University of Southern California Dissertations and Theses

Political participation, public opinion, and the law

(USC Thesis Other)

Political participation, public opinion, and the law

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content Political Participation, Public Opinion, and the Law
by
Kristopher Coombs, Jr.
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(Political Science and International Relations)
August 2020
Copyright 2020 Kristopher Coombs, Jr.
Contents
Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ii
List of Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iv
List of Figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v
Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii
Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix
Chapter 1: Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 Motivations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.3 Article Overviews . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.3.1 The Curious Incident of Citizens United v. Federal Election Commission . . . . . . 5
1.3.2 I’ve Got a Bad Feeling About This: The Utility of Feeling Thermometers in Mea-
suring Approval of Affirmative Action . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.3.3 The Case(s) for Advocating for, and Supporting, LGB Rights . . . . . . . . . . . . . 8
Chapter 2: Do Campaign Finance Regulations Impact Political Efficacy? The Curious
Incident ofCitizensUnitedv. FederalElectionCommission . . . . . . . . . . . . . . . . . 11
2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.2 Campaign Finance in Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.3 Political Efficacy and Political Participation . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
ii
2.4 Methodology & Hypotheses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.5 Results & Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
2.6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
Chapter 3: I’ve Got a Bad Feeling About this: The Utility of Feeling Thermometers in
Measuring Approval of Affirmative Action . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
3.2 Measures of Racism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
3.3 Affirmative Action . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
3.4 Critical Race Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
3.5 Conceptualizing Feelings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
3.6 Feeling Thermometers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
3.7 Methodology & Hypotheses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
3.8 Results & Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
3.9 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
Chapter 4: The Case(s) For Advocating for, and Supporting, LGB Rights . . . . . . . . . 56
4.1 Political Participation in Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
4.1.1 Demographic Identity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
4.1.2 Media . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
4.1.3 Political Polarization and Emotion . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
4.2 Tactics for Achieving LGB Rights . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
4.2.1 LGB Activism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
4.2.2 Litigation Strategy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
4.3 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
4.3.1 Measurement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
4.3.2 Case Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
iii
4.3.3 Poll Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
4.3.3.1 Baehr v. Miike and Baker v. Vermont . . . . . . . . . . . . . . . . . . . . . 75
4.3.3.2 Lawrence v. Texas and Goodridge v. Department of Public Health . . . . . 76
4.3.3.3 Lewis v. Harris . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
4.3.3.4 In re Marriage Cases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
4.3.3.5 Hollingsworth v. Perry and United States v. Windsor . . . . . . . . . . . . 79
4.3.3.6 Obergefell v. Hodges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
4.3.3.7 Masterpiece Cakeshop v. Colorado Civil Rights Commission . . . . . . . . 80
4.3.4 Hypotheses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
4.4 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
4.4.1 Predicting Political Participation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
4.4.2 Predicting Support for Gay Rights . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
4.5 Limitations and Further Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
4.6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
Chapter 5: Conclusion and ongoing work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
Reference List . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
List of Tables
2.1 State Level Summary Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.2 Individual Level Summary Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.3 State Level Political Efficacy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
iv
2.4 State Level Political Participation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.5 Individual Level Political Efficacy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
2.6 Individual Level Political Participation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
3.1 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
3.2 R-Squared Marginal Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
3.3 Results for Feeling Thermometer & Racial Resentment . . . . . . . . . . . . . . . . . . . . 50
3.4 Results for Feeling Thermometer & Old-Fashioned Racism . . . . . . . . . . . . . . . . . . 51
3.5 Results for Racial Resentment & Old-Fashioned Racism . . . . . . . . . . . . . . . . . . . 52
List of Figures
3.1 Support for Affirmative Action . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
3.2 Paired Significance Relationship . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
4.1 The relationship between controversial court cases, the media, and social reactions . . . 74
4.2 Predicting respondents’ overall political participation . . . . . . . . . . . . . . . . . . . . . 84
4.3 Predicting respondents donating to a social or political organization . . . . . . . . . . . . 87
4.4 Predicting respondents contacting an elected representative or official . . . . . . . . . . . 88
4.5 Predicting respondents signing a paper petition . . . . . . . . . . . . . . . . . . . . . . . . 89
4.6 Predicting respondents signing an internet petition . . . . . . . . . . . . . . . . . . . . . . 90
4.7 Predicting respondents being involved in a protest . . . . . . . . . . . . . . . . . . . . . . . 91
4.8 Predicting respondents’ overall support for gay rights . . . . . . . . . . . . . . . . . . . . . 92
4.9 Predicting respondents’ support for LGB individuals adopting childing . . . . . . . . . . . 93
4.10 Predicting respondents’ support for LGB individuals serving in the military . . . . . . . . 94
4.11 Predicting respondents’ support for protecting LGB individuals from job discrimination . 95
v
4.12 Predicting respondents’ support for same sex marriage . . . . . . . . . . . . . . . . . . . . 96
vi
Acknowledgments
I would like to thank my advisor Jane Junn and my committee members James Lo, Jody Armour, Morris
Levy, and Ange-Marie Hancock-Alfaro for their unwavering support in providing constant feedback and
advice as I completed my studies. It was and is an honor to be their student, and their mentorship has
been a priceless part of my development as a scholar and legal practitioner.
Special thanks to Sabina Vaught at Tufts University for introducing me to the line of critical studies
that led me to where I am today. She drew out of me the ability to rigorously and vigorously interrogate
challenging problems and to always challenge results and conclusions, greatly improving my analytical
skills.
Endless love to Darius Izadpanah, Jameelah Morris, Ursula Griffiths-Randolph, AEX, Ryan
Dougherty, and Brandon Archambault for being intellectual sounding boards, providing academic
and life feedback, making fun of my Android phone (sorry for being the green bubble!), and sending
more memes than I knew existed. Evan, Erica, and Ember, you have my eternal gratitude for providing
me with untold amounts of love, food, and laundry services throughout my journey.
Much of my strength (physical, emotional, and intellectual) came—and continues to come—from my
USC 6am Gainz gym crew, made up of Jessica Wu, Nathan Ho, Jayson Ho, John Rahmani, Peter Ta,
Jack Lee, and Akira Nagamori. You all always had my back (thanks for the spots!), you always held
me accountable every single day (thanks for the cheers!), and the ski trips and AYCE KBBQ nights
replenished my soul.
I absolutely have to acknowledge my USC Graduate Student Government and USC Campus Activi-
ties family for pushing me to my limits in both my studies and in campus advocacy even through some
vii
incredibly turbulent times at the university; special thanks to Jakita Morgan who saw me day in and
day out for years and was always there for a joke or to rescue me from tough phone calls and visitors
at the front desk. Sally Pratt and Veridiana Chavarin, you both took a huge risk on, and guided me
through, this uncharted territory by helping me be the first to complete this dual-degree program. To
everyone in the POIR program, you all are the real MVPs; the struggle is real, and you all took time out
of your own work to help me improve my theories and fix coding errors.
To my wonderful dog Jack-Essex Bayard Baldwin, you were always willing to listen to the strangest
and most incoherent thoughts and ravings while watching me pace around the apartment with no
