Policy decision-making, public involvement and nuclear energy: what do expert stakeholders think and why?

Abstract

Despite renewed interest in involving the general public in energy and environmental policymaking, little is known about how policy stakeholders with specialized knowledge and extensive experiences perceive the value of such activity. This study seeks to explore how a series of social, cognitive, and communication factors relates to expert stakeholders’ attitudes toward public involvement in energy policymaking. Using data from a national survey of key stakeholders involved in making high-level decisions on nuclear energy, we find governmental stakeholders are more likely to think public involvement is important when perceiving public opinion as being split. Scientists and non-profit stakeholders, however, attach high-level importance to public involvement regardless of their perception of public opinion. Additionally, the perception of a divided public for the issue of nuclear energy is driven by frequent use of online media. Perceived importance of a salient media agenda – the economics of nuclear facilities – is also significantly related to one's perceived split in public opinion. Implications of the findings with respect to promoting a two-way dialogue among citizens and expert stakeholders in science policymaking are discussed.

Keywords:

• public involvement
• public opinion
• nuclear energy
• online media

1. Introduction

During the past decade, the importance of stakeholder participation in energy policymaking has been increasingly acknowledged, mostly driven by policy trends pursuing sustainable development and collaborative partnership (Reed 2008). For example, the Committee on Radioactive Waste Management (CoRWM), sponsored by the British Department of Energy and Climate Change, included hundreds of lay citizens in its high-level waste management programs and adopted an inclusive approach reflecting the public's ethical concerns (CoRWM 2006). Similarly, the Blue Ribbon Commission on America's Nuclear Future (BRC), established by the Obama Administration in 2009, proposed a consent-based approach to siting future nuclear waste facilities, requiring various government units to protect the interests of affected residents (BRC 2012).

Following these policy initiatives, many researchers and policy practitioners have emphasized the necessity of public outreach and envisioned a transformation of non-expert citizens into informed policy contributors (e.g. Cornwall and Gaventa 2001). For example, it is argued that the public has a democratic right to participate in environmental and energy decision-making (Richards et al. 2006). If the public participation process were perceived to be inclusive and transparent, involved citizens would be more likely to trust policymakers and the decisions they make (Krütli et al. 2010). Public participation also allows expert and non-expert stakeholders to co-construct a set of knowledge, which prompts all involved parties to learn about the legitimacy of each other's views (Reed 2008). Through such a process of social learning, public participation can find new ways for previously disagreeing stakeholders to work together.

In contrast, many other stakeholders, including university scientists and industry engineers, are concerned about lay citizens’ limited scientific sophistication and hence are conservative about their competence in making informed policy decisions (Besley and Nisbet 2013). For example, as shown by a national representative survey of British scientists and engineers at 50 research institutions, the general public has been perceived as less important to engage with than policymakers, private sectors, and educators (Besley and Nisbet 2013). Also, compared with scientists in other disciplines, engineers are less likely to perceive public participation to be important. Engineers are also more likely to view their own research as too specialized to be understood by non-experts (Besley and Nisbet 2013).

Additionally, scientists often view the public as irrational and unsystematic in their thinking, such that people tend to rely on anecdotal evidence, sensations, fears to minor risks, and self-interests to make decisions (see Besley and Nisbet 2013 for a review). Not surprisingly, many of these observations have received empirical support. Researchers have found that lay individuals are likely to rely on heuristics (i.e. mental shortcuts), including value predispositions and mental associations to assess the risks of emerging technologies (Brossard et al. 2008; Brossard and Nisbet 2007). Notably, even scientists and technical experts are prone to heuristic processing, such that scientists tend to rely on their economic ideology as a cognitive shortcut to decide whether research on emerging technologies (e.g. nanotechnology) should be regulated (Corley, Scheufele, and Hu 2009).

