The Effect of the Fukushima Nuclear Accident on Risk Perception,Antinuclear Behavioral Intentions,Attitude,Trust,Environmental Beliefs,and Values

Abstract

Past opinion polls have shown that major nuclear accidents can have a serious impact on public attitudes. Drawing on a values-beliefs-norms (VBN) model of environmental commitment, the authors hypothesized that a major nuclear accident may also affect the most durable cognitive and cultural foundations that underpin public perceptions of nuclear power. For 32 Italian participants, the authors assessed perceptions of nuclear power and values 1 month before and after the Fukushima nuclear accident. Participants reported decreases in nuclear trust, environmental organization trust, and pronuclear attitudes; they reported a significant increase in environmental beliefs assessed by the new ecological paradigm and a marginally significant increase in altruism. Major nuclear accidents may have the potential to influence values and proenvironmental beliefs, probably for the reason that they are the basis of public attitudes toward nuclear power.

Keywords

nuclear attitudes nuclear power trust risk perception proenvironmental beliefs values Fukushima accident

Despite a history of public resistance, recent opinion polls showed that attitudes toward nuclear power were less negative than in the past (Rosa, 2004; TNS Opinion & Social, 2011). However, nuclear power now faces new challenges caused by the Fukushima nuclear accident. One consequence of the nuclear emergency following the Fukushima accident might well be its influence on public attitudes toward nuclear power. The influence on attitudes toward nuclear power of previous nuclear accidents shall be reviewed briefly. A theoretical model shall be examined, describing why and which variables may be affected by the Fukushima nuclear accident.

The Change in Public Views After Nuclear Accidents

Several studies have shown that previous nuclear accidents affected public views toward nuclear power in Europe, the United States, and Japan (Midden & Verplanken, 1990; Peters, Albrecht, Hennen, & Stegelmann, 1990; Rankin, Melber, Overcast, & Nealey, 1981; Renn, 1990; Sjöberg & Sjöberg, 1990; Tsunoda, 2001). These studies are, for the most part, based on secondary analyses of time series data obtained from opinion polls conducted over time. This methodology has significant limitations when used in the evaluation of change in public views after nuclear accidents because these data are collected from independent samples, precluding an analysis at the individual level.

In contrast, Lindell and Perry (1990) conducted a study involving a repeated-measures analysis of a single sample of 69 participants, made up of residents of Southwestern Washington. Data on nuclear power-plant accident risk perceptions were collected 5 months before and 1 month after the Chernobyl nuclear accident. They found no significant difference between accident likelihood and dread risk perceptions before the Chernobyl accident and after the accident. However, increases in the frequencies of thoughts and discussions about a nearby nuclear power plant were observed. Lindell and Perry also discovered that the Chernobyl accident produced a reduction in the perceived personal consequences of a nuclear plant accident. Finally, in this study, small, and in some cases nonsignificant, test–retest correlations were found. Lindell and Perry claimed that public perceptions of nuclear plant accident risks may not be based on stable cognitive representations, and may be indicative of “pseudo-opinions.”

Using a longitudinal within-participants design, Verplanken (1989) investigated beliefs, attitudes, and behavioral intentions toward nuclear energy in the Netherlands 2 months before the Chernobyl accident, 1 month after the accident, 6 months after the accident, and 1 year and 7 months after the accident. In this study, it was revealed that 1 month after the accident, participants reported more unfavorable attitudes, which became less antinuclear 5 months later. Although perceived benefits did not change immediately after the accident, the occurrence of catastrophic accidents in Dutch nuclear power plants was judged far more probable than before the accident.

Differences between countries in the change of public attitude toward nuclear power following a nuclear accident may be dependent on the level of commitment of the public (i.e., the degree to which public attitudes toward nuclear power are well formed). In countries with well-formed attitudes and with a strong nuclear program, the effects might be smaller (Renn, 1990; Van der Pligt & Midden, 1990). In Italy, the majority of people have shown uncommitted opinions (e.g., undecided positions in the nuclear debate) or negative attitudes toward nuclear power. Among the European Union countries, Italian people are among the most dissatisfied with their national nuclear safety legislation (TNS Opinion & Social, 2011). Furthermore, there are no nuclear power plants in operation in Italy. Therefore, we should expect a major negative effect on public opinion toward nuclear power following a nuclear accident.