destination in mind, but you still loved me unconditionally. I know you can’t read this, but the world
should know you are the best pup around.
Finally, I would like to thank my family for always being there whenever, however, for whatever,
whyever, and wherever I needed them. I will continue to hold all the love, wisdom, blessings, and
homecooked meals within me no matter where I find myself, and I will carry them with me no matter
where I go.
viii
Abstract
Political participation and public opinion help to drive electoral and policy decisions. Within these three
articles, I examine whether political efficacy decreased after the passing of Citizens United; whether
feeling thermometers in surveys are a useful tool in predicting support for affirmative action policies;
and whether Supreme Court of the United States cases on same sex relationships impacted public
opinion on various LGB rights. I find that political efficacy increased after Citizens United; that feeling
thermometers do not provide any more information than the other tools for measuring racism; and that
Supreme Court decisions correlate with noticeable shifts in public opinion.
ix
Chapter 1
Introduction
This chapter provides an outline of my research topics. Here, I provide an overview of research into
political participation and public opinion, discussing various ways it is measured and the implications
of gaining additional insight. I provide an summary of the methods used across all three articles, as
well as a brief snapshot of each article itself.
1.1 Motivations
The evolution of democratic societies over several centuries has been rooted in the dispersion of authority
away from a single individual to increasingly larger segments of a society’s population. Decision-making
moderately decentralized such that policies, laws, and regulations were generally based on some kind
of overall consensus or agreement. In order for individuals to effectively ensure that they are able to
influence the future of their society, they must participate in—or try to change—whatever democratic
and political processes that have been arranged. Therefore, it so follows that political participation and
civic engagement are considered to be cornerstones of a healthy and functioning democratic society,
and democracies should not only value all voices but work to ensure that voices receive their due
representation (Acemoglu et al., 2005; Almond and Verba, 1989; Berinsky and Lenz, 2011).
A plethora of research has been conducted to determine and demonstrate different potential corre-
lates for political participation. For example, demographic aspects such as race, ethnicity, religion, and
1
occupation all mediate political participation (Bany, 2012; Cunningham and Regan, 2012; Read, 2015;
Soon and Kluver, 2014; Wright-Phillips, 2015). This is especially true when individuals have shared
senses of identity, which also impacts political opinions and policy preferences (Dawson, 1994; Masuoka,
2006; Rim, 2009; Sanchez and Medeiros, 2016; Simien, 2005). Additionally, scholars have demonstrated
that emotions play a large role in political mobilization and mediating other political interactions
such as news consumption, interpersonal debates, and general reactions to other demographic groups
(Namkoong et al., 2012; Weber, 2013; Woods et al., 2012). Much research has found that “negative”
emotions—such anger, fear, threat, and anxiety—mobilize individuals to a greater extent than “positive”
emotions—such as hope and reassurance—and in general play a central role in politics Banks and
Valentino (2012); Huddy et al. (2015, 2016); Ost (2004); Valentino et al. (2011); Weber (2013).
Throughout the 2000s and 2010s, visibility around political participation—especially protest and
other forms of high-level mass involvement—has increased around the world, thanks in large part to
social media. This includes protest behavior in Latin America, Arab Spring, and in post-Soviet/Eastern
European countries and their inclusion into the European Union (Valenzuela et al., 2012; Kentmen-C ¸ in,
2015; Achilov, 2016; Bodnaruk Jazayeri, 2016; Guasti, 2016; Valenzuela et al., 2016). Overall, many of
the traditional variables—e.g. race, gender, age, political ideology, religion, level of formal education,
etc.— explain political participation across the planet (Desposato and Norrander, 2009; Shaykhutdinov
and Achilov, 2014; Kentmen-C ¸ in, 2015; Bodnaruk Jazayeri, 2016; Tambe, 2016, 2017).
Countries that are more demographically inclusive in the political process also see rises in partic-
ipation, particularly when those groups have been intentionally excluded rather than those groups
having self-selected out of involvement in politics. Moreover, a sense of political efficacy—and the trust
in the political process itself—causes huge impacts on the level of political participation, often having
generational and educational divides. Finally, a sense of competition—whether it is centered around
seeing resource distribution as a zero-sum game by demographic group or around using the political
process as a means to determine the most efficient and effective methods of sociopolitical and economic
governance to benefit the population at large—will generate higher levels of political participation. The
2
results of such participation may, unfortunately, may still lead to differing and potentially negative
impacts on different populations unless so influenced by the state itself (Hirschmann, 1991; Desposato
and Norrander, 2009; Bodnaruk Jazayeri, 2016; Asal et al., 2013; Tambe, 2017; Shaykhutdinov and
Achilov, 2014; Kentmen-C ¸ in, 2015; Takhar, 2011; Tambe, 2016; Achilov, 2016; Bros and Borooah, 2013;
Grasso, 2014; Ceka, 2013).
1.2 Methodology
I will use a similar methodology across all three articles to provide consistency in their results. All data
will come from American National Election Studies (ANES) for the years 2004, 2008, 2012, and 2016.
ANES surveys have been conducted since 1948 to capture data from a random sample of the electorate on
topics such as demographics, political beliefs, political knowledge, and support for particular politicians
or policies. Data collection generally starts approximately one year before US presidential election
years and concludes a few weeks after the presidential election itself. ANES surveys generally contain
1500 to 2000 questions and have sample sizes ranging from approximately 1,000 to 6,000 individuals.
The individual variables that I will use consists of standard demographic information, all of which
are self-reported. Age will be a continuous variable (truncated to the nearest integer) beginning at
age 17. The ANES has varied in the number of racial categories over the years, but tends to have six
racial categories: white, Black, Native American, Asian/Asian-American Pacific Islander, non-white
Hispanic/Latino, and other. I will treat each of these as binary values, such that respondent received a
1 in the category for which they indicated their race, and a 0 for each category they did not. Consistent
with the survey options, gender will similarly be coded as a single binary variable, where “male” is set
to 0 and “female” is set to 1. Income and education are both ranked and ordered factors. Rather than
selecting a specific number of months or years of education, or rather than specifying their exact income,
respondents select predetermined ranges in which they fall. As an example, one range of income is
$15,000—$17,499, which will have its own single integer value (in this case, it would be five). As an
individual’s income bracket or level of education increases, so does its integer value (its rank).
3
The dependent variable for political participation is the same across all three studies. There are five
questions I will use to measure political participation, which ask whether the respondent has:
• Participated in a political protest
• Contacted their representative
• Signed an internet petition
• Signed a paper petition
• Donated to a social or political organization
These questions are all binary, where 0 means that the respondent has not participated in that activity
and 1 means that they have participated in that activity. The surveys differ a bit in that some of the
surveys ask the respondent to think back to the previous four years (2012), the past twelve months
(2004 and 2016), and whether or not someone has ever done one of the above actions (2008).
Additionally, not all of these questions are asked every year, ranging from as few as three questions
to all five questions. In order to approximate general participation across surveys, I will create an index
for each year that will set each respondent’s overall level of political participation to be based on the
distance between zero and one. The index will be created by taking the sum of the respondent’s answers
across all of the questions, which will be their “score”. Then, I will divide that sum by the total number
of “points” possible; in this case, this equals the number of questions. While certainly not perfect, this
allows for a more consistent comparison of results across the selected survey years. For all three papers,
I will conduct my analyses using OLS regression analysis. The results will provide correlations between
variables. The demographic categories will be used as control variables, with political participation as
the final dependent variable.
4
1.3 Article Overviews
The following sections will provide additional details about each project, including other dependent
and independent variables, additional survey questions, and which survey years shall be used. These
sections will also include the relevant theories, hypotheses, and other background information as
needed.