Despite escalating debates regarding the role and effectiveness of public participation, we know little about how elite stakeholders, especially those who have played a critical role in initiating participation activities, perceive the importance of such activity for energy policymaking. Previous research on this subject has revealed the critical roles of various factors (e.g. academic training, previous experience, institutional background, etc.) in shaping individual scientists and other expert stakeholders’ perceptions of public engagement (e.g. Besley and Nisbet 2013). Using the nuclear energy issue as a case study, this study intends to identify all relevant stakeholders involved in making high-level decisions on managing the nuclear fuel cycle in the USA, and examines how stakeholders with specialized knowledge and professional experiences develop their attitudes toward public participation as a function of institutional identity, perception of public opinion, and media use.

2. The US nuclear policy and stakeholder engagement

The US fleet of nuclear power plants generates roughly 20% of the country's electricity output. Throughout its history, the discussion of high-level decisions on nuclear energy and other uses of nuclear power has been restricted to a relatively small group of elite stakeholders (Kunreuther and Easterling 1996). With extensive informational and monetary resources, these elite stakeholders were also responsible for implementing policy decisions. Nonetheless, the top-down approach had been proved ineffective, as the disconnect between elite decision-makers and concerned citizens resulted with the nuclear policy facing a stalemate on multiple occasions.

For example, Samuel Walker, the former historian of the Nuclear Regulatory Commission (NRC), describes in his book how the Atomic Energy Commission (AEC) failed to obtain local and state agreement during its first attempt to site a high-level waste repository in Lyons, Kansas in 1970s (Walker 2009). By the middle of the same decade, the Energy, Research, and Development Administration, which was a surrogate for the AEC, surveyed other potential locations for storing commercial nuclear waste. In 1982, the US Congress passed the Nuclear Waste Policy Act, mandating geological disposal for high-level waste and spent nuclear fuel. This legislation also designated the Department of Energy (DOE), the Environmental Protection Agency (EPA), and the NRC to be responsible for managing, regulating, and licensing disposal sites.

In 1987, the Yucca Mountain, Nevada site was decided to be the only long-term geological repository for nuclear waste disposal. In 2002, the DOE approved the Yucca Mountain repository, overriding the State of Nevada's objections (MacFarlane 2003). In 2011, the Obama Administration ended federal funding for the project and withdrew its licensing application. The policy-decision process, however, has been criticized as receiving no scientific justification (Vandenbosch and Vandenbosch 2012).

Many of these unresolved debates have eroded public trust in regulatory agencies and escalated local opposition (Slovic, Layman, and Flynn 1991). As the BRC reflected in its final report, ‘any attempt to force a top-down, federally mandated solution over the objections of a state or community – far from being more efficient – will take longer, cost more, and have lower odds of ultimate success’ (BRC 2012, ix). Inspired by similar experiences, many countries around the globe have developed participation programs to enhance public confidence in the safety of nuclear power facilities and other applications of nuclear technologies.

Specifically, the International Atomic Energy Agency (IAEA) has identified a number of typical stakeholders involved in the overall life cycle of nuclear facilities, ranging from governmental agencies, scientific bodies, and industry to the media, public, and neighborhood states (IAEA 2011). According to this model, stakeholders who are likely to be influenced by the policy decision should be actively involved as well as those who can influence the decision. Also, those who are required by law to participate (e.g. regulatory agencies) need to engage with the so-called non-statutory stakeholders (e.g. state, tribal, and local governments) when making critical decisions. Most importantly, stakeholders with limited technical knowledge but are inevitably influenced by the decision, such as local residents or members of the lay public, should participate in decision-making processes, along with expert stakeholders who either have highly specialized knowledge or extensive professional experiences (IAEA 2011).

3. Predicting expert stakeholders’ perception of public participation

3.1. Role of institutional background

Previously, it has been shown that institutional background and the culture that associates with it may play an important role in shaping policy stakeholders’ perceptions about public participation. For example, expert stakeholders affiliated with governmental agencies are often optimistic about the normative and pragmatic benefits of public participation and have devoted significant efforts in developing constructive programs (Krütli et al. 2010; NEA 1999). Many of these programs aim to integrate a wide range of perspectives and assure that local needs are successfully met (Dougill et al. 2006).

