 - 16) 
Variables 
Consistency Strength 
All 8.51 5.34 
Adherents 8.10 5.38 
Opponents 10.01 6.09 
Undecided 6.30 4.13 
Kerpen 8.38 4.72 
JUl1ch 9.15 5.01 
Beverungen 8.04 5.06 
Hamm 8.29 5.73 
Kalkar 8.76 5.79 
It is obvious that nuclear energy opponents and the JUlich 
and Kerpen interview subjects were more conaistent Ln their 
response behaviour, although most of the adherents lived 
in theae two towns Al 
• ao, Opponents exhibit less compro-
mising attitudes i n many nuclear energy topics, while 
the adherents and und id d ec e subjects prefer moderate res-
ponse categories. A high intereat 
topics and an adequate informa~ion 
in nuclear energy 
level have a positive 
• 255 • 
effect on the variables of consistency and strength. 
This tends to confirm the initial hypothesis that the 
cognitive stress in the case of the adherents, in this 
case initiated by non-consistent response behaviour and 
lesser attitude strength, is considerably higher than 
it is in the case of the opponents who, on the whole, 
exhibit a more homogeneous response structure. 
However, the most interesting result drawn from 
the analysis of energy options is that the opponents reveal 
a clear discrepancy between their own attitude and their 
perceived reality. This discrepancy was not noted in the 
case of the adherents. Thus, individuals who have. 
developed fairly positive attitudes toward nuclear energy 
and take the corresponding positions, are often contra-
dictory in their cognitive structure, meaning that their 
belief system may also include information which contra-
dicts their overall attitude. But at the same time, they feel 
that the general development of society confirms their 
decision. Conversely, nuclear energy opponents consider 
the development toward expansion of this energy source 
to be an unavoidable process, but have a consistent and 
well-developed negative (cognitive and affective) attitude 
structure. This suggests the conclusion that a positive 
attitude toward nuclear energy is usually associated with 
a cognitive internal stress, and a negative attitude 
with an external realization stress. 
The last attitude structure component to be discussed isthe 
behavioural intention. It was already mentioned in the 
theoretical concept that this propensity cannot be 
derived from the cognitive and affective components but 
is only partly attributable to these two factors. In 
the present study,behaviouraliRtention was measured by means 
of a scale which combines different statements on possible 
actions in the nuclear energy conflict. These state-
ment. were listed in a sequence so as to obtain an 
_ 256 -
tarting with the collection of excalation of actions, s 
d din" with active violence. From the signatures an en , 
1 index was computed which was label-values on this sea e an 
1 i t tion The median values for this led behavioura n en • 
variable have been compiled in Table 27, below. 
Table 27. Median Values for the Behavioural Intention 
Variable, Subdivided by Towns Surveyed 
and Nuclear Energy Attitudes 
Arithmetic ~1J8 All Kerpen J(jl1ch Beverungen 
median 
Behavioural 
intention 0 - 30 9.24 10.86 5.01 8.06 
Median H ..... Kalkar Adherents Opponents Unde-
values cided 
Behavioural 
1ntMltlOn 13.69 11.46 6.6 12.84 4.77 
Correct interpretation of the.e data requires a brief ex-
planation of the scale intervals. at a range frOID 0 to 30, 
the value zero means no commitment at all, the value 5 
means readinels to sign. declaration or, possibly, go to 
a meeting, the value 15 includes participation in a demon-
stration, a scale value of 25 meanl that the interview 
subject was in favour of probable participation in a site 
occupation or removal frem a aite by the police, and the 
maximum of 30 meanl that the 8ubj ect was in favour of 
active Violence. The median value of 9.24 for the total 
number of interview aubjects -meanl that the majority was 
prepared to o1gn a petition pro or contra nu::lear energy, 
to go to meetingl oecaaionally and, posaibly, to write 
letter8 to the editors of newapapers or periodicala. In 
Hamm and Kerpen the interview aubjects were prepared to 
take even stronger action, while in J(jlich most citizens 
- 257 -
were reluctant to translate their attitudes into concrete 
action. However, if the basic nuclear energy attitude is 
held constant, a relatively uniform distribution of behavioural 
intention is found in all five survey areas. 
The differences in behavioural intention are revealed by 
the scalar medians when the positions are differentiated. 
The average nuclear energy adherent would sign a petition 
for a pro-nuclear energy campaign and would perhaps go 
to a meeting, but he refuses to go any further. The 
average opponent, however, would go beyond and would 
write letters to editors, and some are prepared to work 
for citizens' initiatives and possibly even to take part 
in demonstrations. The undecided individuals in a 
referendum could at best be induced to sign a petition 
pro or CQ'lt ra nuclear energy. 
This discrepancy between adherents and opponents is atill 
more evident when the risk-benefit estimate is included 
as a control parameter. The more extreme the positive or 
negative perception of nuclear energy as a risk source, 
the greater will be the propensity to take action. However, the 
strength of this relation i. only half a8 great for a 
positive risk estimation a. it is for a negative risk 
estimation. Fig. l8 shows this relation in graphical form. 
At first glance, this confusing variety of cognitive, 
affective and action-relevant parameters fails to reveal 
any consistent pattern of relationships. However, path 
analysis, a statistical procedure which allows causal 
interpretations on the basis of the distribution of 
correlation coefficients, can be used to explain the basic 
structure of nuclear energy attitudes. The results of 
this path analysis have been plotted in Fig. 29 where 
the numerical data entered above the arrows describe the 
strength of each relationship. 
- 258 -
The model is characterized by three starting points: the 
perception of the economic necessity of nuclear energy, 
knowledge of nuclear energy and the risk-benif1t estimate. 
These three variables are not independent of each other, 
but their influence takes different paths. While the 
belief in the economic necessity of nuclear energy shows 
very high correlation with the pOlition stated for a 
hypothetical referendum, the risk-benefit est1mate 
influences both attitude components, the cogn1t1ve belief system 
and the affect1ve evaluat10n. Nuclear energy knowledge 
has only a m1nor direct relat10nsh1p with all of these 
var1ables. The informat10n level affects the evaluation 
aud the belief scale only via the circuitous route of 
self-evaluat10n and posit1ve bias. The strength of the 
weighted att1tude is obtained from the affective component, 
the stated bias and the values on the belief scale as 
w"ll as, indirectly, from the consistency of the verbal responlel 
in the questionnaire. The consistency in turn is 
influenced by the evaluation level. The final result,behav-
ioural intention, is mostly due to the strength of the 
stated attitude and, to a le.ser degree, to the position 
in a referendum on nuclear energy_ 
With respect to its content, thi8 model can be interpreted 
as follows: individuals who believe in the necessity of 
nuclear energy for economic reasons will vote in favour 
of the nuclear energy source in a referendum, 8~ulated 
in this study, even when they perceive a greater risk. 
And their preference for nuclear energy i. amplified if 
they are convinced of having a good information level 
and, in addition, have a number of po.itive biaBes for 
nuclear energy. 
The belief system, on the other hand, has a greater de-
pendence on the perception of the risk-benefit relation. 
It is true that a po iti s ve risk estimation i. a pre-
requisite for a positive attitude in a referendum, 
•
 
•
 
II 
c. 01 ...
 C 
.E
_O
 
=
~
 
.
.
 
:; 0 
u
 
~
 
.
.
 c 
c o 
;; 
CO 
E 
~ w :!:: - ., c ., m .;. .!! a: 

Percepllon of 
Economic Neec •• -
ally of Nucle.r 
Energy 
Knowledge 
about Nuc· 
I •• r Energy 
Rlak-Beneflt 
eatlmatlon 
Self-Perception o. 
Own Knowledge 
pOIIUon In a 
Referendum on 
Nucl •• r Energy 
Positi.e 
Preduficea 
BeUef-
Scale 
Evaluation of 
Nuclear Energy 
Strength of 
Attitude 
Conalstency 
of 
Attitude 
Willingn ••• to 
Acl (Conation) 

• 263 -
however, it is possible to compensate neqative values on 
this scale by takinq the bitter pill of economic necessity, 
althouqh this will not affect the qeneral beliefs and 
and evaluations of nuclear enerqy. 
When the risk and the economic necessity are both qiven 
a neqative evaluation, the result is total rejection which 
leads to an especially stronq neqative attitude. At the 
same time, neqative risk-benefit estimations and neqative belief 
systems require a hiqh internal consistency of the 
attitude-forminq variables, where coqnitive stress cannot 
even develop. The strenqth of an attitude, combined 
with the stated selection preference (pro or contra nuclear 
enerqyl, can predict a larqe portion of the variance of 
the behavioural intention variable. 
MOreover, polls conducted by other institutions suqqest 
the conolusion that the attitude component. aoquired via 
risk-benefit estimations have a relatively lonq life, 
althouqh events suoh as the Three Mile Island aooident do have 
a destabilizinq effect, but without oausinq any basic 
ohanqes. The stated basio attitudes or the 
simulation of a referendum are dependant on 
the politioal events of the day, sinoe the perceived 
degree of eoonomic neoessity varies, dependinq on the 
world-wide politioal situation. 
The relatively high risk estimation of nuolear enerqy 
and the preference for alternative energy sources, even 
on the part of nuclear energy adherent., make it clear 
that the position oallinq for an expansion of nuolear 
enerqy as a future option oan be maintained only if 
there is no doubt that the eoonomio necessity persist. 
in the peroeption field. It is likely that in times of 
eoonomic orisis many individuals will vote in favour of 
nuclear enerqy althouqh, based on their feelinqs and 
• 264 • 
t t they would be more likely to have cognitive s rue ure, 
a sceptical attitude toward this technology. However, 
theoanverse conclusion does not apply. Committed 
opponents of nuclear energy who not only perceive the 
risks of nuclear energy but also dispute its economic 
necessity will hardly become adherents of this technology 
even in times of crisis because they have a self-contained, 
consistent attitude system. 
5. Significance of Personal Value Preferences and 
Socia-Political Beliefs for the Formation of 
Nuclear Energy Attitudes 
In recent years an increasing shift from materialistic to 
post-materialistic, from quantitive to more qualitative 
values has been noted by social research (Inglehart, 'SO, 
lS3, p. 144ff, Kmieciak, '76, 20S). Although the 
polarity between materialist and post-materialistic 
attitudes is not adequate to describe the value changes 
which have taken place during the peat decade, the trend 
shows that, following quantitative aaturation phenomena 
in the consumer area, objectives aiming at more quality 
of life are gaining significance. 
In this conjunction the thesis i. often presented that 
the rejection of nuclear energy is a value-shift symbol 
and an expression of diasatisfaction with the consumer 
culture (Paschen, '79, 307). 
And, indeed, an earlier study carried out in 1975 revealed 
a high correlation between the desire for restrictions to 
economic growth and a negative nuclear energy attitude 
(van Buiren, '75, 46, p.250). Subaequent .tudies, however, 
yielded a more differentiated reault (Battelle II, '77, 
25, p.l07ff, Renn, '77, 331, p. 90ff). the majority of 
the population attempt to take a middle course between 
materialiatic and poat-materialiatic valuea without 
allocating any clearly defined priorities. 
265 --
In the present study. 17 general problems of society such 
as unemployment. terrorism or health care. and 10 per-
sonal values such &8 a clean environment or harmonious 
family life were specified in a scale. These terms had 
not been selected arbitrarily but determined through 
open-end questions in a pretest. In the actual interviews 
the subjects were requested to state the urgency and 
importance of these problems and values. The responses 
mostly confirmed the results of more recent studies in 
this field that there is a high degree of competition 
between materialistic and post-materialistic values in the 
eyes of the population and that priorities are established 
in a differentiated and ambivalent manner. 
Abstract problems of SOCiety such as 
unemployment, terrorism and inflation are classified as 
being seriOllS while, At the same time, personal values 
such as health. clean environment and harmonious family life 
are estimated highly. Purely material values such as 
raising the standard of living or high income are more 
likely to be found at the bottom end of both 8cales. 
Environmentalism. an important post-materialistic value. 
is given the greatest urgency with respect to both 
society and the personal environment. Other typical 
categories in this area such as social justice and enjoy-
ment of one's work. occupy intermediate positions. 
Comprising factor analysis allowed plotting of the 
SOCiety-related and personal value beliefs in a two-factor 
diagram whose axes represent the degree of post-materialistic 
qualitative value commitment and the degree 
of personal involvement (cf. Fig. 30). 
However. the structure of value beliefs is of little 
interest for the question as to nuclear energy attitudes. 
More significant is the relationship existing between 
nuclear energy attitude and value system which shall be 
illustrated with the aid of some correlation analyses. 



- 269 -
Both tables demonstrate that the stated value beliefs 
with the exception of the clean environment variable 
have only a minor effect on nuclear energy positions 
and attitude.. Again, it is interesting to note the 
greater tie. between economdc concepts and voting 
behaviour in a referendum (position), and between en-
vironment- or risk-specific values and the attitude 
components. That the energy crisis and nuclear energy 
have a discriminating effect on both positions is not 
surprising, however, the correlation coefficient is 
unexpectedly low. Nuclear energy adherents prefer 
economLcally quantitative values, and nuclear energy 
opponents prefer post-materialistic and qualitative 
values, but this relationship is weak with the excep-
tion of quite obvious relations (with respect to nuclear 
energy and the energy crisis). The only exception is 
that a spectal preference for a clean environment haa 
a great effect on the belief system on nuclear energy. 
Negative perceptions of nuclear energy are especially 
well-developed in those individuals who attribute 
special value to a clean environment. An interesting 
feature is that the general attitude toward the environ-
ment protection problem provides a much poorer predic-
tive value for the nuclear energy attitude than the 
individualized desire for a clean environment. Quite 
generally, environmental protection 1. an ~portant 
society-related problem for the adherents of nuclear 
energy a8 well . 
In addition to differentiation between opponents and 
adherents, the category of the undecided individuals 
must be discussed in more detail here, since it is this 
group in particular which takes a pro-materialistic 
and pro-economic position in the survey, which would 
have been more in keeping with the attitude of a nuclear 
- 270 -
energy adherent. The attitude of the undecided indivi-
duals becomes clearer if we look once again at the 
risk-benefit evaluation (medians. adherents +1.7, 
opponents -0.9 and undecided individuals -0.5). Due 
to their negative risk perception they cannot state a 
positive attitude toward nuclear energy, although this 
would match their economic value beliefs. 
Yet, these economic beliefs prevent a complete shift 
to the opposite side. Of cour.e, thi. dualism of nega-
tive risk estimation and po.itive economic evaluation 
does not apply to all undecided individuals I however, 
the medians in each case indicate that a majority of 
this group i. involved in an inner conflict between risk 
evaluation and economic value-orientation. This thesis 
is further confirmed by the fact that, in the question 
A8 to the economic nece •• ity of nuclear energy, the 
undecided individuals are much closer to the position 
of the adherents than to the poa1tion of the opponents 
(medians. adherents 2.1, undecided individuals 1.7 and 
opponents 0.2). 
In a number of studies the clos. proximity of socio-
political beliefs and nuclear energy estimations has 
been determined in addition to the value commitment. 
In order to study thi. hypothesis in more detail, the 
follOWing attitudes described in theoretical or empiri-
cal studies reported in the literature were meaaured 
with the aid of suitable scale •• 
- a confidence scale for science, politics and technology 
(models. Douglin, '76, 811 Goerke '78, 132)1 
- a conservatism scale (model. Dumanil, '77, 85)1 
- a ' participation scale (model. Douglin, '76, 81)1 

• 272 • 
The results of this reqression analysis have been 
compiled in Table 31, below. 
Table 311 Multiple Reqreasion Analysi. of Socio-Political 
Attitudes with the Belief System (Belief Scale) 
R R2 r Independent Multiple Declared Simple Siqn1fi-
variables correlation variance correlation canCe 
1- COnfidence scale 0.49 0.24 0.49 
2. Environmentalism 0 •. 56 0.31 0.35 
scale 
3 . Participation scale 0.60 0 . 36 0.41 
4. COnservatilm Beale 0.62 0.38 0.29 
5. Political apathy 0.63 0.39 0.41 
acale 
Conaequently, almost 40' of the belief aystem'. variance 
can be explained by the combined effects of these five 
scales. Of special siqnificance are confidence in 
science and technoloqy and the atrenqth of environ-
mental awareness . In addition, the desire for partici-
pation has an amplifyinq effect on a neqative nuclear 
enerqy evaluation • 
. The magnitude of the correlation values quite supporta 
the assumptiona that qeneral aocio-political attitudes 
0.00 
0.00 
0.00 
0 .03 
0.02 
are aiqn1ficant for nuclear enerqy perception. The.e 
findinqa are alao lupported by the hiqh correlation values 
with the riak-benefit estimatea. All five relationshipa 
to this variable have been plotted in Fiqure 3' . 
Consequently, low confidence in the statements of scien-
tists and technoloqiats, combined with a hiqh priority 
for environmental protection mean a more neqative nuclear 
enerqy risk perception from the outset. 
flO, Confidence In Science and Technology 
Anti-Fatalistic Point of View 
Tendency toward Con.ervatlve Social Velue. 
Environmental Awarene •• 
Desire for Participation 
Risk-Benefit Estimation 

- 275 -
COnversely, confidence in Bcience and technology and 
a low degree of environmental awareness represent a 
startinq attitude which tends to develop positive 
expected values for the nuclear energy attitude. 
In spite of theae convincinq results, over 60' of 
non-explained variance remains. This demonstrates 
that the deterministic relationship between the socio-
political attitude field and nuclear energy attitudes, 
postulated by Dumenil et aI, can hardly match real 
conditions, althouqh it was impossible to verify all 
possible, related attitudes within the scope of the 
present study. 
Effect of Social and Demoqraphic Characteristics 
on Nuclear Energy Attitudes 
Relationships which may be revealed in studies of the 
interrelationship of attitudes and external criteria such 
as sex, age, class membership etc. do not, at this stage, provide any 
information on their cauEal structure. Rather, correlations are 
indicative of related patterns of the response behaviour 
which, depending on the significance of the relationship, 
can no longer be dismissed as randall phenomena. This 
preliminary remark is especially important in the case 
of the 80clal structural characteristics because, as a 
rule, they can be interpreted neither as causative factors 
nor as consequences. In most eases, third factors can be 
dete~lned (intervening variables) which have an effect 
on both the social characteristics and the phenomena 
requiring explanation. Thus, sex in itself is hardly ever 
the orlqinal cause of any behaviour, but the role aSSOCiated 
with sex, specific types of education or training, or 
even stereotype self-perception. Therefore, althouqh 
the following discussion describe. some cross-relations 
between nuclear energy attitudes and social characteristics 

- 277 -
Sex-specific differences are also found in correlation 
analysis with attitude-forming variables. Women evaluate 
the nuclear energy risk higher than men and, consequently, 
also have more negative beliefs. Their attitude structure 
is less well developed and their behavioural intention slightly 
lower. This may be due to the fact that they occupy 
neutral positions more frequently. 
An interesting feature is the relationship to the "economic 
necessity of nuclear energy"variable where there is no 
difference between men and women and where the correlation 
coefficient is even slightly negative (- 0.07, not sig-
nificant) • At the same time, women are especially open-
minded with respect to alternative energy sources as desirable 
options for the future. OVer 45 • of the female nuclear 
energy adherents opted for solar energy in the first 
place, ahead of nuclear energy. The discussion of these 
results is highly speculative. It is likely that women 
react with much greater sensitivity to risks which have 
detrimental effects on health, and are guided more by 
archetypal beliefs such as the Sun as a life-giving power 
(WUnschmann, '80, 460). However, the value beliefs and 
the socio-political attitudes included in the questionnaire 
revealed almost no significant relationships which would 
further support this interpretation. 
While it is true that a large proportion of women are 
convinced of the economic necessity of nuclear energy, 
this Aspect apparently bas a lesser effect on the evaluation 
of the different beliefs. Consequently, risks to human 
11fe and to the environment have priority over economic 
objectives. Perhaps men are more receptive to economic 
necessities due to their direct involvement in the work 
environment and can therefore compensate risks which 
they perceive as negative. Women, on the other hand! 
are more remote from business and, while perceiving 
economic developments and constraints, subordinate them 
to their personal environment. 
• 278 • 
By~!c!£_~n!t9X-~ttl~~Cl-!D~_~9! 
The relationship between age and nuclear energy attitudes 
i8 lel8 clearly defined than the relationship to 8ex . A 
linear trend cannot be identified. A more negative 
attitude 1s found in the 25-36 age group, a positive high 
in the 18-25 age group, average values in the middle age 
groups and, again, a 8lightly negative attitude in the 
group ovsr 56. These data largely match the re8ults of 
the 1977 8tudy where a preponderance of negative attitudes 
was found in the 26-35 age group (Renn, '77, 331, p. 109). 
It i8 pollible that thil distribution is caused by 
8uperimposing conventional and modern innovation 
protest, a phenomenon already diacus8ed in the theoretical 
part. Traditional individuala of advanced age who are 
frequently opposed to innovation in their living environ-
ment tend to exhibit the same rejection of the use of 
nuclear energy as the younger generation atill affect&d by the 
unre8t of the sixtiea, while the very young prefer a positive 
nuclear energy attitude because they believe in progre88 
and the 35 to 40 age group prefer power stattons for reasons of 
aocial advancement. 
IIl1clelll:_tuel:a¥_litt1tIlI1CI_IIII1_!1!;rU!!!!:!H2!! 
The cla88 membership characteristic comprioe8 data on 
income, occupational prestige and educational level. 
Again, it is of intere8t to determine the degree to which 
the aocial preatige of an individual can be related to 
hi8 or her nuclear energy attitude. Starting out from the 
simple correlation values between etratum and attltude-
forming variables, it i8 found that there i8 a marked 
effect on nuclear energy knowledge and a lesser effect on the 
b&haMioural intention, general belief syotem and negative 
bias. Th .. connection between knOWledge and cla.. is not 
ouprising, since it io· expected that individuals of higher 
education level and occupational · prestige would e180 have 
- 279 -
a higher information level. Of greater theoretical interest 
is the fact that negative nuclear energy attitudes and propensity 
to take actions increase slightly with higher class 
membership. This demonstrates that the positive correla-
tion between class membership and nuclear ener~ attitude 
which was found years ago can no longer be confirmed. 
This relationship becomes even more evident when the class 
variable is subdivided into its basic quantities. 
Income still exhibits a positive correlation with the 
nuclear energy attitude (r - 0.17), while occupational 
prestige and education have a negative effect (- 0.19 and 
- 0.24, respectively) . All of these values are Sig-
nificant on the 99 , confidence level. Apparently, there-
fore, an inversion in the soclal make-up of nuclear energy 
opponents and adherents has taken place in recent years, 
an effect also reported in the United States and Canada. 
This shift demonstrates an inversion of the innovation 
protest from a conservative, more ignorant and soc1a1ly 
defensive rejection attitude to a progressive, educationally 
elitist and socially offensive movement. Initially, this 
change is taking place on a cognitive basis only, while 
the affective classification lags behind, since the 
general evaluation failed to reveal any significant 
differences between the members of the different social 
classes. 
This pattern of class-specific innovation protest can be 
drawn with still greater clarity if the correlation values of the 
stratification index are diffe~entlated by attitudes. These 
data have been compiled in Table 33, below, and plotted 
in a graph (cf. Fig. 33). With increasing class level 
the opponents exhibit an overproportional growth in 
nuclear energy knowl~ and behavioural intention. Also, 
they show a greater firmness (strength) 
of attitude than the adherents who tend to 
become more sceptical in their views and risk-benefit 
• 200 • 
Table 33. Effect of 
Attitudes 
Opponenta 
Stratification on Nuclear Energy 
(Subdivided into Nuclear Energy 
and Adherents' 
Variable Correlation 
coefficients 
Pearson r 
Nuclear energy 
risk estimation - 0.12 
Xnowledge 0.27 
-
Negative bias =-Q.:.11 
Belief system - 0.18 
------
Nuclear energy 
evaluation - 0.08 
Nuclear energy 
safety - 0.11 
coal 
evaluation - 0.17 
-----
Solar energy 
evaluation + 0.12 
Attitude 
strength 0.06 
Attitude 
consistency ~ 
Action 
propensity 0.17 
----
n 422 
------- p - 0.95 significance 
p - 0.99 significance 
Pearson r Pearson r 
adherents opponents 
- 0.26 - 0.04 
0.16 0.34 
-
0.03 - 0.16 
- 0.14 + 0.13 
- 0.02 - 0.11 
- 0.14 - 0.07 
- 0.15 - 0.18 
------ ------
0.14 0.11 
- 0.05 Q!.l~ 
0.18 0.28 
- -
0.11 Q"ll 
187 144 
In percent Knowledge on Nuele.r Queotlono (Oppanenta 01 NucIHr Enernl 
Knowledge on Nuele.r Queollono (Proponenta 01 Nuclear EMfIIJI 
Willing .... to Take Aellon (Opponental 
Wllllngn •• to Toke Aellon (Proponental 
Strength 01 Attltudln.1 Commitment (Opponent., 
Strength 01 Attitudinal Commitment (Proponento) 
Low Lower 
Stratum 
High Lower 
Stratum 
Low Middle 
Stratum 
High Middle 
Stratum 
High Stratum 

• 283 -
estimations with higher social position and, accordingly, 
became less extreme in their attitudes and behavioural 
intentions. 
These results obviously show that opponents tend to 
reinforce their attitudes with higher social position, 
becoming still more extreme, still more prepared to take 
action than they would be on the average, 
while the adherents tend to become more careful in their 
statements with higher social position, and are only 
slightly more prepared to take action. At the same time, 
opponents in higher social positions exhibit better 
knowledge on nuclear topics. It is likely that, due to 
the public discussion and due to the stress caused by 
the incongruity between their own attitude and their 
perceived reality, opponents in the upper social positions 
have concerned themselves more intensively with nuclear 
energy problems, while the adherents from the upper 
social strata attempt to overcome their internal incongruity 
in this respect by suppressing the topic. This assumption 
cannot be fully verified within the scope of the present 
study, but the values obtained for the two groups suggest 
such an interpretation. 
It should be noted in passing that interview subjects of 
all classes were equally interested in environmentalism 
as a personal or society objective and exhibited 
relatively similar values on the environmentalism scale 
(r - 0.10; eta - 0.14, sig - 0.05). This i. contradictory 
to the hypothesis that the relative de-
privation of the middle classes is the cause for their 
commitment to environmentalism. Rather, it appears to 
become evident that behavioural intention and commitment 
correlate highly with class membership, causing a greater 
representation of the middle classes in the organized 
environmentalist groups. This has also been the result 
of the present study where, unlike the members of the lower 
claa.e8, more than double the members of the upper cla88 
and upper middle cla.s stated that they gave some kind of 
support to citizens' initiative •• 
1I1l1:1IlU_lilllrlU_MtUIIIlt:I_DII,LEalitiI:Dl_En!;1I_Eut.lltlllll:lI 
One of the salient characteristics of the nuclear conflict 
is that it form. a front straight across the institutions 
of society and the political parties. Just recently, the 
political success of the environmentalists has initiated 
a new polittcal polarization which, however, cannot over-
come the current leftist-rightist dualism of today's 
politics but at best expand it by yet another dimension. 
Both the split across the political parties and the 
opening of a second front can be seen from the data on 
votIng preferences compiled in Table 34. 
The voters of the parties represented in the Bundestag 
(West German Parliament) are clearly differentiated into 
opponent., adherents and neutral.. The trend that COO 
(Christian Democratic Party) voters have a more positive 
attitude toward nuclear energy than the voters of the 
Government coalition parties, has become stronger. This 
may be due to the fact that the CDO has taken a clear-cut 
decision in favour of nuclear energy which has been sub-
ject to little dispute within the party, ·and that the 
affinity of COO voters to business-related values 
supports the basic tendency toward attributing positive 
weight to nuclear energy. This is supported by the high 
correlation between CDU party preference and busine.a-
related objectives (lambda - 0.23/ SPD (Social Democratic 
Party) party preference. lambda - 0.14) and between 
CDO party preference and the belief in the economic 
necessity of nuclear energy. On the other hand, the 
risk-benefit eltimation does not differ Significantly 
between SPD and COO voters. 

