December 2008

AAmmeerriiccaann SSeeccuurriittyy PPeerrssppeeccttiivveess

Public Views on Energy, Environment, Nuclear Weapons, and Terrorism: 2008

Hank C. Jenkins-Smith

Kerry G. Herron

Conducted with funding and institutional support from Sandia National Laboratories and The University of Oklahoma.

American Security Perspectives

Public Views on Energy, Environment, Nuclear Weapons, and Terrorism: 2008

9

Hank C. Jenkins-Smith, Ph.D.

Professor and Associate Director Center for Applied Social Research

University of Oklahoma

Kerry G. Herron, Ph.D. Research Scientist

Center for Applied Social Research University of Oklahoma

December 2008

Center for Applied Social Research University of Oklahoma

2 Partners Place, 3100 Monitor, Suite 100 Norman, Oklahoma 73072

2

Acknowledgments

The authors wish to express appreciation to the following organizations and individuals whose support made this project possible.

Sandia National Laboratories, Albuquerque, NM

Alton D. Romig, Jr., Ph.D. Senior Vice President and Deputy Laboratories Director for Integrated Technologies and Systems

John M. Taylor Manager, ITS Strategic Office

Richard R. Preston, Jr., Ph.D. Retired

David Cunnington Graphic Artist

University of Oklahoma, Norman, OK

Carol L. Silva, Ph.D. Associate Director, Center for Applied Social Research Assistant Professor, Department of Political Science

Amy S. Goodin Director, OU POLL

Matthew Henderson Consultant, College Station, Texas

Abstract

We analyze and compare findings from matching national surveys of the US general public on US energy and environmental security administered by telephone and Internet in mid-2008. Key areas of investigation include: energy supplies and reliability; energy vulnerabilities and threats, and rela-tionships among security, costs, energy dependence, alternative sources, and research and investment priorities. We analyze public assessments of nuclear energy risks and benefits, nuclear materials management issues, and preferences for the future of nuclear energy in the US. Additionally, we in-vestigate environmental issues as they relate to energy security, to include evolving perspectives on global climate change and relationships among environmental issues and potential policy options.

We also report findings from an Internet survey of the general public con-ducted in mid-2008 that investigates assessments of threats to US security; valuations of US nuclear weapons and nuclear deterrence; perspectives on nuclear proliferation; and support for investments in nuclear weapons capa-bilities. Our analysis of public views on terrorism include assessments of the current threat, progress in the struggle against terrorism, preferences for responding to terrorist attacks at different levels of assumed casualties, and support for domestic policies intended to reduce the threat of terrorism.

3

4

Intentionally Blank

Contents

Front Matter Abstract 3 Contents 5 Executive Summary 7

Chapter One: Introduction and Overview 13 Section 1.1: Research Goals and Objectives 13 Section 1.2: Methodological and Conceptual Considerations 15 Section 1.3: Organization of the Report 22

Chapter Two: Energy and Environmental Security 25 Section 2.1: Issues and Priorities 25 Section 2.2: Perceptions and Preferences of Energy Sources 30 Section 2.3: Views on Nuclear Energy 38 Section 2.4: Preferences for Managing Spent Nuclear Fuel 43 Section 2.5: Views on Global Climate Change 52 Section 2.6: Willingness to Pay for Energy Research and Development 61

Chapter Three: Nuclear Security 71 Section 3.1: Valuing US Nuclear Weapons Capabilities 71 Section 3.2: Public Preferences for the US Nuclear Stockpile 79 Section 3.3: Modernizing the US Nuclear Stockpile 88

Chapter Four: Security from Terrorism 95 Section 4.1: Assessing and Meeting the Threat of Terrorism 95 Section 4.2: Responding to Terrorist Attacks 104 Section 4.3: Balancing Security from Terrorism and Civil Liberties 109

Appendix 1: Research Methodology 115 Section 1: Sampling 115 Section 2: Data Collection 118

5

6

Appendix 2: Energy and Environment Data Summaries 121

Appendix 3: Nuclear Security & Terrorism Data Summaries 157

References 195

Executive Summary

Chapter One: Introduction and Overview

We report findings from parallel telephone and Internet surveys conducted in mid-2008 of US general public views on selected energy security and related environmental issues. We also report findings from an Internet tracking survey of selected nuclear security and terrorism issues conducted in the same time frame. Each of the three surveys build on previous founda-tional studies in this series to show opinion change over time.

Chapter Two: Energy and Environmental Security

Q: How are concerns about key issues evolving, and how are levels of confidence in US energy policies changing? (pg. 25)

Within the set of five persistent issue areas surveyed, mean concerns about security, healthcare, and the environment remain high in absolute terms (all well above midscale), but each appears to be leveling or declining slightly, while mean concerns about energy and the economy are increasing signifi-cantly both in absolute and relative terms. Confidence in meeting future US energy needs is near midscale, but trending lower, while mean satisfaction with current US energy policies overall is well below midscale and declining.

Q: How are different energy sources perceived in terms of attributes, risks, and preferences for meeting future requirements? (pg. 30)

The order from most preferred to least preferred for eight specified sources of energy are as follows: (1) solar; (2) wind; (3) hydroelectric; (4) natural gas; (5) crop based fuels; (6) nuclear generation; (7) oil; and (8) coal. Mean risks of fossil fuels and nuclear energy are rated above midscale and similar in magnitude. Mean risks of renewable sources are rated well below mid-scale. Respondents in five surveys administered via phone and Internet over three years consistently rate the importance of reducing US dependence on foreign sources of energy above 8.5 on a scale from zero (not at all impor-tant) to ten (extremely important). When asked how to balance energy con-servation with energy development, most participants prefer that roughly

7

equal emphasis be placed on each. As to how they would like to see future energy needs sourced, on average their preference is to increase energy from renewable sources from the current six percent to about 50 percent; increase energy from nuclear generation from the current eight percent to 20–25 per-cent; and decrease energy from fossil fuels from the current 85 percent to 25–30 percent.

Q: How supportive are respondents of nuclear energy, and how are perceptions of risks and benefits, including environmental factors, evolving? (pg. 38)

Trends in mean perceptions of nuclear energy risks (as reflected by our com-posite risk index) are holding steady among Internet respondents and declin-ing among phone participants, but means range from 5.5–6.3 on a scale from zero (no risk) to ten (extreme risk). Trends in mean perceptions of nuclear energy benefits (as reflected by our composite benefits index) are holding above a value of seven on a scale from zero (not at all beneficial) to ten (ex-tremely beneficial). When asked to weigh risks against benefits of nuclear energy directly, benefits consistently are judged to outweigh risks among all samples since 2006. Mean support for new reactors at existing nuclear power plants and at new locations are modestly above midscale for all samples, and the gap between the two options is narrowing.

Q: How does the public prefer to manage spent nuclear fuel? (pg. 43)

Most respondents believe the management of spent nuclear fuel poses sub-stantial risks and view the issue as an important factor in considering the future of nuclear energy. But most phone and Internet respondents are mis-informed about how spent nuclear fuel currently is being managed in the US. Fewer than one in four participants know that spent nuclear fuel is be-ing temporarily stored in cooling pools and special containers above ground at designated nuclear power plants. A plurality of respondents think spent nuclear fuel is being stored deep underground at a central facility. When told of present practices and offered balanced arguments for and against four potential policy choices (continue current practices; open a centralized long-term storage facility; open privately managed temporary storage sites; or reprocess spent nuclear fuel), most favor reprocessing and reusing spent nuclear fuel. A proposal to limit the enrichment and reprocessing of nuclear materials to a small number of designated countries who would provide fuel

8

at market prices to others for their nuclear reactors receives only tepid sup-port near midscale.

Q: How are views on global climate change evolving, and how do they relate to support for nuclear energy? (pg. 52)

Most respondents believe average global temperatures are increasing and that the trend poses serious risks to people and the environment. Roughly three out of four participants believe greenhouse gases contribute to global warming, but the certainty with which that belief is held is weakening. Most also believe it is important for the world to reduce greenhouse gas emis-sions. Among those who believe greenhouse gases cause global warming, mean support for nuclear energy increases significantly with understanding that nuclear generation does not produce greenhouse gases.

Q: What drives willingness to pay (WTP) for research and develop-ment of alternative energy sources, and how is WTP evolving? (pg. 61)

Public WTP for energy research and development has declined markedly since the beginning of our survey series in 2006, but Americans neverthe-less remain willing to pay an estimated $23 annually per household. Over the estimated 210 million households in the US in 2008, this would amount to over $2.5 billion per year in revenues. The dramatic decline in WTP ap-pears to have resulted from a combination of factors, including the rise in the relative priority given to economic issues; a decline in trust for govern-ment; and a reduction in the perceived importance of investment in crop-based fuels.