However, while recognizing their responsibilities to support public debates, scientists from universities and national laboratories tend to emphasize the need to educate the public so that people can make rational choices in line with scientific evidence (Neal, Smith, and McCormick 2008). In contrast, stakeholders from for-profit sectors, such as utility companies, lobbying firms and other consulting organizations, often do not feel obligated to engage the public as they usually work directly with regulators and governmental policymakers (Burningham et al. 2007). For this group of stakeholders, the public is often conceived as ‘consumers’ and ‘neighbors’ whose concerns should be allayed instead of being informed (Burningham et al. 2007). In sum, stakeholders from different institutions may have very different attitudes toward a deliberative approach to resolving policy dilemmas.

3.2. Role of perception of public opinion

In addition to institutional background, expert stakeholders’ attitudes toward public participation are also likely to hinge on their assumptions about public opinion climate. The US public opinion on nuclear energy has been split for decades. As of 2012, 44% of the US public favored the increased use of nuclear power, while 49% opposed (Pew Research Center 2012). The opinion split also mirrors the stark partisan divide on nuclear policy, as Republican leaders are generally supportive for new power plants, while Democrats are skeptical about the promise of nuclear power as a sustainable energy source (U.S. House Committee on Science Space & Technology 2011; The Washington Post, March, 2011).

Also, recent national surveys show that a considerable proportion of the US public are unfamiliar with the nature and extent of nuclear risks (Jenkins-Smith 2011). For example, 40% of respondents of a national representative survey believed that spent nuclear fuel can accidentally explode like a nuclear bomb (Jenkins-Smith 2011). Also, 35% believed that nuclear power plants can produce significant amounts of greenhouse gases (Jenkins-Smith 2011).

Given the not-so-ideal opinion climate on nuclear energy, it raises an empirical question of how expert stakeholders affiliated with different institutions view the importance of public involvement as a function of perceived split in public opinion. On the one hand, some expert stakeholders may want to disregard the opinions of a divided public. On the other hand, it also presents significant necessity to inform a divided public with evidence-based information. In addition, as the efficiency and rationality of collective behavior has been increasingly recognized, many researchers have begun to characterize public opinion as a valuable source of information (Page and Bouton 2008). It is worth investigating if expert stakeholders who traditionally endorse public participation still think it is important to involve the public when perceiving the lay opinion as being split.

3.3. Public opinion perception and mediated communications

Moreover, the formation of perception of public opinion is related to one's mediated information environment. Traditional forms of media, including newspapers and television, have provided various cues for audience members to judge public opinion climate, such as the slant of news coverage (Christen, Kannaovakun, and Gunther 2002). In particular, mass media can potentially aid expert stakeholders anticipate changes in public opinion through the news agenda (Mutz and Soss 1997). For example, it has been shown that individuals tend to rely on the most salient issues covered in local media to assess the salience of issues in a community (Mutz and Soss 1997).

Content analyses of national and local newspapers have shown that the use of nuclear power is frequently portrayed through three thematic lenses: the need for alternative energy, the economics of nuclear facilities, and the environmental and health implications of nuclear technologies (Culley et al. 2010; Lazic 2011). Also, local newspapers have focused on the lack of consensus regarding the role of nuclear power in mitigating climate change (Culley et al. 2010). As each of these issues has provided a unique context for news consumers to anticipate the public opinion surrounding nuclear energy, it would be interesting to investigate if mental salience of these issues influence how expert stakeholders perceive public opinion.

With the advent of online media, individuals can infer public opinion based on real-time discussions. In particular, the emergence of self-publishing platforms has provided policymakers and other stakeholders with opportunities to access heated online debates that they may not experience in daily life. For example, a systematic analysis of nuclear-related discussions on Twitter showed that more and more neutral and pessimistic opinions were expressed after the Fukushima Daiichi nuclear accident occurred in 2011 (Li et al. 2013). Frequent access to these discussions may shape how individuals perceive the public opinion in general.

Additionally, the political controversy surrounding nuclear energy and related online discussions can expose audiences to polarized opinions across lines of partisanship. Although political discussions on blogs and Twitter often occur among like-minded users, these boundaries can be easily crossed by increasing use of the ‘Share’ and ‘Favorite’ functions on these platforms (Adamic and Glance 2005). Even for those who stick to specific online sources for information that aligns with their political ideology, the Internet provides many incidental opportunities to encounter opposing opinions (Brundidge 2010). Compared with those who do not often rely on online sources for information, frequent users of online media may perceive more of an ideological divide in public opinion.