- 286 -
~he proportion of opponent I is unu8ually high among the 
FOP (Liberal Democratic Party) voters. However, the 
possibIlity of a measuring artefact cannot be discounted 
in spite of the 8tati8tically proven significance of the 
difference to COO and SPD voter., lince the data base 
includes only 29 casea. 
Surprisingly high is the proportion of environmentalist 
voters. Since thia cateqory included 4' persona, the 
data base i. somewhat broader and yields more information. 
The frequent nomination of environmentalist groups a. 
the preferred party vote is probably due to desires for con-
sistency (being opposed ,one mIght as well say one votes for 
the environmentalista) or to demo.caple distortion (en-
vironmentaHsts hardly ever refused to take part in the 
interviews). Since this study did not concern a voting 
forecaat but relationshlps between voting preference and 
attitude-relevant variablea, the high frequency of environ-
mentalist voters 18 even welcome from the analytical polnt 
of view • 
. The median values of a number of variables for the po-
tential voters of an environmentalist party have been 
compiled in Table 35, below, in order to obtain a better 
profile of environmentalilt voters. 
Baled on the value. 1n this table, the environmentalist 
voter. are clearly opposed to nuclear enerqy and in a 
number of attitude-forming variables take even more 
extreme poal tiona than the average for the nuclear energy 
opponent.. They exhibit a great environmental awarenea. 
and partiCipation proponlity and have loat most of their 
confidence in the expertl of acience and technology. 
They are not better informed than the other interview 
subjects but bave ·1e.8 bias on nuclear energy than 
.011· the inaividuala who would vote againot nuclear energy 
Table 35: Structural data on environmentalist voters (median value comparison) 
lledJ.an Belief Affective Nuc. enerqy Solar energy Confidence Conserva- Participa- Environmen-
values Beale evaluation safety evaluation scale tin Beale tion scale tallam scale n 
Total 
average -1.89 3 . 22 1.26 12.7 5 .62 27.9 28;6 16. 9 503 
Environmen-
taliat votera -49.4 -11.4 -8.4 16.8 -3.5 13 . 1 21:4 30.4 41 
T-Te.t 
Siq. (99)) ye. ye. yeo yea ye. yea yea ye. 
lCnowJ..edge Negative P08itive Behavioural Strength Consistency Aqe Clas. Sex n bias bias intention of attitude of attitude Sl 
.... 
Total 
averagle 4.7 1.6 1.1 9.2 5.3 8.5 42 2 . 6 1.3 503 
Environaen-
ta1iat voters 5. I 1.8 0.2 17,8 7. 4 9 . 2 31 3.1 1.4 41 
T-Teat 
no no Yea yeo yea Siq. (99') no yea yeo no 
• 288 • 
in a referendum. Their nuclear energy attitude is firm, 
consistent and action-relevant. They are prepared to 
take far-reaching action above and beyond the average 
level stated by all opponents. Finally, the environ-
mentalist voters include very many young people and 
members of upper social strata. 
These data have been confirmed by analyses of voting be-
haviour in the Federal State of Baden-WUrttemberg where 
younger voter. and persona in the upper social structure 
more frequently voted for environmentaltst parties than 
older or socially less privileged persons. The inter-
pretation of these data is self-evident: when very 
negat1ve evaluations of nuclear enerqy exist and, at the 
same time, confidence in the technical competence of 
scientists, engineer. and planners has been lost, and 
when, moreover, there 18 a high propensity to translate 
one's attitude. into actlon, a threshold value 18 exceeded 
where the previous party preference 18 abandoned in 
favour of & shift to an environmentalist party. This 
threshold, of course, 18 lowest where the patterns of 
voting behaviour have not yet become very fixed, meaning 
that younger people are probably more inclined to opt 
for environmentalilt parties than older peraona who have 
voted for one of the big parties (pos.ibly different 
parties at different times) over many year. . Since there 
i. a high correlation between ~aviouralintent.1on and stratifi-
cat,ionand, at the same time, behaviouralintentlon is one of the 
prerequisite. for changing to an environmentalist party, 
peraon. fram upper 80cial clal.ea are a180 found more 
frequently among the potential voters wr environmentalist 
parties. 
• 289 -
~~~!!!t_§~!~_~~~!~yg!!_!~~-2~b!t_§2£!!!.£!~~2t!!! 
In addition to the indifference of nuclear energy attitudes 
toward the political spectrum which, however, has been the 
subject of greater polarization in the meantime - as shown 
above - almost all studies on this aubjecthave revealed that 
social category characteristics are independent of nuclear 
energy attitudes. Neither religious confession, marital 
status, occupation, nor household size or other variables 
of this type have been revealed as discriminatory 
quantitiea in empirical studies. This lack of influence 
has been fully confirmed in the present study. 
only education, income and occupational prestige have an 
effect on nuclear energy attitudes. These, however, 
have already been taken into consideration jOintly through 
the stratification variable. 
Confidence in Institutions and Reference Groups 
Man's attitude toward nuclear energy can be based only to 
a slight extent on his personal experience, rather, 
information on the nuclear conflict 1s received via the 
communication media and processed subjectively. Since 
most individuals are unable to verify the correctness of 
this, frequently contradictory, information, the credibility 
image gains a special weight. It is no longer the infor-
mation content which i. the yardstick for cognitive 
judgement, but the perceived social situation involved 
in the information transfer and the estimation of the 
information source. 
Based on the previous results of the analysis of the 
confidence scale, a marked difference between nuclear 
energy opponents and adherents would be expected in the 
attribution of credibility to the institutions of society. 
·m· 
And, indeed, nuclear energy opponents are more sceptical 
toward moat institutions, but not nearly as much a8 we 
~ould have assumed. The numerical data have been compiled 
in Table 36, below. 
This table provides information on a number of circumstances. 
Where credibility in nuclear energy questions is 
attributed to anyone at all, it is most likely to be 
the representatives of science and technology or of 
technically responsible political institutions (such 
as the Federal Ministry of Research and Technology). 
With respect to assessment of cred1b1l1ty of sc1entif1c 
inst1tut10ns there are only minor differences between 
nuclear energy opponents and adherents. Both groups 
have the greatest confidence in these institutions. 
Central political institutions and technically competent 
business institutions predominantly occupy intermediate 
positions on the scale, while general institutions of 
society such aa trade unions and churches, and non-
specific social position. such as journalist, manager 
or local politician, determine the bottom end of the 
scale. 
Opponents and adherents hardly differ in the basic, 
tripartite scale attribution: science first, followed 
by politics in general, and finally the institutions 
of SOCiety. On the whole, however, the opponents 
assigned lower grades to almost all institutions than 
the adherents. 
With respect to the credibility of specialized 8cience, 
opponents and adherents attribute the greatest credibility 
to universities and nuclear research centres. This is 
• 291 -
Table 36. Attributed credibility of institutions in 
nuclear enerqy questions 
(Medians; ~anqe: 1 - hiqh credibility 
4 - zero credibility) 
Institution Total _ian Med1an valu •• 
value Opponenta Adherent. 
Profe.sor at a 
nuclear research centre 1.44 1.88 1.47 
Scientist at a 
university 1.72 1.88 1. 71 
Reactor safety 
coamU8.1on 1.95 2.42 1.80 
Federal M1nister of Re-
learch and Technology 2.09 2.34 1.99 
Scientist in a citizens' 
action qroup 2.32 2.18 2. SO 
Representative of the 
medical profesBion 2.39 2.42 2.38 
Federal Chancellor 2.40 2 .68 2.36 
Spokesman f or a citizens' 
action group 2.60 2.39 2. 73 
Chairman of the board 2.69 2 . 91 2.43 
of a utility company 
TV commentator 2. 70 2.77 2.67 
Spokesman for Social 
Democratic party (SPO) 2.95 3.06 2.84 
Spokesman tor Christian 
Democratic Party (CDU) 2.96 3.03 2. 86 
Politician 2.99 3.07 2.95 
Spokesman for Liberal 
Democratic Party (FOP) 3.00 3.00 3.01 
Journalist 3.03 2.97 2. 96 
D1rector of an lnCluBtrial 
company 3 . 10 3.32 2.89 
Local politician 3.12 3.17 3.06 
Trade union leader 3.20 3. 24 3. 12 
Clergyman 3.56 3.37 3.40 
Chairman of the German 
Football Aa8oclation 3. 87 3.84 3.68 
Averaqe value 2.71 2.80 2. 64 
Confidence interval :!9.23 -+0.64 -+0.38 
- -
T-Teat 
0 . 00 
0.03 
0 . 00 
0.00 
0.00 
0.60 
0 .00 
0.00 
0.00 
0.18 
0.01 
0 .04 
0.09 
0.93 
0.87 
0 .00 
0. 15 
0.10 
0.64 
0.01 
0 . 01 
(p) 
• 292 • 
surprising because the opponents are convinced that 
the overwhelming majority of the members ' of these 
two institutions are in favour of nuclear energy. 
The competent scientists of the citizens' initiatives 
are given only third place by the nuclear energy 
opponents (and eighth place by the adherents). This 
would not have been expected, considering the 
homogeneous belief structure of the opponents. 
How should these data be interpreted? All the interview 
subjects agree that statements on nuclear energy by 
technically competent institutions have the greatest 
credibility. Political and SOCiety institutions are 
given lesser credibility ratings with decreasing attributed 
technical competence. The greater the position-bound 
attitude of scientific institutions, such as citizens' 
initiatives or the Reactor Safety Commission, the greater 
becomes their discriminatory effect on positions and 
attitudes. An almost identical result was found in 
Goerke's study where the interview subjects predominantly 
explained the confidence loss of scientific and business 
institution. by their dependence on fixed institutional 
objectives (Goerke, '79, 132, p.123). Non-affiliated 
institutions such a. the universities 
are equally accepted by both sides, although even the 
nuclear energy opponent. believe that moat professors 
would vote 1n favour of nuclear energy 1n a referendum. 
The nuclear reaearch centres range between the affiliated 
and non-affiliated institutions. 
It appears that the nuclear energy adherents require more 
backing by their confidence group than the opponents 
who are relatively "generous" in attributing a high 
degree of credibility even to institutions holding the 
opposite view, such a. the Reactor Safety Commission and 
the nuclear research centres. It is possible that the . 
- 293 -
opponents can rest more dependably on their consonant 
attitude structure and, based on this inner certainty, 
can be more receptive to information which contradicts 
their personal attitude, while the adherents, uncertain 
in their inner attitude structure, require strong support 
by institutions. Economic functionaries, especially 
private business functionaries, are given the lowest 
credibility potential by the opponents. It appears that 
they have lost their stock of potential persuasive power. 
Beyond the credibility question, it is of interest to 
determine the degree to which potential reference groups, 
i.e. qroups of peraons whose attitudes and behaviour are 
emulated by other peraons, have an effect on individuals' 
nuclear energy attitudes. Direct measurement of reference 
group influence is hardly possible in a sampling survey. 
Therefore, an indirect procedure was used: all the inter-
view subjects were requested to state how their friends, 
family members, members of the big political parties, 
communists, members of citizens' initiatives, university 
professors and nuclear energy experts would vote 1n a 
nuclear energy referendum . In addition, they were asked 
to predict the probable outcome of ~he referendum. The 
results have been summarized in Table 37, below. 
This table show.~ointerest1ng counter-trends: although 
most interview subjects would vote for nuclear energy, 
the majority believes that nuclear energy would be defeated 
in such a referendum. Among the nuclear energy adherents, 
approximately one third would not even dare to make a foreca 
while another third predicted that their own position 
would be defeated in a referendum. On the other hand, 
50 • of the opponents are convinced that their attitude 
would win in a referendum. However, even in this qroup 
one out of four believes that the adherents would win. 
This first counter-trend has been confirmed 

- 295 -
and family, professors and members of the three biq political 
parties. The question immediately arises as to why the.e 
opponents believe that nuclear enerqy would fail in a 
referendum while, at the same time, they classify their 
primary and secondary reference qroups as beinq pro nuclear 
enerqy. In order to study this question in more detail, 
the predictions qiven for the votinq behaviour of friends 
and family were differentiated by positions and these 
values were compared to the subjects' own votinq behaviour. 
Even after this break-down the paradoxical sitUation remain. 
that nuclear enerqy opponents are much less certain that 
their family and friend. share their opinions. Amonq the 
adherents, 77 • believe that their family, and 62 • that 
their friends would vote in favour of nuclear enerqy. 
Amonq the opponents,only half are convinced that the 
members of their primary reference qroups are also 
opponents of nuclear anerqy. The undeCided individuals 
also feel that more adherents than opponents of nuclear 
enerqy are found in their primary environment. The 
difference between the prediction for the whole (nuclear 
enerqy will be defeated in a referendum) and the pre-
diction for the social environment (f~ily and friends 
would vote in favour of nuclear enerqy) is remarkable and 
difficult to explain. Further subdivision of the reaulta 
by sex and class membership also failed to reveal any 
better startinq point fer meaninqful interpretation. 
It is likely that the frequent reports on nuclear enerqy 
protests have so solidified public opinion that unknown 
individuals are predsmtnantly perceived as opponents, 
while positive attitude. toward nuclear enerqy frequently 
dominate 1n one's own aocial environment. It 1s a180 
possible that, similar to the credibility assessment, 
the inconsistency of the attitude structure is compensated 
via an inner dependence on the attitudes of friends and 
family, thus reducinq coqnitive stress. 
.. -- - -.-
- 296 -
The pattern of reference qroup effects becomes even more 
complex when the secondary reference groups, here repre-
sented by the political parties, are included. Since the 
voting preferences of the interview subjects are available, 
they can be compared to the presumed voting behaviour of 
party members. 
It is found that the nuclear energy opponents also feel 
that the majority of the members of their preferred po-
litical party (with the exception of the environmentalist 
parties) are clearly pro nuclear energy. In this respect 
the party for which they vote apparently does not represent 
their position, and this i. clearly perceived. This 
inter-party separation line between nuclear energy 
opponents and adherents is even more clearly revealed if 
the data on the credibility of party spokesmen are 
included. 
Most nuclear energy adherents tend to classify the spokes-
man for their pol1 tical party as credible in nuclear energy 
questions. Thi. applies especially to SPD (Social Demo-
cratic Party) voters and to a lesser degree to CDU (Chris-
tian Democratic Party) voters, but not to FOP (Liberal 
Democratic Party) voters. However, this deviation is not 
significant due to the small data base of FOP voters. 
Nuclear energy opponents among the CDU voters trust their 
own party spokesman less than adherents among the CDU 
voters would trust the spokesman of the competing SPD. 
This relationship is .imilar in the case of the SPD. 
If one attempts to retain an overall impression from these 
data, the idea of • .hift of political fronts practically 
auggests itself. The spl1 t of the pol1 tical parties 
in the nuclear energy ·que.tion has led to a solidarity 
movement across party boundarie •• 
seek support from adherent. in all 
Nuclear energy adher.ents 
other parties and in 
this question disregard all the differences existing 
• 297 • 
between the parties. The opponents, on the other hand, 
feel that they are not properly represented by any 
political party and, therefore, withdraw credibility in 
nuclear energy questions even from their own pre-
5m1ed~. The degree to which this new environmentalist 
front exists for this particular topic only and does not 
appear with respect to other problems, cannot be deter-
mined from the available data . However, there are 
indications that this third dimension has gained access 
to the political perception of the population. 
Attitude of the Population to Specific Questions of 
Nuclear Energy Use 
In addition to questions on attitude structure and its 
influencing parameters, aome specific problems of nuclear 
energy use were oonsideredin this sampling survey. 
These problems, for ' instance the waste disposal problem 
caused by nuclear power stations, are also relevant for 
the overall nuclear energy attitude, but cannot be con-
sidered dependent or independent variables/ rather, 
they are supplementary elements in the cognitive image 
of nuclear energy in the eyes of the population. The 
evaluation of these interview data yields some interesting 
information on the apan of the perception field, but 
beyond that aspect has only minor importance for the 
overall theoretical frame of reference . Therefore, the 
results ahall be reported in summary· form only. 
!~2ngm!~_g£~h_~n~_!n!£~_~2n!~E~!2n 
Previous analyses had revealed a clear trend toward a 
positive evaluation of nuclear energy whenever the interview 
subjects exhibited a more busineaa-related value attitude. 
This resulted, from the value priorities 
of opponents and adherents and from 
the belief in the economic necessity of a further expansion 
- 298 -
h f r it was aatura1 to of nuclear energy. T ere 0 e, 
include a specific question on the possibility of 
e11minatinq the ties between economic qrowth and enerqy 
consumption, in ' addition to the general value priorities 
and economic aspects of nuclear energy. After all, 
there had been very considerable discussion in the media 
in 1979 as to whether economic qrowth processes must, 
of necessity, be tied to a growth in energy consumption. 
The answers qiven by the interview subjects have been 
compiled in Table 38, below. 
Table 38: 
Replies in 
Elimination of Ties Between Economic Growth 
and Energy Conaumption 
percent All Kerpen JUlich Beverunqen Hamm Kalkar 
Ties can be 
eliminated 51 44 50 59 60 46 
Ties cannot 
be eliminated 44 50 48 35 34 51 
No opinion 5 6 2 6 7 3 
. 
Up to a few year. ago, nobody would have doubted a linear 
relationship between economic qrowth and the qrowth of 
energy consumption. The discussion initiated by scientific 
publications on this topic haa found auch a broad response 
in public opinion within a minimum of time, that over 50 • 
are convinced today that ,economic qrowth can be achieved 
without increasing energy C()nIWllpE1..o!l'. ·Of course, 
the objection to thia will be that the interview subjecta 
were overtaxed with auch a question so that the result 
must be considered 118 reaponse set. However this is contra-
dicted by the clear-cut phrasinq of the question, by the 
relatively rare use of the "don't know"response and by 
the similarity of the diatribution with respect to 
related variables. The nuclear energy opponent'. position 
- 299 -
is even more vividly revealed by this result. based on the idea 
that a reasonable living standard and economic progress can be 
achieved even without energy growth, he can let himself 
be induced by his negative risk perception to forgo 
the use of nuclear energy without having to fear serious 
disadvantages. This is all the more easy for him 
because the quantitative values of economic growth mean 
little to him in any event and because he assigns first 
priority to the protection of the environment. 
~~~!~g~!_!~!~~_£!~!!!~!~_!n!~!!~!Y!! 
Analysis of citizens' initiatives is given great attention 
in the social sciences literature on protest movements 
against nuclear energy. Within the scope of the present 
study this topic was only marginally included in order to 
compare attitudes toward citizens' initiatives with 
nuclear energy attitudes. 
In addition to the repeatedly cited study by the Battelle 
Institute which dates back to 1977, two more recent 1979 
studies are available which, in conjunction with nuclear 
energy attitude measurements, were also devoted to the 
attitude of the population toward citizens' initiatives 
(Instltut fUr Demoskople, Allenabach, '79, and Goerke, 
'79, 132). According to these two sources, most of the 
interview subjects believed that. 
_ Citizens' initiatives acted in the interest of the 
general public and less in the interest of the persons 
involved; 
_ Citizens' initiatives did not aim at achie~ing material 
advantages for their members; 
_ Citizens' initiatives were not predominantly influenced 
by leftist or rightist fractional groups (although this 
was believed, nonetheless, by about 30 'l. 
- 300 -
These data are based on representative samples, 1n one 
case covering the entire Federal Republic of Germany 
(Allensbach), and in the other case several towns (Goerke). 
On the whole, there is a trend over time toward increasing 
identification of citizens' initiatives with the public 
interest. However, the diapersion spread has been 
relatively high since 1975 (cf. Allensbach, '79). 
In the present study the interview subjects were presented 
three statements on citizens' initiatives and requested to 
select the correct answer. Table 39 provides a review of 
the statements and percentage values for all survey units. 
Table '39: Confirmation of Statements on Citizens ' 
Initiatives (in percent) 
Statement All Kerpen JUl1ch Beverunqen Hamm Xalkar 
1. Cltlzens' ini-
tiati ves re-
present only 10 14 5 12 3 13 
their own 
interests 
2 . Cltlzens' ini-
tiati ves are 
indications 63 61 73 57 70 54 
that abuses 
exist 
3.Cit1zens' ini-
tiatives should 
have participa- 27 25 
tion ri.ghta 22 31. 27 33 
Citizens' initiatives are attributed a broad functional area_especially 
in those _towns .where ci.t1zens' ini.t1atives contra nuclear 
energy exist, or have existed in the past, but at the same 
time they are moat frequently attributed self-seeking 
motives in these towns. In 'JUlich, Kerpen and Hamm (where 
.a weaker citizens' initiative exists) the "soft" middle 
<Dlr8e 1s preferred which t t B a es that citizens' initiatives 
- 301 -
have initiative signalling functions but no decision-making 
functions. It i8 apparent that the citizens' initiatives 
in Kalkar and Beverungen have had a polarizing effect on 
the population's attitude. 
~!~g!2~!2U_2~_~~!_!!!~!_~!!E2!!!_~~2g!~ 
Based on the analysis up to this point it could be assumed 
that negative nuclear energy attitudes are accompanied 
by the belief that radioactive waste disposal is a problem 
which has not been resolved to this day. In addition, 
nuclear energy opponents should be much more outspo~en 
against the prese,tly planned waste disposal system than, 
conversely, the adherents in their positive att1tude. 
In order to test this assumption, three statements on 
the planned reprocessing facility 1n Gorleben,Gerrnany, were pre 
sented to the interview subjects. The possible responses 
and the associated frequenc1es have been summarized in 
Table 40, below. 
It is seen from this table that almost nine out of ten 
interview subjects felt that the waste disposal 
is not resolved, at . least at the present time. 
problem 
Almost 
30 , feel that the problem is almost insoluble, while 
60 , state that it can be resolved in t1me. These data 
already show clearly that even the nuclear energy adherents 
do not feel that the fuel cycle problem is resolved. 
S1milar results were obtained in Goerke's study in 1978 
(cf. Goerke, '79, 132). 
one out of two nuclear energy opponents feels that the 
waste disposal problem is solved in principle. and only 
one out of twenty feels that the waste dispoaal technology has 
been perfected. on the other hand, only 18 , of the adherents 
believe that the waste di.poaal problem has been solved. 
In the overwhelming majority (72 ') they· vote for postpone-

- 303 -
ment of the construction of a waste disposal facility 
until, in their view, all the scientific problema have 
been solved. But they do not consider these problems 
insoluble in principle. The undecided individual. -
as i. often the case - range in the middle between the 
two extreme positions. The perception of the waste disposal 
problem reveals the position of the nuclear energy adherents. 
Most of the inconsistency of their belief system is 
caused by the fact that they believe 
that nuclear energy is too ri.ky and technologically 
immature at the present time but they also 
believe that the problems can be solved in the future 
and that nuclear facilities are economically necessary. 
The opponents, however, are not convinced either of the 
present or of the future necessity of nuclear energy. 
Like most adherents, the undecided individuals share the 
opinion that nuclear energy 1s economically necessary 
and necessary for our future, but like the opponents 
they are also convinced that nuclear energy is very risky 
and that problems such as waste disposal are insoluble 
(22 'lor at least still unresolved (67 'l. 
- 304 -
summary of all Factors In·fluencing NUclear Energy Attitudes 
Up to this point the analysis included, step for step, new 
levels of variables and discussed their effects on nuclear 
energy attitudes. Occasionally the joint effects of the 
different variables were discussed as well. In the 
following summary all the external variables shall now 
be introduced jOintly into the evaluation and discussed. 
For this purpose the belief scale and the position in a 
nuclear energy referendum were selected AS independent 
variables, since both of these variables exhibited 
different causal relationships during path analysis. 
For statistical reasons, the multiple regression procedure 
was selected for the analysis of the belief 8cale, and 
the di8criminance procedure for the position analysis. 
The results have been summarized in Tables 41 and 42, 
below. 
The state of tension between economic necessity and 
positions in the nuclear energy referendum on the one hand, 
and risk estimation and belief system on the other, is 
again clearly seen from the two tables. In the case of both de-
pendent variables, distorted information - especially 
positive bias - has a high influence strength. The 
importance of the attitude consistency variable and behavioural 
intention variable for the development of a nuclear energy atti-
tude haa Already beend18cu8sed and, once again, demonstrates 
the difference between nuclear energy opponents and 
adherents. While a negative nuclear energy attitude is 
accompanied by a consistent belief structure and a strong behav-
ioural intention, positive attitudes are more frequently 
accompanied by ambivalent beliefs on nuclear energy and 
by a more reluctant action propensity. 