Chapter Three: Nuclear Security

Q: How are public assessments of the relevance and value of US nu-clear weapons capabilities evolving? (pg. 71)

Our respondents continue to believe that nuclear weapons are important for deterring the use of weapons of mass destruction, discouraging proliferation of nuclear capabilities to terrorists, and maintaining US military superiority and international influence. Support for retaining US nuclear weapons in-creased 18 percent between 1993 and 2001, then stabilized about five per-

9

cent lower than its peak immediately following the terrorist attacks of 9/11. Today, the importance of retaining US nuclear weapons averages about 7.5 on a scale from zero (not at all important) to ten (extremely important). Mean support for investing in nuclear weapons infrastructure has declined from its peak in 2001, but remains slightly above midscale and is about 14 percent higher than when first measured in 1993.

Q: What is the perceived desirability and feasibility of nuclear aboli-tion, and what directions are preferred for the future of the US nuclear weapons stockpile? (pg. 79)

While most respondents consider it desirable to eliminate all nuclear weap-ons worldwide, a large majority does not consider nuclear abolition to be feasible and rejects unilateral US nuclear disarmament. After being in-formed that the US and Russia have agreed to reduce the number of opera-tionally deployed strategic nuclear weapons for each side to between 1,700 and 2,200 by the year 2012, most participants are open to reducing below 1,700 if Russia agrees to matching and verifiable reductions, but most do not want the US to have fewer than any other country. When asked how low the US should go, the mean response is 1,425, and the median response is 1,700. Only four percent prefer zero; 28 percent prefer the maximum al-lowed at 2,200. These views were recorded prior to the Russian incursion into Georgia in August 2008.

Q: How are prospects for modernizing US nuclear weapons viewed, and what kinds of modifications to existing weapons constitute percep-tions of “new” nuclear weapons? (pg. 88)

When presented with balanced arguments for and against modernizing the US nuclear stockpile, 50 percent support developing new nuclear weapons, 23 percent are undecided, and 26 percent oppose new weapons. Support is cost sensitive, and increases with expectations of long-term savings. Most respondents consider only modifications to existing nuclear weapons that change operational applications or weapons’ effects to constitute “new” nu-clear weapons. Modifications to enhance reliability, safety, or weapons’ se-curity that do not alter weapons’ effects are not considered by most respon-dents to constitute “new” nuclear weapons. Mean support for spending to develop and test new nuclear weapons has increased about 37 percent since 1993 to near midscale in 2008.

10

Chapter Four: Security from Terrorism

Q: How do respondents assess specific dimensions of the threat of ter-rorism today, and how confident are they in our abilities to prevent fu-ture attacks in the United States? (pg. 95)

Mean assessments of the overall threat of terrorism in the US remain well above midscale but have been declining since 9/11. Biological terrorism is perceived to pose the greatest threat today. On average, the contemporary threat of terrorism worldwide is judged significantly higher than the threat of terrorism in the US. However, the outlook for terrorism in the US over the next decade is judged higher, on average, than today’s US threat. Mean as-sessments of security at US borders and seaports remain below midscale but have increased since 2006. Mean confidence in preventing terrorist attacks in the US has been slowly increasing since 2005 to near midscale. Mean confi-dence in eventually “winning” the war on terrorism remains below midscale, but increased in 2008.

Q: How does public support for differing levels of response to terrorist attacks vary with assumed levels of US deaths? (pg. 104)

At three assumed levels of US deaths (10; 1,000; or 10,000), mean support for diplomatic and economic non-lethal responses averages above 5.6 on a scale from one (strongly oppose) to seven (strongly support). At the same three levels of assumed US deaths, mean support for the use of military force averages below midscale. But mean support for conventional air strikes is significantly higher than mean support for military invasion at each of the three levels. Mean support for lethal response measures in-creases with assumed levels of US deaths. Mean support for using US nu-clear weapons is well below midscale and significantly lower than for con-ventional air strikes or military invasion. However, support for a nuclear response does increase with higher levels of assumed US deaths from ter-rorist attacks, and mean support for the nuclear option is significantly higher in 2008 than when first measured in 2005.

11

12

Q: How do respondents relate protections of civil liberties with security from terrorism? (pg. 109)

Tolerance for intrusive measures designed to reduce the threat of terrorism remains steady and near midscale. Respondents are least tolerant of gov-ernment collecting information about citizens’ behaviors and forcing DNA samples. They are most tolerant of restricting immigration, requiring na-tional identification cards, and monitoring phone conversations of Ameri-cans suspected of involvement in terrorism.

Most respondents prefer that liberty and security are balanced evenly, but mean perceptions are that government currently is placing more emphasis on providing security than protecting civil liberties. Mean preferences for liberty over security are significantly higher among men, college graduates, strong liberals, strong Democrats, those who emphasize egalitarian values, and those who prefer foreign policy that emphasizes cooperation with inter-national institutions and greater reliance on diplomacy.

Chapter One Introduction and Overview

his report presents findings from parallel Internet and telephone surveys of 1,767 Internet participants conducted May 28–30, 2008 and 608 phone respondents conducted between April 11 and June 26, 2008.

These surveys investigate US general public views on selected energy security and related environmental issues.

TWe also report findings from an Internet tracking survey of 1,923 individuals conducted June 12–14, 2008, on selected nuclear security and terrorism issues.

Each of the three surveys builds on comparative baselines established in 2005 (nuclear security and terrorism), 2006 (energy and environmental security), and continuing surveys in 2007. We also build on prior foundational research con-ducted between 1993 and 2005.1 Financial and institutional support for this study was provided by Sandia National Laboratories and the University of Oklahoma.

Section 1.1: Research Goals and Objectives

esearch goals are organized along two research tracks involving four related dimensions of security. The energy and environmental secu-rity track consists of parallel phone and Internet surveys conducted in

even-numbered years, supplemented by an Internet only tracking survey in odd-numbered years. Similarly, the nuclear security and terrorism track con-sists of parallel phone and Internet surveys conducted in odd-numbered years, plus an Internet only tracking survey in even-numbered years. All are designed to provide coordinated research and are intended to measure and analyze evolving public understandings of four interrelated dimensions of security: energy security, environmental security, nuclear security, and secu-rity from terrorism.

R

1 For the baseline study on nuclear security and terrorism, see Herron and Jenkins-Smith 2006a; for the baseline study on energy and environmental security, see Jenkins-Smith and Herron 2007. Each is available on-line at: http://casr.ou.edu/nsp. Findings from previous surveys on related is-sues published between 1994 and 2004 are summarized in Herron and Jenkins-Smith 2006b.

13

http://casr.ou.edu/nsp

Energy and Environmental Security

Our primary research goals for this track are to analyze public views about contemporary energy security and associated environmental issues and to identify trends in public perceptions and preferences relevant to the evolution of related US policies. Specific research objectives include the following:

• Employ a split survey design that compares telephone and Internet data collections to meet two methodological objectives.

– Where appropriate, map backward to selected baseline questions asked in previous surveys in this series for continued trend analyses and de- velop new questions intended for repeated application in future surveys.

– Compare responses collected by telephone with responses to the same questions collected via the Internet to monitor the comparability and validity of telephone and Internet survey methods.

• Identify and analyze public perceptions of US energy security, to include: energy supply and reliability; energy vulnerabilities and threats; and relative risks and benefits of fossil fuels, nuclear energy, and renewable sources. Specifically investigate relationships among energy security, costs, energy dependence, alternative sources, and research and investment priorities.

• Investigate environmental issues as they relate to energy security, to include expected implications of global climate change, willingness to pay for en-ergy research and development and reducing greenhouse gas emissions, and relationships among environmental issues and potential policy options.

• Analyze emerging changes and trends in public views on nuclear energy, to include risks, benefits, policy preferences, research and investment priori-ties, and public trust. Specifically investigate understandings and prefer-ences regarding nuclear materials management and disposal issues.

• Analyze public perceptions of nuclear fuel cycle and waste management options, including regional temporary storage, reprocessing, permanent disposal, and transportation to temporary or permanent storage facilities.

• Analyze belief systems among members of the US general public and their relationships to views on energy and environmental security.