In sum, this study investigates how expert stakeholders, defined as those with specialized knowledge and professional experiences in dealing with (energy) policy issues, perceive the importance of public involvement in making decisions related to nuclear energy. In addition, we examine how expert stakeholders develop their perceptions of public opinion as a function of online media use and perceived salience of media agenda.

4. Methods

4.1. Sampling strategy and procedures

Data were collected with a mail survey of 557 US expert stakeholders involved in making high-level decisions on nuclear energy and other uses of nuclear power. The survey respondents were strategically identified from multiple sources to maximize the relevance and heterogeneity of the sample.

First, 404 respondents were identified through a content analysis of Congressional public hearings related to everything ‘nuclear’.¹ Second, in order to recruit key stakeholders at state and local levels, we analyzed the minutes of public meetings held by the Blue Ribbon Commission between 2010 and 2011. In 2010, the BRC held five public meetings across the country to acquire feedback on its recommendations to the Secretary of Energy. Most attendants of these meetings are directly involved in managing nuclear-related programs at local, tribal, and state levels. Third, to gain insights from federal stakeholders, we identified and recruited 63 retired congressional staff members dealing with energy issues from two sources: the official directories of the 109th, the 110th, and the 111th US congress, and the Insider's Guide to Key Committee Staff of the US Congress (2008–2010). Serving as informants for federal legislators, this group of respondents is supposed to maintain unique expertise regarding the development of nuclear energy.

Contact information for all respondents was collected from publicly available records online. Following Dillman's Total Design Method, the mail survey was administered in three ways between July and September 2013. With 137 valid responses, the final corrected response rate of 27.4% was achieved.

4.2. Measures

Perceived importance of public involvement was measured on a five-point scale (1 = ‘strongly disagree’, 5 = ‘strongly agree’), asking respondents to indicate their agreement with each of the following statements: ‘Consent-based approaches for policy decisions related to nuclear energy are important,’ ‘It is important to involve members of the public in decisions about nuclear energy,’ and ‘It is important to involve members of the public in decisions about advanced nuclear fuel cycles.’ The three items were averaged to form an index, with scores ranging from 1 to 5 (M = 4.23, SD = .79, Cronbach's alpha = .81).

Perception of public opinion was measured on a five-point scale (1 = ‘Strongly disagree’, 5 = ‘Strongly agree’) asking respondents to indicate their agreement with the statement ‘US public opinion has been split on the issue of nuclear energy,’ (M = 4.31, SD = .86).

Institutional background was measured by asking respondents to choose the organization that currently employs them from a list of fourteen organizations. All answers were coded into four categories: governmental agencies, including both federal and state ones (N = 54)2²; non-profit organizations, including non-profit policy research institutions and other not-for-profit organizations (N = 38); scientists, including university scientists and those who work for national laboratories (N = 25); and private sector stakeholders, including those who work for consulting firms, lobbying firms, utility companies, and nuclear industry (N = 20). Three dummy variables were created, labeled as ‘university and national laboratory scientists’, ‘non-profit stakeholders’, and ‘private sector stakeholders’ respectively, with governmental stakeholders serving as the reference group.

Perceived importance of media agenda was made up of three variables tapping the level of importance respondents assigned to different aspects of nuclear energy: climate change mitigation, economics of nuclear facilities, and environmental health and safety. The three variables were measured using a four-point scale (1 = ‘not at all important’, 4 = ‘extremely important’) asking respondents how important they thought each of the issues was to the development of nuclear energy: ‘Climate change mitigation’ (M = 3.02, SD = .87), ‘Economics of nuclear power-related facilities’ (M = 3.38, SD = .75), and ‘Environmental health and safety’ (M = 3.24, SD = .76).