The socio-political attitude and value preferences included 
in the questionnaire are also important factors for both 
dependent variables. While the environmentalism attitude 
and the two value beliefs of social justice and high 
income (as counter-poles) have the greatest effect on 
the position taken in a referendum, confidence in science 
and technology in general and the high income and environ-
mental protection values play an important role in the 
case of the belief scale. However, 
the correlation coefficients indicate that accompanying 
attitude and value beliefs cannot be considered determinant 
variables (as postulated bJr Dumenil). 
The two dispositive characteristics of risk orientation 
and benefit orientation are included in both influence 
patterns and, consequently, support the results of the 
comprehensive survey on the perception patterns of risks. 
Other external variables include sex, represented in the 
belief system, and sex, age and class, represented in the 
position. The strength of influence remains within 
limits, however. Yet it is interesting to note that 
attitude-forming components are less dependent on SOCial 
structural characteristics than the position differentiation. 
In evaluating the information content of these two 
influence models it must be kept in mind that both internal 
and external dependencies have been represented. Con-
sequently, the squared R-value of multiple regression 
does not as yet imply that 88 , of the variance of the belief 
system could be explained by external variables. There-
fore, another regression was uaed to exclude all internal 
variables such as nuclear energy risk estimation, bias 
2 
etc. This yielded a multiple R -value of 0.64. This 
value is still high enough to prove conclusively the 
influence of external factors. Elimination of the internal 
variable changes little in the sequence of the variables 
- 308 -
included: only the risk orientation, a dispositive charac-
teristic, moves up to second place behind the confidence 
scale. 
The same applies to position differentiation. Together 
with the internal variables, almost 76 , of the cases can 
be correctly attributed. When all the internal variables 
are eltminated from the analysis, this value drops to 62 ,_ 
This, too, can be interpreted as a good prediction value 
for the independent variables. 
- 309 -
General Conclusion 
Since the beginnings of our history, technology and man 
have been in a state of tension: man has learned to 
make use of instruments to refine and perfect his non-
specialized organic nature in any desired direction. 
Technology helps him to move faster than any animal, to 
see, hear and smell better, to obtain food faster 
and in greater quantities, protect himself more effective-
ly against hazards and natural risk8, prolong hi8 life 
span to the biological limits and ensure an intensive 
propagation of hi. species. However, a price mU8t be 
paid for this progress: increa8ing specialization and 
differentiation in 80ciety result in anonymous and 
incomprehensible social structures which the individual 
can no longer understand; technological 
systems of ever greater perfection and cost increase 
the risk of being killed by the machine which man himself 
has created. The acceleration of prOduction 
ef.forts create more and more environmental pollution which, 
in the final analysis, can destroy the foundations that 
support human life. Last butnot least, technology can 
a180 amplify the negative developments in human SOCiety: 
the more efficient our technology, the greater is the 
potential scope of cataatrophic events when aggression 
takes place, be it in the for.m of crime, terrorism, 
civil unrest or war. 
There is no doubt that the ambivalence of technology 
exists. Many types of societies have deliberately forgone 
any forced advancement of technological progress because 
the social dynamics associated with the development of new 
instrument. would have threatened the static conditions 
in these societies. Therefore, in the 19th Century, 
technological progres8 meant overcoming rigid hierarchies 
and post-feudal power condition8. Until the past decade, 
- 3Ia -
the advancement of science and technology was, considered 
a priority task of society, associated with high positive 
value and a progressive image. However, vithin a short 
period of time the problema caused by major industrial 
facilities, an increasing awarene •• of the environment, 
disillusionment as to the limltations of research and 
technology in important area. (such as cancer research), 
employment problems due to automation, and saturation 
phenomena 1n the cOnsumer area have initiated a reinter-
pretatlon process whlch has polnted the spotllght of 
perception at the amblvalence of technology. Last but not 
least, the concepts for ~ott or alternative ·counter-
technoloqies· have stimulated the lncreasingly critical 
discussion on technology and its consequences. 
In thls situation, where coqnitlve contradictlons and 
a successive change of value bellefs and ethics both 
introduce uncertalnty into man's bellef system, studies 
on the perceptlon of technology achleve their special 
significance. The task consisted of using the tools 
of emPirical soclal r.search to trace the risk perception 
process and show how attitude. are formed . This, of 
course, was not po.sible for technology as a whole. 
Rather, an espeCially controversial topic, 'use of nuclear 
energy, vas selected as an example for the selective 
study of the determ1nants of risk perception. 
How can ve reconstruct the risk perception process on 
the basis of the empirical results? Are there any 
universal patterns for processlng risk-related 1nformation? 
Using the baSic risk perception end ecceptance model 
described in Part II of this, book, the perception proce.s 
can be subdivided into ' the follOWing steps. 
- Information receptionl 
- Development of cognitive beliefsl 
- 311 -
- weighting of these beliefs by attributing probabilities: 
- Evaluation of these beliefs on the basis of attributive 
and qualitative, dispositive and situative factors, 
. . 
- Internal clarification of beliefs (development of a 
con'siatent attitude), 
- Acceptance decision: 
- Rationalization of this decision. 
Now the results of the present study can be allocated one 
by one to these steps, shedding a little light into the 
"black box· of perception and acceptance. 
Beliefs on complex technological risk sources are formed 
predominantly via information received from third parties. 
The les8 our competen9~ to verify the ,correctness of a 
statement, the more we ar.e. dependent .on the subsidiary 
aspects of information which result from the interactive 
situation. In the case of nuclear energy it was shown 
. . . .' 
that the adherents attribute special credibility .to those 
information sources which on the basis of their personal 
image are considered pro nuclear energy experts, The 
opponents, on the other hand, consider even those experts 
.. ,' . 
credible arid reliable whom they perceive to be adherents, 
but they have lost all confidence in functionaries 
.. 
representing business interests. 
. . 
In a similar pat"tern, 'nuclear energy adherents perce! ve in 
their direct and indirect sooial environment a much greater 
proportion of pro nuclear energy attitudes than the. 
opponents perceive contra attitudes. These findings apply 
only to identifiable refere~c~ groups. When it concerns 
quite generally the perceived public opinion, most of 
the interview subjects believe that a negative attitude 
toward nuclear energy 
Republic of Germany. 
is predominant in the Federal 
The desire to find strong backing 
• 312 • 
by attitude-reinforcing expert., and the support sought 
by perceiving like-minded per.ona in the .ocial environ-
ment and in the institutions of society, are more 
characteristic for nuclear energy adherents than for 
opponents. It was attempted to explain this fact with 
the theory that the adherents are dependent on social 
support for their attitude. in order to compensate their 
inconsistent beliefs (more di.sonant elements), while 
the opponents, due to their more homogeneous attitude 
structure, require less social support. 
If these findings on nuclear energy attitudes could be 
transferred to other risk .ource., the result would be 
an interesting compens.tion mechanism to overcome cognitive 
dis.onancel the problema involved in fOrming one's own 
judgement are reduced by a subconscious delegation of 
decision-making powers to SOCial authorities and by 
.eeking protection within a perimeter of identical 
attitudes. The greater the homogeneity of attitudes, 
the less will be the necessity of this type of social 
re-lnlurance. 
2. ~!Y!122!!Ut_2!_~!!tt!! 
After absorbing the information and weighting the infor-
mation source, the next step i. to integrate the informa-
tion into the belief system. The results of this atudy 
suggest the conclusion that the perception of risk sources 
d6es not take place in accordance with a uniform reception 
pattern but that each source i. a.sociated with object-
specific characteri.tics. However, in the case of all 
risk sources covered by thi. .tudy a tendency was revealed 
to .eparate the perceived information into the categories 
of "pereonal involvement" and "econOmic and SOCiety-related 
a.pects". Within the.e major block., .pecific patterns 
developed. The benefit-ri.k relation was always positive 
when personal advantages were subjectively perceived 
without any doubt. 
- 313 -
Conversely, perceived personal disadvantages result in a 
negative attitude. In the case of risk sources given an 
ambivalent evaluation, such as nuclear energy or pest i-
cides, the general society-related, economic and tech-
no logical problems occupy most of the attention, while 
the personally perceived advantages and disadvantages 
are approximately balanced. 
The degree of knowledge about • risk source does not have any 
effect on the general attitude, at least in the case of 
nuclear energy. However, this does not apply to the 
self-evaluation of the interview subjects' information 
level: here the adherents prove to be especially self-
confident, while the opponents give themselves a more 
sceptical evaluation. Also, the adherents believe that 
other citizens are much less informed than they are, 
while the opponents, agaln, do not see any differences 1n 
the information level between themselves and others. 
3. Attribution of Probabilities 
----------------------------
Probabilities are absorbed together with the information, 
but are allocated intuitively. Within the scope of the 
present study it has not been possible to confirm the 
assumption that the size of a possible catastrophe is 
always given a greater weight than the probability of its 
occurrence. At least within the range of the comprehensible 
numerical relationships, the interview subjects attributed 
similar weights to both parts of the scientific risk 
equation. However, the abstract numerical relations 
appear to have only a small influence on the imagination 
of individuals, since excessive risk dimensions are 
considered pOSBible in the case of risks frequently dis-
cussed in the media where the catastrophes can be directly 
- 314 -
imagined. Thus, most interview subjects believed that 
they weuld live to see one reactor accident which weuld 
cause 600,000 deaths (median value). Even the moat 
pessimistic risk calculations would show up this value 
as being in the realm of fantasy. Other, easily imaginable 
hazards (such as being caught in an avalanche while skiing) 
are also overestimated in terms of time and dimensions of 
a catastrophe. It 18 seen clearly that the interview 
subjects can state fairly reliable averages for hazards 
per annum, but are not able to predict the probability 
and the dimensions of extreme accidents. In this respect 
it doesn't matter whether the risk dimensions or the 
time frame are given as constants. The last result to 
be mentioned is that the failure probabilities of redundant 
safety devices, such as double barriers, are not multiplied 
but quite Obviously summated. 
4. Evaluationof Beliefs 
Evaliationof beliefs by their per80nal importance is the 
most important step in risk acceptance and, therefore, 
the major area of interaat for the present study. The 
relationship to the subject (what have I got to do with 
it?), the potential exposure (will I have to bear the risk?) 
and qualitative characteristics (which positive or 
negative side effects can be expected?) were emphasized 
as internal evaluation patterns. In the case of the 
qualitative risk characteristic. (.uch aa voluntar1nea8, 
personal control), a verifiable effsct on the acceptancs 
of the risk source was found only when certain orders of 
magnitude of the.e characteristics were exceeded. Thus, 
voluntary ri.k acceptance and the possibility of indi-
vidualizing the harmful consequences increase the pro-
pensity to tolerate risks, while the p08sibility of 
long-term harmful effects and severe localized consequence. 
lead to a negative evaluation of the risk source in 
question It was clearly demonstrated in both experiment. 
- 315 -
that risk sources are more easily accepted when they are 
accepted voluntarily or associated with a possibility 
of pers~nal control. However, the 
assumption was disproved that the qualitative risk 
characteristics have a determinant effect on the risk-benefit 
estimation. They are one factor among many. 
The question as to the distributive effect of risks was 
not verified explicitly. However, the study of qualitative 
characteristics of the analysis of the experiments revealed 
a clear tendency that risk sources with equally distributed 
benefit and detriment are more easily accepted than risk 
sources with imbalanced benefit and risk distribution. 
Among the subject-referred characteristics the following 
quantities were included: risk and benefit orientation, 
general value beliefs, socio-po1itica1 attitudes as well 
AS demographic and social structural characteristics. 
General risk and benefit orientation has a surprisingly 
high effect. Measured in terms of the conSistency of 
high or low risk evaluations, this variable correlates 
strongly with the independently measured risk source 
evaluation and with the question as to the desired pro-
cedure for handling the risk (from support to prohibition). 
This relationship was identified in all three surveys and 
for most risk sources. Additional studies will be required 
to determine whether this a110wa the conclUsion that a 
phenomenon such as a dispositive tendency toward risk 
overevaluation or underevaluatlon exists. 
Both, the general value system and the socio-po1itica1 
attitudes are important influencing quantities which have 
an effect on the estimation of the nuclear energy risk 
source. The same applies to the evaluation of the 
chemical plant as a risk source which also depends on 
the value system and attitude environment scales included 
- 316 -
in the questionnaire, while the risk sources of coal energy, 
solar energy and other alternative energy sources are 
largely independent of these two classes of variables and 
are given homogeneous evaluations by all the interview 
subjects. This suggests that risk sources which have 
political and society-relate~ aspects, such as nuclear 
energy, pesticides and chemical industries, are also 
affected by the socio-political attitude environment, 
while considerations of expedience and enjoyment pre-
dominate in the case of risk sources which are mostly per-
ceived as technological or personal risk sources (coal, 
smoking). Society-related values and attitudes playa 
minor role. 
Among the social and demographic characteristics,stratification 
and sex are of special interest. With higher class 
membership the importance of personal aspects of risk 
perception ~nesand they are replaced by socio-political 
arguments. Consequently, it is not surprising that, 
following the discussion on the disadvantages to society 
caused by nuclear energy, negative attitudes toward this 
energya=uDeincrease with higher class membership. This 
relationship is also remarkable from a theoretical point 
of view because diffusion research haa in most cases 
associated opposition to innovation with traditional 
value attitudes, conservative ideology or lower class 
situation. It is seen, therefore, that an inversion has 
taken place over time. 
Women react with special sensitivity to health risks 
and are less receptive to business-related values and 
objectives. In the nuclear energy question these two 
effects intensify each other. The perception of the health 
risk can be compensated only inadequately by the per-
ception of the economic necessity of nuclear energy 
because these values are of lesser importance in the 
belief system of women. The same effect is found in the 
case of pesticides and food preservation. 
- 317 -
Age has only a minor effect on risk perception, and 
different patterns of influence are found from one risk 
source to another. In the CAse of nuclear energy evalua-
tion it 1B especially the youngest people (16-25) and the 
middle-aged (from 46 up) who tend to have a more positive 
attitude toward nuclear energy, while the age group fram 
26 to 35 has the most negative attitude. This relation-
ship has been identified exclusively for nuclear energy 
and not for any other risk source. 
On the whole, therefore, it is found that almost all 
the belief evaluation factors discussed in theory have 
an effect. In terms of total effectiveness the sub-
ject related characteristics predominate, but for these 
only a few typical patterns were identified. Qualitative 
characteristics are more easily systematized, but are 
decisive for risk perception only when certain threshold 
values in their development are exceeded. Dispositive 
and situative characteristics are also relevant influ-
encing quantities which, however, require differentiated 
analysis for each risk source. 
5. ~yd21!l!1!:nL2L!_~Qn!!!!t~!!LlItHt!!!!~ 
In the conceptional model of risk acceptance the weighted 
beliefs which are relevant for a decision are classified, 
and contradictory elements are eliminated. This process 
step cannot be recreated in an interview. In the case 
of nuclear energy attitudes we noted a relatively well-
reinforced and homogeneous attitude 8t~ucture ·on the 
part of the opponents, and a more ambivalent system with 
many dissonant elements on the pa~t of the adherents. 
It is possible that the subjective impression that most 
of the direct social environment has a positive ' attitude 
toward nuclear energy has required a greater elimination 
of dissonant elements on the part of the opponents. 
Following this attitude reinforcement, however, the 
- 318 -
opponents have became more open-minded toward the arguments 
of the experts, even though, on the whole, they have less 
confidence in the statements of scientists and engineers 
(even those who belong to their own camp). 
6. ~££~~!~£~_~~£~!~~~ 
The acceptance decision was defined as striking a balance 
after risk weighing, directly measured as . · risk-
benefit estimation and, indirectly (and 
applicable to nuclear energy exclusively), determined 
via the voting behaviour in a hypothetical referendum 
on the further expansion of nuclear energy. The ~18k­
benefit estimation is subject to almost the same relation-
ships as the sum total of the belief system. Both 
quantities correlate strongly (0.71). A simulated 
referendum on nuclear energy revealed the influence of 
bUsiness- and society-related valuations even more 
clearly than in the case of risk evaluation. Those who 
are convinced of the economic necessity of nuclear energy 
will usually vote for this energy option even though they 
feel otherwise that nuclear energy involves more detriments 
than benefits. Therefore, acceptance decisions are more 
affected by the political power play (energy crisis, 
accidents) than attitudes. 
7. Rationalization 
---------------
The degree to which, after developing an attitude, elements 
of the belief system are modified in order to achieve a 
consistent overall pattern, is also difficult to determine 
through empirical research. However, a number of indirect 
observations seem to indicate that individuals did indeed 
rationalize their own position. Thus, a considerably 
greater number of interview subjects stated that they would 
vote for environmentalist parties than was actually 
the case. A highly plausible explanation for this 
phenomenon is that committed nuclear energy opponents 
- 319 -
wanted to lend adequate weight to the decisiveness of 
their attitude. Even in the case of very topical subjects 
such as eliminating the ties between economic growth and 
energy consumption, and the nuclear waste disposal problem 
in Gorleben, at least the opponents exhibited an extremely 
attitude-related response behaviour. 
8. ~!!~_~!r~~E~!22_!Ug_§2~!!!_~~!~!r~h 
The sequence of steps involved in the risk acceptance 
process, outlined only briefly in this chapter, provides 