14

Nuclear Security and Terrorism

For this track, our primary research goals are to analyze public views about the evolving nature of nuclear security and terrorism and to identify trends in public perceptions and preferences relevant to the evolution of related US security policies. Specific research objectives include the following:

• Where appropriate, map backward to selected baseline questions asked in previous surveys in this series for continued trend analyses and develop new questions intended for repeated application in future surveys.

• Identify emerging trends in public perceptions of US nuclear weapons policies and selected national and international security issues. Examine evolving US public assessments of risks, benefits, policy preferences, and research and investment priorities associated with nuclear weapons and strategic security.

• Measure public views on modernizing the US nuclear stockpile and inves-tigate public perceptions of what kinds of modifications to existing nuclear weapons constitute perceptions of “new” nuclear weapons. Monitor trends in public beliefs about the desirability and feasibility of nuclear weapons abolition.

• Investigate concepts of multidimensional security and under what conditions threats to national security warrant varying levels of public sacrifice. Specifi-cally investigate casualty tolerance as it relates to combating terrorism.

• Identify and analyze trends in public concerns about homeland security, including multidimensional public assessments of the threat of terrorism and US policies to prevent and respond to terrorism. Include public views on the security dimensions of illegal immigration.

• Analyze belief systems among members of the US general public and their relationships to views on nuclear security and terrorism. Specifically in-vestigate public beliefs about balancing security and liberty.

Section 1.2: Methodological and Conceptual Considerations

W e design all phases of the larger research project to support mul-tidimensional analyses, including quantitative methods such as descriptive, relational, and trend analyses. The split survey de-sign for energy and environmental security in 2008 includes telephone in-

15

terviews conducted between April 11 and June 26 with 608 respondents randomly chosen nation-wide and 1,767 surveys employing the same ques-tions administered via the Internet May 28–30. Investigations of nuclear security and terrorism in 2008 consist of an Internet tracking survey con-ducted June12–14 with 1,923 participants responding to the same baseline questions asked by phone and Internet in 2007. Sampling methods, collec-tion procedures, and cooperation rates are discussed in Appendix 1.

Conceptualizing Security

The term “security” is associated with contextual meanings that are so broad and variable that some scholars consider it to be an “essentially con-tested concept” (Buzan 1991, Freedman 1992, Gallie 1962, Rothschild 1995). Like other complex ideas such as power, justice, peace, and freedom, the concept of security includes an ideological dimension that reduces the utility of empiricism for resolving differences in definitional and conceptual explanations (Buzan 1991; Little 1981). Even those who specialize in secu-rity studies cannot agree on the boundaries of the concept or of the field of study. To some who take a more classically narrow approach, security re-lates to matters of the state and its military capabilities—particularly the use of force (Buzan, Waever, and de Wilde 1998). But since the end of the Cold War, the concept of security has broadened to include conventions associ-ated with many aspects of globalization and humanitarian concerns, such as hunger, health, human rights, economics and trade, global climate change, and international system stability (Fierke 2007). Some, such as Buzan (1991) and Fierke (2007) caution that the proliferating conceptual applica-tion of the term “security” to new fields and new concerns may locate agency in states rather than in institutional or individual actors in specific fields, and some issues may become militarized even though a political so-lution may be more appropriate.

While a detailed examination of the concept of security is beyond the scope of this brief discussion, it is useful note a few key points. Essentially, per-ceived security is about feeling safe from harm or danger, and actual secu-rity is about being safe. When measuring and analyzing public opinion, we are dealing with perceptions and beliefs, and thus at the individual level of analysis, security is a feeling that is inherently subjective to individual con-texts and beliefs. At a social level, security is a normative political con-

16

struct. It is assessed by governmental agencies and political leaders, and is partially a function of policy processes. While some empiricism may be ap-plied, there remain large areas of subjective interpretation of public security that become the bases for official judgments and policies. These areas of subjectivity are the focus of intense public debate in which the views of ex-perts and those of the general public must be considered by policy makers.

One of the most critical aspects of defining and understanding the meaning of security is to recognize that it is heavily dependent on risk or threat. Theoretically, in the absence of some real or imagined risk or threat, secu-rity would be maximized, but actually, under such a theoretical construct (which is not realistically plausible), security would have no meaning at all. Edkins (2003) contends that the human desire for perfect security from all threats to our existence is illusory, and some degree of insecurity is inherent to all life—including human existence. Fierke (2007, 8) argues that: “The search for perfect security is not merely illusory, but becomes part and par-cel of the problem, that is, it contributes to the production of insecurity and the construction of threats.”

If it is the imagined and real sources of risks and threats that give the concept of security meaning, it follows that one of the most useful ways of conceiving security is in relation to perceived and actual risks and threats. Following the insightful conceptualization of security by Arnold Wolfers (1952), perhaps security can be best understood as the inverse of risk/threat. Because there are some risks and threats over which no individual or government has control (such as the threat of eventual death), comprehensive and enduring security is impossible. Because security takes its meaning from the absence of risk/threat, and because it is impossible to prove why something did not occur, attributing the sources and causes of security is problematic. We may presume the rea-sons a threatening event, such as interstate nuclear war, has yet to occur relate to deterrence based on mutually assured destruction, but we cannot know that is the sole or even primary reason. Similarly, we cannot know for sure why large-scale acts of terrorism have not occurred in the United States from Sep-tember 11, 2001 to the time of this writing. We can make assumptions about the effectiveness of preventive measures and about terrorist capabilities and motivations, but we cannot prove why another act of the scale of 9/11 has not yet occurred. From this line of reasoning, we conclude that the concept of se-curity is based on individual feelings and political assumptions and assess-

17

ments of risks and threats. This becomes key when considering how to meas-ure and track security.

Because of the essentially contested nature of the concept of security, be-cause our understanding of it is based on assumptions about risks and threats, and because of the growing application of the concept of security to more fields and policy domains, we need to carefully delineate those di-mensions being studied in this project. As previously noted, we are limiting our investigation and analysis to public understandings of four interrelated dimensions of security.

• Energy security includes energy dependence, adequacy of energy sources and supplies, threats and vulnerabilities to energy access, nuclear energy risks and benefits, alternative energy sources, and research and development into future energy requirements and options, including willingness to pay for energy research and development.

• Another dimension of security is the growing importance of environmental issues as they relate to traditional concepts of physical security, economic se-curity, and energy management. Of particular interest in this dimension is global climate change (another contested concept) and how public assess-ments of its dynamics are evolving.

• Nuclear security encompasses nuclear weapons and their development, management, modernization, and uses; nuclear materials and their produc-tion, applications, and safeguards; nuclear proliferation and associated im-plications; and public perceptions of and support for policies relating to each of these aspects of nuclear security.

• Terrorism and its implications for all levels of security includes public under-standings of the various threats posed by terrorism, assessments of ongoing efforts to prevent and combat terrorism, and the effects of terrorism on key societal values such as freedom and liberty.

Interrelationships

We consider these four dimensions of security to be closely related and in-teractive, and one of our long-term goals is to better understand how fellow citizens relate concepts and beliefs associated with multiple dimensions of security. Given the baselines established in each of our two research tracks investigating four dimensions of security, we are now able both to probe

18

more deeply into their perceived connectedness and to monitor trends in relative public views. Some areas seem obviously to be closely related, such as nuclear weapons and the potential for their use in terrorism. Others may be somewhat less clear, such as the relationships among energy independ-ence, fossil fuels, and global warming. Still others are much more subtle, such as the relationships of porous borders and illegal immigration with se-curity from terrorism and with the social and economic implications of the associated labor pool. Through repeated and refined measurements, we pur-sue more detailed examination of how Americans relate these four dimen-sions, the degree to which they see crosscutting security implications, and how long-term trends evolve.

Measuring Security

Given the previously noted complexities involved in defining and conceptu-alizing security, direct questions about security and how secure people feel are problematic. First, one must specify what level of analysis is being asked: individual, national, regional, global? Next, it is quite possible for individuals to feel secure from terrorism or interstate war, and yet feel inse-cure about economic well-being, health, the costs of energy, environmental issues, or any number of other aspects affecting security at the individual level. And even when respondents are asked to assess security at a specific level of analysis, it is difficult for them to separate personal feelings and concerns at the individual level from those of the social and political groups to which they belong at higher and more aggregated levels of analysis. To address direct questions about security means that respondents must make several assumptions. First is the level of analysis; is it security for me per-sonally, or for my family, or for my community, or for my nation, or for everyone? The next cognitive demand is to evaluate security from what or from whom, which requires assumptions of specific sources of risks or threats and their nature. Then there is the temporal dimension of security, which raises issues of immediate and contemporary security versus longer-term and future security. This dimension is particularly relevant to discus-sions of energy and environmental security. There also is the important question of whose security is being threatened or strengthened and who is bearing what proportion of the costs. Security measures for one group often impinge on the security of other groups. This is particularly relevant in con-sidering security among different nation states or when dealing with issues

19

such as immigration. Some people may be very accepting of intrusive or re-strictive measures if the associated costs seem likely to be born by others. While the designs of questions can specify some of these requirements, di-rect inquiries about security still demand cognitive compartmentalization and individual assumptions on the part of respondents, all of which can af-fect response validity.