Environmental Beliefs, Values, and Antinuclear Behavioral Intentions After the Fukushima Accident

On March 11, 2011, the tsunami following the Tōhoku earthquake provoked the Fukushima nuclear accident. In this accident, several reactors within the Fukushima Daiichi nuclear power plant were severely damaged, and radioactive materials were released. Reports from the print and television media on this accident dominated headlines for weeks and had an impact on energy policy of different countries (Wittneben, 2012).

The hypothesis of this study was that the Fukushima nuclear accident influenced, negatively, public views regarding nuclear power, such as nuclear power attitudes, antinuclear behavioral intentions, trust and risk perceptions, but also other factors such as environmental beliefs and general values. Nuclear power attitudes are different from environmental beliefs and general values because the latter are not related specifically to nuclear power. This hypothesis was derived from the values-beliefs-norms (VBN) model of environmental commitment (Stern, Dietz, Abel, Guagnano, & Kalof, 1999; Stern, Dietz, & Kalof, 1993), which was successfully used by Whitfield, Rosa, Dan, and Dietz (2009) to predict public views of nuclear power. More specifically, in their model, traditional and altruistic values along with environmental beliefs were considered the underlying dynamics of public views about nuclear power, such as nuclear power risk perceptions, attitudes toward nuclear power, trust in environmental organizations and in nuclear management institutions (nuclear trust), and antinuclear behavioral intentions. Given the relationship between values or environmental beliefs and public views toward nuclear power (Whitfield et al., 2009), we hypothesize that the occurrence of an event that promotes negative attitudes toward nuclear power may also create more positive general values and environmental beliefs. To our knowledge, no studies have examined these hypotheses. However, there are theoretical reasons to expect these relationships. According to the social amplification of risk framework (Kasperson et al., 1988), risk signals are processed by social, as well as individual, amplification stations. This process of amplification influences public perceptions and first order behavioral responses, which, in turn, result in secondary impacts such as changes in enduring perceptions, beliefs, and social and cultural values. Changes in public views toward nuclear power might be considered first order responses, while the changes in values and environmental beliefs might be considered secondary impacts because they represent enduring beliefs and are related to social and cultural changes.

Aim and Hypotheses of the Study

The aim of this study is to extend previous studies by investigating the effects of a nuclear accident not only on nuclear power risk perceptions, trust, attitudes toward nuclear power, and antinuclear behavioral intentions, but also on environmental beliefs and values. We use a repeated-measures analysis of a single Italian sample measured before and after the Fukushima accident. We hypothesized that, post-Fukushima, participants would report higher nuclear risk perception, more antinuclear behavioral intentions, more environmental concerns (beliefs), lower trust, and lower pronuclear attitudes compared with the scores before the accident. Recall that past studies have shown that pronuclear attitudes are associated with stronger traditional values and weaker altruistic values (Whitfield et al., 2009). Thus, we expected, post-Fukushima, greater environmental concerns (beliefs), a stronger adherence to altruistic values, and a weaker adherence to traditional values.

Method

Participants

A total of 32 Italian citizens participated in the study. In all, 20 (62.5%) participants were women. The average mean age was 39.93 (SD = 12.13, range = 18-65). Most of the respondents (n = 21, 65.6%) completed high school (corresponding to Grade 13), 25.0% (n = 8) completed university, 9.4% (n = 3) completed middle school (corresponding to Grade 8). Participants were recruited using a snowball approach.

Instrument

We administered the same questionnaire before and after the Fukushima nuclear accident. The only difference was that in the second wave, we added a section named “Information Related to the Accident.”

Human value

Nine items from Stern, Dietz, and Guagnano (1998) were used to measure traditional and altruistic values. These items were derived from the Schwartz’s (1992) value scale and were adapted by Stern et al. (1998) for the VBN model. Response options ranged from 1 = not at all important to 10 = extremely important. Cronbach’s alphas for traditional and altruistic values were .77 and .98 at Time 1, and .74 and .98 at Time 2, respectively.

Environmental beliefs

We used the revised new ecological paradigm (NEP) scale (Dunlap, Van Liere, Mertig, & Jones, 2000; Dunlap & Van Liere, 1978) to measure environmental beliefs. A previously translated version into Italian was used (Prati, Zani, & Pietrantoni, 2011). Items were measured on a 1 to 5 scale, from 1 = strongly disagree to 5 = strongly agree. We decided to use all 15 items of the revised NEP as recommended by Hawcroft and Milfont (2010). The five factor names and the respective reliabilities were as follows: balance of nature (α = .61), ecocrisis (α = .79), antiexemptionalism (α = .40), limits to growth (α = .26), and antianthropocentrism (α = .73). Although Amburgey and Thoman (2011) recommended using the separate factors, this recommendation did not extend to the case where they are not internally consistent. Therefore, we decided to use an average score across all items. Cronbach’s alphas for this scale were .72 at Time 1 and .69 at Time 2.