just an impression of the variety and complexity of 
risk perception among the population. Many mechanisms of 
intuitive evaluation can be revealed by empirical means 
and compared to model concepts. The result has been 
that the presently existing concepts on the processes 
involved in risk perception are too narrow. Although it 
has not been possible to clarify fully the overall risk 
perception process, the results achieved in this study 
seem to allow the hope that this better insight into 
the structure of risk perception could contribute to 
an understanding of the structure of opinions and 
positions held by the public and, consequently, toward 
overcoming unnecessary communication barriers. 
It is not the ambivalence of technology which was under 
discussion here, but its perception. It is immaterial 
whether it is justified, objectively true, rational or 
irrational. Such a comparison inevitably 
leads to the development of elitist theories where, based 
on a false antagonism, the Rnalve w attitude of the 
population would be discredited in comparison to the 
experts' judgements. This is not in keeping with either 
the purpose or the meaning of psychological studies on 
risk perception problems. In particular, the results of 
this study have shown that the spectrum of factors in-
volved 1n an acceptanme decl.10n 1s much more complex 
and varied in the perception process of the population 
- 320 -
than it is in the scientific and technological approach 
with whose results the public is quite familiar. If 
studies of this type fulfill their purpose of pointing 
out the relevant .factors of risk perception, the concerns 
of the population and the cognitive problems involved 
in the evaluation and handling of technology, then the 
effort expended on this research has already been 
worthwhile . For, in the future it will be more important 
than ever that we can find our way back to a 
strategy which will find a consensus of opinion when 
grave technological decisions must be made. 
- 321 -
A P pen d 1 x 
- 322 -
Index of Technical Terminology with Definitions 
Acceptance 
Positive judgement about an object or a procedure 
Acceptance Threshold 
Internal limit above which objects become acceptable to an 
individual or group 
Aggregate Evaluation 
cf. Evaluation 
Allensbach, Instltut fUr Demoskople 
National Opinion Poll Establishment 
Alienation 
Category describing the discrepancy between a man's work and the 
resulting product in modern manufacturing 
Anxieties 
Diffuse fears of threat 
- Suppressed anxieties 
Emotions which are not admitted as being felt 
- Anxiety-Envy syndrome 
Theory developed by RBglln according to which modern societies 
are characterized by a comblnat·lon of anxiety and envy as a 
psychological channeling of lack of orientation 
Archetype 
Typical subconscious orientation patterns with high affective 
tension (acc. to Jung's theory); cf. pp 22/23 
Attitude 
See definition on Page 87 
- Attitude components 
In our opinion, attitudes consist of three components; cognitive 
beliefs, affective evaluation and behavioural Intention 
Attributes 
Allocation of properties to certain objects 
- Attributive biases 
Intuitive Incorrect allocation of properties to objects or of 
incorrect conclusions on the basis of Information received 
- Attributive theory 
Systematic assessment of Intuitive attribution processes 
Ava 11 ability 
Form of attributive bias where events are assessed as being 
more probable, the better they can be remembered 
- 323 -
Belief 
Idea as to the characteristic properties of an object or a 
person 
- Belief system 
Sum of all organized ideas with respect to an object 
- Salient beliefs 
The decisive ideas for the attitude taken toward an object 
- Weighted beliefs 
The ideas held with respect to an object multiplied by the 
affective significance of the content of the idea 
Bolstering 
Removal of self-doubt as to the correctness of a decision once 
it has been made 
Capitalistic Production Crisis 
cf. Crisis 
Citizens' Action Group (Initiatives) 
Short-term organization of citizens affected, with the aim of 
preventing, modifying or initiating projects in their immediate 
environment . " 
In the Federal Republic of Germany, umbrella groUps covering 
larger areas have developed from several of these local groups, 
particularly in the environmental protection movement 
Citizen's Initiatives 
cf. Citizens' Action ·Group 
Cognitive 
Characteristic of the infonnation content of beliefs. 
- Cognitive component 
One of the three components of the attitude structure 
- Cognitive dissonance 
Contradictory information content with respect to a single 
object. It is assumed that everyone attempts to remove dis-
sonances 
- Cognitive stress 
Psychological tension caused in the perception process by. contra-
dictory information content or by attitude-destroying information 
Collective Subconscious 
Unconscious, internalized attitude pattern within a societal group 
Common Mode Failures 
Several errors in a technological system occurring Simultaneously 
Common Sense Processing 
Intuitive processing of external stimuli or infonnation 
- 324 -
Communicative Consistency 
Parallel nature of communication content (cf. p. ~2) 
Canpensation 
Econanic: 
Psychological: 
Canpensation Payments 
Transfer payments 
Canponents of Risk 
cf. Risk 
Conflict Theory 
Canpensatory payment 
Overcaning psychological deficits or setbacks 
by means of performance in other areas 
as canpensation for damage suffered 
Sociological theory. according to which the basic features of 
social development are determined by the conflicts occurring 
between the "actors· in a society and the resulting con-
flict resolution strategies 
Consistency of Attitudes 
Congruence between the cognitive and affective elements of an 
attitude structure (cf . Attitude) 
Correlation (Coefficients) 
Measure of the strength of a linear connection between two 
variables. A correlation coefficient of 0 means that there 
is no connection. a coefficient of 1 means total agreement. 
- MUltiple correlation coefficient 
Connection between one dependent and several independent variables 
- Squared correlation coefficient (r2) 
The square of the correlation coefficient r gives the proportion 
in per cent of the declared variance for the dependent variable 
- Types of correlation coefficent 
In statistics there Is a large number of correlation coefficients. 
the use of which depends on the measuring level of the 
variables (naninal. ordinal. at graduated intervals or rational) 
In this work. the following coefficients are used: 
Pearson r. Lambda. Gamma. Eta. Spearman. Tau. Phi; the formulas 
for calculating these coefficients can be found in any 
statistics textbook. 
Cost-Effectiveness Method 
Econanic method of optimizing the efficiency of expenses for 
projects 
- 325 -
Crises 
Subjective impression of a malfunction in a continuous mental 
or action-related development, where the final objective no 
longer seems attainable 
- Capitalistic production crisis 
According to Marxist theory, the imminent collapse of the 
capitalist economic system which is inevitable because of the 
antagonism between profit interest (maximalization of greater 
value) and worker interests 
- Identity crisis 
Disturbance in the individual maturing process of acqUiring 
a self-identity, usually as a result of lack of orientation 
and insecurity of values 
- legitimacy crisis 
Impression that the official national powers are no longer in a 
position to handle the problems of the future 
- Value reflection crisis 
lack of orientation with respect to the walidity and logic 
of rival value beliefs 
Decision Analysis 
Field of science concerned with form, types and processes of 
decision-making 
- Descriptive decision analysis 
Descriptive analysis of the actual decision-making process 
by individuals or institutions 
- Nonnative decision analysis 
Compilation of processes in order to be able to make decisions 
as rationally as possible 
- Typological decision analysis . 
Compilation of typical decision-making process patterns 
- Explicative decision analysiS 
Compilation of the logical and mathematical bases for decision-
making processes 
Descriptive Decision Analysis 
cf. Decision Analysis 
Displacement 
Psychoanalytical term: unpleasant memories or beliefs · are 
forced down from the conscious mind into the subconscious 
Diffusion Research 
Branch of sociological research concerned with the introduction 
and dispersion of innovations in a society 
Divisibility 
See Page 41 
Distributive Equivalence 
Equal distribution of benefits and risks over the population 
(cf. p. 419) 
Dispositions 
Properties Inborn or acquired in the human psyche 
- 326 -
()Jnmy Variables 
Nominal variable (e.g . religion) which can be converted into 
metric variables (data at graduated intervals) by splitting 
it into several 'yes-no' variables (Catholic yes-no, P,rotestant 
yes-no) so that it can be more easily evaluated statistically 
Discriminance Analysis 
Statistical method of detemining the connection between one 
dependent and several independent variables, not at graduated 
intervals, only the Joint effect of all independent variables 
being taken into consideration 
Ego-Stabilizing Orientation 
Establishment of a value system which justifies and safeguards 
the individual personality and action 
Elimination by Aspects 
Method of declslon-making for multi-dimensional problems (see 
Page 14) 
Environmental Assimilation 
The perception and processing of environment-related information 
and experl ences 
Environmentalist Movement 
Organized but not yet officially channelled influence control 
movement to bring about changes to improve the environment by 
means of mass mobilization 
Error Tree Analysis (Fault Tree Analysis) 
Statistical method to,determinesynthetlcally the probabil ity of failure 
of a system In the case of rare occurrences using the combina-
tion of failure probabilities of the individual system components 
Eta cf. Correlation 
Evaluations 
Affective evaluations (weights) of beliefs with respect to an 
object or an Individual 
- Evaluation scale 
Measuring process for obtaining the affective weighting of beliefs 
- Aggregate evaluation 
The unification of affective weightings for several objects 
Expected Values 
The losses or gains calculable per unit of time using the 
probability function 
Explicative Decision AnalYSis 
cf. Decision Analysis 
Exploratory Research 
Research which attempts to probe the scope and depth of phenomena 
to be explained, without as yet using a strict methodical 
concept 
- 327 -
Fa 1s1ffab1ll ty 
Term from the scientific theory of critical rationalism (Popper). 
Working fro~ the realization that there is no real possibility 
of proving the truth of a statflllent,statll1lents lllade on the basis 
of scientific conventions are considered as being only provision-
ally proven until they can be proven false. For this purpose, 
however, it is necessar,y to be able to subject the statement 
to thorough empjrical examination 
F1shbein Model 
Attitude concept according to Fishbein, where beliefs about 
an object are ~ltiplfed by the affective weightings of the 
content of the belief (beliefs x evaluations) 
Functionalistic Approach in Sociology 
The sociological school of thought which classifies actions 
and institutions within a society according to their degree 
of functionalism or dysfunctionalism for the smooth running 
or maintenance of basic social structures is described as 
functionalistic or structurally functionalistic. 
Functional Determinacy 
Direct and perceivable relation between the benefits of an 
object and the person receiving the benefits (see Page 42) 
Factor Ana lys 15 
Statistical method of tracing back a number of variables to 
the major. basic factors and of acquiring the dimensions of 
the multiplicity of variables 
- Factor loadings 
The strength of the influence of each variable on the factor 
representing .it . 
- Factor score c08ffi ci ents 
The weight of each individual variable in the summation to a 
factor 
- Three-dimensional factor. analysis 
Factor analysis as a function of a third variable (e.g. sex, 
age, etc.) 
GaJrma 
cf. Correlation 
Game Theory 
Part of decision-making theory where the reactions of the 
'game participants' affected by the decisions made are in-
corporated into the calculations 
Hierarchic Needs Structure According to Maslow's theory, there Is a hierarchy of needs 
In which attempts are only made to satisfy a higher-level 
need when the one Immediately below It bas been satisfied. 
Tbe 5 stages are as follows: physical needs, safety, affection. 
self-esteem and self-actualization 
IAEA International Atomic Energy Agency in Vienna 
- 328 -
Identification 
Process for the internalization of ideas or beliefs as part 
of the individual's own self-value concept 
Identity Crisis 
cf . Crisis 
IIASA 
International Institute of Applied Systems Analysis in 
Laxenburg. Austria 
Inadequate ReCiprocation 
Imbalance between two dependent systems (one gives more than 
it receives). An inadequate reciprocation requires a transfer 
legitimation (e.g. ideology. altruism. power) in order to become 
stable 
Institut fUr Oemoskopie. Allensbach 
cf . Allensbach 
Intentional Value Congruence 
Congruence between changes in SOCiety caused by innovations 
and the existing value system (see Page 41) 
Internalization 
Internalization of values and norms 
Internalized Risk Threshold 
Individually established threshold above which risks are con-
sidered to be unacceptably high 
Intervening Variables 
Variables forming part of a causal chain between the causative 
and the dependent' variables 
Intuitive Generalization 
Types of process prevalent among the general public for drawing 
conclusions from specific data and applying them to general 
circumstances (see Page 71) 
Intuitive Regression 
Attributive bias when intuitive conclusions are drawn. where 
statistical exceptions are assessed as criteria for the 
probability of future behaviour 
Invaccination Effect 
Attitudes which havo been exposed to low-level loads 
over a long period of time. still remain stable if more 
extreme fOnDS of attitude-destroying experiences are undergone 
Lambda 
cf. Correlation 
Latin Square 
Specific arrangements of test groups In experiments In order 
to assess separately the influence of several stimuli 
- 329 -
Legitimacy 
Confidence in the problem-solving cognition of political 
decision-makers 
Legitimacy Crisis 
cf. Crisis 
Marginal Utility Theory 
Economic theory according to which the value of a product is 
measured against the subjective benefits of the last used unit 
' Materialized Rule 
Power relationships are disguised by means of the interpolation 
of non-power-related authorities or objects (e.g. time clocks) 
Mixed Scanning 
Theoretical decision-making process where the selection from 
a series of optimal or sub-optimal solutions is left entirely 
to the social and political powers (see Page 15) 
Muddling Through 
Decisions are made according to the'free interplay of the political 
powers and spheres of inffuence (see Page 15) 
Multiple Correlation Coefficient 
cf. Correlation and Multiple Regression 
Multiattributive Utility Measurement 
Process for calculating the overall utility of projects or 
products which cause benefits and damage in more than one 
'dimension 
Multiple Regression 
Statistical process for estimating the joint influence of a 
large number of variables on a single dependent variable. 
The multiple regression presupposes data sets which are graduated 
at intervals and which have a normal distribution curve 
- Multiple correlation 
cf. Correlation 
- Squared multiple correlation 
Declared variance of the dependent variable 
New Economic Theory (Economics of Politics) 
Transfer of economic terms and process analyses to the explana-
tion of the political behaviour of institutions and members 
of the public, for example considering voting to be ~ 
market decision for political programmes 
Normative Decision Analysis 
cf. Decision Analysis 
Normative Democratic Structure 
Belief system with respect to the way a society oriented to 
democratic ideals ought to be organized 
- 330 -
Opinion leaders 
People whose opinions carry a large amount of weight in their 
local surroundings and who have an opinion-forming influence 
on the individuals in their environment 
Overt Anxiety 
Externally observable behaviour which suggests anxiety feelings 
Participation (participatory) 
Participation in political decision-making processes (ranging 
from participation in the passing of a resolution to co-
determination) 
Pearson r 
cf. Correlation 
Perception . 
Mental representation of images about objects or persons 
- Associative perception 
see Page 61 
- Communicative perception 
see Page 60 
- Intuitive perception 
see Page 61 
- Self-perception 
Self-image of own personality 
Phi cf. Correlation 
Political Apathy 
ReSignation to or disinterest in political issues 
Political Economy 
cf . New Economic Theory 
Political Exchange System 
cf . Social Exchange System 
Portfolio Theory 
Mathematical and axiomatic variant of the descriptive decision-
making theory (see Page 11) 
Pragmatic Political Approach 
Analyses as to how political deficits may be balanced out by 
new forms of political decision-making without involving 
changes in the overall system 
Probability 
see Pages 57/58 
Projection 
- Economic term: perceived advantages of a project which will 
take place in the future 
- Psychological term: transfer of personal mistakes. which one 
does not want to recognize. to other people or objects 
- According to Raglin (see Page 25): the problem-solving 
capacity of a functional object 
- 331 -
'.ychological Inconsistencies 
Contradictory perception patterns and processing categories 
in the psychological handling of external stimuli 
Psycho-Affective Reinforcement 
Intensification of beliefs and opinions which most closely 
correspond to internalized values and feelings 
Qualitative Risk Characteristics 
cf. Risk 
Quota Sampling 
Selection procedure for random sampling, where the respondents 
are selected according to particular features (e .g. age, sex, 
income) from representative or systematiC viewpoints 
r or r2 
cf . Correlation Coefficient 
Reality Stress 
Tension arising when the perceived development of a society 
does not agree with the wishful thinking as to how this develop-
ment ought to appear 
Rationalization 
Phase following attitude formation when decisions once made 
are safeguarded in retrospect with arguments 
Reduction of Complexity 
The most urgent task of political and social systems, according 
to the political Systems Theory, is to simplify complex problems 
and to find solutions which are acceptable to all. 
Reduction of Technological Complexity via Moralization 
Premise made by E. Scheuch, according to which environmental 
structures which have grown too complex to be understood by 
common sense or formal education can be cognitlvely· handled 
by means of applying moralization processes to the relevant 
problem 
Reference Group 
Group within society to which one would most like to belong 
Referred or Expressed Preference Analysis . 
Method in descriptive decision-making theory of determining 
preferences for solutions via surveys 
Reinforcement 
Reinforcement of behaviour or thought patterns by internal 
or external stimuli (term from theory of learning) 
Relative Deprivation 
Deprivation of people or groups of people in particular areas, 
although in other areas equality or even privileges exist 
- 332 -
Representativeness 
Attributive bias according to which redundant information and 
pre-structured beliefs suggest a belief in the representative 
nature of this information or these beliefs for everyone 
Revealed Preference Analysis 
Method in descriptive and normative decision-making theories 
where the acceptability of risks is derived from the previous 
scope of acceptance of familiar risks 
Role Attribution in 
Allocation of 
cOll'lOOni ca ti ng 
lnteractive Situations 
expectations to people with whom one is 
Risk 
Definition: see Page 
- Risk acceptance: see Page 
- Risk assessment 
The systematic assessment and weighing up of all possible 
positive and negative consequences of a risk source 
- Risk dimensions: see Pages 
- Risk Estimation 
The intuitive assessment of risk consequences 
- Risk evaluation 
Assessment of identified risk consequences 
- Risk identification 
Assessment of risk consequences without analysis of their 
probability and without weighing up or evaluating the damage 
- Risk perception: see Page 
- Risk socialization 
The gradual familiarization with risks in the life-long process 
of learning to come to terms with the environment. Development 
of speeifie attitude patterns towards risk sources 
- Components of risk 
The eonsequences of a risk souree ean be elassified into the 
following eategories: direet. indireet. intangible and symbolie 
hazards or benefit. (.ee Page 99) 
- Qualitative risk charaeteristie. 
Feature. whieh are eharaeteristie of the eoneomitant eircum.tanee. 