For these and related reasons, we think it is preferable methodologically for purposes of opinion survey research to conceive of security and measure it as the inverse of risk or threat. As argued above, one of the many ways to con-ceive of security is as the absence or minimization of threat. Because threats can be more discretely defined and specified, we hypothesize that respon-dents are better able to compartmentalize and separately assess threats of dif-ferent types acting at different levels of analysis than they are to assess con-ceptual questions about the more difficult to specify and more variable concept of security. While we have included a few direct inquiries about se-curity assessments, we focus more intently on exploring public perceptions of threats and risks, the inverse of which can be used more reliably to repre-sent feelings of security.

Phone vs. Internet Surveys

There are two major trends in opinion survey research that seem especially relevant to our long-term goals in this project. First, the representativeness of and access to mass publics in the developed world via wired telephony is de-clining as more households take advantage of wireless communications and depend less on wired landlines. The number of US households with wired phone connections is declining even while our population continues to grow. The most recent estimates available from the federal government concerning telephone usage indicate that during the last six months of 2007, nearly one out of every six American homes (15.8 percent) did not have a landline tele-phone, but did have at least one wireless telephone. During the same period, approximately 14.5 percent of all adults (more than one out of seven) lived in households having only wireless phones (Blumberg and Luke 2008). It is not reliably known what percentage of households have both wired and wireless connections, and the issue of telephone penetration is becoming even more complex with the introduction of cable systems that carry wired digital phone services that, in some cases, may be channeled through wireless routers and

20

handsets. Technical developments are occurring so rapidly as to blur the tra-ditional understandings of wired, wireless, and cellular telephony.

Regardless of definitional issues, an increasing proportion of households are depending exclusively on cell phones and wired and wireless Internet con-nections for voice and text communications. Since random surveying via telecommunications that incur costs per call that are born by the respondent present legal and expense sharing issues (Brick, et al. 2007) that make sur-veying cell phones impractical (raising safety concerns and requiring the use of monetary incentives), the proportion of the US population that can be reached for phone interviews is declining and becoming relatively less repre-sentative of the population at large (compared to previous decades when wired telephony was the standard for most US households). For example, important age differences are apparent in phone availability and usage pat-terns. Following are the approximate percentages of varying age groups liv-ing in homes having only wireless phone service: ages 18–24, 31%; ages 25–29, 35%; ages 30–44, 16%; ages 45–64, 8%; ages 65 and over, 2%. Men, adults living in poverty, Hispanic, and African American adults are more likely than women, higher income groups, and non-Hispanic white adults to live in households having only wireless service. Some US regional differ-ences also are apparent, with adults living in the south and midwest being more likely to have only wireless service.

The second trend is that access to the Internet continues to grow and is in-creasingly making the proportion of US households that can be surveyed via the Internet larger and more representative of the parent population. Neil-son/Net Ratings estimates that as of June 2008, 72.5 percent of the US popu-lation had access to the Internet (Internet World Stats 2008). The PEW Inter-net and American Life Project (Horrigan 2008) estimates that in early 2008, approximately 55 percent of all adult Americans had a high speed Internet connection at home, an eight percent increase from the previous year. But Internet access continues to be uneven, with rural areas having lower levels of access than metropolitan areas. There also are differential access patterns among various population subgroups, such as racial and ethnic minorities and lower socioeconomic groups.

These trends can be thought of as two lines of public representativeness, one of which is trending downward while the other is trending upward. Where and when they will cross (or have crossed) cannot be known precisely, but

21

the implication for opinion survey research is that neither phone-only nor Internet-only surveys should be relied upon exclusively for understanding mass public opinion in the United States. These trends have important impli-cations for sampling, response rates, and survey validity, and in today’s tran-sitional environment, it is prudent to comparatively use both survey collec-tion methods. Accordingly, we are employing the previously described cross-modal methods to establish and compare views among both subsets of the population—those who continue to be accessible via wired phone connec-tions, and those who increasingly choose to communicate via other means, including the Internet. We believe doing so supports more rigorous findings whose validity is reinforced.2

Section 1.3: Organization of the Report

Chapter Two analyzes multiple dimensions of energy and environmental secu-rity by addressing the following inquiries:

• How are concerns about key issues evolving, and how are levels of confi-dence in US energy policies changing?

• How are different energy sources perceived in terms of attributes, risks and preferences for meeting future requirements?

• How supportive are respondents of nuclear energy, and how are percep-tions of risks and benefits, including environmental factors, evolving?

• How does the public prefer to manage spent nuclear fuel?

• How are views on global climate change evolving, and how do they relate to support for nuclear energy?

• What drives willingness to pay (WTP) for research and development of al-ternative energy sources, and how is WTP evolving?

Chapter Three analyzes issues relating to nuclear security by addressing the following three inquiries:

2 For a more detailed discussion comparing phone and Internet survey methods, see Jen-kins-Smith and Herron 2007.

22

23

• How are public assessments of the relevance and value of US nuclear weapons capabilities evolving?

• What is the perceived desirability and feasibility of nuclear abolition, and what directions are preferred for the future of the US nuclear weapons stockpile?

• How are prospects for modernizing US nuclear weapons viewed, and what kinds of modifications to existing weapons constitute perceptions of “new” nuclear weapons?

In Chapter Four, we focus on security from terrorism by addressing the fol-lowing questions:

• How do respondents assess specific dimensions of the threat of terrorism to-day, and how confident are they in our abilities to prevent future attacks in the United States?

• How does public support for differing levels of response to terrorist attacks vary with assumed levels of US deaths?

• How do respondents relate protections of civil liberties with security from terrorism?

Appendix One describes sampling, data collection, and associated research methods. We also provide illustrations of the demographic representativeness of respondents compared to US national population parameters.

Because there are many more survey questions than can be discussed in this report, we provide two appendices listing all the questions contained in our latest surveys. In Appendix Two, we provide a comprehensive listing of ques-tions asked in the parallel phone and Internet surveys on energy and environ-mental security. Response frequencies and central tendencies are displayed.

Appendix Three provides a comprehensive listing of questions asked in our Internet tracking survey on nuclear security and terrorism. Here too, we de-scribe distributions of responses and central tendencies.

Chapter Two Energy and Environmental Security

In this chapter, we investigate the following questions about energy security and related environmental issues.

• How are concerns about key issues evolving, and how are levels of confi-dence in US energy policies changing?

• How are different energy sources perceived in terms of attributes, risks and preferences for meeting future requirements?

• How supportive are respondents of nuclear energy, and how are percep-tions of risks and benefits, including environmental factors, evolving?

• How does the public prefer to manage spent nuclear fuel?

• How are views on global climate change evolving, and how do they relate to support for nuclear energy?

• What drives willingness to pay (WTP) for research and development of al-ternative energy sources, and how is WTP evolving?

Section 2.1: Issues and Priorities

iven the complex interactions among changing conditions in the struggle against terrorism in Iraq and Afghanistan, rising energy costs, and a troubled US economy, we wanted to better understand

how public concerns about such dynamic issues are evolving. While a com-prehensive listing of all major competing issues is not feasible, we have tracked relative concerns about five key issue areas since 2005 using the following inquiry.

GLead-in: For each of the following issues, please rate your level of concern about the issue using a scale from zero to ten, where zero means you are not at all concerned, and ten means you are extremely concerned. How concerned are you about:

E4: Threats to national security, including terrorism?

24

E5: The delivery and cost of healthcare in the US?

E6: The availability and cost of energy in the US?

E7: The quality and the stability of the environment?

E8: The state of the economy, including jobs and inflation?

We compare trends in mean concerns for each in Figures 2.1—2.5.