Trust

We asked participants to rate how much trust they have in national environmental organizations, scientists, the International Atomic Energy Agency, the Italian Atomic Energy Authority, and the Italian Government. Response options ranged from 1 = very little trust to 10 = a great deal of trust. Following Whitfield et al. (2009), we computed an index of nuclear trust (α = .82 at Time 1 and .84 at Time 2; International Atomic Energy Agency, Italian Atomic Energy Authority, and Italian Government). The variables “trust in national environmental organizations” and “trust in scientists” were analyzed separately, due to their low intercorrelation (r = .40).

Risk perception

A single item measured nuclear risk perception—“Overall, how risky is nuclear power?”—which was rated on a 1 to 10 scale, ranging from not at all risky to extremely risky. Then, we used the psychometric model of risk perception (Fischhoff, Slovic, Lichtenstein, Read, & Combs, 1978; Slovic, 1987) to measure the following risk attributes related to nuclear power: dread, old risk, personal exposure, voluntariness, number of exposed people, observability, severity of consequences, threaten future generations, personal knowledge, known to science, social benefit, personal benefit, acceptability, and controllable. Risk attributes were to be rated on a 10-point scale. The items were derived from a previous study conducted in Italy (Savadori et al., 2004).

Pronuclear power attitudes

We used a revised version of the five-items scale used by Peters and Slovic (1996) to assess attitudes toward nuclear power. Questions assessed the extent to which respondents supported nuclear power. The formulation of the items was slightly adapted to fit the Italian context, in which nuclear power is not currently used. Response options ranged from 1 = strongly disagree to 10 =strongly agree. Cronbach’s alphas were .94 at Time 1 and .93 at Time 2.

Antinuclear behavioral intentions

We measured antinuclear behavioral intentions by selecting three items. Questions assessed the extent to which participants intend to mobilize themselves (to protest, to support activist movement, or to sign a petition) against the plan to build a power plant in their community. Cronbach’s alphas were .80 at Time 1 and .79 at Time 2.

Information related to the accident

We asked participants if they had heard about the Fukushima nuclear accident, and, if yes, we asked for the source (television, radio, newspapers, and friends/family members) and the extent to which they were interested in news related to this topic. Response options ranged from 1 = I was not interested to 10 = I was very interested. Then, participants rated their emotional involvement in the Fukushima nuclear accident. Response options ranged from 1= I was not emotionally involved to 10 = I was deeply emotionally involved. We asked them to rate the frequency of discussion about the Fukushima nuclear accident with friends and family members. Response options ranged from 1 = I have never talked about this to 10 = I have always talked about this. Finally, participants evaluated the extent to which friends and family members were worried about the Fukushima nuclear accident (1 = not at all worried to 10 = extremely worried) and the degree to which they feel personally in danger because of the consequences of the Fukushima nuclear accident (1 = not at all in danger to 10 = extremely in danger).

Procedure

During the weeks immediately prior to the Fukushima nuclear accident (end of February and the first week of March 2011), we collected some data using a questionnaire on nuclear attitudes and their determining factors according to the VBN model (Stern et al., 1993, 1999). Participants were contacted through a network of personal acquaintances (one researcher recruited people she knew and from whom she could subsequently request further contacts). This situation provided an opportunity to evaluate the influence of a nuclear accident on risk perception, antinuclear behavioral intentions, attitude, trust, environmental beliefs, and values. About 1 month after the Fukushima nuclear accident (during the 2nd week of April 2011), participants who filled out the questionnaire before the Fukushima nuclear accident were contacted again. All the people contacted agreed to complete the questionnaire a second time.