of risk. but whieh are not related to the level of the ri.k (e.g. voluntariness. eontrol possibility. ete.) 
Risky Shift 
Phenomenon in group experiments: the group as a whole shows a 
higher degree of risk propensity than do its members when making 
individual deeisions . The eonverse. known as the "Coutious 
Shift" can, of course, also occur 
Salient Beliefs 
cf. Belief 
Satisfieing Strategy 
Deeision-making process whieh is not eoneerned with optimization 
but with fulfilling minimum eonditions 
- 333 -
Self-Perception 
cf. Perception 
Semantic Differential 
Technique for measuring the affective evaluation of bellefs 
regardlng persons or objeets 
Semantic Reductlon 
Simplification of complex condltions by the allocation of 
associatlve attributes whlch are affectlvely loaded (e.g. 
male, female) . 
Senslbl1lty Studles 
Important technique In systems analysis for describing changes 
In output variables when the input variables are shifted 
withln credible limits 
Signlflcant Other 
Person whose opinion about oneself one values 
Soelal (Political) Exchange System 
The soelal and political powers of a soelety are regarded os 
a system which depends for Its malntenanee and funetloning 
capacity on the existenee and functioning copocity of the 
remainder of the system. In splte of the mutual dependenee, 
the exchange need not be symmetrieal, I.e. there are possibilities 
e.g. privileges , for profitlng more than one is obliged to give (refer 
also to Inadequate Reclprocal Exchange) 
Soeial Resourees 
Power of disposal of a group or an institution of means cf 
social influenee (e.g. access to mass media) 
Soeio-Reeeptive Relnforeement 
Reinforeement of behaviour or thought patterns by the pereeption 
of soelal rewards (such as more prestige, recognltlon) 
Spearman 
cf. Correlation 
Squared Correlatlon Values 
cf. Correlation 
Stratlfleatlon . 
In this study, stratum is understood to mean the sociol p~estlge 
of aperson. This is measured by ineome, level of edueatlon 
and professional prestige 
Subconseious Meehanlsm 
Forms of cognltlve coneluslon-drawlng process where the 
declslon-making proces. Is not voluntarlly controlled 
Subjectlvely Expected Utl1lty (SEU) 
Central coneept of declslon-maklng theory. It means the s~b­jectlvely determlned beneflt (or detrl~ent) whlch an indlvldual 
belleves he will galn or lose from an action 
Suppressed Anxieties 
cf. Anxieties 
- 334 -
Symbolic Reduction Theory 
Psychological theory according to which people perceive objects 
in their environment more easily and rearrange them with less 
difficulty if they are able to allocate them key attributes (e.g. nuclear power · big business) 
System Analysis Theory 
Meta-theory used in many branches of science according to which 
phenomena to be explained are seen as elements of systems (or 
as systems themselves) which have a set order within their 
. system and which, as a whole, can affect other systems 
Three-Dimensional Factor Analysis 
cf. Factor Analysis 
Typological Decision Analysis 
cf. Decision Analysis 
Tau cf. Correlation 
T-Test 
Statistical process for finding out whether there is a significant 
difference between the mean values of two random samples, i .e. 
whether their difference can with all probability no longer 
be explained by coincidence 
Values 
General orientation points for the behaviour of individuals 
and groups 
- Materialistic values (quantitative) 
Values which refer in particular to the acquisition of material 
goods 
- Post-materialistic values (qualitative) 
Values which refer in particular to the quality of the environ-
ment and to living conditions 
Value Reflection Crisis 
cf . Crisis 
Value Tree Analysis 
Process whereby the value beliefs of groups participating in 
decision-making or affected by the decisions made are elicited 
and are spread out to such an extent that conclusions can be 
drawn regarding consequences relevant to the decision 
Value Pluralistic Society 
Society in which value systems compete with one another as 
universal solutions 
Weighted Beliefs 
cf. Beliefs 
- 335 -
R E F E REN CES 
1) Aachener Nachrichten: Windräder passen nicht in die Eife1 
Meldung vom 24. Januar 1980 
2) Adkins. M. u.a.: Pub1ic Invo1vement in Oecision Making Re1ated to 
Advanced Techno109ies such as Nuc1ear Power. OECO-Papier 
unveröffentlichtes Manuskript. Paris 1972 • 
3) Agraphiotis. 0.; Pages J.: Le Nuc1eaire et 1es th~mes d'actua1it~ 
essai de classification. In: Col1oQue sur les implications 
psycho-sociologiques du de~eloppment de l'industrie 
nuc1eaire. Paris 13-15. Januar 1977 
4) Agraphiotis U.a.: Le pub1ic et 1. nuc1~aire. IAEA-CN-36/254 Report. 
International Atomic Ener9Y Agency. Wien 1977 
5) Alemann. U.v. (Hrg): Partizipation-Oemokratisierun9-Mitbestimmun9 
Eine Einführung. Op1aden 1975 
6) Alemann. U.v.: Partizipation - überlegung Zur normativen Diskussion 
und empirischen Forschung. In: Bürgerbeteiligung und 
Bürgerinitiativen. hrg. von H. Matthöfer. Vi11ingen 1977 
7) A1fven. H.: Energy Po1icy and Pub1ic Acceptance. IAEA-CN-36/588. Vor-
trag auf der International Conference on Nuclear Power and 
Its Fue1 Cyc1e. 5a1zburg. 2.-13. Mai 1977. 
8) Allport. G.W.: Attitudes. In: A Handbook of 50cia1 
University Press 1935. 5.798-844 
Psycho1ogy. C1ark 
9) A1tner. G.: 
10) A1tner. G.: 
11) Amery, C.: 
12) Amery C.: 
Der Philosoph und die Macht. In: Zeit zum Umdenken. 
hrg. von 5. de Witt und H. Hatzfeld. Reinbek 1979, 5.54-60 
Kirche und Kernkraftwerke - Grundsätze und Erfahrungen 
In: Kernenergie. Mensch, Umwelt, hrg. von K. User u.a. 
Köln 1976. 5.15ff 
Natur als Politik. Die ökologische Chance des Menschen. 
Reinbek 1976 
Das zweite Netz - Ein Vorschlag in die Energielücke 
hinein. In: überleben ohne W;rtschaftswa~hstum? hrg . . 
von der evangelischen Akademie Arnoldshelm und der Stlftung 
Mittlere Techno109ie. Kar1sruhe 1977, 5.51-67 
13) Andritzky, E. 
-2-
W., Wah1-Ter1inden, U.: Mitwirkung von Bürgerinititativen an 
der Umweltpolitik. Forschungsbericht 77, hrg. vom Institut 
für Zukunftsforschung im Auftrag des Umweltbundesamtes. 
Ber1in 197B 
14) Andritzky, W.: Erstmals wissenschaftlich untersucht: Bürgerinititativen. 
In: Bild der Wissenschaft, Nr. 2, 197B, S.84-94 
15) Arbeitsgemeinschaft für Umweltfragen e.V. (Hg.): Bürgerinititative Umwelt-
schutz - Analyse einer Fragebogenaktion . Bonn 1975. 
16) Armbruster, B. , Leisner, R. : Bürgerbeteiligung in der Bundesrepublik. 
Zur Freizeitaktivität verschiedener Bevölkerungsgruppen 
in ausgewählten Beteiligungsfeldern (Kirchen, Parteien, 
Bürgerinitiativen und Vereinen). Göttingen 1975. 
17) Arminger, G.: Faktoranalyse. Studienskripte zur Sozio10qie, Bd.3. Stutt-
gart 1979. 
18) Atkinson, J.: Handbook for Interviewers. London 1971. 
19) Autorenkollektiv: Zum richtigen Verständnis der Kernindustrie - 66 
Erwiderungen. SAIV, Uni.Bremen . Bremen 1975. 
20) Back, H.W., Gergen, K.J.: Apoca1yptie and Seria1 Time Orientations and 
the Strueture of Opinlons. In: Public Opinions Quarterly, 
Vol . XXVII, Nr.2, 1963. 
21) Bahr, D.-D.: Kritik der "Politischen Technologie". Frankfurt 1970. 
22) Ba1dewicz, W. u.a.: Historical Perspectives on Risk for Large-Sca1e Tech-
nologiea1 Systems . UCLA-ENG-7485. Schoo1 for Engineering 
and App1ied Science. Los Angeles 1974 . 
23) Barnett, H.G.: Innovation: The Basis of Cu1tura1 Change . New York 1953. 
24) Barnett, H.G.: Wer nimmt Neuerungen an und wer lehnt sie ab? In: Sozio-
logie der Entwicklungsländer, hrg. von Peter Heinz. Köln, 
Ber1in 1962, S.73-109. 
25) Battelle-Institut (Hrg): Einstellung und Verhalten der Bevölkerung 
gegenüber verschiedenen [nergiegewinnungsarten. Band 1 
und Band 2, Bericht für das Bundesministerium für For-
schung und Technologie. Frankfurt/Main 1977. 
-3-
26) Battelle-Institut : Untersuchungen über die technischen, organisatorischen 
und gesellschaftlichen Vorsaussetzungen für Risikostrate_ 
gien im Bereich technologischer Entwicklunq. Präsentation 
des Projektes im 8undesinnenministeriurn. 11. Mai 1978 
RS12-510 120/4-3-1/1. 80nn 1978. ' 
27) Beckmann, P. : The Health Hazards of Not Going Nuclear. Boulder (Colora-
do, USA) 1976 . 
28) Benninghaus, H. : Soziale Einstellungen und soziales Verhalten. In: Sozio-
logie-Sprache, Bezug zur Praxis, Verhältnis zu anderen 
Wissenschaften, hrg. von G. Albrecht u.a •• Opladen 1973, 
S.671-707 . 
29) Berger, H.: Untersuchungsmethode und soziale Wirklichkeit. Frankfurt! 
Main 1974. 
30) Bernt, H. : Kernenergietechnik und Monopolkapital. In: IPW-8erichte, 
Bd.1, Berlin 1974, S.39-43 
31) Bethe, H.A.: The Controversy About Nuclear Power. IAEA-CN-36!582. Vortrag 
auf der International Conference on Nuclear Power and Its 
Fuel Cycle. Salzburg, 2-13 . Mai 1977. 
32) Bick, Th., Kasperson, R.E.: Pitfalls of Hazards Management. 
rnent, Vol.20, Nr.8, Oktober 1978, S.30ff. 
In: Environ .. 
33) Bieber, H.-J.: Zur politischen Geschichte der friedlichen Kernenergie-
nutzung in der Bundesrepublik Deutschland. Materialien 
zum Gutachten der FEST, "Alternative Möglichkeiten für 
die Energiepolitik" • Band 3. Heidelberg 1977. 
34) Bierbrauer, 
35) Bierbrauer, 
36) Blau, P. : 
37) Bloch, E. : 
G.: Intuitive Erklärungen und Prognosen. Eine sozialpsycho-
logi sche Analyse zur Legiti~ation po~it~scher Planung . 
In : Bürgerbeteiligung und Burgerin1t1at1ven, hrg. von 
H. Matthöfer. Villingen 1977, S.66-87 
G.: Die Zuschreibung von Verantwortlichkeit. Eine attri~u­
tionstheoretische Analyse . In: Sozialwissenschaften 1m 
Studium des Rechts . Strafrecht und Kriminologie, hrg. von 
W. Hassener und L. Luderssen. München 1977. 
Exchange and Power in Social Life . New York 1964 
Das Prinzip Hoffnung. Frankfurt 1959 
-4-
38) Borsdorf-RUhl, B.: Bürgerinititativen im Ruhrgebiet. Essen 1973 
39) Borseh, P., 
40) Borseh, P., 
Münch, E.: Die Sicherheit von Kernkraftwerken. In: Nutzen 
und Risiko der Kernenergie . Jül-Conf-17 der Kernfor-
schungsanlage Jülich GmbH. Jülich 1978. 
Münch, E.: Die Sicherheit von Kernkraftwerken. In: Tatsachen 
über Kernenergie, hrg . von E. Münch , Essen 1980. 
41) Bowmann, C.H. u.a.: The Prediction of Voting Behaviour in a Nuclear Energy 
Referendum. International Institute for Applied Systems 
Analysis. RM-78-8. laxenburg (österreich) 1978 . 
42) Brown, H.: learning how to live in a technological society. Teil I: 
The Crisis of Affluence. Teil 11: Toward a World Community. 
In : The Bulletin of the Atomic Scientists, Vol.35, Nr.8, 
September 1979, S.13ff und Oktober 1979, S.23ff. 
43) Buchholtz, H.-Chr. u.a. (Hrg): Widerstand gegen Atomkraftwerke . Informa-
tionen für Atomkraftge9ner und solche, die es werden 
wollen . Wuppertal 1978. 
44) Büchel, W. : Gesellschaftl;che Bedingungen der Naturwissenschaft. 
München 1975. 
45) Burdge, R.J . : ludtke, R.l.: Social Separation Among Oisplaced Rural 
Families: The Case of Flood Control Reservoirs . In: 
Social Behavior. Natural Resources and the Environment, 
hrg . von W.R.Burch u.a •• New York 1972, S.85-107. 
46) Buiren. S.van: Bürgerinitiativen im Bereich von Kernkraftwerken. Bericht 
des Deutschen BatteIle Instituts fUr das Ministerium für 
Forschung und Technologie. Bonn 1975. 
47) 8urnham, J .B. u.a.: Quantified Social and Aesthetic Values in Environ-
mental Oecision Making. In: Siting of Nuclear Facilities. 
IAEA-SM-188/43. Wien 1975. 
48) Buse, M., Nelles, W.: Formen und 8edinqunqen der Partizipation im poli-
tisch/administrativen Bereich. In: Partizpation-Oemokra-
tisierung-Mitbestimmung, hrg . von U. von Alemann. Opladen 
1975, 5.41-111. 
49) Campbell, O.T.; Stanley, J.C.: Experimental and quasi-experimental designs 
for research on teaching. In: Handbook of Research on 
Teaching. Chicago 1963, S.171-246. 
50) Campbell, O.T. u.a.: Experimental Methods . Englewood Cliffs 1972 
-5-
51) Cantril, H.: The Patterns of Human Concern . New Brunswidk 1965 
52) Chandler, R.: Public Opin!on: Chan9ing Attitudes on Contemporary Polit!cal 
and Social Issues . New York, London 1972. 
53) Chester, H. : Perception and Cognitive Theory. London 1964 . 
54) Cohen, B.L.: Perspect!ves on the Nuclear Debate . In: The Bulletin of the 
Atomic Scientists. Nr . 3D, 1974, S.25-39. 
55) Cohen, B.L. : Some Issues in the Nuclear Power Controversy. In : Public 
Utility Fortni9ht 9B, Nr .4, August 1976. 
56) Cole, W.S. : Atomic Energy: Its Implications for Society. In: High 
Points, Nr. 37, 1955. 5.5-11. 
57·) Combs, B. ; Slovic, P. : Causes of Death: Biased Newspaper Coverage and 
Biased Judgements . Manuscript : Journal ism Quarterly, 
Decision Research Report 78-8. Eugene (Oregon) 1978. 
58) Comey, 0.0.: 
59) Conrad, J. : 
Social Stability and Nuclear Energy. Vortrag auf Int. 
Symposium: Nuklearindustrie in der Krise. Rueschlikon (Schweiz) 26-28.2.76. 
Zum Stand der Risikoforschung. Bericht des Battelle-Insti-
tuts im Auftrag des Bundesministerium des Innern . Bericht 
Nr. 100/3. Frankfurt 1978. 
60) Coombs, C.H.: A Review of the Mathematical Psychology of Risk and Risk 
Taking . Michigan-University-Report 72-6. Michigan 1972. 
61) Coser, L.: Conflict-Social Aspects. In : International Encyclopedia 
of the Social Sciences, Vol.3. New York 1968, 5.232-236. 
62) Courvoisier,P . : Vom Sinn der Information der öffentlichkeit über hoch-
komplexe Technologien. Schweizerische Vereinigung für 
Atomenergie. Beilage zu Bulletin Nr. 5, März 1974. 
63) Dahl, J.: 
64) Dunazedier, 
Auf Gedeih und Verderb. Zur Metaphysik der Atomenergie-
Erzeugung . In: Scheidewege, 5, 1975, 5.201-226 . 
J.: Leisure .nd Post-Industrial Soc!eties. In: Technology, 
Human Values und Leisure, hrg. von M. Kapl an and Ph . Bosser-
man . Nashville, New York, 1971. 
-6-
65) Dawes, R.M.: Formal Models of Dilemmas in Social Decis ion-Making 
In: Human Judgement and Decision Processes. New York 
1975. S.87-129. 
66) Dawes, R.M . : Grundlagen der Einstellungsmessung. Weinheim/Basel 1972. 
67) Davies, 
68) Oavies, 
J.E .O. : Canadian Attitudes to Nuclear Power. IAEA-CN-36/5BO. Vor-
trag auf der International Conference on Nuclear Power 
and Its Fuel Cycle . Salzburg, 2.-13. Mai 1977 . 
J .E.O.: Recent Public Attftudes to Nuclear Power. Lecture at the 
15th Annual Conference of the Canadian Nuclear Association. 
Ottawa 1976. 
69) Oeclaration of the European Nuclear Society vom 6. Mai 1977: Offener Brief 
an die öffentlichkeit . Salzburg 1977. 
70) Oe Fleur, M.L.; Westie, F.R.: Verbal Attitudes and Overt Acts: An Experiment 
on the Salience of Attitudes. American Sociological Review, 
Nr. 23, 1958, S.667-673. 
71) Oenenber9, H.S . : Nuclear Power: Uninsurable . Congressional Report 
November 25, Washington, D.C.: U.S.Government Printing 
Office 1974. 
72) Oeutsche-Presse-Agentur (OPA-U"",eltfragen): Die "Grünen" jetzt als Bundes-
partei - Aber noch kein Pro9ramm. Umweltfragen, Nr.l, 
17.01.1980, S.1-5. 
73) Deutscher Bund für Lebensschutz: Kritischer Bericht zum Expertengespräch 
über Reaktorsicherheitsforschung . In: Energiediskussion, 
hrg . vom Bundesministerium für Forschung und Technologie. 
Nr.6/7, Dezember 1978, S.6Dff. 
74) Dicksan, 0. : Alternative Technologie. Strategien der technischen 
Veränderun9. München 1978. 
75) Dienel, P.C.: Zur Entwicklung eines Verfahrens geordneter bürgerschaft-
licher Planungsbeteiligung. In: Analysen und Prognosen, 
Nr .45, hrg. vom Zentrum Berlin für Zukunftsforschung. 
Berlin 1976. 
76) Diepold W., Krebsbach, C.: Versuch zur Erfassung von Risikomaßstäben und 
Definitionen. Teil A aus technischer, Teil B aus gesell-
schaftlicher Sicht. Bericht für das Bundesministerium des 
Innern, hrg. vom Battelle-Institut Frankfurt. Teilbericht 
100/2. Frankfurt 1977. 
77) D~derlein, J.M . : The Nuclear Controversy International. Vortragsmanuskript 
der KFG-Veranstaltung über "Kernenergie und Umwelt". 
Geesthacht 14.3.1975. 
-7-
78) D~derlein, J.M.: Nuclear Power and Society - Generalizations from the 
European Scene. In : Socioeconomic Issues for Nuclear 
Plants. Invited Paper. Konferenz der American Nuclear 
Society. Washington D.C. 14.11.1976, 5.98 . 
79) D_derlein, J.M.: Nuclear Power as a Public Issue. IAEA-CN-36/451. Vortrag 
auf der International Conferenee on Nuelear Power and Its-
Fuel Cyele. Salzburg, 2.-13. Hai, 1977. 
80) D_derlein, J.M: Soeiety, Risk and Nuclear Power. Vortrag vor dem Hahn-
Heitner-Institut in Berlin, März 197B. 
81) Douglin, J .J.: The Anti-Nuclear-Movement and Its Critics . The Social Base 
of Support. Vortrag auf dem Seminar for Industry Spakesmen . 
Toronto, g. November 1976. 
B2) Douvan, E., Withey, 5.: Public Reaetions to Nonmilitary Aspects of Atomic 
Energy. In: SCienee, Nr.119, Jahrgang I, 1954, 5.1-3. 
83)· Downs, A.: 
84) Downs, A.: 
85) Dumenil, G.: 
An Economic Theory of Democracy. New York 1957. 
Up and Down with Ecol09Y - The 'Issue-Attention-Cycle'. 
In: The Publie Interest, Nr .28, 1972, S.3B-50. 
Energie nucleaire et opinion publique. Analyse des r~sults 
de quelques. Report NO.2. ADISH/IREP. Paris/Grenoble 1977. 
86) Dunster, H.J.; Vinck, W.: The Assessment of Risk - Its Value and Limi-
tations. Vertrag auf der: Conference of the European 
Comission on Technological Risk. Berlin, 1-3. April 1979, 
abgedruckt in: Monitor-Nuclear EnQineerin9 International, 
August 1979, S.23ff. 
87) Duttweiler, G.: Social Stability and Nuclear Energy. Lecture at the 
Symposium: Nuklearindustrie in der Krise . Rüschlikon, 
Februar 1976. 
88) Duve, F.: Zur Einführung - Demonstranten oder Dissidenten. In: Bürger 
gegen Kernkraftwerke, von H. H. Wüsten hagen , Hamburg 1975. 
89) Ebbighausen R.: Die Krise der Parteiendemokratie und die Parteien-
SOZiologie. Berlin 1969 
90) Eberwein, W.-D.: Krise und Konflikt - Zum Stand der Theorie. Fachgebiet 
Politikwissenschaft der Universität des Saarlandes. For-
schun9spapiere Nr.4. Saarbrücken, Juni 1973. 
91) Edwards, A.L.: Versuchsplanung in der psychol09ischen Forschung. Weinheim 
1971. 
92} Edwards, W.: 
-8-
The Theory of Declslon Making. In: Psychologlcal 8ulle-
tin, Nr. 51, 1954, S. 380-417. 
93} Electricit~ de France: Themes d'actualit~ et themes d'expresslon, 
8d. 1-4. Paris 1977 
94} Ellwein, T. , 
95} Ellul, J.: 
Lippert, E., Zoll, R.: Politische 8etelligung 
desrepublik. Göttingen 1975. 
In der 8un-
The Technological Society. Vintaqe Books. New York .964 
(orl9inal 1954) 
96} Engelmann, P.A. , Renn, 0.: On the Methodology of Cost-Benefit-Analysis 
and Risk Perceptlon. In: Directlons in Energy Policy. 
A Comprehensive Approach to Energy Resource Declsion-Ma-
kln9, hr9. von B. Kursunoglu und A. Perlmutter. Cambridge 
(Mass) 1980 
97} Eichholz, G.: Cost-Benefit and RIsk-Benefit Assessment for Nuclear 
Power Plants. In: Nuclear Safety, Vol.17, Nr.5, Sep-
tember/Oktober 1976, S.525ff. 
98} Eisenbart, C.: Außenpolitik und Kernener9ie. In: Alternative Möglich-
keiten für die Energiepolitik. Haterialien zum Gutachten. 
Band 2, hrg. von W. Lienemann u.a •• Heldelberg 1977, 
S.291-322. 
99} Eisenhart, G.: Krebsbach, C.: Stereotype Wahrnehmung. Ihr Einfluß auf 
Interaktion und Kommunikation Im Bereich politischer Pla-
nung von großtechnischen Einrichtungen. Auftrag des Bun-
desmlnlsterium des Innern. hr9. vom Battelle-Institut 
Frankfurt. Bericht 400/3. Frankfurt 1978. 
1OO} Erbslöh, E. : Das Interview. Techniken der Datensammlun9, 8d.1. Stutt-
gart 1972. 
101} Eroufman, L.H., Mattin9ly, T.J.: Critical Mass: Polltlcs, Technology 
and the Publlc Interest. In: Nuclear Safety, 8and 17, 
Nr. 5, 1976. 
102} Erskine, H.G.: The PolIs: Atomic Weapons and Nuclear Energy. In: The 
Publlc Opinion Quarterly, Nr.2, 1963 
103} Esser, H.: Soziale Regelmäßigkelten des Befragtenverhaltens. Meisen-
heim 1975. 
104} Esser, H.; Klenovits, K.; Zehnpfennig, H.: Wissenschaftstheorie. Band 1 
und Band 2, Teubner Studienskripte, hrg. von E.K. Scheuch 
und H. Sahner. Stuttgart 1977. 
-9-
105) Farmer, F.R.: Risk Quantification and Acceptabi1ity. In: Nuc1ear 
Safety, Vo1.17, Nr.4, Juli-August 1976, S.418ff. 
106) Fe1 ix, F.: The Socioeconomic Cha11enges of the Energy/lncome Re1-
ationship. Lecture at the Conference on Energy Deve10p-
ment. Washington, Februar 1976. 
107) Festinger, L.: A Theory of Cognitive Dissonance. Evanston, I11.{USA) 
1957. 
108) Flora, P.: Modernisierungsforschung. Op1aden 1974. 
109) Fischerhoff, H.: Das Kernenergierisiko - Seine Einschätzung und seine 
Annahme. Vortrag auf dem AIDN-INCA-Kongress: Nuc1ear 
Inter Jura, 2-5. Oktober 1977. 
110) Fischhoff, B.: Cost-Benefit-Ana1ysis and the Art of Motorcyc1e Maintenance 
In: Po1icy SCiences, Bd.8, 1977, S.177-202. 
111) Fischhoff, B.: Behaviora1 Aspects of Cost-Benefit-Ana1ysis. Vortrag auf 
dem International Review Seminar on "Impacts and Risks 
of Energy Strategies". Stockholm, 26-29. September 1978. 
112) Fischhoff, B.u.a.: How Safe is Safe Enough1 A psychometric study of 
attitudes towards techno10gica1 risks and benefits. 
In: Po1icy SCiences, Nr.9, 1978, S.127-152. 
113) Fischhoff, B.u.a.: Handling Hazards. In: Environment, Vo1.20, Nr.7, 
September 1978, S.16ff. 
114) Fischhoff, B.; Slovic, P.; Lichtenstein, S.: Weighing the Risks. Which 
Risks Are Acceptab1e1 In: Environment, Vo1.21, Nr.4, 
Mai 1979, S.17ff. 
115) Fishbein, M. (Hrg): Readings in Attitude Theory and Measurement. New 
York 1967. 
116) Fishbein, M.: 
117) Fishbein, M.; 
The Relation between Beliefs, Attitudes and Behavior. 
In: Cognitive Consistency, hrg. von S. Feldmann. New 
York 1966, S.199-233. 