Figure 2.1: Mean Concerns About US National Security, Including Terrorism

7.91

7.71 7.86

7.96

7.75

7.0

7.5

8.0

8.5

9.0

2006 2007 2008

Web

Phone

E4: 0 = not at all concerned—10 = extremely concerned

Figure 2.2: Mean Concerns About Delivery and Cost of US Healthcare

8.47

8.29 8.41

8.43 8.50

7.0

7.5

8.0

8.5

9.0

2006 2007 2008

Phone

Web

E5: 0 = not at all concerned—10 = extremely concerned

25

Figure 2.3: Mean Concerns About Availability and Cost of Energy in US

Figure 2.4: Mean Concerns About Quality and Stability of Environment

Figure 2.5: Mean Concerns About the Economy

7.62

8.54

7.92

7.80

8.50

7.0

7.5

8.0

8.5

9.0

2006 2007 2008

Phone

Web

E8: 0 = not at all concerned—10 = extremely concerned

7.50

9.0

2006 2007 2008

Phone

Web

7.45

7.527.63

7.81

7.0

7.5

8.0

8.5

E7: 0 = not at all concerned—10 = extremely concerned

8.09

8.41 8.31

7.0

7.5

8.0

8.5

9.0

2006

8.38

8.61

Web

Phone

E6: 0 = not at all concerned—10 = extremely concerned

2007 2008

26

Mean concerns about these five major issue areas have remained well above midscale in absolute terms in each of our three measurement periods, but while concerns about national security, healthcare, and the environment have remained relatively stable, on average, concerns about energy and the environment have grown substantially (p < .0001), with increases in means from 2006 to 2008 being statistically significant among respondents to both survey modes. Concerns about energy increased 2.4 percent among Internet respondents and 3.6 percent among phone participants, while concerns about the economy increased more markedly, registering 7.3 percent among Internet respondents and 12.1 percent among phone participants.

In terms of ordinal rankings, Table 2.1 shows how mean concerns about energy and economic issues have trended upward, while concerns about the environment have remained lowest among these five areas, and concerns about national security have moved relatively lower. Notice that concerns about energy and the economy equal or exceed concerns about healthcare in our 2008 surveys.

Table 2.1: Ranking Mean Issue Concerns

Issues

Phone 2006

Web 2006

Web 2007

Phone 2008

Web 2008

National security, including terrorism 3 4 3 4 5 Healthcare 1 1 1 3 2 Energy 2 1 2 2 1 Environment 5 5 5 5 4 Economy 4 3 4 1 2 Given growing concerns about the availability and cost of energy in the US, we chart in Figures 2.6 and 2.7 responses to the following two questions about confidence in meeting future energy needs and satisfaction with cur-rent US energy policies.

• E9: Using a scale from zero to ten, where zero means you are not at all confident and ten means you are completely confident, how confident are you that there will be adequate sources of energy to meet the energy needs of the US during the next 20 years. Please think about US energy needs overall, including transportation, heating, electricity, and other energy re-quirements when considering your answer.

27

• E10: As you may know, US energy policies generally deal with such is-sues as the sources and adequacy of energy supplies, the costs of various types of energy, and the environmental implications of using energy. Us-ing a scale from zero to ten, where zero means not at all satisfied and ten means completely satisfied, how satisfied are you with current US energy policies overall?

Figure 2.6: Mean Confidence in Meeting Future US Energy Needs

5.97

5.46

5.16 5.38

4.85

3.0

4.0

5.0

6.0

7.0

2006 2007 2008

Phone

Web

Midscale

E9: 0 = not at all confident—10 = completely confident

Figure 2.7: Mean Satisfaction With Current US Energy Policies Overall Mean confidence in meeting future US energy needs is near or above mid-scale but trending downward while mean satisfaction with current US en-ergy policies is substantially below midscale and trending further down-ward. In both cases, the declining trends are statistically significant (p < .0001) for both phone and Internet participants.

3.97

3.54 3.77 3.80 3.36

3.0

4.0

5.0

6.0

7.0

2006 2007 2008

Phone

Web

Midscale

E10: 0 = not at all satisfied—10 = completely satisfied

28

Short Answer

Q: How are concerns about key issues evolving, and how are levels of confidence in US energy policies changing?

Within the set of five persistent issue areas surveyed, mean concerns about security, healthcare, and the environment remain high in absolute terms (all well above midscale), but each appears to be leveling or declining slightly, while mean concerns about energy and the economy are increasing signifi-cantly both in absolute and relative terms. Confidence in meeting future US energy needs is near midscale, but trending lower, while mean satisfaction with current US energy policies overall is well below midscale and declining.

Section 2.2: Perceptions and Preferences of Energy Sources

iven recent developments in energy prices and the growing political debate about fossil fuels and alternative energy sources, we wanted to sketch a picture of how perceptions of the attributes of competing

sources of energy are evolving. In our 2008 Internet survey, we asked respon-dents to choose from among a list of six pairs of polar opposite attributes for each of eight different sources of energy. Participants could identify as many or as few of the attributes as they wished. Following are the instructions provided.

GLead-in: Now we want to know what kinds of qualities you associate with dif-ferent sources of energy. Please check the descriptions you associate with each of the following sources of energy. You can check as many boxes as you want.

For each of the eight sources of energy, we provided the following six pairs of polar attributes.

Clean — Dirty Safe — Dangerous

Renewable — Perishable Plentiful — Scarce

Affordable — Expensive Preferable — Not Preferable

Respondents could select only one of each of the two paired attributes but were not required to record a response to any pair. For example, a partici-

29

pant could select either “Clean” or “Dirty” (but not both), or could choose not to check either attribute of that pair. The list of attributes was provided for each of the following eight energy sources presented in random order:

• E32: Energy from coal

• E33: Energy from the wind

• E34: Energy from the flow of water

• E35: Energy from oil

• E36: Energy from solar panels heated by sunlight

• E37: Energy from natural gas

• E38: Energy from crop-based fuels

• E39: Energy from nuclear power plants

In most cases, respondents chose to check one of each of the paired attrib-utes for each energy source, and in Table 2.2 we summarize the percentages of respondents who chose each attribute.

Table 2.2: Comparing Perceptions of Energy Attributes (Web 2008)

Attributes (%)

Solar

Wind

Hydro

NaturalGas

Crops

Nuclear

Oil

Coal

Clean 99 97 96 78 79 54 10 9 Dirty 1 3 4 22 21 46 90 91

Affordable 96 96 90 36 77 59 14 19 Expensive 4 4 10 64 23 41 86 81

Renewable 58 75 80 41 43 40 10 54 Perishable 42 25 20 59 57 60 90 46

Safe 96 96 95 54 85 23 36 36 Dangerous 4 4 5 46 15 77 64 64

Plentiful 85 86 79 52 59 62 21 52 Scarce 15 14 21 48 41 38 79 48

Preferred 92 91 90 60 58 39 20 19 Not Preferred 8 9 10 40 42 61 80 81

30

Note that the energy sources are listed in order of respondent preference, with solar, wind, and hydroelectric sources being the most preferred overall and oil and coal being the least preferred sources. These groupings are quite robust, and though there is some variation in the order of preference for the eight energy sources, the groupings of most preferred sources (solar, wind, hydro) and least preferred sources (coal and oil) do not vary with gender, education, age, or even political partisanship (strong Republicans rate hydro first, followed by wind and solar).

While these distributions represent impressionistic reactions to paired ex-tremes, they suggest how competing energy sources currently are perceived by members of the general public we surveyed, and they suggest certain as-pects of the energy equation that may not reflect the technical judgments of more expert observers. For example, solar and wind energy are perceived as more affordable than nuclear generation and coal, but purely economic com-parisons of costs per unit of energy would suggest that for electricity, solar and wind generation currently are more expensive than either nuclear or coal generation. Note also that nuclear generation is perceived to be more danger-ous than that of coal, but objective analyses would show that many more deaths are attributable to the burning of coal than to nuclear generation (Cra-vens 2007). There also appears to be some perceptual confusion about the term “renewable,” with fewer than half of respondents perceiving crop-based fuels to be renewable and more than half perceiving coal to be renewable. But the purpose of this inquiry is to better understand public perceptions of the attributes associated with various energy sources, and these initial find-ings suggest that most respondents perceive solar and wind to be the most preferred sources of energy and oil and coal to be the least preferred, even when prompted to consider energy affordability.

Energy Risk Perceptions

We asked the following questions to measure differential risks perceived to be associated with groups of energy sources.

Lead-in: The next set of questions concerns all kinds and uses of energy, in-cluding electricity for homes and businesses; gas, oil, and coal for heating; and transportation fuels, such as gasoline and diesel. Considering the effects of both normal operations and potential accidents, how do you rate the risks to so-

31

ciety and the environment from each of the following sources of energy using a scale from zero to ten, where zero means no risk and ten means extreme risk.