Data Analysis

To test whether the distribution of the variables is normal, we performed the Kolmogorov–Smirnov test and the Shapiro–Wilk test. Both tests indicated that at least one of the paired variables (pre- and post-Fukushima) investigated in this study deviated from a comparable normal distribution. Therefore, we used the Wilcoxon’s signed-rank test to compare the scores on pre-Fukushima to those on post-Fukushima. When multiple univariate tests are conducted, it is important to revise the standards for significance levels, to limit the Type I error. In this study, we used the sequential approach to controlling the false discovery rate in multiple comparisons proposed by Benjamini and Hochberg (1995). This procedure yields much greater power than the widely used Bonferroni technique (Williams, Jones, & Tukey, 1999). According to the Benjamini–Hochberg procedure, the extent of the difference is significant, when the observed p value is less than the Benjamini–Hochberg critical value.

We used Cohen’s (1988) suggestions for interpreting the magnitude of the effect size: A small effect size is r ≥ .10, a medium effect size is r ≥ .30, and a large effect size is r ≥ .50.

A priori power analysis, conducted with the general power analysis program GPOWER 3 (Faul, Erdfelder, Lang, & Buchner, 2007), revealed that a sample of 35 participants was sufficient to detect a medium effect size and a minimum of 80% statistical power.

Results

Missing data analysis showed that there were one or two missing values in some of the items of the following variables: altruistic values, traditional values, severity of consequences, nuclear trust, trust in scientists, and environmental organizations trust. In dealing with missing data, the multiple imputation procedure (n = 10 data sets) was used as recommended by Graham (2009).

All the participants reported that they had heard about the Fukushima nuclear accident. The most frequent source of information was the television (96.9%), followed by friends/family members (90.6%), radio (78.61%), and newspapers (65.6%). Participants reported that they were very interested in news related to this topic (M = 8.53, SD = 1.10), they were emotionally involved with the accident (M = 8.81, SD = 1.00), they have frequently discussed the accident with friends and family members (M = 8.88, SD = 0.83), they found friends and family members to be particularly worried about the accident (M = 8.81, SD = 0.97), and they felt they were in danger because of the consequences of the Fukushima nuclear accident (M = 8.75, SD = 1.05).

Table 1 shows the descriptive statistics and the Wilcoxon’s test results for the differences between pre- and post-Fukushima in the variables considered in the study. All the effects in Table 1 were in the predicted direction, although three of them were nonsignificant. Nuclear trust, pronuclear attitudes, antinuclear behavioral intentions, environmental beliefs (NEP scale), and trust in environmental organizations changed to a large extent after the Fukushima accident. More specifically, after the Fukushima accident, participants were more likely to endorse antinuclear behavioral intentions and environmental beliefs, whereas they reported fewer pronuclear attitudes, less trust in environmental organizations, and nuclear trust. A medium effect size was obtained for altruistic values (participants were more likely to endorse these values after the Fukushima accident). However, the observed p value for altruistic values is higher than the Benjamini–Hochberg critical value; therefore, this effect should not be considered statistically significant. Trust in scientists and traditional values did not change after the Fukushima accident.

Table 1.

Mean Ratings, Standard Deviations, Test–Retest Correlations, and Wilcoxon’s Test Results for Pre- and Post-Fukushima Human Values, Environmental Beliefs, Trust, Pronuclear Attitudes, and Antinuclear Behavioral Intentions.

	Pre-Fukushima		Post-Fukushima
Variables	M	SD	M	SD	r _T1T2	z	p	r_es:z
Altruistic values	8.57	2.11	8.91	1.78	.86	−2.109	.038	−.37
Traditional values	8.64	1.84	8.99	1.38	.81	−1.858	.065	−.33
Environmental beliefs (NEP)	3.01	0.55	3.34	0.47	.57	−3.677	.000^a	−.65
Nuclear trust	4.08	2.14	2.30	1.48	.58	−4.367	.000^a	−.77
Trust in scientists	7.01	1.88	6.60	1.79	.45	−0.952	.345	−.17
Environmental organizations trust	5.95	2.52	4.69	1.96	.36	−2.661	.008^a	−.47
Pronuclear attitudes	4.13	2.56	2.99	1.73	.83	−3.761	.000^a	−.66
Antinuclear behavioral intentions	7.10	3.01	8.51	1.76	.77	−3.435	.001^a	−.61

Note: NEP = new ecological paradigm; T1= after Pre-Fukushima; T2= after Post-Fukushima.

Rows in which the observed p value is less than the Benjamini–Hochberg critical value.