1.3.1 The Curious Incident ofCitizensUnitedv. FederalElectionCommission
Political efficacy has been defined in many ways, and is often split into multiple concepts. An overall
definition is that political efficacy is the extent of one’s belief that they can understand and participate in
the political process with the expectation that one’s representatives will respond to the demands asserted.
Several landmark Supreme Court of the United States (SCOTUS) decisions have addressed methods of
regulating how political campaigns are financed, including issues such as limits on the amount of money
that can be donated by individuals to donor identity disclosure. One concern constantly raised by the
courts, legislatures, citizens, and the media is whether different forms of campaign finance regulations
will impact political speech—which, to some extent, is protected under the First Amendment—and how
much of a voice or impact on the political process—in other words, a sense of political efficacy—the
“average” American will have. This sense of lowered efficacy, as portrayed by politicians and the media,
would stem from the sense that wealthy Americans and large corporations would have increased
influence since, as the saying goes, “money talks”. A lowered sense of political efficacy would, in turn,
lower individuals’ political participation as a result of feeling as though their voices would not be heard
anyway.
This project will consist of two stages of analysis. First, I will examine whether the decision of
Citizens United—considered the most important case on campaign finance thus far in the 21st century—
actually altered citizens’ sense of political efficacy. This will be tested first by examining all respondents’
sense of political efficacy before and after Citizens United, followed by comparing respondents’ sense of
political efficacy in states that already had campaign finance regulations similar to those struck down
5
by Citizens United to those that had already deregulated campaign finance. Second, I will investigate
whether these changes in political efficacy actually had an impact on political participation at both the
state level and the aggregate level.
In order to complete my analysis, I will need to include variables beyond the standard control
variables and those related to political participation. These variables will be the battery of questions
under “political efficacy”:
• How much of the time do you think that the officials in Washington can be trusted to do what is
right/make decisions in a fair way?
• How much of a say do people like me have in what the government does?
• How many people running the government are crooked?
• Do you think the government is run to benefit a few big interests or for the benefit of all people?
• How satisfied are you with the way democracy works in the United States?
Similarly to measuring political participation, I will create a 0 to 1 index of these questions in order to
ensure all variables are on the same scale.
There are several possible impacts of, or lessons learned from, this project. First, it can shed light
on whether political efficacy can actually be impacted by the different roles of money in politics. This
is particularly important given many candidates for political office will run on platforms that include
if they are taking money from certain large business or organizations. Additionally, the results could
highlight specific relationships between campaign finance, political efficacy, and political participation.
This could change mobilization methods for future political campaigns. Finally, the findings from this
project could reveal the importance of specific components of political efficacy and their impact on
participation.
6
1.3.2 I’ve Got a Bad Feeling About This: The Utility of Feeling Thermometers
in Measuring Approval of Affirmative Action
Emotions can greatly impact public opinions and levels of public participation. Laws and policies often
target members of particular demographic groups, which often leads to backlash from other populations
who—while believing that access to, and possession of, resources etc. is a zero-sum game—may believe
they are losing out. One such policy is affirmative action, which seeks to remedy past exclusion of
marginalized groups from institutions of higher education. One potential way of understanding the
level of support for affirmative action—outside of factors such as political ideology, race, and party
identification—would be to understand the emotional response. That response, then, might guide an
individual’s voting behavior, political ideology, level of political participation, and more.
I will examine how well three different measures of an individual’s belief or opinion about other
races predicts support for affirmative action. These measures are “old-fashioned” racism, which is based
on disgust and feelings of superiority; “racial resentment”, which is based on an unfair and undeserved
distribution of resources; and “feeling thermometers”, which is a 101-point scale to measure how warm
or cool a respondent feels for a particular demographic group. Since each measure has a different
number of potential responses, they will all be set to indices of 0 to 1. Additionally, for this study I will
only be using the responses of white survey participants towards Blacks. Not only will I examine how
well each individual method predicts support for affirmative action, but I will also compare the relative
effect size—if any are statistically significant—between each measure.
This paper will first establish different manners for conceptualizing and differentiating emotion,
feeling, affect, and mood in order to provide a framework for discussing these different responses. Then,
I will use Critical Race Theory to ground the legal response of affirmative action to its history of the
development and entrenchment of race within American society. Most importantly, this background will
examine support for, and backlash against, affirmative action and link it directly to views about race.
The main impact of this project will be to illuminate whether “feeling thermometers” are useful tools
in measuring political opinion or political participation. It is possible that feeling thermometers will be
7
co-linear with old-fashioned racism, racial resentment, or even political ideology as support for many
policies falls along party or ideological lines. If each measure is statistically significant in the same
direction, then the usefulness of the tool—at least in this respect—can be questioned if the other two
measures give similar responses but with larger effect sizes (and thus, seem to better capture what is
going on). Finally, if they do tell a different but meaningful story, then the tool could be further studied
to see what additional information can be gathered by measuring feeling.
1.3.3 The Case(s) for Advocating for, and Supporting, LGB Rights
This project will examine how media and SCOTUS decisions impact changes in public opinion regarding
specific policies regarding those identifying as LGBTQ+, and how litigation involving LGBTQ+ organi-
zations generated or influenced media around LGBTQ+ issues. In addition to providing an overview of
political participation, this project will begin by providing background on various activist and litigious
strategies for achieving LGB rights. This will provide context for when and how different tactics
are used to generate different amounts of media attention around LGB rights. That exploration will
illuminate the relationship between LGB rights movement, public opinion, and legal/judicial responses
and reactions (for example, why certain cases or incidents were selected for litigation).
I will also provide a brief review of several landmark SCOTUS cases that might impact respondents’
positions on LGB rights. While prior important cases will be discussed, the cases selected cases on
which I will focus will be from 2000 to 2016, as those are the cases that might have generated significant
media attention and lead to differing positions on LGB rights. The cases used as the backdrop to the
measures of public opinion will be as follows:
• Baehr v. Miike (December 9, 1999)
• Baker v. Vermont (December 20, 1999)
• Lawrence v. Texas (June 26, 2003)
• Goodridge v. Department of Public Health (November 18, 2003)
8
• Lewis v. Harris (October 25, 2006)
• In re Marriage Cases (May 15, 2008)
• United States v. Windsor (June 26, 2013)
• Hollingsworth v. Perry (June 26, 2013)
• Obergefell v. Hodges (June 26, 2015)
• Masterpiece Cakeshop v. Colorado Civil Rights Commission (June 4, 2018)
These cases generated large amounts of media attention both leading up to, and after, SCOTUS handed
down their decision. In many instances, the cases received attention even at the lower courts and before
SCOTUS decided to hear the case.
Additionally, this project utilizes several additional independent and dependent variables in order
to conduct the proper evaluation. These variables include the respondent’s lesbian, gay, or bisexual
(LGB) identity and respondent’s stances on various LGB rights and policies. When it comes to a person’s
sexual orientation, I will only use LGB, primarily because ANES surveys do not ask about respondents’
gender identity, only sexual orientation. Thus, I will combine the bisexual identity with the lesbian
and gay (homosexual) identities. This will be done because a bisexual identity is still distinct from
a heterosexual identity. Moreover, some surveys do not ask about bisexual identity, and none of the
surveys include “bisexual” in the battery of questions regarding the civil and political rights and liberties
of non-heterosexual, non-cisgender respondents.
The questions used to determine a respondent’s level of approval ask about the following issues:
• Protection from job discrimination
• Military service
• Adoption
• Legal marriage
9
While this list does include the issue of same-sex marriage, the goal of this paper is also to examine
the impact of these cases on the other issues. Seeing the impact on opinions on same-sex marriage will
assist in understanding and measuring the relative degree of change in public opinion.
I will also use data from polls taken close before and shortly after SCOTUS made their decision.
Such polls will come from sources like YouGov, Pew, and media outlets. Although these polls are not
always structured the same and measuring the responses of the same participants, I am limited to
the accessible polls. The measured reactions will be compared to the responses given on the ANES to
approximate any overall shifts in public opinion.
The results from this project could provide additional guidance or whether to pursue litigation as
a means of achieving advocacy goals, whether to increase media attention around a particular case,
and which changes in public opinion would be sustained over time. While the focus of this project is
on policies affecting same-sex relations, the results could be tested for other policies and demographic
groups in their fights for justice.
10
Chapter 2
Do Campaign Finance Regulations Impact Political Efficacy?
The Curious Incident ofCitizensUnitedv. FederalElection
Commission
2.1 Introduction
The United States Congress has broad, well-established power to regulate federal elections, derived
primarily from the Constitution and solidified through caselaw (United States Declaration of Indepen-
dence, 1776; Buckley v. Valeo, 1976). This power allows Congress to regulate election logistics—such
as the time and day of Presidential elections and alter state-made regulations for U.S. Senators or
Representatives—and to enact measures aimed at maintaining the overall integrity of the electoral
process (”Burroughs, ”Burroughs). This otherwise plenary power may be constrained only if the regu-
lations run afoul of the rights to freedom of political speech and association guaranteed by the First
Amendment, which is the crux of a contentious conflict in modern campaign finance law: how to strike a
balance between the essential need for electoral integrity on one end of the scale, and the fundamental
right to free political speech and association on the other. On both sides of the scale rest interests which
“are integral to the operation of the system of government established by our Constitution”, making the
balancing act a precarious one (Buckley v. Valeo, 1976).
11
The Supreme Court has viewed—and continues to view—all government-imposed restrictions on
speech with a suspicious eye, whether said speech is subject-matter or speaker-based (Citizens United
v. Fed. Election Com’n, 2010). Thus, they have developed an analytical framework to determine the
legality of any such restriction. Therefore, regulations which burden a citizen’s ability to speak or
associate in the political realm are subject to “the exacting scrutiny required by the First Amendment”
(Nixon v. Shrink Mo. Gov’t PAC, 2000). Buckley v. Valeo separated the First Amendment analysis into
two separate (albeit related) frameworks: political association and (political) speech.
Although regulations on speech require a higher burden than those on association—one that is
rarely, if ever, met—both require a positive showing from the government that the regulations further
an important state interest. The Supreme Court of the United State’s (hereinafter ”SCOTUS” or ”the
Court”) recognition of sufficiently important interests seems to have fluctuated since Buckley, however
all of the interests seriously considered since that seminal decision have aimed to advance the same
underlying interest: maintaining the integrity of the electoral and political processes in order to uphold
public confidence—and (electoral) involvement—in representative government. This article seeks
to examine whether the loosening of campaign finance laws has an impact on an individual’s sense
of political efficacy, and thus on their level of political participation (Fed. Election Comm’n v. Colo.
Republican Fed. Campaign Comm., 2001).
2.2 Campaign Finance in Review
The United States was founded upon the idea that sovereignty lies with the people, and the country’s
constitution clearly created a system of government—including and especially the people within it—that
derives its power by consent of the people. The Framers understood that the erosion of popular consent
might call into question the system of representative government, and structured the Constitution to
prevent what they saw as the leading cause of that erosion: corruption (Teachout, 2009). Corruption
was such a pressing and important issue that it was discussed in both the Federalist and Anti-Federalist
papers.
12
While the late eighteenth century definition of corruption might differ somewhat from today, the
Framers’ near obsession has guided over 100 years of jurisprudence, specifically as it relates to election
law (Citizens United v. Fed. Election Com’n, 2010). This idea was so powerful that by the late nineteenth
century, Congress had already begun placing limits on corporate participation in elections. This was due
to corporate spending in elections became more prevalent as a result of rising individual and corporate
wealth due in part to the Industrial Revolution. Briffault (2011) traces this history showing that by the
mid-twentieth century, over twenty-seven states had completely banned corporate contributions with
an additional nine barring it from certain kinds of corporations (e.g. insurance companies and banks).
Unsurprisingly, these bans sparked decades of lawsuits that invoked First Amendment rights,
necessitating a strict scrutiny analysis in which the challenged law, in order to stay in place, must
be proven to have been created due to a compelling government interest and that the law itself is
narrowly tailored such that the infringement upon Constitutional rights is no more than what is needed
to achieve the interest. The majority of litigated cases revolve specifically around issues of free speech
as it pertains to spending on political campaigns (and whether money is, in fact, a form of speech).
However, challenges to statutes that limited political spending generally failed as, while the Court
recognized it put a damper on certain kinds of political participation, the purpose of the regulations
was to ensure the integrity of elections and governmental processes. In other words, Congress was not
attempting simply to unduly limit lobbying, but rather to ensure—and perhaps signal to the American
people—that the process was honest. Such policies were aimed at arrangements that could appear to be
quid pro quo or, even worse, things that are actually quid pro quo. Thus, it was within the government’s
interest to enact legislation to limit even the appearance of impropriety in order to maintain the trust of
the people (Buckley v. Valeo, 1976; ”Burroughs, ”Burroughs).
In Buckley v. Valeo, the Court upheld limitations on political association in the form of campaign
contributions, and struck down limitations on political speech in the form of independent expenditures.
In doing so, the Court recognized two interests compelling enough to justify restrictions on political
association and speech through campaign finance regulations: quid pro quo corruption, and “the
13
appearance of corruption stemming from public awareness of the opportunities for abuse inherent in
a regime of large individual financial contributions” (Buckley v. Valeo, 1976). Although Buckley did
not foreclose the existence of other sufficiently important interests, recent jurisprudence—much to the
chagrin of Constitutional scholars and past and present Supreme Court justices–all but has (Citizens
United v. Fed. Election Com’n, 2010; Hasen, 2011). One of the most important contributions of Buckley
was its clear distinction between campaign contributions and campaign expenditures (Abraham, 2010).
SCOTUS acknowledged that “virtually every means of communicating ideas in today’s mass society
requires the expenditure of money”, thus providing the groundwork to see expenditures as a form of
freedom of expression for political actors (Buckley v. Valeo, 1976; Abraham, 2010).
After Buckley, years of subsequent litigation ensued, with the Court consistently and continuously
affirming the important governmental interest in preventing the appearance of corruption as well
as actual corruption as the rationale behind limits and bans on individual, corporate, and aggregate
campaign contributions and expenditures (Abraham, 2010; Buckley v. Valeo, 1976; Fed. Election
Comm’n v. Nat’l Right to Work Comm., 1982; Nixon v. Shrink Mo. Gov’t PAC, 2000). One of the trickiest
parts of grappling with the issue of campaign finance and corruption, however, is that SCOTUS has
never specifically defined what constitutes “corruption” Hasen (2011). Nevertheless, the Court has
continued their evaluation of campaign finance laws and also examined other governmental interests
proffered in support of increased campaign finance regulations, including: reducing the overall cost
of conducting political campaigns, preventing donors from bypassing contribution ceilings by limiting
independent expenditures, preventing undue donor access to elected officials, and preventing the
drowning out of ordinary speakers by wealthier ones (the “anti-distortion rationale”).
The Court has held that reducing the overall amount of money in campaigns, closing loopholes that
allow campaigns to bypass contribution ceilings, and access to representatives were not legitimate