• E40: The risks from fossil fuels, such as coal, oil, and natural gas

• E41: The risks from nuclear power plants

• E42: The risks from renewable sources of energy, such as from hydroelec-tric dams, solar power, and wind generation

We compare trends in mean risk assessments in Figures 2.8–2.10.

Figure 2.8: Mean Risks of Fossil Fuels

Figure 2.9: Mean Risks of Nuclear Energy

6.99

5.90

6.50 6.14

6.86

.0

.0

.0

.0

.0

.0

.0

2006 2007 2008

E41: 0 = no risk—10 = extreme risk 2

3

4

5

6

7

8

Midscale

Phone

Web

6.53 6.45 6.73

.0

.0

4.0

5.0

6.0

.0

.0

2006 2007 2008

E40: 0 = no risk—10 = extreme risk

8

6.40 6.36

2

3

7

Midscale

Phone

Web

32

Figure 2.10: Mean Risks of Renewable Sources of Energy

3.38 2.82

2.81 2.35 2.55

2.0

3.0

4.0

5.0

6.0

7.0

8.0

2006 2007 2008

E42: 0 = no risk—10 = extreme risk

Midscale

Phone

Web

The trend in perceived risks is mostly level for fossil fuels and downward for renewable sources, while assessments of risks associated with nuclear generation yield a more mixed picture. Among Internet participants, mean risks of nuclear energy are about six percent higher in 2008, but among phone respondents, mean risks of nuclear energy are significantly lower (15.6 percent) than those recorded in 2006. The most clear picture in terms of policy relevancy is that mean risks of fossil fuels generally are compara-ble to those perceived for nuclear energy, with Internet respondents judging nuclear risks somewhat higher, on average, and phone participants judging them somewhat lower than those for fossil fuels. All means for fossil fuels and for nuclear energy are above midscale, whereas mean risks for renew-ables are well below midscale.

Preferences for Meeting Future Energy Needs

We investigate respondent preferences for meeting future energy needs with three related lines of inquiry. In the first, we measure the priority placed on reducing US energy dependence. In the second, we compare preferences for balancing conservation and energy development. In the third, we examine preferences for how future energy sources should be apportioned among fossil fuels, nuclear energy, and renewable sources.

33

Reducing Energy Dependence

Previous surveys in this series have indicated strong dissatisfaction with US dependence on foreign sources of energy. Beginning in 2006, we posed the following question in each annual survey, and we chart trends in means re-sponses in Figure 2.11.

E49: Using a scale from zero to ten, where zero means not at all important and ten means extremely important, how important is it to reduce US depend-ence on foreign sources of energy of all types?

Figure 2.11: Mean Importance of Reducing US Dependence on Foreign Energy

These kinds of consistent means above 8.5 on a scale from zero to ten indi-cate a clear consensus among our respondents that current levels of US de-pendence on foreign energy sources are not satisfactory. We find deep and broad public support for reducing US energy dependence.

Balancing Conservation and Development

There are two general remedies for energy dependence. One is to promote energy conservation measures that reduce overall demand for energy; an-other is to develop additional sources of energy (either through increased production of US sources or development of alternative sources). A bal-anced energy policy includes both approaches, but the relative weight given to conservation and development can vary substantially. To help assess

8.79

8.61 8.60

7.0

8.0

9.0

2006 200

9.04

8.65

10.0

7 2008

E49: 0 = not at all important—

Phone

10 = extremely important

Web

34

preferences among our respondents for how energy conservation and energy development ought to be balanced, we posed the following inquiry.

Lead-in: Some people argue that the best way to meet our future energy needs is to emphasize conservation. Others think we should emphasize the devel-opment of new energy sources. Still others prefer that we balance conserva-tion and development equally.

• E66: Which of the following three options do you most prefer for meeting our future energy needs?

1 – Place about two-thirds of our efforts on conserving energy and about one-third on developing new sources of energy.

2 – Place about two-thirds of our efforts on developing new sources of en- ergy and about one-third on conserving energy.

3 – Place about half of our efforts on conserving energy and about half on developing new sources of energy.

As shown in Table 2.3, while energy development is favored somewhat more among Internet participants than phone respondents, most participants in both groups prefer a balanced approach in which roughly equal emphasis is given to energy conservation and energy development.

Table 2.3: Preferences for Balancing Energy Conservation and Development

Phone 2008 (%)

Web 2008 (%)

2/3 efforts on conserving energy and 1/3 on developing new sources 14 14

2/3 efforts on developing new sources and 1/3 on conserving energy 27 36

1/2 efforts on conserving energy and 1/2 on developing new sources 58 50

Future Energy Mix

We conclude our examination of preferences for the energy future by in-forming participants approximately how much of total US energy require-ments currently are derived from each of three sources: fossil fuels, nuclear

35

generation, and renewable sources. We then ask respondents to indicate how they would like to see the proportions change over the next twenty years. We summarize percentages and illustrate trends in Table 2.4.

Lead-in: Now think about the overall mix of energy sources for the US. We cur-rently get about 85 percent of our energy from fossil fuels, eight percent from nuclear energy, and six percent from renewable sources. The following three questions concern how you would like to see this mix of energy sources change over the next 20 years. Approximately what percentage of the total US energy supply would you like to see come from each of these three energy sources?

• E50: What percent of our energy should come from fossil fuels, which cur-rently provide about 85 percent of our energy?

• E51: What percent of our energy should come from nuclear energy, which currently provides about eight percent of our energy?

• E52: What percent of our energy should come from renewable sources, which currently provide about six percent of our energy?

Table 2.4: Preferred Total Mix of Energy in Next 20 Years (Mean Percentages)

Phone 2006

Web 2006

Web 2007

Phone 2008

Web 2008

E50: Fossil fuels (currently 85%) 31.3 26.6 25.3 28.9 26.5 E51: Nuclear energy (currently 8%) 22.2 22.0 23.6 24.4 21.9 E52: Renewable sources (currently 6%) 46.3 51.4 51.0 47.2 51.9

Average preferences across the five surveys over three years are as follows: fossil fuels: 27.7 percent; nuclear energy: 22.8 percent; and renewables: 49.6 percent. While preferences for nuclear energy have remained relatively steady, preference for fossil fuels is declining and preference for renewable sources is increasing by about the same amount. In approximate terms, our respondents indicate they would like to see about half our future energy come from renewable sources, about 20–25 percent from nuclear, and about 25–30 percent from fossil fuels.

36

Short Answer

Q: How are different energy sources perceived in terms of attributes, risks, and preferences for meeting future requirements?

The order from most preferred to least preferred for eight specified sources of energy are as follows: (1) solar; (2) wind; (3) hydroelectric; (4) natural gas; (5) crop-based fuels; (6) nuclear generation; (7) oil; and (8) coal. Mean risks of fossil fuels and nuclear energy are rated above midscale and similar in magnitude. Mean risks of renewable sources are rated well below mid-scale. Respondents in five surveys administered via phone and Internet over three years consistently rate the importance of reducing US dependence on foreign sources of energy above 8.5 on a scale from zero (not at all impor-tant) to ten (extremely important). When asked how to balance energy con-servation with energy development, most participants prefer that roughly equal emphasis be placed on each. As to how they would like to see future energy needs sourced, on average their preference is to increase energy from renewable sources from the current six percent to about 50 percent; increase energy from nuclear generation from the current eight percent to 20–25 per-cent; and decrease energy from fossil fuels from the current 85 percent to 25–30 percent.

Section 2.3: Views on Nuclear Energy

ublic support for and opposition to nuclear energy is heavily influ-enced by perceived risks and benefits associated with nuclear genera-tion. Since 2006 we have been tracking trends in perceptions of nu-

clear energy risks and benefits and their tradeoffs using composite indexes composed of responses to multiple questions about different dimensions of nuclear energy risks and benefits. To measure risk perceptions, we pose the following four questions presented in random order.

PLead-in: Please consider both the likelihood of a nuclear event occurring and its potential consequences when evaluating the risk posed by each of the fol-lowing on a scale from zero to ten where zero means no risk and ten means extreme risk.

37

• E67: An accident at a US nuclear power plant within the next 20 years that results in the release of large amounts of radioactivity.

• E68: An accident during the transportation or storage of spent nuclear fuel from nuclear power plants in the US within the next 20 years that results in the release of large amounts of radioactivity.

• E69: A terrorist attack at a US nuclear power plant within the next 20 years that results in the release of large amounts of radioactivity.

• E70: The diversion of nuclear fuel from a nuclear power plant in the US within the next 20 years for the purpose of building a nuclear weapon.