In Table 2, the descriptive statistics and the Wilcoxon’s test results for the differences between pre- and post-Fukushima in risk perception and attributes are presented. To a large extent, participants rated the risks of nuclear power as more observable, less controllable, with more severe consequences, older, more dreadful, more as threat for future generations, and more dangerous. Moreover, they perceived this technology to be less beneficial for society. However, the observed p value for social benefit is higher than the Benjamini–Hochberg critical value. Ratings of personal exposure, voluntariness, number of exposed people, personal knowledge, scientific knowledge, personal benefit, and acceptability did not change after the Fukushima accident. Test–retest correlations were all above .4, with the exclusion of the following risk attributes: threat to future generations, voluntariness, number of exposed people, and known to science.

Table 2.

Mean Ratings, Standard Deviations, Test–Retest Correlations, and Wilcoxon’s Test Results for Pre-·and Post- Fukushima Risk perception and Attributes.

	Pre-Fukushima		Post-Fukushima
Risk perception and attributes	M	SD	M	SD	r_T1T2	z	p	r_es:z
Riskiness	7.66	2.66	8.75	2.05	.50	−2.483	.013^a	−.44
Dread	7.81	2.53	9.25	1.14	.42	−3.189	.001^a	−.56
Old risk	7.72	2.75	9.03	1.67	.54	−3.238	.001^a	−.57
Personal exposure	7.41	2.69	7.72	2.04	.58	−0.754	.451	−.13
Voluntariness	3.91	2.80	3.72	2.23	.12	−0.077	.939	−.01
Number of exposed people	7.44	2.76	7.69	2.09	.30	−0.284	.776	−.05
Observability	5.69	3.05	8.38	2.57	.41	−4.106	.000^a	−.73
Severity of consequences	8.09	2.29	9.53	1.09	.42	−3.400	.001^a	−.60
Threat to future generations	7.97	2.66	9.50	1.34	.31	−3.131	.002^a	−.55
Personal knowledge	6.19	1.99	6.09	1.38	.71	−0.491	.624	−.09
Known to science	7.69	1.71	7.38	1.43	.30	−0.943	.346	−.17
Social benefit	4.34	2.84	3.66	2.22	.83	−2.351	.019	−.42
Personal benefit	3.72	2.76	3.50	2.21	.87	−1.027	.304	−.18
Acceptability	3.31	2.60	3.06	2.21	.75	−0.940	.347	−.17
Controllable	4.69	2.88	3.03	2.02	.52	−3.43	.001^a	−.61

Rows in which the observed p value is less than the Benjamini–Hochberg critical value.

T1= after Pre-Fukushima; T2= after Post-Fukushima.

Discussion

The Fukushima nuclear accident caused by the Tōhoku earthquake and tsunami on March 11, 2011, may exert an important influence on public opinion and policy makers. The current research examined whether this accident had an impact on risk perception of nuclear power, pronuclear attitudes, trust, and antinuclear behavioral intentions. Furthermore, we explored whether the accident has influenced their underlying factors, namely, human values and environmental beliefs. Drawing on a VBN model of environmental commitment (Stern et al., 1999, 1993), as well as on the results of this model applied to nuclear attitudes (Whitfield et al., 2009), we tested our predictions using a prospective longitudinal study. To our knowledge, this is the first report of a prospective effect of a nuclear accident on human values and environmental beliefs. Our hypothesis was that, since attitudes toward nuclear power are rooted in particular human values and environmental beliefs, a sudden and large change in these attitudes may have a reflection on their roots. The results of this study showed that, after the Fukushima nuclear accident, participants were more likely to endorse proenvironmental beliefs.

The change in altruistic values after the nuclear accident was close to significance. It is likely that a larger sample size would be needed to detect a significant effect. Altruistic values represent concern for the welfare of human beings, nonhuman species, and the biosphere, for reasons other than self-interest. Several studies showed that altruistic values have been a strong and consistent predictor of proenvironmental beliefs (Dietz, Fitzgerald, & Shwom, 2005). The results of the present study may suggest that, although human values and environmental beliefs are considered relatively stable over the course of a person’s life, a significant event might lead to a different degree of adherence. Indeed, the effect size was medium for values. However, we found a large effect size for environmental beliefs. Altruistic values and proenvironmental beliefs may be shared within a socially committed community concerned for others and the environment. Taken together, these results demonstrate that when these concerns are confirmed by events, such as the Fukushima nuclear accident, people may be more likely to support or be part of this community. In line with the social amplification of risk framework (Kasperson et al., 1988), this study showed that secondary impacts of social amplification may involve not only nuclear power risk perception, attitudes toward nuclear power, and trust, but also more enduring factors such as environmental beliefs and, probably, human values.