Ajzen, J.: Belief, Attitude, Intention and Behaviour: An 
Introduction to Theory and Research. Reading (Mass) 1975. 
118) Fisher, B.; M,'tzner R' Darsky B.J.·. Peacet,'me Uses of Atomic Energy. • • • • Ann Harbor 1951. 
119) 
-10-
Forschungsgruppe an der FU In Ber1in: Zur Rolle und Funktion von Bür-
gerlnititativen In der Bundesrepublik und West-Ber1in. 
Analysen von 61 Bürgerinitiativen. In: Zeitschrift für 
Parlamentsfragen 4, 1973, S.247ff. 
120) Frantzen, 0.; Schmld-JÖrg.: Sozialwissenschaftliehe Untersuchungen über das Wahrnehmungsfeld der Bevölkerung im Bereich technolo-
gischer Risiken (hier Kernkraftwerke). Bericht für das 
Bundesministerium des Innern, hrg. vom Batte11e-Institut 
Frankfurt. Zwischenbericht 4001. Frankfurt 1976. 
121) Freyer, H.: 
122) Friedrich, 
Die Technik a1s ·Lebensmacht, Denkform und Wissenschaft. 
Malnz 1970. 
W.: Methoden der marxlstlsch,lenlnistischen 
Ber1in (Ost) 1972. 
Sozialforschung. 
123) Friedrichs, J.: Methoden der empirischen Sozialforschung . Relnbek 1973 
124) FürniB, B. u.a.: Optlmlerun9 eines regionalen Energieversorgungssystems 
bei mehrfacher Zielsetzung. In: Energiewirtschaftliehe 
Tagesfragen, Heft 3, März 1980, S.155ff. 
125) Fürstenberg, F. : Die Sozialstruktur der Bundesrepublik Deutschland. 
Ein soziologischer Oberblick, 2. Auflage . Op1aden 1972. 
126) Garrisson, J.: The Re1igious Background of the Nuc1ear Controversy. 
Dissertation, Trinity College. Cambridge 1977 . 
127) Gehlen, A.: Die Seele im technischen Zeitalter. Sozialpsychologische 
Probleme in der Industriellen Gesellschaft . Hamburg 1957. 
128) Gesellschaft für Reaktorsicherheit (Hrg): Die Deutsche Risikostudie. 
Kurzfassung. Bonn, 8. August 1979. 
129) Gi11ian, H.: Jobs versus Environment. In: San Francisco Examiner and 
Chronic1e, 10, Dezember 1972, 5.41. 
130) Gitt1er, J.B . : Socia1 Trends and Atomlc Energy. In: Socla1 Sclence 28, 
1953, S.67-71. 
131) Goerke, 0.: Untersuchung und Ausarbeitun9 von Maßnahmen zur Offent-
11chkeltsarbelt bei Erörterungsterminen über den geplan-
ten Bau kerntechnischer Anlagen. Auftrag des Bundesmi-
nisteriums des Innern. Vechta 1975. 
-11-
132) Goerke, 0.: Meinungsbildung in der öffentlichkeit über energiepoliti-
sche Fragen, insbesondere im Bereich der Kernenergie-Si-
cherheit. AUftrag des Bundesministeriums des Innern. 
Vechta 1978. 
133) Gorz, A.: ökologie und Politik. Beiträge zur Wachtsumkrise. Reinbek 
1977 
134) GOlant, S., Burton, B.: Advoidance Response to the Risk Environment. 
Research Working Paper Nr.6, Dept. of Geography. University 
of Toronto 1969. 
135) Golden, J., Watson, J.: Socio-Economic and Ecological Impacts of Nuclear 
Power Plants. Mc Lean (Virg.USA) 1975. 
136) Gonschior, P. : Die Berücksichtigung des technischen Fortschritts in 
Projektionen des wirtschaftlichen Wandels. In: Analysen 
und Prognosen, hrg. von Zentrum für Zukunftsforschung e.V., 
Heft 45, Mai 1976 
137) Gottstein, K.: Die nukleare Kontroverse in den USA. Vortrag bei der 
Mitgliederversammlun9 des Deutschen Atomforums e.V •• Bonn 
November 1975. 
138) Gottstein, K.: Technikkritik und Technikkontrolle. In: Physikalische 
Blätter, Band 35, Heft 6. Juni 1979, S.237ff. 
139) Graumann, G.F.: Eigenschaften als Problem der Persönlichkeitsforschung. 
In: Handbuch der Psychologie, hrg . von Ph. Lersch und 
H. Thomae. Band 4. Göttingen 1960. 
140) Greenhalgh, G.: The Opposition to Nuclear Power. An Appreciation of 
the Uranium Institute . London 1978. 
141) Greer-Wootten, B., Mitson, L.: Nuclear Power and the Canadian Public. ANational and Regional Assessment of Public Attitudes 
142) Gripp, H.: 
143) Gruhl, H.: 
and Perceptions of the Use of Nuclear Power for the Pro-
duction of Electricity. Institute for Behavioural Research, 
York University. Toronto 1976. 
Die Suche nach einem Neuen Lebensstil - ein sinnvoller 
Ausweg aus einem gesamtgesellschaftlichen Dilemma? 
In: Neuer Lebensstil, Verzichten oder Verändern?, hrg. 
von K.E. Henke und H. Zilleßen. Opladen 1978, S.248-289 . 
Kernener9ie ohne soziale Legitimation: In: Kernenergie -
offen bilanziert, hrg. vom Gottlieb Duttweiler-Institut 
für wirtschaftliche und soziale Studien. Frankfurt a.M. 
1976, S.113-121. 
144) Gruh1, H.: 
145) Grupe, H.: 
146) Guggenberger, 
147) Gutmann, G.: 
-12-
Ein Planet wird geplündert - Die Schreckensbilanz unserer 
Politik. Frankfurt/M. 1975 
Gedanken zur nuklearen Kontroverse in der Bundesrepublik 
Deutschland. In: Kerntechnik-Atompraxis, 18. JhrQ. Nr.7, 
Juli 1976, S.317-320 
B.; Kempf, U.(Hrg): Bürgerinitiativen und repräsentatives 
System. Op1aden 1978 ' 
Sozialisation von Risikoeinste11un~en, bearbeitet vom 
Batte11e-Institut Frankfurt für das Bundesministerium 
des Inneren, Nr. 401. Frankfurt 1977. 
148) Haas, J.E.; Boggs, K.S.; Bonner, E.J.: Science,Techno10gy and the Pub1ic. 
The Case of P1anned Weather Modification. In: Socia1 
Behavior, Natural Resources and the Environment, hrg. 
von W.R. Hurch u.a •• New Vork, Evanston, San Francisco, 
London 1972, S.151-173 . 
149) Habermas, J.: Legitimationsprobleme im Spätkapitalismus. Frankfurt/Main 
1973. 
150) Habermas, J.: Technik und Wissenschaft als Ideologie. Frankfurt/Main 
1968. 
151) Habermas, J . , Luhmann, N.: Theorie der Gesellschaft . Frankfurt a.M. 1975 
152) Hackmann, J.R.; Anderson, L.R.: The Strength, Re1evance and Source of 
Be1iefs about an Object in Fishbein's Attitude Theory. 
In: Journal of Socia1 Psycho10gy, Nr.76, 196B, S.55-67 
153) Höfele, W.: 
154) Höfele, W.: 
Hypothetica1ity and the New Cha11enges, The Pathfinder 
Ro1e of Nuc1ear Energy. Laxenburg. International Insti-
tute for App1ied Systems Analysis (IASA). RR-73-14. 
Laxenburg 1973. 
Die historische Entwicklung der friedlichen Nutzung der 
Kernenergie. In: KernenerQie und internationale ,Politik, 
hrg. von K. Kaiser und B. Lindernann. Reihe: Internationale 
Politik und Wirtschaft, Bd.37. München, Wien, 01denburQ 
1975. 
155) Haenschke, F.: Modell Deutschland? Die 8undesrepub1ik in der technologi-
schen Krise. Reinbek 1977. 
-13-
156) Hamilton, L.O.; Manne, A.S.: Health and Economic Costs of Alternative 
Energy Sourees. IAEA-CN-36/448. Vortrag auf der Inter-
national Conference on Nucle.r Power and Its Fuel Cycle. 
Salzburg, 2.-13.Mai 1977. 
157) Hansmeyer, K.H .; RUrUp, R.: Staatswirtschaftliche Planungsinstrumente. 
Wiso StudienbUcher, 2. AUflage. TUbingen 1975. 
158) Harder, Th.: Werkzeuge der Sozialforschung. ·8ielefeld 1970. 
159) Harnik, P.: The Ethics of Energy Production and Use: Oebate within 
the National Council of Churches. In: The Bulletin of 
the Atomic Scientists, Nr.2, Februar 1979, S.5-8. 
160) Harriss, R.C. u.a.: Our Hazardous Environment. In: Environment, Vol.20, 
Nr.7, September 1978, S.6ff. 
161) Hartkopf, G.: Risiko der Kernenergie. In: liberal, Heft 6, 18. Jhrg. 
162) Harvey, M.: 
Juni 1976. 
Project Summary: Improving the Societal Management of 
Technological Hazards. National Science Foundation. 
Juni 1979. 
163) Heintz, P. (Hrg): Soziologie der Entwicklungsländer, Köln 1962. 
164) Heintz, P.: Oie Zukunft der Entwicklung. Wien 1974. 
165) Helfen, 0., Laga, G.: Messung sozialer Schichtung. In: Kölner Zeitschrift 
für Soziologie und SozialpsycholoQie. hrQ. von R. König, 
Jg.26, Heft I, 1974. 
166) Herz, Th.: 
167) Hewitt, J.S.: 
168) Hinz, M.O.: 
169) Hirsch, H.: 
Der Wandel von Wertvorstellungen in westlichen Industrie-
geseIlschaften. In: Kölner Zeitschrift für Soziologie 
und Sozialpsychologie, hrg. von R. König, Jg.31, Heft 2, 
1979, S.282-302. 
Educational Responsibility in the Canadian Nuclear 
Controversy. In: Trans. American Nuclear Society. 
November 1976, S.73-74. 
Widerstand gegen das Atomprogramm - Krise des Rechtsstaa-
tes1 Referat auf dem Kernenergie-Seminar des Bundes fUr 
Lebensschutz. LUnP.burg,17 . u. 18.09.1977. 
The 'Information Campaign on Nuclear Energy' of the 
Austrian Government. PreTiminary Surve.v. IAEA-CN-36/589. 
Vortrag auf der International Conterence on Nuclear Power 
and Its Fuel Cycle. Salzburg, 2.-13.Mal 1977. 
170) Hirsch, J.: 
-14-
Wissenschaftlich-technischer Fortschritt und politisches 
System. Frankfurt/Main 1971. 
171) Hirshleifer, J. u.a.: Applyinq Cost-Benefit Concepts to Projects which 
Alter Human Mortality . Los Anqeles 1974. 
172) Hofstätter, P.R.: Oer Teufel aus der Maschine. Warum wir heute noch die 
Technik fürchten. In: Oie Welt, 7.Juli 1979. 
173) Hohenemser Chr. u.a.: The Distrust of Nuclear Power. In: Science, Vol.196, 
April 1977, S.25-34. 
174) Holm, K.(Hrg . ): Die Befraqung 1. Der Fraqebogen - Die Stichprobe. 
175) Huber, J.: 
176) Hülsmann, H.: 
177) Hülsmann, H.: 
178) HUlsmann, H.; 
179) Hunecke, H.: 
180) Illich, J . : 
181) Inglehart, R.: 
182) Inglehart, R.: 
München 1975. 
Technokratie oder Menschlichkeit. Zur Theorie einer huma-
nen und demokratischen Systementwicklung. Achberg 1978. 
Verantwortung von Wissenschaft - von Manhatten zu Wyhl . 
Meisenheim/Glan 1977 
Die Machtergreifung - Zur Verantwortbarkeit von Wissen-
schaft, Forschung und Technologie . In: Die atomare Ge-
sellschaft. Grundlage für den Dialog um die Kernfrage. 
Darmstadt und Neuwied 1978, S.147-162. 
Tschiedel, R.(Hrg): Kernenergie und wissenschaftliche Ver-
antwortunq. München 1978. 
Experimentelle Untersuchunqen zum Problem des Zusammen-
hangs zwischen erfragten und beobachteten Verhalten. 
Dissertation . Universität Heidelberg 1972. 
Die sogenannte Energiekrise oder Die Lähmunq der Gesell-
schaft. Das sozial kritische Quantum der Energie. 
Reinbek 1977. 
The Silent Revolution in Europe. Intergenerational Chanqe 
in Post-Industri.l Societies. In: American Political 
5cience Review, 65, 1971, S.991-1017. 
Values, Objective Needs and Subjective Satisfaction 
Among Western Publies. tomparative Political Studies 9, 
1977, 4, 5.428-458 
-15-
183) Inglehart, R.: Zusammenhang zwischen sozioökonomischen Bedinqunqen und 
individuellen Wertprioritäten. In: Kölner Zeitschrift 
fUr Soziologie und Sozialpsychologie, Bd.1 März 1980 
S.144ff. ." 
184) Inhaber, H.: Scientists and Economic Growth.In: Social Studies of 
SCience, Vol.7, Nr.4, November 1977, S.517-524. 
185) Inhaber, H.: Risk of Energy Production . Atomic Enerqy Control Board 
1119 REV-1, 2. Auflage. Ottawa (Canada) 1978. 
186) Jungk, R.: 
187) Jungk, R.: 
188) Kaase, M.: 
Der Atomstaat. Vom Fortschritt in die Unmenschlichkeit. 
München 1977. 
Vom 1000jährigen Atomreich. In: Der Spiegel, Heft-Nr.11, 
1977, S. 44ff. 
Bedingungen unkonventionellen politischen Verhaltens 
in der Bundesrepublik Deutschland. In: Leqitimationspro-
bleme politischer Systeme, hrg. von P. Kielmansegg. 
Politische Vierteljahresschrift, 17, 1976, Sonderheft 7, 
S.179-216 
189) Kafka, P.; Müneh, E.: Risikokonzept. Zur Einführung und Diskussion. 
In: KTG-Fachseminar. Störfallverhalten an Leistungsre-
aktoren. Jülieh, 23.-26.11.1976. 
190) Kahneman, 0., Tversky, A.: Inituitive Prediction: Biases and Correetive 
Proeedures. Manuskript. Heb"""-University. Jerusalem 1978. 
191) Kahneman, 0., 
192) Kahneman, D., 
193) Kahneman, D.; 
194) Kaplan, K.J.; 
195) Kaspar, R.: 
Tversky, A.: Subjeetive Probability: A Judgement of 
Representativeness. In: Cognitive Ps.yeholoqy, Band 2, 
1972, S.430-454. 
Tversky, A.: On the Psychology of Predietion. In: 
Psychological Review Nr. 80, 1973, S.237-251. 
Tversky, A.: Availibility: A heuristie view for judging 
frequency and probability. In: Cognitive Ps.yehology, 
Nr. 5, 1973, S.207-232. 
Fishbein, M.: The Souree of Beliefs, their Salieney 
and Prediction of Attitude. In: Journal of Personality 
and Social Psychology, Nr.19, 1969, S.92-99. 
"Real" versus Perceived Risk: Implications for Poliey. 
Manuscript. In: Impacts and Risks of Energy St~ategies. 
,ThßJri Analysis and Role in Management. Proeeed,ngs of 
the Conferenee. Stockholm 1978. 
196) Kates, R.W.: 
-16-
Hazard and choice perception in f100d p1ain management 
(Research Paper No. 78) Chicago: University of Chicago. 
Oepartment of Geography 1962. 
197) Kates, R.W.: Risk Assessment of Environmenta1 Hazard. SCOPE-Report, 
Nr.8. International Counci1 of Scientific Unions. 
Paris 1976. 
198) Katz, E.; Lazarsfe1d, P.: Personal Inf1uence. G1encoe (111.USA) 1955 
199) Katz, E. u.a.: Traditions of research on the diffusion of innovation. 
200) Keiser, G.: 
201) Kempf, U.: 
202) Kernenergie 
In: American Socio10gica1 Review, Nr. 28, April 1963, 
S.237-252. 
Die Energiekrise und Strategien der Energiesicherung. 
München 1979. 
Bürgerinitiativen - der empirische Befund. In: Bürger-
initiativen und repräsentatives System hrg. von 
B. Guggenberger und U. Kempf. Op1aden i978, S.35B-374. 
und internationale p01itik j hrg. von K. Kaiser und B. Lindernann. MünChen, 0 denburg 1975. 
203) Kernenergie - offen bilanziert. Schriftenreihe: Brennpunkte Bd.4, 
Frankfurt 1976. 
204) Keyes, G.E., Ponce1et, C.G.: Adressing the Nuc1ear Controversy on 
University Camouses . IAEA-CN-36/450. Vortrag auf der 
International Conferente on .Nuc1ear Power and Its Fue1 
Cyc1e. Salzburg, 2.-13.Mai 1977. 
205) Kie1mansegg, Graf P.: Politik in der Sackgasse? Vortrag auf der wissen-
schaftlichen Tagung der CDU zu Umwelt und Wachstum, abge-
druckt in: Energiediskussion, hrg. vom Ministerium für 
Forschung und Technologie. Heft 3, Juni 1979, S.31ff. 
106) Klausner, S.Z.: Some Problems in the Logic of Current Man Environment 
Studies. In: Socia1 Behavior, Natural Resources and the 
Environment, hrg. von W. Burch u.a •• York 1972, 
S.334-363. 
107) K1ineberg, 0.: Socia1 Psycho10gy. New York 1966. 
208) Kmieciak, P. : Wertstrukturen und Wertwandel in der Bundesrepublik 
Deutschland - Grundlagen einer interdisziplinären empi-
rischen Wertforschung mit einer Sekundäranalyse von Umfra-
gedaten . Veröffentlichunq der Kommission für wirtschaf-
lichen und sozialen Wandel Nr.135. Göttinqen 1976. 
-17-
209) Knirsch, H.P.; Nickolmann, F.: Die Chance der BÜQerinitiativen. Ein 
Handbuch. Wuppertal 1976. 
210) Kodolitsch, P. von: Gemeindever.altunQen und BürQerinitlatlven. In: 
Archiv für Kommunalwissenschaften, Heft 2, 1975, S.264ff. 
211) König, R.: Neue Strömungen der GemeindesozloloQie. In: Handbuch der 
empirischen Sozialforschun~. hrq. von R. Köniq. 3.Auflaqe, 
Bd.2. Stuttgart 1973. 
212) Kommission der EG: Wissenschaft In der öffentlichen Meinung Europas. 
213) Korff, W.: 
Brüssel 1978. 
Kernenergie und Moraltheologie. Der Beitraq der theoloQi-
sehen Ethik zur Frage al1Qemeiner Kriterien ethischer Ent-
scheidun9sprozesse. Frankfurt/Main 1979. 
214) Krech, 0. , Crutchfield, R.S., Ballachey, E.L.: Individual in Society. 
Ne. York, London, Sydney 1962. 
215) Küchler, M.: Multivariable Analyseverfahren.Studienskripten zur 
Soziol09ie. Stuttgart 1979. 
216) Kutsch, Th.: 
217) Lange, W.P.: 
218) Lapp, R.E.: 
219) Lenzer, Ch.: 
220) Lerner, 0.: 
221) Levine, G.; 
Die Welt im Jahre 2000. Analyse einer internationalen 
Umfrage über die Einstellunq zur Zukunft. Kronberg (Taunus) 
1974. 
Zur Rolle und Funktion der Bürgerinitiativen in der BRD. 
In: Zeitschrift für ParlamentsfraQen. Opladen, Bd.4 
Juni 1973. S.247ff 
The Heart of the Nuclear Controversy. Atomic Industral 
Forum. AED-73-55B-018. San Francisco, November 1973. 
Probleme forschungspolitischer Planung beim BundestaQ. 
In: Technikfolgen, Technik-Abschätzunq, hrg. von H.Haas. 
München, Wien 1975. 
Die Modernlsierung des Lebensstils: Eine Theorie 
In: Theorien des sozialen Wandels, hrg. von W. Zapf. 
Köln, Berlin 1971. 
Modell J.: American Public Opinion and Fallout-Shelter Issue. 
In: Public Opinion Quarterly, Nr.2, 1965, S.270-279. 
-18-
222) Lindackers, K.-H.: Die Bedeutung technischer Risiken. In: Atomwirt-
schaft-Atomtechnik, 19, 1974, 5.284-288 
223) Lindblom, Ch.E.: Politics and Markets. The World's Political-Economic 
Systems . New York 1977. 
224) Linnerooth, 
225) Linnerooth, 
226) Lob, 5.: 
J.: A Critique of Recent Modellin9 Efforts to Evaluate Human 
Life. RR-75-67. International Institute for Applied Systems 
Analysis. laxenburg 1975 . 
J.: Methods for evaluatin9 mortality risk. In Futures, 
Nr. 8, 1976, 5.293-309. 
Atomenergie - Zur Struktur eines gesellschaftlichen Kon-
fliktes. In: Kernenergie und wissenschaftliche Verantwor-
tung, hrg. von H. Hülsmann und R. Tschiedel . Kronberg 1977. 
5.123-154. 
127) Lovins, A.B.: Soft Energy Paths: Toward a Durable Peace. San Francisco, 
New York 1977 . 
128) Lowrance, W.W. : Of Acceptable Risk. Science and the Determination of 
Safety. Los Altos (Cal.) 1976. 
228) Lübbe, H.: Herrschaft und Planung. Die veränderte Rolle der Zukunft 
in der Gegenwart. In: Festschrift für Max Müller. 
Freiburg 1966, S.188ff. 
230) Lübbe, H.: Technischer und sozialer Wandel als Orientierunqsproblem. 
Vortragsmanuskript. 29.05.1979. 
231) Luhmann, N.: Soziol09ische Aufklärung. Aufsätze zur Theorie sozialer 
Systeme. Band 1, 3. Auflage. Opladen 1972. 
232) Luhmann, N. : Offentliehe Meinun9. In: Politische Planung. Aufsätze zur 
Soziologie von Politik und Verwaltun9; hr~. von N. Luhmann. 
Opladen 1971, 5.9-34. 
233) luhmann, N. : Le9itimation durch Verfahren. Neuwied-Berlin 1969 
234) Luhmann, N.: Macht. Stuttgart 1975 
235) Luhmann, N.: Vertrauen. Ein Mechanismus der Reduktion sozialer Komplexi-
tät . Stuttgart 1973. 
-19-
236) Maderthaner, R.; Guttmann, G.; 5waton, E. ; Otway, H.J.: Effect of 
Oistance upon Risk Perception. In: Journal of Applied 
Psychology, 63 (3) 5.380-382. 1978. 
237) Marcuse, H. : Der eindimensionale Mensch. Neuwied/8erlin 1967. 
238) Matthöfer, H. : Kernenergie und öffentlichkeit. Rede vor der Reaktorta-
gung in Nürnberg am 10. April 1975, veröffentlich als 
50nderdruck des Presse- und Informationsamtes der Bun-
de,regierung, aus Nr.48, 5.455-458. Bonn 1975 . 
239) Mauss, A.l.: 
240) Maxey, M.N. : 
241) Maxey, M.N.: 
242) Mayer-Tasch, 
243) Mayer-Tasch, 
50cial Problems as 50cial Movements. New York 1975 
(insbesondere Chapter 15: The Environment as a 50cial 
Problem, 5.556-701). 
Nuclear Ener9Y Oebates : liberation and Oevelopment. In : 
The Christian Century, Heft 93 , Juli 1976, 5.656ff . 
Exorcising Nuclear Oemonry: Ethics Versus Ideological 
Politics. lecture at the AIF-Workshop, Nuclear Informa-
tion Strategies, unveröffentlichtes Manuskript, 
las Vegas 1976. 
P.C.: Kernenergie und Bürgerprotest. In : Energiepolitik 
ohne Basis. Vom bürgerlichen Ungehorsam zu einer neuen 
Energiepolitik, hrg. von C. Amery, P.C. Mayer-Tasch, 
K.M.Meyer-Abich. Frankfurt/Main 1978, 5.7-39 . 
P.C . : Die Bürgerinitiativbewegung. Der aktive Bürger als 
rechts- und politikwissenschaftliches Problem. Rein-
bek 1976 . 
244) Maynard,W.5.u.a:Public Values Associated with Nuclear Waste Disposal . Battelle Memorial Institute. BNWl-1997. 5eattle, Wash . 
1976 . 
245) Mayntz, R. ; 
246) Mazur, A.: 
247) Mc Evoy, J.: 
Holm, K.; Hübner, P.: Einführung in die Methoden der empi-
rischen 50zialforschung. 2.Auflage. Opladen 1971. 
Opposition to technological innovation . In: Minerva, 
Vol.13, london 1975, 5.58-81. 
The American Cancern with Environment. In: Social 
Behavior, Natural Resources and the Environment, hrg. 
von W. R. 8urch u.a •• New York 1972, 5.214-236. 
24B) Me Guire, 
-20-
W J . The Nature of Attitudes and Att1tude Change. In: Hand-
• ·book of Sec1a1 Psycho10gy, hrg. von G. Lindzey und E. Aron_ 
son, Bd.III. Reading, Mass. 1969, 5.136-314. 
. 249) Means, R.L.: Pub1ic Opinion and P1anned Changes 1n Socia1 Behavior: 
250) Me1ber, B.O. 
The Eco10gica1 Crisis. In: Socia1 Behavior, Natural Re-
sources and the Environment, hrg. von W.R . Burch u.a •• 
New York 1972, 5.203-213. 
u.a . : Nuc1ear Power and the Pub1ic: Ana1ys1s of Co11ected 
Survey Research. Batte11e Memor1a1 Institute. Seatt1e, 
Washington. November 1977. 
251) Meinefeld, W.: Einstellung und soziales Handeln. Reinbek 1977 
252) Metzen, H.: Mensch und Technik aus philosophischer und psycho10Qischer 
Sicht. In: Technik contra Humanität. LenQqrieser Taoe 
6.-B.Ju11 1979. München 1979. 
253) Meyer-Abich, K.M. (Hrg): EnergieeinsparunQ als neue Energieque11e/Wirt-
schaftspolitische Möglichkeiten und alternative Technolo-
gien. Wien 1979. 
254) Mez., L. (Hrg): Der Atomkonflikt. Atomindustrie, Atompolitik und Anti-
AtombewegunQ im internationalen Vero1eich. Ber1in (West) 
1979. 
255) Mez, L.; Wi1ke, M. (Hrg): Der Atomfi1z. Gewerkschaften und Atomkraft. 
Ber1in 1977. 
256) Mi11er, A.G., Gi11en, B., Schenker, C., Radlove, 5.: The prediction and 
perception of obedience to authority. In: Journal of 
Persona1ity Nr. 42, 1974, 5.23-42. 
257) Moldenhauer, B.: Politische und ökonomische Entstehunosbedingungen der 
zivilen Atomindustrie. In: Blätter für deutsche und inter-
nationale Politik Nr. 7, 1975, 5.741-754. 
25B) Moncrief, L.M.: The Cu1tura1 Bas1s of our Environment.1 Crisis. In: 
Western Man und Environment.1 Eth1cs _ Att1tudes toward 
N.ture and Techno10gy, hrg. von O.G. B.rbour. London, 
Don Mi11s, Ont.rio 1973. 
259) Morrison, O.E. u.a.: The Environmenta1 Movement. Some Pre1iminary 
Observations and Pred1ct10ns. In: Soc1.1 Beh.viour, N.-
tur.1 Resources and the EnVironment, hrg. von W. Burch 
u.a •• New York 1972, $.259-279. 
-21-
260) Mortimer, R.C.: Moral Aspects of Nuclear Energy. In: Economic and 
Social Consequences of Nuclear Energy, ed. by Lord 
Sherfield. Oxford 1972, S.69-75. 
261) Müller, A.M.K.: Rechnerische oder wirkliche Sicherheit? Zum Dilemma der 
Risikoeinschätzunq in nuklearen und nuklear induzierten 
Mensch-Maschine-Systemen. In: Alternative Möqlichkeiten 