Unweighted responses to each are averaged (ignoring missing values) to form a nuclear energy risk index, and in Figure 2.12 we show trends in mean perceptions for that index.

Figure 2.12: Trends in Mean Composite Nuclear Energy Risk Index

6.21

5.49

6.27 6.23 6.32

5.0

6.0

7.0

8.0

2006 2007 2008

E67—E70: 0 = no risk—10 = extreme risk

Phone

Web

Midscale

Note that mean composite assessments of risks associated with nuclear en-ergy have remained steady among Internet participants, while mean risk as-sessments declined significantly in 2008 among phone respondents. Risk perceptions are sensitive to demographic characteristics. For example, using combined data from all five surveys since 2005, we find that risk percep-tions decrease systematically with increasing age, education, and income, and are higher among women and minorities. Some of the differences in mean assessments in 2008 may be associated with differences in demo-graphic characteristics between our phone and Internet samples.

38

To gage perceptions of benefits associated with nuclear generation, we pose the following four questions in random order.

Lead-in: Now I want to ask about your beliefs about some of the possible benefits associated with nuclear energy use in the US. Please evaluate the benefits associated with each of the following on a scale form zero to ten, where zero means not at all beneficial and ten means extremely beneficial.

• E71: Fewer overall greenhouse gas emissions because nuclear energy pro-duction does not create greenhouse gases.

• E72: Reliable power because nuclear energy generates large amounts of electricity and is not affected by weather conditions, such as low rainfall or no wind.

• E73: Greater US energy independence because nuclear energy production does not require oil or gas from foreign sources.

• E74: Reduced environmental damage because of less need for mining coal or extracting oil and gas.

Again, we average unweighted responses to form a composite nuclear en-ergy benefit index, and we compare trends in index means in Figure 2.13.

Figure 2.13: Trends in Mean Composite Nuclear Energy Benefit Index Note that mean composite benefits are rated significantly higher than mean composite risk assessments shown in Figure 2.12. Unlike the differences in mean risks across survey modes in 2008, mean composite benefits of nu-

7.00

7.33 7.33 7.46

7.16

.0

.0

.0

.0

2006 2007 2008

E71—E74: 0 = not at all beneficial—10 = extremely beneficial

Phone

Web

Midscale

8

7

6

5

39

clear energy do not differ significantly between phone and Internet respon-dents in 2008.

In addition to the separate lines of inquiry about different dimensions of nu-clear energy risk and benefit perceptions, we also ask participants to com-pare risks and benefits and express their integrated views in response to the following inquiry.

Lead-in: Now I want you to consider the overall balance of possible risks and benefits of nuclear energy in the US.

• E75: Using a scale from one to seven, where one means the risks of nu-clear energy far outweigh its benefits, four means the risks and benefits are equally balanced, and seven means the benefits of nuclear energy far out-weigh its risks, how do you rate the overall balance of the risks and bene-fits of nuclear energy in the US? Remember, you can choose any number from one to seven.

We compare mean responses in Figure 2.14.

Figure 2.14: Balancing Risks and Benefits of Nuclear Energy

Notice that all means for both groups in each year have are above midscale, indicating judgments that the benefits of nuclear energy outweigh associated risks. These responses are consistent with those reported for our individual and grouped measures, and together, they illustrate central tendencies about how respondents perceive and weigh nuclear energy risks and benefits.

4.64

4.95

4.32

4.574.38

4.0

4.5

.0

5.5

.0

2006 2007 2008

6

5

E75 (Means): 1 = risks far outweigh benefits—4 = risks and benefits balanced—

7 = benefits far outweigh risks

Phone

Web

Midscale

40

Our next two questions allow us to see if preferences for expanding nuclear generation capacity are consistent with beliefs that the benefits of nuclear energy outweigh associated risks.

• E76: Using a scale from one to seven, where one means strongly oppose and seven means strongly support, how do you feel about constructing additional nuclear reactors at the sites of existing nuclear power plants in the US?

• E77: Using the same scale from one to seven, where one means strongly oppose and seven means strongly support, how do you feel about con-structing additional nuclear power plants at new locations in the US?

We chart trends in mean responses to each in Figures 2.15 and 2.16.

Figure 2.15: Mean Support for Additional Reactors at Existing Nuclear Plants

Figure 2.16: Mean Support for Additional Reactors at New Locations

3.92

4.294.16

4.40

4.19

3.5

4.0

4.5

5.0

5.5

2006 2007 2008

E77: 1 = strongly oppose—7 = strongly support

Phone

Web

Midscale

4.40

4.34

4. 4.58 54

4.29

.5

4.0

.5

5.0

.5

2006 2007 2008

5E76: 1 = strongly oppose—7 = strongly support

Phone

Web

4

Midscale

3

41

With the exception of phone respondents in 2006, all other means are above midscale for both policy choices. While support for building additional nu-clear reactors at existing nuclear power plants is somewhat higher, on aver-age, building additional nuclear reactors at new locations also is supported. Together, our measures of beliefs about risks and benefits of nuclear energy and moderate support for building additional generation capacity suggest cautious but substantial support for nuclear energy among phone and Inter-net respondents since 2006.

Short Answer

Q: How supportive are respondents of nuclear energy, and how are perceptions of risks and benefits, including environmental factors, evolving?

Trends in mean perceptions of nuclear energy risks (as reflected by our com-posite risk index) are holding steady among Internet respondents and declin-ing among phone participants, but means range from 5.5–6.3 on a scale from zero (no risk) to ten (extreme risk). Trends in mean perceptions of nuclear energy benefits (as reflected by our composite benefits index) are holding above a value of seven on a scale from zero (not at all beneficial) to ten (ex-tremely beneficial). When asked to weigh risks against benefits of nuclear energy directly, benefits consistently are judged to outweigh risks among all samples since 2006. Mean support for new reactors at existing nuclear power plants and at new locations are modestly above midscale for all samples, and the gap between the two options is narrowing.

Section 2.4: Preferences for Managing Spent Nuclear Fuel

he management of spent nuclear fuel is a key issue affecting the fu-ture of nuclear energy. We know from previous research in 2006 and 2007 that it is one of the most important concerns our respondents

identify as helping shape their attitudes about nuclear energy.1 When asked to rate the risks associated with transporting and storing spent nuclear fuel,

T

1 See Herron and Jenkins-Smith (2007, 72–73) for a comparison of the importance of seven issues when making judgments about nuclear energy. Managing spent nuclear fuel, includ-

42

mean responses are above midscale. Together, these results suggest that re-spondents have a sense that how spent nuclear fuel is managed is both impor-tant to the future of nuclear energy and is a continuing source of risk. But we also know that the debate about spent nuclear fuel can be confusing to some respondents. To better judge factual knowledge of current practices, we ask participants the following question about how spent nuclear fuel is being managed today.

E78: As nuclear fuel is used to generate electricity, it becomes contaminated with radioactive byproducts. When it can no longer efficiently produce elec-tricity, it is called spent nuclear fuel. To the best of your knowledge, what is currently being done with most of the spent nuclear fuel produced in the US? Is it: (random order)

a. Stored above ground in special containers at specified nuclear power plants throughout the US?

b. Shipped to Nevada and stored in a facility deep underground?

c. Chemically reprocessed and reused?

d. Shipped to regional storage sites?

Table 2.5 compares the percentages of respondents who chose each re-sponse option in our surveys since 2006.

Table 2.5: Current Disposition of Spent Nuclear Fuel in the US

Policy

Phone 2006

Web 2006

Web 2007

Phone 2008

Web 2008

Stored above ground at specified US nuclear power plants 20 20 22 23 22

Shipped to Nevada and stored deep underground 43 33 33 47 32

Chemically reprocessed and reused 10 13 13 7 17

Shipped to regional storage sites 26 34 31 23 30

ing its transportation and storage, was rated second in importance only to the safety of the operation of nuclear reactors.

43

Two points seem important to note. First, fewer than one in four respondents know that spent nuclear fuel currently is being temporarily stored in special containers at the sites of specified nuclear power plants. In each of five sur-veys, a plurality of respondents thought that spent nuclear fuel is being shipped to Nevada and stored in a facility deep underground. The second point is that the misperceptions reflected in Table 2.5 do not appear to be decreasing substantially. While the debate about how to manage spent nu-clear fuel may have raised public awareness of the problem and some of its attendant risks, the debate does not appear to be clarifying current practices and future options among our respondents. One speculation would be that because neither nuclear energy advocates nor opponents have found it useful to their interests to publicize current practices and locations of spent nuclear fuel temporary storage sites, those practices are not widely understood. The policy debate surrounding Yucca Mountain as a potential long-term storage site may have caused some respondents to assume the project already is re-ceiving spent nuclear fuel. But whatever the reasons (and we cannot know them with certainty), most of the people we have surveyed over the past three years do not know how spent nuclear fuel currently is being managed. Most have a sense that it is an important and risky problem, but most are not factually informed of current practices and options.