Media attention to scientific issues consisted of four variables measuring respondents’ attention to scientific issues across four media platforms with a five-point scale (1 = ‘no attention’, 5 = ‘a lot of attention’): print media, including national newspapers, local newspapers, and magazines (M = 3.09, SD = .66, Cronbach's alpha = .69); television (M = 2.72, SD = 1.15); radio (M = 3.24, SD = 1.21); and online media, including online-only magazines, independent blogs, and online social networks (M = 2.21, SD = 2.21, Cronbach's alpha = .78).

Educational attainment was measured on a four-category scale, ranging from ‘bachelor degree and below’ (coded as ‘1’, 16.8%) to ‘doctoral degree’ (coded as ‘4’, 29.9%). The sample median was ‘2', indicating ‘master degree (M.S./M.A./M.B.A.)’ (38.7%). Discipline was measured by asking respondents to report the field of their highest degree. Two dummy variables were created, including ‘science’ (29.9%) and ‘engineering’ (24.1%) with all other disciplines, including social sciences and business, as reference group. Years of professional experiences was measured with a five-category scale, ranging from ‘less than a year’ (coded as ‘1’) to more than 15 years (coded as ‘5’). Sixty-eight percent of respondents reported having more than 15 years. Gender was a dichotomous variable with female coded as ‘1’ and male coded as ‘0’ (16.3% females).

4.3. Analytical framework

A hierarchical ordinary least squares (OLS) regression model was run for investigating the effects of institutional background and perception of public opinion on stakeholders’ perceived importance of public involvement. Another OLS regression model was run for examining the effects of media attention and issue salience on stakeholders’ perception of public opinion.

For both models, the independent variables were entered based on their assumed causal order. Block 1 consisted of demographics; Block 2 included disciplines; Block 3 included institutional background; Block 4 measured attention to scientific issues across different media outlets; Block 5 assessed perceived importance of media agenda related to nuclear energy; and Block 6 was measuring perception of public opinion. In the model that predicts perceived importance of public involvement, Block 4 and Block 5 were entered as controls for Block 6.

Block 7 for the public participation model influenced three interaction terms capturing how the relationship between public opinion perception and public involvement attitude varies for (1) governmental stakeholders versus non-profit stakeholders; (2) governmental stakeholders versus university/national laboratory scientists; and (3) governmental stakeholders versus for-profit stakeholders. All interaction terms were created by multiplying the standardized value of public opinion perception with each of the three dummy variables for institutional background. This procedure allows us to avoid issues of multicollinearity between interaction terms and their components (Cohen et al. 2003).

5. Results

5.1. Predicting expert stakeholders’ perceived importance of public involvement

In general, 94.2% of respondents somewhat agree or strongly agree that it is important to involve members of the public in decisions about nuclear energy. However, only 73% agree that it is important to involve members of the public in decisions about advanced fuel cycles. Also, 86.9% somewhat or strongly agree that consent-based approaches for policy decisions related to nuclear energy are important. Meanwhile, 88.3% of respondents agree that the US public opinion has been split on the issue of nuclear energy.

Table 1 reports the results for our first model that predicts expert stakeholders’ perceived importance of public involvement. Overall, the model accounts for 32.3% of the variance in the dependent variable. As Table 1 shows, none of the institutional background variables is significantly related to perceived importance of public involvement. In other words, compared with governmental stakeholders, all the other groups of stakeholders did not significantly differ in the amount of importance they attached to public involvement in making decisions related to nuclear energy.

Table 1. OLS regression predicting expert stakeholders’ perceived importance of public involvement in nuclear decision-making.

	Final B
Block 1: Demographics
Years of professional experience	−.05
Gender (female = 1)	.27
Highest degree	.07
Block 2: Disciplines
Science	.04
Engineering	−.06
Block 3: Institutional Background
University and national laboratory scientists (Scientists)	.16
Non-profit stakeholders (Nonprofit)	.24
Private sector stakeholders (Private)	−.24
Block 4: Media attention to scientific issues
Print media	.15
Radio	.10
Television	−.11
Online media	.02
Block 5: Perceived importance of media agenda
Climate change mitigation	.11
Economics of nuclear facilities	.04
Environment health and safety	.11
Block 6: Perception of public opinion
Perceiving public opinion to be split (Split)	.50***
Block 7: Interactions
Scientists × Split	−.45**
Nonprofit × Split	−.38*
Private × Split	−.19
Total R^2	32.3%**

Notes: N = 121.