fUr die Energiepolftik, hrg. von W. lienemann u.a., 
Band 2. Heidelberg 1977, S.128-158. 
262) Müller, P.: Energie. Von der Staumauer Zum Kernkraftwerk. Wien, 
ZUrich, München 1978. 
263) Müller, S.: Untersuchungen zur Messung peSSimistischer und oDtimis-
tischer Zukunftserwartungen. Köln/80nn 1973. 
264) Müllert, N.R . : VeränderunQschancen in der Industriegesellschaft. In: 
Technologie und Politik, 8d.11, Sanfte Technik, hrq. 
von F. Duve. Reinbek 1978. 
265) Münch, E.; Borseh, P.: Gibt es eine wissenschaftliche Kernenergie-Kon-
troverse. In: Kernenerqfe und öffentlichkeit, hrQ. von 
W.D. Müller und R. Hossner. Düsseldorf 1977, S.28-42. 
266) Münch, E.(Hrg.): Tatsachen über Kernenergie. Essen 1980 
267) Murphy, A.H.; Winkler, R.L.: SUbjective Probability Forecastinq: Some 
Real World Experiments. In: Utility, Probability and 
Human Decision Making, hrg . von A. Wendt u. C. Vlek. 
Dordrecht (NL), Boston (USA) 1975, S.177-198. 
268) Nader, R.; Abotts, J.: Tödlicher Fortschritt - Atomenergie und ihre 
Folgen. München 1978. 
269) Naßmacher, K.H.: Politikwissenschaft I. Politische Systeme und politi-
sche Soziologie . 2. AUflage, DUsseldorf 1973. 
270) Narr, W.-D.: Theoriebe9riffe und Systemtheorie. Stuttgart/Berlin/Köln/ 
Mainz 1972. 
271) Narr, W.-D .: Zur Genesis und Funktion von Krisen - Einige systemana-
lyt ische Marginalien. In: Herrschaft und Krise, hrg. 
von M. Jänicke. Opladen: 1973, 5.224-236. 
272) National Wildlife Federation: A Study of the Attitudes of the American 
Public Toward Improvement of the Natural Environment. 
Report of the ·National Survey conducted by L. Harris 
and Associates . Washington, D.C. 1970. 
-22-
273) Nelkin, 0.; Pollak, M.: Consensus and Conflict Resolution. The Politics 
of Assessing .Risk. Vertrag auf der Conference of the 
European Commission on Technological Risk. Berlin, 1.-3. 
April 1979 . 
274) Nelkin, 0.: 
275) Nelkin, 0. : 
276) Niehaus, F. : 
277) Nisbett, R.E., 
278) Noelle-Neumann, 
Nuclear Power and its Critics . Ithaca and lundon 1971. 
Technological Oecis ions and Democracy. london, New York 
1978. 
Identification de d~terminants technologiQues, psycho-
109iQues et sociaux des altitudes vi~-~-vis de l'ener9ie 
nucleaire. In: COlloQue sur les implications psycho-so-
zil09iQues du d~veloppment de'l industrie nucleaire. 
Paris 13-15. Januar 1977, 5.454-486. 
Borgida, E. : Attribution and the Psychology of Prediction . 
In : Journal of Personality and Social Psychology, Nr. 32, 
1975, 5.932-943. 
E. : Mehrheit für Kernenergie? In: Atomwirtschaft-Atom-
technik, Bei lage : Kernenergie und Umwelt, November 1978, 
5.111. 
279) Noel le-Neumann, E.: Jahrbuch der öffentlichen Meinung 1968-1973. Allens-
bach . Bonn 1974. 
2BO) Noelle-Neumann, E.: Werden wir alle Proletarier? Wertewandel in unserer 
Gesellschaft. Osnabrück 1978. 
281) Nowotny, H. : 
282) Nowotny, H. : 
Social Aspects of the Nuclear Power Controversy. 
national Institute for Applied Systems Analysis . 
RM-76-33. laxenburg 1976 . 
Inter-
Kernenergie : Gefahr oder Notwendigkeit . Anatomie eines 
Konfliktes. Frankfurt/M. 1979. 
283) Oberbacher, B.u.a.: Nutzen der Kernenergie - eine ökologisch-ökonomische 
Betrachtung. Bericht für das Bundesministerium des Innern, 
hrg. vom Battelle-Insti tut Frankfurt. Nr .300/2. Frank-
furt 1977 . 
284) Oberbacher, B.u.a. : Vergleich der Gesundheitsgefährdung bei versch iedenen 
Technologien der Stromerzeugung und erster Versuch der 
Einordnung des Risikos der Kernenergie . Bericht für das 
Bundesministerium des Innern, hrg. vom Battelle-Institut 
Frankfurt . Zwischenbericht 200/1 . Frankfurt 1976. 
285) Oerter, R.: 
286) Offe, C.: 
287) Offe, C.: 
-23-
Struktur und Wandlung von Werthaltungen. München 1972. 
Qualität des Lebens - politikwissenschaftliche Aspekte 
eines politischen Schlagworts. In: Lebensqualität - Von 
der Hoffung, Mensch zu sein; hrg. von der Landeszentrale 
für Politische Bildung, NRW. Köln 1974, S.129-136. 
Oie Logik des kleineren Obe1s . Ein Vorschlag an die Sozial-
demokraten. In: Die Zeit, Heft 46, 9.11.1979 . 
288) Ohe, von der W. u.a. : RisiKoperzeption und Risikoakzeptanz. Zielsetzung, 
theoretische Annahmen, methodisches Vorgehen und Erqebnis-
se einiger empirischer Untersuchungen zur Bestimmunq der 
Akzeptanz der Kernenergie durch die Bevölkerunq. AuftraQ 
der Gesellschaft für Reaktorsicherheit. Köln 1979. 
289) Ohori, H.; 
290) Osgood, C.; 
291) Okrent, 0.: 
292) Okrent, 0.: 
Vamada, T.: Public Acceptance of Nuclear Power Development in 
Japan. IAEA-CN-36/158. Vortrag auf der International 
Conference on Nuclear Power and Its Fuel Cycle. Salzburg, 
2.-13. Mai 1977. 
Tannenbaum, P.H.: The Principle of Congruity in the Predic-
tion of Attitude Change. In : Psycho10gical Review, V01.52, 
1955, Nr.3, S.42-55. 
Risk-Benefit Methodology and Application. Los Angeles 1975 
A General Evaluation Approach to Risk-Benefit for Large 
Technological Systems and its Application to Nuclear 
Power. Los Angeles 1977. 
293) Otway, H.J . : Risk Estimation and Evaluation . In : Proceedings of the 
IIASA Planning Conference on Energy Systems. 
International Institute for Applied Systems Analysis. 
IIASA-PC-3. Laxenburg, österreich 1973. 
294) Otway, H.J. : 
295) Otway, H.J. 
Risk Assessment and Societa1 Choices.International Institute 
for Applied System Analysis . RM-75-2. Laxenburg 1975. 
u a • Social Values in Risk Acceptance . International Institute 
• f~r Applied Systems Analysis. IIASA RM-75-54. Laxenburg 1975. 
296) Otway, H.J. 
297) Otway, H.J . : 
298) Otway, H.J.: 
299) Otway, H.J., 
300) Otway, H.J.; 
-24-
u. a. : Advoidance Response to the Risk Environment. A Cross 
Cultural Comparison . IIASA RM-75-14. International In-
stitute for Applied Systems Analysis . Laxenburg 1975. 
A Review of Research on the Identification of Factors 
Influencing the Social Response to Technolooical Risks . 
·IAEA-CN-3614 . Vortrag auf der International Conference 
on Nuclear Power and Its Fuel Cycle. Salzburg, 2.-13.Mai 
1977. 
The Perception of Technological Risks: A PS'ycholo~ical 
Perspective. Vortrag auf der International Conference 
of the European Community on Technological Risks . 
Berlin, 1.-3. April 1979. 
COhen, J.J.: Revealed Preferences. Comments on the Starr 
Benefit-Risk Relationship . International Institute for 
Applied Systems Analysis. IIASA-RM 75-5. Laxenburg 1975. 
Pahner, P.O . , Niehaus, F.: Risk Assessment. In: Second 
Status Report of the IIASA Project on Energy Systems. 
International Institute for Applied Systems Analysis. 
RR-76-1. Laxenburg 1976 . 
301) Otway, H.J., Pahner, P.O.: Risk Assessment . In : Futures, 8, 1976, 
S. 122-134. 
302) Otway, H.J., Fishbein, M. : The Oeterminants of Attitude Formation. An 
Application to Nucle!r Power. RM-76-So. International 
Institute for Applied Systems Analysis . Laxenburg 1976. 
303) Otway, H.J.; Maurer, 0.; Thomas, K.: Nuclear Power. The Question of 
Public Acceptance. In: Futures, April 1978, S.109ff. 
304) Pahner, P.O.: The Psychological Oisplacement of Anxiety . An Application 
to Nuclear Energ.v. In: Risk-8enefit Methodoloqy and 
Application, hrg . von O.Okrent, Los Angeles 1975. 
305) Pahner, P.O.: A Psychological Perspective of the Nuclear Ener~y 
Controversy. IIASA-Research Memorandum RM-76-6i. Inter-
national Institut for Applied Systems Analysis . Laxen-
bur9 1976. 
306) Papp, R. u.a.: A New Concept in Risk AnalYSis for Nuclear Facilities. 
In : Nuclear News, 14, 1974, S.62-65. 
-25-
307) Paschen, H.: Konsequenzen des großtechnischen Einsatzes der Kernener-
gie in der Bundesrepublik Deutschland. Teil V: Die Ak-
zeptanzprob1ematik der Kernenergie, bearbeitet von 
G. Friedrichs und M. Loeben. Kernforschunqszentrum 
Kar1sruhe ASA/ZE-14/78. Kar1sruhe 1979. . 
308) Patermann,. 0.: Das antinukleare Vo1ksbegehren in Ka1ifornien. Ein neuer 
Abschnitt der nuklearen Kontroverse in den USA. In: Atom-
wirtschaft/Atomtechnik. Mai 1976 21, 5, S.237-238. 
309) Pearce, D.W.: Decision Making for Ener9Y Futures. London 1979. 
310) Pearce, D.W.: The Preconditions for Achievin9 Consensus in the Context 
of Techno109ica1 Risk: Executive Summary. Vortrag auf 
311) Pesch, J.: 
312) Pelicier, Y. 
313) Penczynski, 
314) Peters, W.: 
der Conference of the European Commission on Techno10gi-
ca1 Risk. Ber1in, 1.-3. April 1979. 
Staatliche Forschungs- und Entwicklungspolitik im Span-
nungsfe1d zwis·chen Reqierung, Parlament und privaten Ex-
perten, untersucht am Beispiel der deutschen Atompolitik. 
Oiss .• Freiburg 1975. 
u.a.: Expos~ des recherches de l'equipe des psycho19ie. 
In: Colloque sur les imp1ications psycho-socio10giques 
du dev~loppement de 1 industrie nuc1eaire. Paris 13-15. 
Januar 1977, S.208-222. 
P.: Which Energy Strategy is Viab1e1 Invited Paper for the 
European Nuc1ear Conference and Foratom VII Congress. 
Hamburg, 6.-11. Mai 1975. 
Befragung in Bib1is. Haben Sie Angst vor "Ihrem" Kernkraft-
werk. In: Bild der Wissensch.ft, Heft 11, Jhrg.16, Novem-
ber 1979, S.96ff . 
315) Pickshaus, K.: Das Beispiel Wyh1 - Der Kampf gegen den Bau von umwelt-
gefährdenden Kernkraftwerken. In: Marxistische Blätter, 
Band 3, 1975, S.71ff. 
316) Picht, G.: Oie Wertordnung einer humanen Umwelt. In: Alternative 
Möglichkeiten für die Ener9iepo1itik. Materialien zum 
Gutachten. Band 1, hrg. von W. Lienemann u.a •• Heide1-
berg 1977. 
317) Pochin, E.E.: 
318) Pochin, E.E.: 
319} Popp, M. : 
320} Prüß, K.: 
321} Albrecht, P. 
322} Pyron, 8.: 
-26-
The Acceptance of Risks . In: 8ritisch Medical 8ulletin, 
Nr.3, 1975, S.184-190. 
Importance relat i ve des differents r isques . In: Colloque 
sur les implicatioDs psycho-sociologiques du d~veloppement 
de 1 'industrie nucleaire. Paris 13-15 . Januar 1977, 
S.44-55. 
The Public Discussion About the Peaceful Utilization 
of Nuclear Energy in the federal Republic of Germany 
and the Information and Discussion Campaign of the fede-
ral Government. IAEA-CN-36/81. Vortrag auf der International 
Conference on Nuclear Power and Its fuel Cycle. Salzburg, 
2.-13. Mai 1977. 
Kernforschungspolitik in der Bundesrepublik Deutschland. 
frankfurt 1974. 
u.a.: Public Acceptance of Nuclear Power - Same Ethical Issues, 
hrg. von der Ener9Y Advfsory Group of the Workin9 Commi ttee 
on Church and Society. IAEA-CN-36/383 . Vortrag auf der Inter-
national Conference on Nuclear Power and Its fuel Cycle. 
Salzburg, 2.-13. Mai 1977. 
form and Diversity in Human Habitats . Jud9emental and 
Attitude Responses. In: Environment and Behaviour, Nr.3, 
März 1972, S.87ff. 
323} Rammstedt, 0;: Krisenverhalten. In: BUrgerbeteiligung und BUrgerinitia-
tiven, hrg . von H. Matthöfer. Argumente in der Energie-
diskussion, Bd .3. Vil1ingen 1977, S.40-65. 
324} Rammstedt, 0.: Soziale Bewegung - Modell und forschungsperspektiven. 
Ebenda, S.447-475 . 
325} Rammstedt, 0.: Soziale Bewegung . frankfurt 1972. 
326} Rankin, W.L.; Nea1ey, S.M. : Public Attitudes about Nuclear Waste Manage-
ment. Batte11e Memorial Institute . Human Affairs Research 
Centre. Seatt1e 1978. 
327} Ravetz, J.R. : Criticisms of Science. In: SCience, Techno10gy _nd Society, 
hrg. von I.Spiege1-Rösin9 und D. de Soll_ Price. London/ 
Bever1y Hi11s 1972, S.71ff. 
328} Ravetz, J.R.: Pub1ic Perceptions of Acceptab1e Risks - As Evidence for 
Their Cognitive, Technic_l and Socia1 Structure. Vortrag 
auf der: Conference of the European Commission on Tech-
nological Risk . Berlin, 1.-3. April 1979. 
-27-
329) Reactor Safety Study: An Assessment of Accident Risks in U.S. Commercia1 
Nuc1ear Power P1ants. WASH-1400, NUREG-74/014. Washington, 
D.C. Nuc1ear Re9u1atory Commission. Washington 1975. 
330) Renn, H. : 
331) Renn, 0. : 
332) Renn, 0.: 
333) Renn, 0.: 
Nichtparametrische Statistik. Studienskripten Zur Sozio-
logie, Band 4. Stuttgart 1975. 
Kernenergie aus der Sicht der Bevölkerung. KFA-AKI-IB-I/77. 
Jüli eh 1977. 
Kernenergie und ihre Aufnahme durch die Bevölkerung. 
In: Energie für morgen - Planung von heute; hrg. von 
D. GÖrgmaier. Bayerische Landeszentrale für Politische 
Bildung, München 1978, S.105ff. 
Sozialpsychologische und soziologische Erklärungsversuche 
zur Ris ikoakzeptanz . In: Energie, Zeitschrift für prakti-
sche Energietechnik, Hefte 6, 7, 8/9 Juni bis September 79, 
5.161-166, 195-197, 271-274 . 
334) Rogers, E.M. : The Diffus i on of Innovations . New York 1963. 
335) Rög1in, H.Ch.: 
336) Rög1in, H.Ch.: 
337) Rög1in, H.Ch . : 
338) Rög1in, H.Ch . : 
339) Röth1ein, B.: 
340) ROkeach, M.: 
Sozialpsychologie der Kernenergie-Kontroverse. Kurzfassung 
des Vortrages in der Informationszentrale der Elektrizi-
tätswirtschaft e.V. Frankfurt, 29.11.1976. 
Das aktuelle Interview. "Sind Atomkraftwerke sicherer 
als Kernkraftwerke?" Umschau in Wissenschaft und Technik, 
77, 1977, 2, S.3B-40. 
Sozialpsychologische Aspekte der Kernenergie. In : Kern-
energie und öffentlichkeit, hrg . von W.D . Müller und 
R. Hossner. Nr .9 der atw .. Broschüren "Kernenergie und 
Umwelt". Düsse1dorf 1977, 5.10-27. 
Kommunikation und Audovision. In :Audiovision, 8. Jahrgang 
Nr.12, Dezember 1977, 5.58-65. 
Kernenergie - ein Thema der öffentlichen Meinung Disser-
tation. Universität München. München 1979. 
The Nature of Attitudes . In: Encyc10pedia of the Socia1 
Sciences . Band I, New York 1968, 5.449-458. 
-28-
341) Rakeach, M.: Be1iefs, Attitudes and Va1ues. San Francisco 1969 
342) Rosenberg, 
343) Rosenber9, 
J.J., Hov1and, C.J.: Attitude Drqanization and Change. 
An Analysis of Consistency Among Attitude Components. 
New Haven 1960 (S.1-64). 
M.J.: C09nitive Structure and Attitudina1 Affect. In: Journal 
of Abnormal and Socia1 Psycho1ogy, Nr.53, 1956, S.367-372. 
344) Rönsch, H.D.: Reaktionen auf staatliches Handeln am Beispiel des Wah1ver-
ha1tens. In: 8ürgerbetei1igun9 und Bürgerinitiativen, hrg. 
von H. Matthöfer. Vi11ingen 1977, S.344-395. 
345) Rosi, E.J.: 
346) Ross, L.D.: 
347) Roszak, Th.: 
348) Roth, W.: 
349) Rowe, W.D.: 
Mars and Attentive Opinion on Nuc1ear Weapons Test and 
Fallout 1954-1963. In: Pub1ic Opinion Quarter1y, Nr.2 
1965, S.280-297. 
The Intuitive Psycho109iSt and His Shortcomings: Distor-
tions in the Attribution Process. In: Advances in 
Experimental Socia1 Psycho10gy, hrg. von L. 8erkowitz, 
Vo1. 10 . New York 1977, S.173-220. 
Gegenkultur. Gedanken über die technokratische Gesell-
schaft und die Opposition der Jugend. Düsse1dorf, Wien 
1971. 
Mit den Grünen nIcht ins Bett. Warum die Sozialdemokraten 
die A1ternativ1er am Einzug in den Bundesta9 hindern wol-
len? In: Die Zeit, Heft 49, 30.11.1979. 
An Anatomy of Risk. New York, London, Sydney, Toronto 1977 
350) Rowe, W.D.: Cost-Effectiveness Determination by Structured Va1ue Analy-
sis. In: WP-Review 1, 1980, Me Lean, virginia. The MITRE-
Corporation, Februar 1969. 
351) Rud1off, W.: "Kernener9i e und Offent 1 i chkeit - Nur ein I nformat i onspro-
b1em?" Von der Information zur Komnunikation mit dem Bür-
ger. Referat auf der Jahrestagung der Deutschen Pub1ic 
Relations Gesellschaft, 27. Mai 1976. 
352) Sachsse, H.: Anthropo109ie der Technik. Braunschweig 1978. 
353) Sagan, L.A.: The Human Costs of Nuc1ear Power. In: Science, Vo1.177, 
Nr. 11, August 1972, S.487-493. 
-29-
354) Sahner, H. : Schließende Statistik. Studienskripte Zur Soziologie, 
Band 2. Stuttgart .1971. 
355) Sandstr~m, S. : The Swedish Nuc1ear Debate . Vortrag auf dem AIF-Workshop: 
Nuc1ear Information Strategies : New Settings, New 
Approaches. Las Vegas, 26-29. September 1976. 
356) Santmire, H.P.: Hi storica1 Dimension of the American Crisis. In : Western 
Man and Environmenta1 Ethics, hrg. von J .G. Barbour. 
London, Don Mi11s, Ontario 1973 . 
357) Seetzen, J. : 
358) Seetzen, J . : 
359) Seitz, W.: 
Auf der Suche nach einem neuen Lebensstil. Reaktion auf 
die Probleme der Großtechnologie. Vortrag auf der RWE-
Veranstaltung "Energie- und Umweltfragen". Mainz, 16.-
19.4.1978. 
Entwicklung zu einer allgemeinen Systemtechnik - Die darin 
enthaltenen Probleme des Menschen. In: Kybernetik und 
Theologie. Sonderheft der Zeitschrift für Pastoraltheolo-
gie, Band 56 . Köln 1967, S.160-170. 
Soziale Einstellungen und ihre objekt-spezifische Differen-
zierung. Dissertation an der Universität Würzburg. Würz-
burg 1973. 
360) Se1vidge, J. : A Three-step Procedure for Assigning Probabi1ities to 
Rare Events . In: Uti1ity, SUbjective Probability and 
Human Oecision Making, hrg. von D. Wendt u. C. V1ek. 
Dordrecht (NL), Boston (USA) 1975 . 
361) Shaw M E • Wr,·ght J.M·. Sca1es for the Measurement of Attitudes. New , . . , , 
York 1967 . 
362) Slierfie1d, L.: The sociological consequences of nuclear energy. 
nomic and Social Consequences of Nuclear Energy. 
von L. Sherfie1d. Oxford 1972. 
In: Eco-
hrg. 
363) Silbermann, C.E. : The Myth of Automation . London, Evanston, New York 
1966. 
364) Slovic, P.; Lichtenstein 5.: Relative Importance of Probabi1ities Payoffs i~ Risk-Taking. In: Journal of Experimental 
Psycho10gy. Vo1.78, Nr.3, November 1968, 5.1-18. 
and 
365) Slovic, P., Kunreuther, H., White, G.F.: Dec1sion processes, rationa11ty. 
and adjustment to natural hazards. In: Natural hazards. 
Loca1, national, global, hr9. von G.F.White . New York 
1974, 5.187-205. 
366) 510vic, P., 
-30-
hhoff, 8. , Lichtenstein, 5.: Cognitive processes and 
societal risk taking. In: Cognition and 50cial 8ehavior, 
hrg. von J.5.Carroll and J.W. Payne. New Vork 1976, 
5.165-184 . 
367) 5lovic, P. ;Fischhoff, B., Lichtenstein, S. : Behavioral Oerision Theory. 
In : Annual Review of Psychology. Bd. 28, 1977, S.1-39. 
368) Slovic, P., 
369) Slovic, P.: 
Lichtenstein, S. ;Fischhoff, B.: Images of Disaster: 
Perception and Acceptance of Risks from Nuclear Power. 
In: Impacts and Risks of Energy Strategies. Their Analy-
sis and Role in Management. Proceedings of the Confe-
rence in Stockholm, 26-25. Sept. 1978. In Druck. 
The Psychology of Protective Behavior. In: Journal of 
Safety Research, Vol.10, Nr. 2, 1978, S.58-68. 
370) Slovic, P.;Fischhoff, 8.: How safe is safe enough? Determinants of 
Percieved and Acceptable Risk. In: Too Hot to Handle: 
Social and Policy Issues in the Management of Radioactive 
Wastes, hrg. von L. Gould and C.A. Walker . Vale Universi-
ty Press. In Vorbereitung (Manuskript: Januar 1979). 
371) Slovic, P.; Lichtenstein 5. ;Fischhoff, B.: Rating the Risks. In: Environ-
ment, Vol. 21, Nr.3, April 1979, S.14ff. 
372) Slovic, P.;Fischhoff, B.; Lichtenstein, S.: Perceived Risk. Unveröffent-
lichtes Vortragsmanuskript. Revised Version vom Dezember 
1979. Eugene (Oregon) 1979. 
373) Smidt, 0.: Die Grenzen von Naturwissenschaft und Technik. In : 
Atomwirtschaft, Atomtechnik, Jahrgang 25, 8d.2, Februar 
1970, 5.70-73. 
u.a.: The 5tructure of Energy Attitudes and Beliefs. An 
Interview Study. Stock holm 1978. 
375) Spada, H.; Hoffmann, L.; Lucht-Wraage, H.: Student Attitudes Toward 
Nuclear Power Plants. A Classroom Experiment in the Field 
of Environmental Psychology. In: Studies in Educational 
Evaluation . Vol.3, Nr.2, 1977, S.109-128. 
376) Spiegel-Rösing, I.: The Study of Science. Technology and Society. 
Recent Trends and Future chall.nges . In: Science, Tech-
nol09Y and Society, hrg. von I. Spiegel-Rösing und 
D. de Solla Price. London/Beverly Hills 1972. 
-31-
377) Suchman, E.A . : Preventive Hea1th Behavior: A Model for Research on 
Community Hea1th Campaigns. In: Journal of Health and 
Social Behaviour Nr.B, 1967, S.197-209. 
37B) Summers, 0.1 AShworth, C. : Types of Conflict in Nuc1ear Power Plant 
Oisputs. Battelle Research Center . Unveröffentlichtes 
Manuskript. Seatt1e (Washin9ton) 1976. 
379) Swaton, E. u.a.: The Oeterminants of Risk Perception: The Active-Passive 
Dimension. International Institute for Applied Systems 
Analysis. RM-76-74. Laxenburg 1976. 
3BO) Schäfer, R.E.: What are we talking ab out when we talk about "risk"? 
A Critical Survey of Risk and Risk Preference Theories. 
International Institute for Applied Systems Analysis . 
RM-7B-69. Laxenbur9 (österreich) 1978. 
381) Schäfer, R.E.; Swaton, E.; Niehaus, F.: Measuring Attitudes Towards 
the Use of Nuc1ear Power : An Analysis of a Measuring 
Instrument . WP-81-XX, International Institute for Applied 
Systems Analysis. Laxenburg (österreich) 1981. 
382) Scharioth, J . ; Krebsbach, C.: Ergebnisse sozialwissenschaftlicher For-
schung zur Energiediskussion. Arbeitsunterlage für die 
Sitzung des Bund/Länder-Arbeitskreises in Bonn. 17. Mai 
1977 • 
383) SCharioth, J . : Die nukleare Kontroverse aus gesellschaftlicher und psycho-
logischer Sicht . In: Atomwirtschaft-Atomtechnik, 22, 1977, 
S.338-343. 
384) Schelsky, H.: 
385) Scheueh, E.K.: 
387) Scheueh, E.K.: 
3B8) Scheueh, E.K.; 
Der Mensch in der wissenschaftlichen Zivilisation. Köln/ 
Op1aden 1961. 
Auswahlverfahren in der Sozialforschung. In: Handbuch der 
empirischen Sozialforschung, hrg. von R. König, 3. Aufla-
ge . Band 1. Stuttgart 1973. 
Zehnpfennig, H. : Skalierun9sverfahren in der Sozialforschung. 
In: Handbuch der empirischen Sozialforschung, hrg . von 
R. König. 3. Auflage. Band 3a, Stuttgart 1973, S.97-203. 
Daheim J.P.: Soziale Schichtung. In: Sonderheft 5 der 
Kölner'Zeitschrift für Soziologie und Sozialpsychologie, 
hrg . von D.V. G1ass und R. Köni9, 4. Auflage. Köln/Op1a-
den 1970. 
-32-
389) Scheueh, E.K.: Kulturintelligenz als Machtfaktor: Intellektuelle zwischen 
Geist und Politik. Zürich 1974. 