reasons to justify First Amendment burdens as the Court did not find evidence that the regulations
would impact public confidence in electoral processes (Abraham, 2010; Bopp et al., 2015; Buckley v.
Valeo, 1976; Fed. Election Comm’n v. Mass. Citizens for Life, Inc., 1986; Hasen, 2011; McConnell v. Fed.
14
Election Comm’n, 2003). Anti-distortion rationale, on the other hand, was formally recognized as a
legitimate concern and grounds for the government imposing contribution limits or banning certain
kinds of contributions (Austin v. Mich. Chamber of Commerce, 1990; Fed. Election Comm’n v. Mass.
Citizens for Life, Inc., 1986). Note the different treatment between contributions and expenditures;
limits on contribution continue to be seen as valid, while limits on expenditures generally are not seen
as a compelling government interest.
Challengers to campaign finance laws relentlessly assert that the regulations limit neither actual
corruption nor its appearance, necessitating the use of empirical evidence to prove the validity and
necessity of such restrictions (Citizens United v. Fed. Election Com’n, 2010; Volokh, 2006). The burden
of proof is high, such that governmental conjecture or hypothesizing the impact of a regulation does not
suffice to uphold limits to independent expenditures. Contribution limits, on the other hand, require the
challengers to bring in evidence that the restrictions fail to prevent corruption or its appearance, and
the absence of clear proof have allowed to Court to uphold such restrictions (Austin v. Mich. Chamber of
Commerce, 1990; Nixon v. Shrink Mo. Gov’t PAC, 2000).
The Bipartisan Campaign Reform Act of 2002 (BCRA) attempted to limit how much money affected
politics by trying to ban “soft money” contributions, which went to organizations rather than to particular
individual campaigns and had allowed corporations to fund election communications. The law was
challenged in McConnell, but the Court upheld the provisions due to the idea of “access corruption” in
which those providing large donations would receive special treatment (McConnell v. Fed. Election
Comm’n, 2003). However, this led to groups capitalizing on creating independent political groups and,
eventually, led to the formation of Political Action Committees (Abraham, 2010). These committees
allowed for corporations to more freely spend on elections without limit, the issue at the crux of Citizens
United.
After reviewing the arguments, SCOTUS declared that any and all independent spending limits
were unconstitutional despite any evidence presented before them. This decision has been widely
and heavily criticized in large part due to the belief that such First Amendment rights challenges
15
should focus on the average political donor (Abraham, 2010; Briffault, 2011; Citizens United v. Fed.
Election Com’n, 2010). Under this theory, scholars argue that not only will this limit campaigners from
spending their time raising funds for political communication in lieu of time spent representing their
constituents, but it also allows more candidates—whether self-financed or not—the opportunity to be
heard (Alexander, 2011; Fevurly, 2010). Overton (2004) notes, however, that there still exists a small
group of donors who ultimately dominate federal campaign finance, and that this “donor class” should
not be ignored.
This fear of the small donor class ultimately is what influenced the widespread concern around
the impact of Citizens United (though no survey data results or other data on political participation
was presented in their article). Levinson (2013) highlights the importance of ensuring the diversity
of viewpoints presented to voters, and allowing a small number of private political actors to dominate
the airwaves (so to speak) restricts this diversity of views. Of course, SCOTUS has never entirely
agreed with this view, and has argued that the size of contributions does not actually impact the overall
quantity of expressions and viewpoints; rather, they argued the exact opposite whereby contribution
limits would be the cause of the reduction of communications (Esenberg, 2010).
2.3 Political Efficacy and Political Participation
Elizabeth Beaumont has summarized how decades of research has proven the importance of political
efficacy (also referred to as political confidence or competence) in democratic participation (Almond and
Verba, 1989; Beaumont, 2011). For example, political efficacy has been found to increase political news
media consumption (Bachmann and De Z ´ u ˜ niga, 2013; Gil de Z ´ u ˜ niga et al., 2017) as well as to separately
increase political participation (Corrigall-Brown and Wilkes, 2014). Though political efficacy enjoys no
singular definition, much of the literature indicates that political efficacy is the extent of an individual’s
belief that they can understand and participate in the political process with the expectation that their
representatives will respond to the demands they, as constituents, have asserted.
16
This concept has been divided, narrowed, and studied this concept in several parts based on past
decades of research, looking at internal efficacy, external efficacy, government efficacy, group efficacy,
and individual efficacy (Denver and Hands, 1990; Feldman et al., 2017; Gil de Z ´ u ˜ niga et al., 2017;
Pinkleton et al., 2012). The differences between these concepts revolved around the acting unit—for
example, internal efficacy is often based on one’s individual belief that they can make a difference in
the political process, while external efficacy is often based on one’s belief that the government cares
about, and will alter its policies for, its constituents.
Scholars have delved into the relationship between political efficacy and political participation, which
is ultimately the concern of this paper (Spencer and Wood, 2012). While there has been a proven link
between efficacy and participation, it is unclear the extent to which particular independent variables
ultimately are the cause. For example, Beaumont (2011) found that when a person’s environment
and their relationships to others encourage political engagement, they are more likely to “acquire the
political skills and commitment to play a meaningful role” in the political process. Eveland and Hively
(2009) found similar results, in which frequency of political discussions created more political knowledge
which influenced political participation. Interestingly, discussion diversity was negatively correlated
with participation; thus, it might be fair to say that too much diversity of opinion may leave one less
sure of their views and thus, reduce participation at large. Ikeda et al. (2008) reviewed studies that
specifically tested attitudinal variables such as party identification and political efficacy, finding that
generally, there was a correlation between increased efficacy and increased participation.
On the other hand, Primo and Milyo (2006) and Milyo and Primo (2017) observed that changes
in “measures of perceived corruption, political efficacy, and trust in the federal government do not
track closely with either campaign spending or major federal campaign finance reforms”. Coleman and
Manna (1998) found that campaign spending at the congressional level correlated with an increase in
political knowledge and no decrease in trust in government or increase in overall cynicism. However,
they concede that increases in mistrust and cynicism may be the result of the overall campaign.
17
Throughout all of this research, though, Milyo (2012) noted the lack of research testing campaign
finance regulations as a treatment. Their own research has found “no substantively important effects
of state campaign finance reforms on these measures of political efficacy”, though they specify that
their research didn’t look at the survey respondents’ state governments. When testing the impact of
Citizens United, they found no evidence that any of the hype or concerns around the case materialized
Milyo and Primo (2017). They hypothesized that while the public generally knows little about money in
politics and generally—though skeptically—supports campaign finance reform and its lack of impact
on corruption. Additionally, they contemplate that shifts in trust in government comes more from the
”contentious nature of democratic politics in a pluralistic society”, which is somewhat separate from the
issue of campaign finance.
2.4 Methodology & Hypotheses
There is some debate among scholars over measuring and interpreting the impact of campaign finance
regulations on political efficacy. However, what is clear is the relationship between political efficacy and
political participation. Moreover, there was a large amount of negative news coverage and publicity
around Citizens United before SCOTUS made their decision. Given the discussion in Section 2.2, we
present the following hypotheses:
H1 Political efficacy for individual ANES respondents will see a very slight decrease from before Citizens
United to after Citizens United (this will not include state-ban as an independent variable).
H2 Political efficacy for states that had a ban before Citizens United will see a slight decrease in
political efficacy after Citizens United, while those in states that already had no bans will see no
change.
H3 Increases in political efficacy will correlate with increases in political participation, while decreases
in political efficacy will correlate with decreases in political participation.
18
To address the lack of studies using Citizens United as a treatment, this paper will use the case to
test political efficacy by comparing it both before and after the decision. We will use data from the 2008
and 2012 different years American National Election Studies (ANES), with state aggregate data coming