***p < .01.

**p < .05.

*p < .1.

Cell entries are final unstandardized regression coefficients (B) for all blocks. Unstandardized regression coefficients were reported because they are less sensitive to the relative size of groups and more useful than standardized coefficients for nominal variables (our chief interest) (Cohen et al. 2003).

Perceived split in public opinion is significantly positively related to the dependent variable (B = .50, p < .001). To put it alternatively, the more split respondents perceived in public opinion climate, the more likely they would think public involvement is important. However, this relationship appears to differ between governmental stakeholders and academic scientists, as the interaction between scientists and split is significant. Since the interaction between non-profit stakeholders and split is also significant, the relationship between perception of public opinion and public involvement attitudes varies for governmental and non-profit stakeholders. As Figure 1 shows, while governmental stakeholders were much more likely to believe public involvement is important when perceiving more split in public opinion, such relationship was not evident for either scientists or non-profit stakeholders.

Figure 1. Interactive effects of institutional background and public opinion perception on perceived importance of public involvement in nuclear policy decision-making.

Notes: estimated values, after controlling for all the independent variables, are depicted in the figure (scale only partially displayed). Error bars indicate standard errors.

5.2. Predicting expert stakeholders’ perception of public opinion

Table 2 shows the OLS regression model predicting expert stakeholders’ perception of public opinion. Stakeholders majoring in hard sciences were more likely to perceive public opinion to be split, compared with their counterparts majoring in social sciences and business (B = .38, p < .1). Also, attention to scientific issues on online media was positively related to perceived split in public opinion, indicating that the more online media stakeholders used, the more likely they would perceive public opinion to be split (B = .19, p < .1). Moreover, we are interested in the potential effects of perceived importance of media agenda on stakeholders’ perception of public opinion. Results show that among the three areas, only economics of nuclear facilities was positively related to perceived split in public opinion. In other words, the more importance individuals attached to the economics of nuclear facilities as one prominent agenda covered in mass media, the more likely they would assume public opinion to be split.

Table 2. OLS regression predicting expert stakeholders’ perceived dividedness in public opinion with respect to the issue of nuclear energy.

	Final B
Block 1: Demographics
Years of professional experience	.11
Gender (female = 1)	.07
Highest degree	−.14
Block 2: Disciplines
Science	.38*
Engineering	.11
Block 3: Institutional background
University and national laboratory scientists (Scientists)	−.06
Non-profit stakeholders (Nonprofit)	−.09
Private sector stakeholders (Private)	−.29
Block 4: Media attention to scientific issues
Print media	−.05
Radio	.02
Television	−.05
Online media	.19*
Block 5: Perceived importance of media agenda
Climate change mitigation	−.06
Economics of nuclear facilities	.22*
Environment health and safety	.01
Total R^2 (%)	13.2

Notes: N = 121.

**p < .05.

*p < .1.

Cell entries are final unstandardized regression coefficients (B) for all blocks.

6. Discussion

The importance of public involvement in decision-making related to energy and environmental issues have been increasingly acknowledged among both academic and policymaking communities. Equipped with specialized knowledge and extensive professional experiences, the so-called expert stakeholders have played a critical role in identifying and recruiting all relevant stakeholders into decision-making processes. Expert stakeholders are also expected to inform non-expert stakeholders with evidence-based information and accessible tools for policy decision-making. At the same time, experts and elites can also benefit from the alternative perspectives and lay knowledge offered by non-experts.

Nonetheless, given the somewhat divided opinion climate surrounding the issue of nuclear energy, we knew little about whether or not expert stakeholders would be willing to engage the public in spite of an absence of public consensus. This study systematically identified and surveyed varying stakeholders involved in making high-level decisions on the issue of nuclear energy. We also examined how stakeholders affiliated with different institutions perceive the importance of public involvement in decision-making related to nuclear energy. Our results show that institutional background and perception of public opinion have played important roles in shaping expert stakeholders’ attitudes toward public involvement.