390) Scheueh, E.K . : 
391) Scheueh, E.K.: 
Bedarf an zukünftiger Energie. Brief an das 8undesministeri_ 
um für Forschung und Technologie, z.Hd. Herrn Däunert, 
14. Januar 1975. 
Bürgerinitiativen: Haben die Parteien versagt? Fernseh-
interview in Fragen zur Zeit. ZDF-Sendung vom 27.03.1977, 
13,00 Uhr, Manuskript Nr. 6493/0291. 
392) Schild, R.A.: Bombe und Kraftwerk. Ludwigshafen 1974. 
393) Schiraski, W.: Möglichkeiten und Grenzen vergleichender Mode11untersu-
chungen am Beispiel der Umweltbelastung durch Kernkraft-
werke . In : Technikfo1gen-Technik-Abschätzung, hrg. von 
H. Haas. München, Wien 1975 . 
394) Schmidt, H.D. u.a.: Soziale Einstellungen. München 1975 
395) SChmidtner, F.X. : Oas Kernkraftwerk. In : Energie für morgen - Planung 
von heute, hrg. von D. Görgmaier, Bayerische Landeszen-
tra1e für Politische Bildung. München 1978. 
396) Schoeck, H. : Oekospasmus oder die An9st vor Kernkraftwerken. In : Umwelt, 
Bd.2, 1976, S.125-128. 
397) SChoeck, H.: 
398) Schrenk, M.: 
399) Schrenk, M.: 
Oie Entwick1un9 des Umweltbewußtseins - Konfliktstoff in 
einer dynamischen Wirtschaft. In : Datscope 23 . 1977, S.4-13. 
Sachlichkeit - Emotionalität - Identität. In: Kernenergie-
Mensch-Umwelt, hrg. von K. öser u.a •• Köln 1976, S.99ff . 
Solidarität, Dialog und Dialekt in privaten und öffentli-
chen Konflikten. In: Unabhängiger Biologischer Nachrichten-
dienst . Ausgabe 11, 1978, S.7ff. 
400) SchUller, .R., Fow1er, J .R.: Socia1 Consequences of Power Generation. 
Research Difficu1ties and Opportunities. Lecture on the 
Annua1 Meeting of the Society for General Systems Research . 
New York, 27-30 Jan.1975 . 
401) Schumacher, E.F.: Die Rückkehr zum menschlichen Maß . Hambur9 1977. 
402) Schumacher, H.G.: Legitimation und Partizipation von Bürgerinitiativen . 
Umweltschutz in der Energiediskussion - Möglichkeiten 
und Grenzen . In : Bürgerbetei1i9ung und Bürgerinitiativen, 
hrg . von H. Matthöfer. Vi11ingen 1977, 
-33-
403) Schuster, G.: Die Kontroverse über die Einführung der Atomenergie. 
In: Das Leben, Nr.5, 6. Wiesbaden 1971. 
404) Schuster, G.: Technological Risk: Perception and Handling in the 
EEC. Introductory Remarks. International Conference of 
the European Committee on Technological Risk. Berlin, 
1-3. April 1979. 
405) Stannard, J.N.: Evaluation of Health Hazards to the Public Associated 
with Nuclear Power Plant Operations. In: Nuclear Power 
and the Public, hrg. von H. Foreman. Minneapolis 1970. 
406) Starr, Ch.: Social Benefit Versus Technological Risk. In: Science, 
Vol.165, Nr.169, September 1969, S.1232-123B. 
407) Starr, Ch.: Benefit-Cost Relationships to Socio-Technical Systems. 
In: Environmental Aspects of Nuclear Power Stations. 
IAEA SM-146/47. International Atomic Energy Agency. Wien 
1971, S.900ff. 
40B) Starr, Ch.: General Philosophy of Risk-Benefit Analysis. Vortrag im 
Electric Power Research Institute EPRI, Stanford, IES 
Seminar. Stanford (California), 30. September 1974. 
409) Starr, Ch.: Risk and Risk Acceptance by Society. In: Technische 
Mitteilungen, 70. Jahrgang, Heft 6/7, Juni/Juli 1977. 
S.444-450. 
411) Stazesky, F.M.: The Public Acceptance Issue. Lecture at the Annual 
Conference, Industrial Forum. San Francisco, 16-20. No-
vember 1975. 
412) Steger, U.: Die soziale Verträglichkeit von Energietechnolgien. In: 
Die neue Gesellschaft, Bd.25, 197B, S.992-993. 
413) Steinbuch, K.: Mensch, Technik, Zukunft. Reinbek 1973. 
414) Sternstein, 
415) Der Störfall 
W.: Bürgerinitiativen gegen Atomanlagen. Aus 
Aktionsforschers. Frankfurt/Main 1978. 
der Arbeit eines 
von Harrisburg. Offizieller Bericht der von Präsident Carter 
eingesetzten Kommission über den Reaktorunfall auf Three 
Mile Island. Oüsseldorf 1979. 
416) Strümpel, B.: Wirtschaftpolitische Strategien zur Senkung des Energie-
verbrauchs. In: Energfeversorgung und Lebensqualität, 
hrg. von V. Hauff . Argumente in der Energiediskussion, 
Bd.6. Vfllfngen 1978, S.155ff. 
-34-
417) Strümpel, B.: Oie Krise des Wohlstandes. Stuttgart 1972. 
418) Taviss, I.: A Survey of Popular Attitudes Toward TechnoloQY. In: Technology and Culture, Bd.13, 1072, 5.606-621. 
419) Tenbruck, 
420) Thomas, K. 
F.H.: Deutsche SozioloQie im internationalen Kontext. Ihre 
Ideengeschichte und ihr Gesellschaftsbezu~ . In: D~utsch~ 
Soziologie seit 1945. Sonderheft 21 der Kolner Zeltsehr1ft 
für Soziologie und Sozialpsychologie. Köln 1979, S.71ff . 
u.a.: Nuclear Energy: The Accuracy of Policy Makers Percep-
tions of Public Beliefs. International Institute for 
Applied Systems Analysis. RM-80-18. Laxenburg (österreich) 
1979. 
421) Thomas, K. u.a.: A Comparative Study of Public Beliefs about Five Energy 
Systems. International Institute for Applied Systems 
Analysis. RM-80-15 . Laxenburg (österreich) 1979. 
422) Tognacci, L.H. u.a.: Environmental Quality: How Universal is Public 
Concern? In: Environment and Behaviour. Hr.3, März 1972, 
S.73ff. 
423) Touraine, A. : Workers Attitudes to Technical Change. Paris 1965. 
424) Tschledel, R.: Zur Theorle der Orientierung von Wissenschaft. Disser-
tation an der Philosophischen Fakultät der Westfälischen 
Wilhelms-Universität Münster. 1977. 
425) Traube, K.: Müssen wir umschalten? Von den politischen Grenzen der 
Technik. Reinbek 1978. 
426) Traube, K.: Wachstum oder Askese? Kritik der Industrialisierung von 
Bedürfnissen. Reinbek (bei Hamburg) 1979 . 
427) Triandis, H.C.: Toward an Analysis of the Components .of Interpersonal 
Attitudes. In: Attitude, Ego-Involvement and Change. 
New York 1967, 5.227-270. 
428) Tubiana, M.: One Approach to the Study of Public Acceptance. In: 
Directions in Energy Pollcy. A Comprehensive Approach 
to Energy Resource Decislon-Making, hrg. von B. Kursunoglu 
und A. Perlmutter, 1980, 5.343-356. 
429) Turner, R.H.: The Public Perception of Protest. American Socl010glcal 
Review 1969, 35, 5.815-831. 
-35-
430) Tversky, A., Kahneman, 0.: The bellef ln the "law of sman numbers". 
In: Psychologlcal Bulletln Nr.76, 1971, 5.105-110. 
431) Tversky, A., Kahneman, 0.: Avallabll1ty: A heurlstic for judging 
frequency and probability. In: Cognitive Psychology 
Nr.4, 1973, 5.207-232. 
432) Tversky, A., Kahneman, 0.: Judgement under Uncertainty. Heuristics 
and Biases. In: Utility, Probability and Human Decision 
Making, hrg. von O. Wendt und C. Vlek. Dordrecht (Hol-
land), Boston (USA) 1975, 5.141-162. 
433) Tversky, A.: Kahneman, 0.: Judgement under Uncertalnty: Heuristics 
and Biases. In: Science, Band 1B5, 1974, 5.1124-1131. 
434) Ullrich, Otto: Technlk und Herrschaft. Frankfurt/Main 1977. 
435) Unruh, G.C. 
436) Verba, 5. i 
von: Wie verhält sich die Bürgerinitiative zur Repräsenta-
tion. In: Kontrovers-Bürgerinitiativen, hrg. von der 
Bundeszentrale für politlsche Bildung NRW, bearbeitet 
von U. Kempf. Bonn 1974, 5.74. 
Nle, N.H. : Particlpatlon ln American Polltical 
Oemocracy and Soclal Equallty. New Vork 1972. 
437) Das Veto, Der Atombericht der Ford-Foundatlon (Orlglnal von Keeny 
Spurgeon als Herausgeber). Frankfurt 1977. 
und 
438) Vllmar, F.: 
439) Vllmar, F.: 
340) Vlnck, W.: 
Elemente einer lntegralen Theorie der Partizpation. In: 
Bürgerbeteillgung und Bürgerinitiative, hrg. von H.Matt-
höfer. Argumente in der Energiediskussion, Bd. 3, 
Vl111ngen 1977, 5.294-315. 
Strategien der Demokratisierung. Neuwied 1973. 
Practices and Rules for Nuclear Power Stations; The Role 
of the Risk Concept in Assessinq Acceptability. Vortrag 
auf der International Conference of the European 
Committee on Technoloqical Risk. Berlin, 1-3. April 1979. 
441) von Erlchsen, L. : Friedliche Nutzunq der Kernenerqie. Ihre Vorteile und 
Gefahren. Berlin, Göttinqen, Heidelberq 1962. 
442) Vroom, V.H.: Work and Motivation. New Vork 1964. 
443) Wagner, H.J . : 
-36-
Risikoanalyse ausgewählter Energieversorgungstechno10gien. 
Interner Bericht KFA-STE-IB 2/78. Kernforschungsanlage 
JÜ11ch. Jü1ich 1978. 
444) Weinberg, A.M.: Science and Trans-Science. 
1972, S.1342ff. 
In: Minerva, Vol .10, London 
445) Weinberg, A.M: 
446) Weingart, P.: 
Social Institutions and Nuc1ear Energy. 
179, Nr.4043, 7. Juli 1972, 5.27-34. 
In: Science, Vo1. 
Das "Harrlsburg-Syndrom" oder die Oe-Professionalisierung 
der Experten. In: Kernenergie: Gefahr oder Notwendigkeit, 
von H. Nowotny. Frankfurt/M. 1979, S.9-17. 
447) Weizsäcker, C.F. von: Wege in der Gefahr. Eine Studie über Wirtschaft, 
Gesellschaft und Kriegsverhütung. München 1976. 
448) Weizsäcker, C.F. von: Eine Schlacht der Moral . Plädoyer fUr ein besse-
res Verständnis beim Pro und Contra um die Kernenergie. 
In: Die Zeit, 15.06.1979. 
449) Whltney, H.: Some Socio10gical Consequences of Atomic Power. Anna1s 
of the American Academy of Polltlca1 and Socla1 Science. 
Philadelphia 1953. 
450) WHO-Report: Mensch und Atom - Bericht einer Studiengruppe der Welt-
gesundheitsorganisation. WHO-Report Nr. 13. München 1960. 
451) Wicker, A.W.: Attitude versus Actlons: The Relationship of Verbal and 
Overt Behavioral Responses to Attitude Objects. In: 
Journal of Socla1 Issues, Band 25, 1969, S.41-78. 
452) Wiehn, E.R.: Industrielle Bedürfnisse und soziale Werte. TUblngen 1972. 
453) Wlehn, E.R.: Theorien der sozialen Schichtung. In: Studien der Soziolo-
gie, hrg . von R. Oahrendorf, Band 9. München 1969. 
454) Wiesner, L.: Minimierung der Summe aller Risiken. In: AtomwIrtschaft-
Atomtechnik, Band 3, 1979, 5.145-150. 
455) Wil1i, V.J.: Grundlagen einer empirischen Soziologie der Werte 
und Wertsysteme. Zürich 1966. 
456) Wi1kinson, P.: Soziale Bewegungen. München 1974. 
-37-
457) Wirtz, Karl: Die Errichtung großer Kernkraftwerke in oder in der Nähe 
von Bevölkerungszentren. Gutachten im Auftrag . des Hessi-
sehen Ministeriums für Wirtschaft und Verkehr. Karlsruhe 
1968. 
458) Wiswede, G.: Rationalität und soziales Wertesystem. In: Zeitschrift 
für Wirtschafts- und Sozialwissenschaften, Nr.2, 1972. 
459) Wiswede, G.; Kutsch, Th.: Sozialer Wandel. Zur Erklärun~skraft neuerer 
EntwicklunQs- und Modernisierun9stheorien. Darmstadt 1978. 
460) Wünschmann, A.: Unbewußt dagegen - Zur Psychol09ie der Kernenergie-
kontroverse. Bonn aktuell. Stuttgart 1980. 
461) WUstenhagen, H.-H.: Bürger gegen Kernkraftwerke, Wh.vl _ der Anfang? 
Reinbek 1975. 
462) WUstenhagen, 
463) Wyer, R.S.: 
461) Wyer, R.S.: 
465) Zangger, C.: 
466) Zajonc, R.B.: 
H.-H.: Bürgerinitiativen, Atomenergie und Wissenschaft. 
In: Sozialisierun~ der Verluste, hrg. von K.W. Kapp und 
F. Vilmar. München 1972, S.8ff. 
An Investigation of the Relations Ameng Probability Esti. 
mates. In: Organizational Behavior and Human Performance. 
Nummer 15, 1976, S.1-18. 
The Prediction of Evaluations of Social Role Occupants 
as a Function of the Favourableness. Relevance and Proba-
bility with Attributes of these Occupants. In: Sociametry, 
.Nr. 33, 1970, S.79-96. 
Interaction entre controle et acceptation de l'energie 
nucleaire en Suisse-objectifs et mise en oeuvre. IAEA-CN-36/579. 
Vortrag auf der International Conference on Nuclear Ener9Y 
and lts Fuel Cycle. Salzbur9, 2.-13. Mai 1977. 
Cognitive Theories in Social Psychology. In: The Handbook 
of Social Psychology, hrg. von G. Lindzey und E. Aronson. 
Reading (Mass.) 1968. 
467) Zebraski, E.L.:Attainment of Balance in Risk-Benefit-Perceptions. In: 
Risk-Benefit-Methodology and Application. En~ineering 
Foundation Workshop. Asilomar (Californien), Report ENG-
759B, Dezember 1975, S.633-644. 
-38-
468) Zeckhauser, R.: Risk Spreading and Distribution. In: The Political 
Economy of Income Distribution, hrg. von H. Hochmann 
und W.E. Peterson, New York 1973, 5.206-228. 
469) Zehnder, R.: 
470) Zetterberg, 
Historical 8ackground of Societal Resistance to Tech-
nol09Y. Research-Memorandum des International Insti-
tue for Applied Systems Analysis, Vorabdruck vom 
28.12.1977. Laxenburg 1977. 
H.: Environmental Awareness and Political Change in 
Sweden. Vortragsmanuskript zur Konferenz on Environ-
mental Awareness and Politcal Change. Wissenschafts-
zentrum Berlin, 1978. 
471) Zilleßen, H. : Bürgerinitiativen im repräsentativen Regierungssystem. 
Aus Politik und Zeitgeschehen. Beilage der Wochenzei-
tung "Das Parlament" Nr. 12, 23 .03.1974 . 
472) Zilleßen, H. : Energiepolitik - Dialog mit dem Bürger? In: Aus Politik 
und Zeitgeschichte. Beilage zur Wochenzeitung: Das Par-
lament Nr. B, 27, 5.12-19. 
473) Zilleßen, H. : Energiepolitik und Bürgerinitiative. In: Kernenergie-
Mensch-Umwelt, hrg . von K. Deser u.a., Köln 1976. 
474) Zilleßen, H. : Bürgerinitiativen und repräsentative Demokratie. In: 
Bürgerinitiativen und repräsentatives System, hrg. von 
B. Guggenberger und U. Kempf. Opladen 1?78, 5.112-132 • 
• 
475) Zimmermann, E. : Das Experiment in den Sozialwissenschaften . Studien-
skripten zur Soziologie. Stuttgart 1972. 
476) Zischka, A.: Kampf ums Oberleben . Das Menschenrecht auf Energie . 
Oüsseldorf/Wien 1979. 
477) Zur friedlichen Nutzung der Kernenergie . Eine Dokumentation der Bundes-
regierung. 2. Auflage, hrg. vom Bundesminister für 
Forschung und Technologie. Bonn 1978. 
- 39 -
Ergänzung zur literaturliste 
478) Abplanalp. P.A . ; Hettlage. R.: Wirtschaftspolitjk und 
Kommunikative Umwelt in der "Neuen 
Politischen Okonomie • - Einige sozio-
logische Bedenken. In: Jahrbuch für 
Sozialwissenschaft. Bd 3 • 1979. S.293 
bis 311 
479)Barrados.M. Impact on Public Opinion of the Three 
Mile Accident.Vortrag auf der ENC-FOR-
ATOM VII Conference.Hamburg 6.-11. Mai 
1979. 
480) Bettman. J.R. Perceived Risk and Its Components : A Mo-
del and Empirical Test . In : Journal of 
Marketing Research. Nr.10. 1973. S. 184 
bis 190 
481) 81ack.S.; Niehaus, F. ; Simpson ,0 .: How safe is IItoO" 
safe? Report WP-79-68. International In-
stitute for Applied Systems Analysis. 
laxenburg (Osterreich) 1979 
482) Bledjian. F.; Stosberg.Ch.: Analyse der Massenkommunika-
tion : Wirkungen. DUsseldorf 1972 
483) Cohen.J.; Hansel.M.: Glück und Risiko. Oie lehre von 
der subjektiven Wahrscheinlichkeit . 
Frankfurt 1961 
484) Coombs.C.H . ; Pruitt.D.G.: Components of Risk in Decision 
Making: Probability and Variance Pre-
ferences . In: Journal of Experimental 
Psychology, Nr. 60 , 1960. S. 265 - 277 
485) Coombs,C.H . ; Bowen,J.N.: A Test of EV-Theories of Risk 
and the Effect of the Central limit 
Theorem. In: Acta Psychologica. Nr.35, 
1971. S.15-28 
486)Cunningham.S.M. The Major Dimensions of Perceived Risk. 
In: Risk Taking and Information Handling 
in Consumer Behavior. hrg. von D.F.Cox . 
Boston 1967, S.82-108 
487) Domar,E.; Musgrave, R.A.: Proportional Income Taxation and 
Risk-Taking. In: Quarterly Journal of 
Economics, Nr. 58. 1944. S. 388-422 
488) Oroege, F.; Weisenborn. F.; Haft, R.: Wirkungen der Massen-kommuni kati on. 2. Aufl age, Frankfurt/Mai n 
1973 
489) Edwards,W. Variance Preferences in Gambling . In: American Journal of Psychology, Nr . 67. 
1954. S. 441 - 452 
490) JUgend in Europa -
491) Etzioni ,A 
492) Frey, B.S. 
493) Goude, G. 
494) Hammond, K. R. u.a.: 
495) Henrich,H. 
496) Himmelmann,G . 
497) Hofs tätter, P. R.; 
498) Huang, L. C. 
499) Janis, J . L. ; Mann , 
- 411 -
Ihre Eingliederung in die Welt der 
Erwachsenen, hrg. vom Jugendwerk 
der Deutschen Shell. Bände I bis 111 
Bearbeiter: Viggo Graf B1Uch·er. Ohne 
Ortsangabe 1977 
Mixed Scanning . A Third Approach to 
Decision Ma king . In : Public Administra-
tion Review, Nr . 27, 1967 . S. 385-392 
Eine Theori e demokratischer Wirtschafts-
politik. In: Kyklos, Nr.31, 1978. 
S. 208-234 
Han- A Biological Being or a Technical 
Mistake? A need to decide how to make 
decisions. Repo r t of the Project: Risk 
Generation and Ris k Assessment in a 
Social Perspective. Nr . 4, 1977. Stock-
holm 1977 
Social Judgement Theory. In: Human 
Judgement and Decision Processes: 
Formal and Mathematical Approaches, 
hrg. von M. Kaplan und S.Schwartz . 
New York 1975 
Störfälle an Kernkraftwerken. In : 
Kernfragen,hrg. von der Kernfor-
schungsanlage JUlieh . Redaktion: 
S. Weinhold u.a . JUlich 1978.S.16 
bis 18 
Polit i sche Okonomie - Herkunft,Pro-
bleme, Ansätze : Eine Obersicht. In : 
Politische Vierteljahresschrift, Bd . 18, 
1977. S.178-213 
Wendt,D . : Quantitative Hethoden der 
Psychologie . MUnchen 1966 
Subjective Expected Utility, Subjective 
Expected Risk, and Portfolio Theory. 
Michigan Mathematical Psychology Pro-
gram, MHPP-74-9, University of Michi-
gan 1974 
L. : Decision Making. A Psychological 
Analysis of Conflict, Choice and Com-
mitment. New York 1977 
500) Kaplan, L.B . , Szybillo, G. J . ; Jacoby, J. : Components of 
Perceived Risk in Product Purehase: 
A Cross-Validation . In: Journal of 
Applied Psychology, Nr . 59, 1974. 
S. 287 - 291 
- 41 -
501) Katz E. ; Levin, M.L.; Hamilton, H.: Traditions of Research on the 
Diffusion of Innovation. In: American Socio-
logical Review, Nr. 28, 1963. S. 237 - 252 
502) Kleining, G. ; Moore, H.: Soziale Selbsteinstufung (SSE). Ein Instrument 
zur Messung sozialer Schichten. In: Kölner 
Zeitschrift für Soziologie und Sozialpsycho-
logie, Bd. 20, 1968 
503) Kogan, N.; Wallach, M.A. : Risk Tak ing. Chicago 1964 
504) Kogan, N.; Wallach, M.A. : Risk Taking . In: Behavioral Science Foun-
dation of Consumer Behav;or. hrg. von 
J.B. Cohen . New York 1972. S.133-14O . 
SOS) Kozielecki, J . 
506) Kriz, J. 
507) Krelle, W. 
508) Lakoff, S.A. 
SOg) Lee, W. 
510) Lichtenstein, S. u.a . 
511) Lindbloom, C.E. 
512) Lohmöller, J.-B. 
513) Markowitz, H. 
514) Miller, D.W.; Starr, M.K. 
515) Payne, I .W. 
Environment and Personalitiy Determinants 
in Decision-Making. Poli sh Psychological 
Bulletin, Nr. 5, 1974 . S. 3 - 11 
öffentliche Meinung und politisches Handeln -
Kritik der "Schweigespirale". In: Bürgerbetei-
ligung und Bürgerinitiativen, hrg. von H. 
Matthöfer. Villingen 1977. S. 396 - 416 
Präferenz und Entscheidungstheorie. Tübingen 
1968 
Scientists , Technologists and Political Power . 
In: Science, Technol09Y and Society, hrg . von 
I. Spiegel-Rösing and D. de Solla Price. Lon-
den, Beverly Hills 1972. 5.355-391. 
Decision Theory and Human Behavior. New York 
1971 
Judged Frequency of LethaI Events. In : Journal 
of Experimental Psychology: Human Learning 
and Memory, Nr. 4, 1978. S. 551 - 578 
The Science of Muddling Through . In : Public 
Administration Review, Nr . 19, 1959. S. 79 - 99 
Die trimodale Faktorenanalyse von Tucker: 
Skalierung, Rotationen, andere Modelle. In : 
Arch. Psychol . Vol 131, Nr . 2, 1979. S. 137 -
166 
Portfolio Selection. New York 1959 
: The Structure of Human Decisions. Englewood 
Cliffs, New Jersey, 1967 
Relati on of Perceived Risk to Preferences 
among Gambles . In: Journal of Experimental 
Psychology (Human Perception and Performance), 
Nr. 104, 1975. S. 86 - 94 
516) POllatsek, A. ; Tversky, A.: A Theory of Risk. In : Journal of Mathema-
tical Psychology. Nr. 7, 1970. S. 540 - 553 
517) Renn 
518) Rosenstiel, l. von; 
519) Rothschild, l. 
520) Savage, l.J. 
521) Simon, H. 
522) SjPberg, l. 
523) Stallen, P.J.M. 
524) Stoner, A. 
525) Sevenson, O. 
-42-
Auswertung einer 8efragung von lehrern und Ju-
risten zum Thema Kernenergie während eines Be-
suchs von Fortbildungsseminaren in der Kernfor-
schungsanlage JÜlich. Interner Bericht. KFA-KUU-
IB-1/78. Kernforschungsanlage Jülich 1978. 
Ewald, G.: Marktpsychologie Band I: Konsumverhalten 
und Kaufentscheidung. Stuttgart u.a. 1979. 
Risk. In: Atom, Nr.268, Februar 1979. S.30-35. 
The Foundation of Statistics. New York 1954. 
: Administrative Behavior: A Study of Decision-
Making-Processes in Administrative Organisations. 
Third Edition. New York 1976. 
Strenght of Bel ief and Risk. Report of the Pro-ject: Risk Generation and Risk Assessment in 
a Social Perspective. Gäteborg (Schweden) 1977. 
: Science of Risk and Risk of Science - with 
Special Preference to Nuclear Power. In: Utility, 
Probability and Human Decision Making, hrg. 
von A. Wendt u. C. Vlek. Dordrecht (Nl), Boston 
(USA) 1975. 
A Comparison of Individual and Group Decisions 
Involving Risk. Massachusetts Institute of 
Technology. Boston 1961. 
A Unifying Interpretation of Different Models 
for the Integration of Information when 
Evaluating Gambles. In: Scandinavian Journal 
of Psychology, Nr.16, 1975. S.187-192. 
526) Schweiger, G.; Mazanec, J.; Wiegele, 0.: Das Modell des "erlebten Risi-
kos" (-perceived rlsk-). Struktur und Operationa-
llsierungskonzepte. In: Der Markt, Nr.60, 1976, 
S.93-102. 
527) Tversky, A. 
528) Tucker, l.R. 
: Elimination by Aspects: A Theory of Choice. 
In: Psychological Review, Nr.79, 1972. S.281-299. 
Some Mathematical Notes on Three-Mode-Factor 
Analysis. In: Psychometrica, Nr.31, 1966, 
S .279-311 . 
529) Vlek, Ch.; Stallen, P.J.: Rational and Personal Aspects of Risk. Invited 
Paper of the 7th Research Conference on Subjective 
Probability, Utility, and Decislon Making. Gäte-
borg 27.-31 .August 1979.530) Wilde, G.J.S. 
530) Wilde, G.J.S. : Theorie der Risikokompensation der Unfallverur-
sachung und praktische Folgerungen für die 
Unfallverhütung. In: Hilfe zur Unfallkunde, 
Nr. 130, 1978, S.134-156. 
531) Zuckermann, M. 
532) Kitsche1t, H. 
- 43 -
Sensation Seeking and Anxiety - Traits and 
States. as Determinants of Behavior in 
Nove1 Situations. In: Stress and Anxiety, 
ed. by I.G. Sarason and C.D. Spielberger, 
Vo1. 111. Washington D.C. 1976 
Kernenergiepolitik - Arena eines gesell-
schaftlichen Konflikts. Frankfurt/New York 
1980