In line with the results of several opinion polls (Midden & Verplanken, 1990; Peters et al., 1990; Rankin et al., 1981; Renn, 1990; Sjöberg & Sjöberg, 1990; Tsunoda, 2001), our study found, immediately after a nuclear accident, a large increase in risk perception of nuclear power and antinuclear behavioral intentions, together with a large reduction of pronuclear attitudes and trust in authorities. The extent of these changes was large and suggests that people may be influenced by even a single dramatic event. It should be expected, a partial return to the levels before the accident in the following months or years, in line with the results of surveys conducted after Chernobyl or Three Miles Islands (e.g., Peters et al., 1990; Rankin et al., 1981; Verplanken, 1989). It should be noted, however, that this “rebound” hypothesis has been seriously challenged by the results of long-term surveys (Rosa & Dunlap, 1994).

It is interesting to note that the most significant changes were observed in nuclear trust and the evaluation of the extent to which the potential damage produced by nuclear power is observable. The first finding may be partially explained by the fact that, in spite of the opposition of the majority of the Italian citizens (TNS Opinion & Social, 2011), Italian authorities planned to revive nuclear power and suggested building new reactors. One of the reasons put forward for this plan by authorities was that nuclear power was safe. After the Fukushima nuclear accident, Italian authorities have put a moratorium on plans to revive nuclear power. This course of events may explain, at least partially, the loss of trust in the authorities.

The second finding may be explained by the public’s exposure to the media’s coverage of the accident. The images of the accident may be considered vivid and dramatic; therefore, participants may have inferred that the potential damages produced by nuclear power are real and concrete.

Another interesting finding is the loss of trust in environmental organizations after the Fukushima nuclear accident. Although these organizations are clearly not responsible for nuclear accidents, it is possible that participants may have perceived their limited power to prevent nuclear risks from manifesting. Moreover, participants may have blamed these organizations for their lack of influence on plans to revive nuclear power. On the contrary, trust in scientists did not change after the Fukushima nuclear accident. This finding may be ascribed to the fact that science is perceived to be neutral (nonpartisan) because scientists are considered the most trusted source of information about nuclear power and nuclear safety (TNS Opinion & Social, 2011).

Although we found a large change in risk perception, pronuclear attitudes, behavioral intentions, and trust, the results of this study should not be interpreted as indicating that participants were enormously preoccupied with nuclear power. However, in line with the study of Verplanken (1989), we found the predominance of the perceived catastrophic consequences attributed to nuclear power accidents. The results of this study are different from those of Lindell and Perry (1990), in that we found that participants rated as more dreadful the risk of nuclear power after the accident.

There is also another apparent conflict between the present study and Lindell and Perry’s (1990) study. Lindell and Perry found very low test–rest correlations between the measures used before and after the accident. They interpreted this finding as indicative of pseudo-opinions or beliefs expressed by participants that did not exist prior to the time that a question was asked by the interviewer. In our study, we did find some pseudo-opinions in relation to the presumed number of exposed people and to the degree that the risk associated to nuclear power is faced voluntarily by the exposed population, is known to scientists, and affects future generations. It may be that those questions might not have been well understood by participants or that these judgments have not been quickly accessible to participants. All the other measures showed adequate test–retest correlations and, therefore, may be considered relatively stable.

Although the current research provides new insights into the influence of a nuclear accident, several limitations should be acknowledged. One limitation is that in pre–post design, the first measurement might influence the second one (pre–test sensitization). Although we cannot exclude effects associated with a repeated-measure design, boredom effects were unlikely to have influenced significantly the results for three reasons. First, there was an interval of at least 4 or 5 weeks between the administrations of the questionnaires. Second, all the effects were in the predicted directions. A second limitation is that we could not include a control condition. Third, to our knowledge, there is no evidence that unambiguously proves that, in repeated-measure design, boredom leads to higher estimates in all the variables. The prospective, longitudinal nature of the findings showing a change in attitude, intentions, perceptions, trust, values, and environmental beliefs after the accident is consistent with but does not prove the hypothesis that a nuclear accident may influence those variables. Future studies need to replicate the nature of this longitudinal relationship in other contexts to address cross-cultural generalizability. More important, experimental manipulations that study the effect of exposure to news related to a nuclear accident are needed to demonstrate a casual effect. Although we could not find any other important event related to nuclear power between the first and the second assessment, we cannot completely exclude the influence of history. A fourth limitation is that the sample size was small. Although the small sample size may limit the power to detect a small change before and after the Fukushima accident relationship, we argue that this has no negative effects in terms of the statistically significant differences (i.e., statistical conclusion validity) because, if anything, a small sample would bias the findings in a conservative manner. The small sample size is due to the exceptional nature of the event. The small sample size prevented the calculation of more comprehensive analytical strategy, such as latent change scores.