from the mean of each variable. We will use the following independent variables:
• Age (a continuous variable)
• Race (six categories, each binary with a 1 being for the selected category, excluding the “other”
category)
• Gender (binary, with 1 being female)
• Political Ideology (a 7 point scale from liberal to conservative)
• Party Identification (a 7-point scale from strong democrat to strong republican)
• Education (set in ranges, each level receiving one integer for its rank; a higher rank means a
higher level of education)
• Income (set in ranges, each level receiving one integer for its rank; a higher rank means a higher
income)
We will have two dependent variables: political efficacy and political participation:
• Political Efficacy (set to an index of 0 to 1)
– In 2008, the variables this paper uses are as follows:
1. How much of the time do you think you can trust the government in Washington to
do what is right? OR How much of the time do you think you can trust the federal
government in Washington to make decisions in a fair way?
2. Would you say the government is pretty much run by a few big interests looking out for
themselves or that it is run for the benefit of all the people?
3. Public officials don’t care much what people like me think
19
4. People like me don’t have any say about what the government does OR How much can
people like you affect what the government does?
5. On the whole, are you very satisfied, fairly satisfied, not very satisfied, or not at all
satisfied with the way democracy works in the United States?
6. Do you think that quite a few of the people running the government are crooked, not very
many are, or do you think hardly any of them are crooked?
– In 2012, the variables were as follows:
1. How often can you trust the federal government in Washington to do what is right? OR
How much of the time do you think you can trust the government in Washington to do
what is right?
2. Would you say the government is pretty much run by a few big interests looking out for
themselves or that it is run for the benefit of all the people?
3. Public officials don’t care much what people like me think OR How much do public
officials care what people like you think?
4. People like me don’t have any say about what the government does OR How much can
people like you affect what the government does?
5. On the whole, are you very satisfied, fairly satisfied, not very satisfied, or not at all
satisfied with the way democracy works in the United States?
6. Do you think that quite a few of the people running the government are crooked, not
very many are, or do you think hardly any of them are crooked? OR Do you think that
hardly any of the people running the government are crooked, not very many are, or do
you think quite a few of them are crooked?
• Political Participation (set to an index of 0 to 1)
– In 2008, the variables this paper uses are as follows (all starting with “Have you done
this, or have you never done it?”):
20
1. Joined in a protest march, rally, or demonstration.
2. Attended a meeting of a town or city government or school board.
3. Signed a petition on paper about a political or social issue.
4. Signed a petition on the Internet about a political or social issue.
5. Not counting a religious organization, gave money to any other organization concerned
with a political or social issue.
6. Attended a meeting to talk about political or social concerns.
– In 2012, the variables this paper uses are as follows:
1. During the past 4 years, have you joined in a protest march, rally, or demonstration, or
have you not done this in the past 4 years?
2. During the past 4 years, have you attended a meeting of a town or city government or
school board, or have you not done this in the past 4 years?
3. During the past 4 years, have you signed a petition on paper about a political or social
issue, or have you not done this in the past 4 years?
4. During the past 4 years, have you signed a petition on the Internet about a political or
social issue, or have you not done this in the past 4 years?
5. Not counting a religious organization, during the past 4 years, have you given money to
any other organization concerned with a political or social issue, or have you not done
this in the past 4 years?
6. During the past 4 years, have you contacted or tried to contact a member of the U.S.
Senate or U.S. House of Representatives, or have you not done this in the past 4 years?
Note that the wording of some questions changes ever so slightly between 2008 and 2012, which
were specifically designed to experiment with different wording. Thus, in order to not lose data, we have
combined those variables together as those surveys expressly explained that they are variations of the
same question (though it is understood that different phrasing of questions may lead respondents to
21
answer slightly differently). In order to create indices for political efficacy and political participation,
we ensured that all variables read in such a way that higher values meant higher levels of political
efficacy or participation. Then, we summed up the number of questions for each index and divided that
total by the number of questions; both political efficacy and political participation had six questions,
ensuring that they were on the same scale.
As an additional independent variable, we have included “state ban”, which is a binary measure of
states that had a ban before Citizens United or that did not have a ban. The states that did not have a
ban were the control group (as Citizens United would have no impact on campaign finance regulations
in those states as there were no prior restrictions), while those that had a ban were assigned to the
treatment group (as Citizens United would have loosened or removed those restrictions). States that did
not have ban—those in the control group—were coded as “0”, while those that had a ban—the treatment
group—were coded as “1”. The list of states was narrowed to thirty-three based on available data for
both 2008 and 2012, meaning that twenty-one states were in the control group and twelve were in the
treatment group.
First, we created summary statistics for each level of analysis and each year of analysis. We then
conducted OLS regression analyses to see what correlations, if any, we could find. In these regressions,
we had a set that broke the population down by race, and a set that looked at the population as a whole,
regardless of race. Each set of regressions will be done at both the individual level—that is, looking at
individuals by state—as well as at the state level, which aggregates up the variables by the mean value
of the variable for each state.
2.5 Results & Discussion
First, we examined the summary statistics for each level of data—state level and individual level—by
survey year. Our means for political participation and political efficacy at the state level and at the
individual level were quite similar within survey years. We also see a slight increase in the mean levels
of political efficacy from 2008 to 2012, and a mean level of participation more than doubling from 2008
22
to 2012 at both levels of analysis. At the outset, this large leap suggests that something had a strong
influence on levels of political participation.
Table 2.1: State Level Summary Statistics
Statistic N Mean St. Dev. Min Pctl(25) Pctl(75) Max
2008 Political Efficacy 33 0.432 0.021 0.387 0.420 0.445 0.490
2008 Political Participation 33 0.306 0.082 0.144 0.248 0.359 0.487
2012 Political Efficacy 33 0.513 0.019 0.478 0.500 0.521 0.576
2012 Political Participation 33 0.744 0.041 0.649 0.719 0.764 0.844
Table 2.2: Individual Level Summary Statistics
Statistic N Mean St. Dev. Min Pctl(25) Pctl(75) Max
2008 Political Efficacy 2,322 0.429 0.160 0 0.364 0.535 0.858
2008 Political Participation 2,322 0.296 0.293 0 0 0.5 1
2012 Political Efficacy 5,914 0.511 0.123 0 0.447 0.565 0.917
2012 Political Participation 5,914 0.741 0.319 0 0.6 1 1
After the review of summary statistics, we turned to analyzing our data at the state level. Our
results show that at the mean level of all variables in each state, only two level of education and party
ID had any predictive significance. This only occurred in 2012 when excluding race. With all variables
included, none were statistically significant in predicting the mean political efficacy at the state level.
When looking at levels of political participation at the state level, none of the variables—which include
political efficacy—held any predictive value in 2012. In 2008, both gender and level of education were
statistically significant with roughly the same effect size whether or not race was included.
While state level results did not reveal much, we see more variables having an impact on efficacy
and participation at the individual level. Nothing at the individual level predicted levels of political
efficacy in 2008; however, in 2012, we see that age, party ID, and race were correlated with levels of
political efficacy. The size of the effect for age is almost negligible when compared to the impact of race.
Identifying as Black or Latinx were predictive at a p-value of less that .05. This means that we can
successfully reject the null hypothesis as the chance of it being a false-positive is less that 5%. If race is
excluded, the only predictive variable was party ID, where moving further from “strong democrat” (and
23
Table 2.3: State Level Political Efficacy
Dependent variable:
Political Efficacy
2008 2012 2008 2012
Age 0:001 0:001 0:001 0:001
(0:002) (0:002) (0:002) (0:001)
Gender 0:023 0:024 0:026 0:022
(0:072) (0:057) (0:064) (0:054)
Income 0:007 0:001 0:005 0:001
(0:004) (0:003) (0:004) (0:002)
Education 0:014 0:026 0:007 0:032