Specifically, perception of divided public opinion is positively related to attitudes toward public involvement, indicating that a major proportion of stakeholders are more likely to endorse the necessity of engaging the public when perceiving split in the overall opinion climate. Moreover, this relationship holds the strongest for governmental stakeholders, which is not surprising given their obligation of taking various groups’ interests into account when making policy decisions. With respect to the perception of public opinion, attention to scientific issues on online media, including online-only magazines, independent blogs, and social networking sites, drive individuals to perceive more split in public opinion. Additionally, expert stakeholders appeared to rely on salient media agendas, such as the economics of nuclear facilities, to assess whether public opinion has been divided on this issue or not. These results illustrate the potentially critical role of information environments in contextualizing stakeholders’ assessment of public opinion.

Before elaborating on these findings, it is important to discuss several methodological limitations. The first concern is related to the strategies for categorizing stakeholders’ institutional background. Currently, the stakeholders were coded into four groups, namely governmental stakeholders, non-profit stakeholders, university and national laboratory scientists, and private sector stakeholders. However, it should be pointed out that none of these groups is homogenous in terms of institutional functions, goals, interests etc. For example, scientists working for national laboratories can also be considered as governmental employees. Also, non-profit organizations, such as some of the prominent think tanks, are as frequently involved in advocating activities as lobbying firms and interest groups (Ahmad 2008).

Despite these concerns, the current strategy is justifiable because it allows us to distinguish between specialized experts, such as university and national laboratory scientists, and policy experts who usually do not hold sufficient technical knowledge. Also, those who worked for non-profit sectors were separated from those who worked for for-profit ones, as previous research has shown that stakeholders’ monetary interests are closely related to their conception of the public (Burningham et al. 2007). Therefore, the current strategy is particularly useful with respect of revealing the varying attitudes held by stakeholders affiliated with different institutions.

The second limitation concerns the generalizability of our findings. While our survey respondents were identified through content analyses of a wide range of policy-related documents, we lacked capability to conduct a search of stakeholders across the board. Nonetheless, most of our respondents were actively involved in policy decision-making at various levels as decision-makers, decision-informants, as well as decision-influencers. In addition, these respondents represent a group of stakeholders who not only has legitimate stakes in the outcome of policy decisions, but also care about the negotiating processes. Given these considerations, the practical value of our findings and analyses should not be overlooked. Besides, given their presence at past public meetings, our respondents are likely to take initiatives in engaging members of the public. An analysis of their attitudes can inform future strategies to undertake such activity.

Another potential concern is related to the relatively small number of valid cases, which may limit our ability to reject the null hypothesis. Currently, we use .1 as the baseline alpha value, which results in a relatively small probability of committing a Type II error and thus increase the power (Cohen et al. 2003). This decision can also be justified based on the presumably small amount of variance in the population of expert stakeholders dealing with nuclear energy. In general, we were very successful in collecting data with a hard-to-reach group of policy stakeholders. To the best of our knowledge, our sample and data have not been matched by any previous study in the literature.

In addition, the way how we measure attitudes toward public involvement is somewhat limited. Public participation can take on many different forms depending on specific needs. Some of the alternative forms of public involvement include information, consultation, collaboration, and empowerment (Kemp, Bennett, and White 2006). While our measure covered multiple facets of public involvement and demonstrated high reliability, future studies should develop more nuanced measures tapping different modes of public involvement. Researchers may also explore if stakeholders’ attitudes toward public involvement would change based on how such activity is defined.

Our results, in many ways, have provided empirical grounding to enduring debates about how policymakers and elite stakeholders view the value of public input in policy decision-making. Experiences in many countries have demonstrated that stakeholders judge the value of public participation based on their institutional functions, goals, and interests. For example, while the normative benefits of such activity has been widely acknowledged, suspicions about the public's scientific sophistication and competence in making rational decisions has lingered in energy policymaking community for decades (Reed 2008).