Conclusion

The present research provides support for the notion that a nuclear accident may have a deep impact on public views regarding nuclear power in the short term. Indeed, we found a large change in trust, risk perception, pronuclear attitudes, and antinuclear behavioral intentions following the Fukushima accident. More interestingly, it seems that nuclear power accidents may have the potential to influence more enduring perceptions, such as environmental beliefs and, probably, altruistic values.

Footnotes

Acknowledgements

The authors thank Sara Bragonzoni for her assistance with this research. We also thank the editor, Barbara Brown, for extensive and insightful comments and for suggesting some improvements in presentation.

Declaration of Conflict Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Author Biographies

Gabriele Prati is fellow researcher at the University of Bologna (Italy). He received doctoral degrees in social and organizational psychology from the University of Bologna. His research interests focus on risk perception and environmental issues.

Bruna Zani got her degree in political sciences (major in sociology) in July 1970 at the University of Bologna. In 1989 she became associate professor in social psychology and, since March 2000, she is full professor of social psychology at the Faculty of Psychology in Bologna. Her research interests focus on community psychology, political and civic participation, risk behaviors in adolescence, and risk perception.

References

Amburgey

J. W.

Thoman

D. B.

(2011). Dimensionality of the new ecological paradigm: Issues of factor structure and measurement. Environment and Behavior, Advance online publication. doi:10.1177/0013916511402064

Benjamini

Hochberg

(1995). Controlling the false discovery rate: A practical and powerful approach to multiple testing. Journal of the Royal Statistical Society B, 57, 289-300.

Cohen

(1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Erlbaum.

Dietz

Fitzgerald

Shwom

(2005). Environmental values. Annual Review of Environment and Resources, 30, 335-372. doi:10.1146/annurev.energy.30.050504.144444

Dunlap

R. E. R.

Van Liere

K. D.

(1978). The “new environmental paradigm:” A proposed measuring instrument and preliminary results. Journal of Environmental Education, 9, 10-19.

Dunlap

R. E.

Van Liere

K. D.

Mertig

A. D.

Jones

R. E.

(2000). New trends in measuring environmental attitudes: Measuring endorsement of the new ecological paradigm: A revised NEP scale. Journal of Social Issues, 56, 425-442. doi:10.1111/0022-4537.00176

Faul

Erdfelder

Lang

A. G.

Buchner

(2007). G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behavior Research Methods, 39, 175-191. doi:10.3758/BRM.41.4.1149

Fischhoff

Slovic

Lichtenstein

Read

Combs

(1978). How safe is safe enough? A psychometric study of attitudes toward technological risks and benefits. Policy Sciences, 9, 127-152. doi:10.1007/BF00143739

Graham

J. W.

(2009). Missing data analysis: Making it work in the real world. Annual Review of Psychology, 60, 549-576. doi:10.1146/annurev.psych.58.110405.085530

10.

Hawcroft

L. J.

Milfont

T. L.

(2010). The use (and abuse) of the New Environmental Paradigm Scale over the last 30 years: A meta-analysis. Journal of Environmental Psychology, 30, 143-158. doi:10.1016/j.jenvp.2009.10.003

11.

Kasperson

R. E.

Renn

Slovic

Brown

H. S.

Emel

Goble

. . .Ratick

(1988). The social amplification of risk: A conceptual framework. Risk Analysis, 8, 177-187.

12.

Lindell

M. K.

Perry

R. W.

(1990). Effects of the Chernobyl accident on public perceptions of nuclear plant accident risks. Risk Analysis, 10, 393-399. doi:10.1111/j.1539-6924.1990.tb00522.x

13.

Midden

C. J. H.

Verplanken

(1990). The stability of nuclear attitudes after Chernobyl. Journal of Environmental Psychology, 10, 111-119. doi:10.1016/S0272-4944(05)80122-7

14.

Peters

Slovic

(1996). The Role of Affect and Worldviews as Orienting Dispositions in the Perception and Acceptance of Nuclear Power. Journal of Applied Social Psychology, 26, 1427-1453. doi: 10.1111/j.1559-1816.1996.tb00079.x

15.

