Examining the role of perceived risk and benefit, shared concern for nuclear stigmatization, and trust in governments in shaping citizen risk acceptability of a nuclear power plant

ABSTRACT

This study aims to present a more balanced perspective of risk acceptability through integrating economic individualism and social constructivism and examining the impact of trust in the national and local government on citizens’ attitudes toward risk and nuclear power plants, respectively. The data were gathered from surveys of local residents engaged in fierce debates on whether they agreed to accept a nuclear power plant in their local area, the city of Samcheok in South Korea. To capture the causal links among the determinants of risk acceptability of a nuclear power plant and their relationships, the authors utilized a structural equation model. Demonstrating that a shared worry about nuclear stigmatization has both a positive impact on risk perception but also a negative impact on trust in local government, our findings emphasize that risk studies of nuclear energy should pay attention to the social, cultural, and historical contexts of nuclear energy as well as its economic aspects when examining the determinants of risk acceptability. This study also enriches our understanding that national and local governments can play different roles in promoting citizens’ risk acceptability of a nuclear power plant. Thus, this study contributes towards presenting a better specified causal model of risk acceptability and offers practical implications of how to promote the public’s risk acceptability of nuclear power facilities.

KEYWORDS:

• Risk acceptability
• Nuclear power plant
• Public policy
• Perceived risks and benefits
• Citizen trust in the national and local government

Highlights

• Our findings emphasize that risk studies of nuclear energy should pay attention to the social, cultural, and historical contexts of nuclear energy as well as its economic aspects when examining the determinants of risk acceptability.

• Citizen surveys demonstrate that shared worry about nuclear stigmatization has both a positive impact on risk perception but also a negative impact on trust in local government.

• Our findings demonstrate that the impact of a collectively shared concern for nuclear stigmatization on trust in local government depended on trust in the national government. People’s shared fear of being stigmatized did not influence their trust in local government if their trust in the national government was high.

• Our findings enrich our understanding that national and local governments played different roles in promoting citizens’ risk acceptability of a nuclear power plant, at least in the planning stage of the facility.

Notes

¹ Richard Riordan, the Republican candidate, and Michael Woo, the Democratic candidate, ran against each other in the 1993 mayoral election. According to the LA Times public opinion poll, Riordan was leading by 7% among likely voters, while Woo was leading by 6% among all registered voters. The election results clearly show how the survey of likely voters was more accurate. In this case, conducting a survey among all registered voters was the wrong approach which, in a broader context, reveals how random sample assignment, representative of the whole target population, is not always the most accurate sampling method.

² In Korea, when one wants to mail a letter, (s)he should go to a post office because a mailman does not pick up a letter from a mail box at home. There is no other post office in the city so any persons who use a mailing service should go to the post office. Whenever citizen rallies occur, the location in front of the post office is the most widely used place for them in the city.

³ The city of Samcheok has been a potential candidate for nuclear facility construction for over 20 years. Furthermore, Samcheok has witnessed the dreadful nuclear disaster in Fukushima in the adjacent country of Japan in 2011. The city of Samcheok is geometrically close to Japan as it is located in the east coast of the Korean peninsula. Fishermen testified that some earthquakes and tsunamis that appeared in Japan were indirectly transmitted to the ocean near Samcheock through the shake of strata. The series of events became a stigma to Samcheok, which caused great concern among the local residents with them fearing that the presence of a hazardous facility might damage not only the local tourist industry but also the city’s image, which would result in negative economic effects for the whole community.

⁴ Samcheok City Government persistently advertised “good” things about nuclear energy to the local people to encourage them to accept site proposals. Due to the image designed around boosting the local economy, the nuclear energy agenda in Samcheok became more complicated and controversial. Community residents could expect some economic benefits derived from a nuclear facility site.

⁵ This variable is derived from Flynn et al. (1992), which were originally referring to attitudes toward national government (i.e., The Federal Department of Energy).

⁶ Technically, trust in national government is also an exogenous variable. However, it is later converted into a dichotomous variable to be a moderator for two-group analyses; hence it is excluded in a single group analysis. Miller’s (1974) items were referred to in creating these items.

⁷ Before the analyses were conducted, the distribution of residuals for the dependent variables was examined, which was a fairly normal distribution except for the high peaks. Also, the homoscedasticity of the errors was satisfied. Last, the variance inflation factor (VIF) and tolerance levels among all the indicators were calculated, and problematic variables were removed.

⁸ By conducting a χ² difference test between the final measurement model and the hypothesized structural model, the result was 2.638 (df = 3), which is smaller than the critical value 7.815 (p <.05). Therefore, we can conclude that the structural model is more favoured than the final measurement model (See Appendix C).

⁹ Four items of trust in national government selected by the purification process have acceptable factor loadings. The latent factor is later converted into a moderator (dichotomous variable) under the structural model.

¹⁰ All the traits were above the generally acceptable criterion (≥.7) suggested by Nunnally (1978). All the constructs fell within the commonly accepted range (.6 ≤ α ≤.7) according to the psychological testing rule for describing internal consistency (Kline, 2005).

¹¹ Steenkamp and Baumgartner (1998) suggest the measurement invariance should be assessed when multiple groups are involved in the statistical analyses. They also require both measurement invariance and partial metric invariance to be supported in order for a comparison of standardized path coefficients to be made across groups (Steenkamp & Baumgartner, 1998). Our analysis followed their procedure for carrying out measurement invariance tests. According to the χ² test, two of the measurement items were not invariant (p <.05) between the groups, indicating partial metric invariance for those constructs. Because partial metric invariance is a sufficient condition for a two-group comparison to be made, we proceeded with the multi-group analysis.

¹² Readers might wonder if there is a moderation with respect to trust in national government on two additional pathways, namely the effects of nuclear stigma on perceived benefits and the effects of perceived risks on the perceived benefits, due to the large differences between path coefficients among high trust and low trust groups on those paths (e.g., Low group −.406 and high group .298 of the effect of nuclear stigma on the perceived benefits, low group −.239 and high group −.868 of the effect of perceived risks on the perceived benefits). In response to this, the researchers conducted additional χ² difference tests for these paths (see Appendix C), and the results did not statistically support moderation effects on the two pathways between stigma of nuclear and perceived benefits, and between perceived risks and perceived benefits.