This study, however, has implied a more complex picture regarding how stakeholders perceive the value of participation activities as a function of their beliefs about public opinion. Specifically, non-profit stakeholders and scientists working for various institutions value public involvement regardless of their perceived split in opinion climate, whereas governmental stakeholders are more likely to embrace the input of a divided public than a united one. Governmental stakeholders usually serve as the conduits between lay citizens and other expert stakeholders and are primarily responsible for initiating any form of public form. As it appears that scientists are less likely to value the opinions of a divided public, governmental policymakers, along with non-profit stakeholders, should play an active role in conveying the potential promise of lay audiences’ perspectives to scientists and other technical experts.

Inspired by previous research on the Internet and the polarizing information environment it creates (Brundidge 2010), our research sheds light on the implications of the online environment on expert stakeholders’ understanding of public opinion. The Internet and emerging information technologies have provided extensive opportunities for expert-public interaction (Lukensmeyer, Goldman, and Stern 2011). However, little is known about how uses of these platforms would impact the dynamics of such activity. Our study offers a baseline understanding of the effects of online media attention on expert audiences. For elite audiences, such as policymakers and stakeholders dealing with energy issues, frequent access to online media might enable them to be aware of the focus of public discussion and fierce debates that they would not have commonly experienced in everyday life (Wojcieszak and Mutz 2009).

The Internet allows its users to actively seek or incidentally encounter polarized opinions along partisan lines, and expert stakeholders might rely on the intensity of discussions and partisan stances as mental shortcuts to estimate the distribution of public opinion. In this regard, the dominant online discourse has a potential to influence the outcome of policy decisions through indirectly affecting how expert stakeholders engage the public. Although traditional forms of media, including print media, radio, and television, may not contain sufficient cues implicitly implying the existence of a divided public, the media agenda featured on these outlets have contextualized expert stakeholders’ assessment of public opinion. Specifically, for those who attach greater importance to the economics of nuclear facilities, they are more likely to believe that public opinion is split. The stalemated negotiations on financing nuclear power plants have received extensive news coverage in the USA. (Culley et al. 2010). It is not surprising to see that expert stakeholders automatically connect this issue to a divided public.

In conclusion, expert stakeholders’ attitudes toward public involvement in decision-making related to nuclear energy is shaped by a range of social, cognitive, and communication factors. The issue of involving members of the lay public in science policymaking merits further exploration. Given the close relationship between public opinion perception and online media use, it is worth exploring the causal links between online discourse of energy issues and elites’ perceptions of public opinion or other dimensions of public sphere. Also, how these factors would ultimately influence the quality and outcome of policy decisions is a more intriguing question to answer.

Notes on contributors

Nan Li, Ph.D., is currently a Joan Bossert postdoctoral fellow at the Annenberg Public Policy Center of the University of Pennsylvania. She receives her Ph.D. in mass communications at the University of Wisconsin-Madison.

Leona Yi-Fan Su is a Ph.D. candidate in Life Sciences Communication at the University of Wisconsin-Madison.

Xuan Liang is a Ph.D. candidate in Life Sciences Communication at the University of Wisconsin-Madison.

Michael A. Xenos, Ph.D., is Professor and Chair of the Department of Communication Arts at the University of Wisconsin-Madison.

Dominique Brossard, Ph.D., is Professor and Chair of the Department of Life Sciences Communication at the University of Wisconsin-Madison. She is also the leader of the Societal Implications of Nanotechnology group in the National Science Foundation (NSF)-funded Nanoscale Science and Engineering Center (NSEC).

Dietram A. Scheufele, Ph.D., is the John E. Ross Professor in Sciences Communication at the University of Wisconsin-Madison, and a Distinguished Visiting Scholar at the Annenberg Public Policy Center of the University of Pennsylvania.

Notes

1. A search term of ‘nuclear’ was used to find public hearings indexed in the ProQuest Congressional Publications database (May 2009 to January 2013).

2. Governmental agencies include Department of Homeland Security, Department of Defense, Department of Energy, Department of State, Environmental Protection Agency, Government Accountability Office, Nuclear Regulatory, state governments, and other governmental agencies.