Peters

H. P.

Albrecht

Hennen

Stegelmann

H. U.

(1990). “Chernobyl” and the nuclear power issue in West German public opinion. Journal of Environmental Psychology, 10, 121-134.

16.

Prati

Zani

Pietrantoni

(2011). Quanto siamo intenzionati a comprare un prodotto OGM? Indagine su un campione di italiani [To what extent are we willing to buy a GM product? A survey on a sample of Italian people]. In Villano

Zani

(Eds.), Parlare di OGM in Italia [Talking about GMOs in Italy] (pp. 79-99). Bologna, Italy: Clueb.

17.

Rankin

W. L.

Melber

B. D.

Overcast

T. D.

Nealey

S. M.

(1981). Nuclear power and the public: An update of collected survey research on nuclear power. Seattle, WA: Batelle Human Affairs Research Centers.

18.

Renn

O. S. M.

(1990). Public responses to the Chernobyl accident. Journal of Environmental Psychology, 10, 151-167. doi:10.1016/S0272-4944(05)80125-2

19.

Rosa

E. A.

(2004). The future acceptability of nuclear power in the United States. Paris, France: Institute Francais des Relations Internationales.

20.

Rosa

E. A.

Dunlap

R. E.

(1994). The polls-poll trends: Nuclear energy: Three decades of public opinion. Public Opinion Quarterly, 58, 295-325. doi:10.1086/269425

21.

Savadori

L. R. E.

Savio

Nicotra

Rumiati

Finucane

Slovic

(2004). Expert and public perception of risk from biotechnology. Risk Analysis, 24, 1289-1299. doi:10.1111/j.0272-4332.2004.00526.x

22.

Schwartz

S. H.

(1992). Universals in the content and structure of values: Theoretical advances and empirical tests in 20 countries. Advances in Experimental Social Psychology, 25, 1-65.

23.

Sjöberg

B. D.

Sjöberg

(1990). Risk perception and worries after the Chernobyl accident. Journal of Environmental Psychology, 10, 135-149. doi:10.1016/S0272-4944(05)80123-9

24.

Slovic

(1987). Perception of risk. Science, 236, 280-285. doi:10.1126/science.3563507.

25.

Stern

P. C.

Dietz

Abel

Guagnano

G. A.

Kalof

L. A.

(1999). Social psychological theory of support for social movements: The case of environmentalism. Human Ecology Review, 6, 81-97.

26.

Stern

P. C.

Dietz

Guagnano

G. A.

(1998). A brief inventory of values. Educational and Psychological Measurement, 58, 884-1001. doi:10.1177/0013164498058006008

27.

Stern

P. C.

Dietz

Kalof

(1993). Value orientations, gender and environmental concern. Environment and Behavior, 25, 322-348. doi:10.1177/0013916593255002

28.

TNS Opinion & Social (2011). Europeans and Nuclear Safety Report [Special Eurobarometer]. European Commission. Retrieved from http://ec.europa.eu/energy/nuclear/safety/safety_en.htm

29.

Tsunoda

(2001). Public response to the Tokai nuclear accident. Risk Analysis, 21, 1039-1046. doi:10.1111/0272-4332.216172

30.

Van der Pligt

Midden

C. J.

(1990). Chernobyl: Four years later: Attitudes, risk management and communication. Journal of Environmental Psychology, 10, 91-99. doi:10.1016/S0272-4944(05)80120-3

31.

Verplanken

(1989). Beliefs, attitudes, and intentions toward nuclear energy before and after Chernobyl in a longitudinal within-subjects design. Environment and Behavior, 21, 371-392. doi:10.1177/0013916589214001

32.

Whitfield

S. C.

Rosa

E. A.

Dan

Dietz

(2009). The future of nuclear power: Value orientations and risk perception. Risk Analysis, 29, 425-437. doi:10.1111/j.1539-6924.2008.01155.x

33.

Williams

V. S. L.

Jones

L. V.

Tukey

J. W.

(1999). Controlling error in multiple comparisons, with examples from state-to-state differences in educational achievement. Journal of Educational and Behavioral Statistics, 24, 42-69. doi:10.3102/10769986024001042

34.

Wittneben

B. B. F.

(2012). The impact of the Fukushima nuclear accident on European energy policy. Environmental Science & Policy, 15, 1-3. doi:10.1016/j.envsci.2011.09.002