(0:009) (0:019) (0:009) (0:015)
Liberal-Conservative Placement 0:008 0:014 0:002 0:008
(0:017) (0:020) (0:015) (0:018)
Party ID 0:036 0:012 0:002 0:019

(0:026) (0:014) (0:016) (0:008)
White 0:131 0:071
(0:364) (0:133)
Black 0:077 0:017
(0:362) (0:150)
Asian-American/Pacific Islander 0:321 0:130
(0:422) (0:272)
Latinx/Hispanic 0:093 0:021
(0:361) (0:128)
Native American 0:364 0:178
(0:455) (0:251)
State Ban 0:002 0:004 0:001 0:008
(0:008) (0:006) (0:008) (0:006)
Constant 0:365 0:408

0:352

0:449

(0:403) (0:142) (0:112) (0:079)
Observations 33 33 33 33
R
2
0.426 0.595 0.138 0.479
Adjusted R
2
0.082 0.352 -0.103 0.333
Note:

p<0.05;

p<0.01;

p<0.001
24
Table 2.4: State Level Political Participation
Dependent variable:
Political Participation
2008 2012 2008 2012
Political Efficacy 0:864 0:268 0:564 0:546
(0:541) (0:461) (0:466) (0:420)
Age 0:006 0:003 0:007 0:004
(0:004) (0:003) (0:004) (0:003)
Gender 0:383

0:146 0:335

0:091
(0:175) (0:118) (0:151) (0:114)
Income 0:003 0:003 0:002 0:004
(0:010) (0:006) (0:009) (0:005)
Education 0:077

0:064 0:076

0:030
(0:024) (0:040) (0:020) (0:035)
Liberal-Conservative Placement 0:065 0:027 0:068 0:044
(0:042) (0:041) (0:035) (0:037)
Party ID 0:023 0:004 0:034 0:040
(0:066) (0:028) (0:037) (0:019)
White 0:865 0:257
(0:882) (0:276)
Black 0:955 0:434
(0:877) (0:309)
Asian-American/Pacific Islander 1:492 0:165
(1:036) (0:564)
Latinx/Hispanic 0:880 0:312
(0:874) (0:265)
Native American 1:446 0:037
(1:117) (0:523)
State Ban 0:025 0:014 0:012 0:009
(0:021) (0:013) (0:020) (0:013)
Constant 0:030 0:150 0:836

0:323
(0:995) (0:347) (0:309) (0:250)
Observations 33 33 33 33
R
2
0.789 0.630 0.703 0.501
Adjusted R
2
0.645 0.376 0.604 0.335
Note:

p<0.05;

p<0.01;

p<0.001
25
closer to “strong republican”) saw a minor decrease in political efficacy. Given the reaction by many
conservatives to President Obama winning the 2008 election alongside the rise of the right-of-right
Tea Party (a small but loud group), it may be somewhat unsurprising given that more individuals
felt that the government was favoring liberal values and that more conservative views weren’t being
acknowledged or heard.
Political participation on the individual level, however, brings several statistically significant results.
First, political efficacy is highly predictive of political participation across all four tests. This falls in
line with previous research that has proven that the more an individual believes they have a voice in
what occurs in the government, the more likely it is that they will participate in some way; in 2012,
the size of the effects was the highest among any of the statistically significant results by a large
margin. Age and income had very small statistically significant predictive power, while education and
liberal-conservative placement had somewhat larger effects. However, none of these other demographic
effects were even one-tenth the size of the effect of political efficacy.
The goal of this paper, however, is to examine the potential effect of Citizens United on political
efficacy and political participation. We found that for states that had a ban of some kind on campaign
finance before Citizens United (that is, in the 2008 election), individuals were actually less likely to
participate in any political process at a statistically significant level. This occurred regardless of the
inclusion of race as variables, with nearly identical effect sizes. However, in 2012, the lifting of bans
increased levels of political participation; this also was present at the same effect sizes at the same level
of statistical significance regardless of the inclusion of race. Of greatest interest, though was that the
size of the absolute value of the effects in 2008 was double the size of the effects in 2012.
This suggests that certain limits on campaign finance may have been more stifling, rather than the
lifted bans being more encouraging of participation. This seems to run counter to the oft-purported
idea that lifting the bans on campaign finance would lead to lower rates of political efficacy. However,
given that the 2012 questions asks individuals if they have done something in the past four years—as
opposed to the 2008 questions that ask only in the past twelve months—this could be explained in part
26
Table 2.5: Individual Level Political Efficacy
Dependent variable:
Political Efficacy
2008 2012 2008 2012
Age 0:0005 0:0003