Because of these misperceptions, pursuing public preferences on this issue requires providing a minimum of basic factual information about competing policy options before asking for reactions. We do that with condensed, bal-anced arguments for and against five options for managing spent nuclear fuel. Four of the options are specific to US spent nuclear fuel disposition, and one of the options approaches the issue from a broader international perspective. We first describe each option and present associated pro and con arguments, then we ask participants to indicate their support or opposi-tion. For each option, the pro and con arguments are randomly ordered, and responses to each question are recorded before subsequent options are ad-dressed. Following are the arguments, questions, and trends in mean re-sponses for each option.

Lead-in: As you may know, spent nuclear fuel is highly radioactive. There are two basic options for managing spent nuclear fuel. It can be stored and protected for thousands of years, or it can be reprocessed and reused. First I will ask you about storage options, then I will ask you about reprocessing. After hearing key arguments for and against each of the options, I will ask you to rate each as an alternative way to deal with the spent fuel from nuclear power plants.

44

Storage option one is to continue the current practice of storing spent nuclear fuel above ground in special containers at some of the existing nuclear power plants. (order of following arguments randomized)

Opponents argue that many of these plants are near rivers, oceans, and large population centers, and permanent storage is needed where the waste can be better secured against possible terrorist attacks.

Supporters argue that transporting spent nuclear fuel to a central underground storage facility by truck, train, or barge would be too risky, and that the cur-rent practice of storing spent nuclear fuel at nuclear power plants buys time for finding future solutions.

• E79: Using a scale from one to seven, where one means strongly oppose and seven means strongly support, how do you feel about the current prac-tice of storing spent nuclear fuel at existing nuclear power plants?

Figure 2.17: Mean Support for Continuing On-Site Storage of Spent Nuclear Fuel

After being informed that spent nuclear fuel currently is being stored above ground in special containers at specified nuclear power plants, most respondents are opposed to that practice. Differences between phone and Internet respon-dents over three years are small, with mean responses all falling just below mid-scale. Notice the consistency of means whose differences within each group over the three year period are not statistically significant.

Our next inquiry addresses the option for a central underground facility.

Storage option two is to ship spent nuclear fuel, primarily by train, to a central facility where it would be stored in special containers deep underground and

3.73

2007 2008

5.0

3.513.65 3.60

3.58

2.0

3.0

4.0

E79: 1 = strongly oppose—7 = strongly support

Phone

Web

Midscale

2006

45

permanently monitored. The potential site being studied is in southern Ne-vada inside Yucca Mountain. (order of following arguments randomized)

Opponents argue that nuclear materials could eventually leak into under-ground water, and the risks of transporting radioactive materials to a central facility would be too high.

Supporters argue that a central facility would remove radioactive materials from their current locations near large population centers, rivers, and oceans, and would allow more careful monitoring and control.

• E80: Using the scale from one to seven, where one means strongly oppose and seven means strongly support, how do you feel about the option of opening a long-term underground storage facility where spent nuclear fuel from all over the US would be stored?

Figure 2.18: Mean Support for Opening Centralized Underground Storage Facility

Again, mean responses from both survey modes across all three years are consistently near midscale. Differences in means are not statistically sig-nificant. Though not shown, distributions of responses cluster around the three middle scale values (3–5). These kinds of response patterns suggest that most participants do not hold strong or polarizing views for or against underground storage of spent nuclear fuel. When combined with responses to the related knowledge question previously discussed, they also suggest that for many ordinary citizens, the ongoing debate over Yucca Mountain

4.11 3.93

4.12 4.213.99

2.0

.0

.0

.0

2006 2007 2008

E80: 1 = strongly oppose—7 = strongly support

Phone

Web

Midscale

5

4

3

46

may still be in what Daniel Yankelovich (1991) refers to as the “conscious-ness raising” or “working through” stages.2

Our third option for managing spent nuclear fuel involves opening tempo-rary storage facilities operated by private industries.

Lead-in: Storage option three is to ship the spent nuclear fuel by truck and train to privately owned temporary storage sites where it would be monitored for up to 50 years. At that point, a more long-term decision could be made about how to manage it. (order of following arguments randomized)

Opponents argue that private firms might not be safe stewards of the spent nuclear fuel, and that the risks of transporting radioactive materials to and from a temporary facility would be too high.

Supporters argue that private facilities for temporary storage of spent nuclear fuel would remove the materials form their current locations near large popu-lation centers, rivers, and oceans, and would allow more time for developing longer-term management options.

• E81: Using the scale from one to seven, where one means strongly oppose and seven means strongly support, how do you feel about the option for opening privately owned facilities where spent nuclear fuel would tempo-rarily be stored?

Figure 2.19: Mean Support for Opening Privately Owned Storage Facilities

2.59 2.46

3.12 3.19 3.21 2.0

3.0

4.0

5.0

2006 2007 2008

E81: 1 = strongly oppose—7 = strongly support

Phone

Web

Midscale

2 Yankelovich (1991) describes three stages in the process of evolving from mass opinion to public judgment. (1) consciousness raising; (2) working through; and (3) resolution. As a contrasting example to spent nuclear fuel management, abortion is an issue about which many members of the public have progressed to the resolution stage in which firm (and polarizing) judgments have been reached.

47

Notice that all means for this option consistently are below midscale and lower than mean support for the two previous options. Our data show little support for the use of privately owned and managed storage facilities for spent nuclear fuel.

In 2008 we added the option for reprocessing spent nuclear fuel using the following inquiry. Keep in mind that should reprocessing be employed, it would need to be coupled with a waste disposal strategy of the kind evalu-ated in the prior three questions.

Lead-in: Now we want to ask you about nuclear reprocessing. Spent nuclear fuel can be chemically separated to allow most of it to be reused to generate more electricity. Doing so reduces but does not eliminate the need to store smaller quantities of radioactive waste materials. (order of following argu-ments randomized)

Opponents argue that nuclear reprocessing is expensive; it poses risks of acci-dents during transportation and reprocessing that could release radiation; and some of these materials might fall into the hands of terrorists.

Supporters argue that the expense of reprocessing is offset by reduced costs for storage; it is more efficient to reuse most of the spent nuclear fuel to pro-duce more electricity; and it reduces the amount of radioactive waste requir-ing storage.

• E82: Using the scale from one to seven, where one means strongly oppose and seven means strongly support, how do you feel about the option for reprocessing spent nuclear fuel?

Having added this question only in our most recent surveys, we are unable to examine trends in means, but in Figure 2.20 we illustrate distributions of responses and show means for our phone and Internet respondents in 2008.

48

Figure 2.20: Mean Support for Reprocessing Spent Nuclear Fuel

0

5

10

15

20

25

30

1 2 3 4 5

Phone 08: 5.01

Mean support among both samples for the option of reprocessing spent nu-clear fuel is well above midscale and significantly higher than for any of the prior three options.

Our final policy option in this series focuses on proliferation issues by broadening the inquiry from spent nuclear fuel disposition at the national level to enrichment and reprocessing of nuclear fuels at the international level. We inquire into support for restricting the production and supply of fuel for nuclear reactors to a small number of specified countries authorized to enrich and reprocess nuclear materials.

Lead-in: Worldwide, the spread of nuclear materials that might be used to make nuclear weapons is a growing concern. Some countries may attempt to use nuclear energy programs to produce enriched uranium that can be used for nuclear weapons. Others may try to recover plutonium from spent nuclear fuel and use it in nuclear weapons. Today, North Korea and Iran illustrate these kinds of concerns.

One suggestion to control the spread of nuclear materials is to designate a small number of countries such as the US, Russia, France, Britain, China, and possibly others as the only countries authorized to enrich nuclear materials and reprocess spent nuclear fuel. These countries would provide fuel for nu-clear power plants at market price to other countries. (order of following ar-guments randomized)

Opponents of such a plan argue that this arrangement would place developing countries at the mercy of more highly developed countries who might with-hold nuclear fuel needed for producing electricity.

6 7

Web 08: 5.05

E82: Means

Strongly Support Strongly Oppose

%

49

Supporters of such a plan argue that this arrangement would help prevent the spread o