0:0004 0:0001
(0:0003) (0:0001) (0:0003) (0:0001)
Gender 0:008 0:002 0:008 0:001
(0:010) (0:004) (0:010) (0:004)
Income 0:0004 0:0003 0:0005 0:0001
(0:001) (0:0002) (0:001) (0:0002)
Education 0:003 0:002 0:002 0:001
(0:002) (0:002) (0:002) (0:002)
Liberal-Conservative Placement 0:003 0:001 0:003 0:002
(0:004) (0:002) (0:004) (0:002)
Party ID 0:003 0:007

0:002 0:011

(0:005) (0:001) (0:005) (0:001)
White 0:086 0:002
(0:109) (0:010)
Black 0:077 0:038

(0:109) (0:011)
Asian-American/Pacific Islander 0:067 0:012
(0:113) (0:017)
Latinx/Hispanic 0:078 0:039

(0:109) (0:011)
Native American 0:093 0:022
(0:116) (0:025)
State Ban 0:002 0:001 0:002 0:005
(0:010) (0:004) (0:010) (0:004)
Constant 0:474

0:504

0:402

0:541

(0:113) (0:013) (0:036) (0:009)
Observations 931 4,264 936 4,273
R
2
0.005 0.055 0.004 0.034
Adjusted R
2
-0.008 0.052 -0.004 0.032
Note:

p<0.05;

p<0.01;

p<0.001
27
Table 2.6: Individual Level Political Participation
Dependent variable:
Political Participation
2008 2012 2008 2012
Political Efficacy 0:415

0:796

0:410

0:789

(0:055) (0:040) (0:055) (0:039)
Age 0:002

0:001

0:002

0:001

(0:001) (0:0003) (0:001) (0:0003)
Gender 0:030 0:013 0:027 0:013
(0:017) (0:009) (0:017) (0:009)
Income 0:007

0:002

0:007

0:002

(0:002) (0:001) (0:002) (0:001)
Education 0:034

0:059

0:038

0:058

(0:004) (0:004) (0:004) (0:004)
Liberal-Conservative Placement 0:023

0:012

0:024

0:012

(0:007) (0:004) (0:007) (0:004)
Party ID 0:003 0:005 0:014 0:006

(0:009) (0:003) (0:008) (0:003)
White 0:149 0:026
(0:180) (0:026)
Black 0:235 0:019
(0:181) (0:028)
Asian-American/Pacific Islander 0:212 0:094

(0:188) (0:044)
Latinx/Hispanic 0:239 0:005
(0:181) (0:027)
Native American 0:031 0:086
(0:192) (0:064)
State Ban 0:052

0:020

0:053

0:021

(0:017) (0:009) (0:017) (0:009)
Constant 0:211 0:483

0:495

0:495

(0:189) (0:039) (0:064) (0:031)
Observations 931 4,264 936 4,273
R
2
0.274 0.141 0.254 0.139
Adjusted R
2
0.264 0.139 0.248 0.138
Note:

p<0.05;

p<0.01;

p<0.001
28
by those rallying to maintain the bans on campaign finance. An additional explanation could be that
the media and attention around Citizens United simply made the public more aware of the issue.
2.6 Conclusion
This paper set out to see if there was a relationship between senses of political efficacy and political
participation both before and after SCOTUS decided Citizens United. Overall, the only correlations
of note—that is, involving “state ban” or “political efficacy”, our variables of interest—we found were
at the individual levels of analysis for 2012. Our summary statistics suggests an increase of both
political efficacy and participation at both levels, and our OLS regression of individual level political
participation demonstrates that being in a state that had a ban that was struck down by the case may
have led to an increase in levels of political participation. However, our findings should be taken with
caution. First, given our limited data and the size of the country, we only had 33 data points—in other
words, 33 states—from which to run an analysis. This is hardly a powerful n value, but we are bound by
the number of states in the country and the decentralized nature of record-keeping across and within
the states.
Second, the manner in which the questions were asked for political participation may have clouded
the data. The ANES asked about participation in the past twelve months in 2008, whereas the 2012
survey asked about the past four years. This gives a much longer time period for respondents to consider,
thus providing more opportunities for individuals to have been politically active. Moreover, depending
on when respondents were questioned, there is a very small possibility that the timing overlapped or
that individual respondents included some behavior from 2008. Finally, the removal of race was done
to prevent tiny parsings of data and remove any potential noise. This also ensured that our n values
were of a substantial size. However, this may have influenced our data in unknown or unseen ways.
Regardless, the inclusion of race had no impact on the predictive power or effect size of political efficacy
and the presence of a state ban prior to Citizens United on political participation at the individual level.
29
Future studies could examine the impact of the state ban on specific types of political participation
and particular kinds of senses of political efficacy (rather than as indexes). Another variable to include
would be political knowledge, as the issue of campaign finance is not particularly clear cut. Moreover, it
is possible that not many Americans knew, understood, or cared about the issue or the court case. A
measure of political knowledge might reveal how in-tune an individual is with the campaign finance,
though this would be difficult as no question on the ANES specifically asks about campaign finance
laws. Finally, this kind of study and aggregation could be done at the congressional or county level.
This could complement the other research that has already examined this in some way and hopefully
add to the body of literature so we can better understand how people and why Americans feel about
their involvement—and ability to get involved—in political and democratic process of the republic of
the United States of America.
30
Chapter 3
I’ve Got a Bad Feeling About this: The Utility of Feeling
Thermometers in Measuring Approval of Affirmative Action
3.1 Introduction
Emotions play an important part in political mobilization, political beliefs, and influencing public
opinion around specific policy goals(Namkoong et al., 2012). Much research has found that “negative”
emotions—such anger, fear, threat, and anxiety—mobilize individuals to a greater extent than “positive”
emotions—such as hope and reassurance—and in general play a central role in politics (Banks and
Valentino, 2012; Huddy et al., 2015; Ost, 2004). Many policies in the United States have racialized
connotations, and affirmative action is perhaps the most notorious example, eliciting strong emotional
reactions both in favor and against its implementation.
Large social science surveys employ multiple batteries of questions in order to provide insight
into the identities, opinions, beliefs, attitudes, and behaviors of different populations. Many sets of
questions—and their format—are somewhat standardized across surveys, regardless of the entity
conducting the survey. One such popular tool is the “feeling thermometer”, which has respondents
place how they “feel” towards a particular demographic, individual, or idea on a scale of 0 (meaning
“cold”) to 100 (“warm”) with 50 being neutral. Feeling Thermometers are used on the American National
31
Election Survey (ANES), and manny are employed to have respondents rate how they feel about other
demographic groups, including race.
Affirmative Action policies have been used to attempt to bridge the gaps in education and employment
between whites and marginalized populations; in particular, these policies are aimed at helping African-
Americans/Blacks as a way to make up for whites excluding them from higher education and the
workplace (Hudson, 1995; Salazar Arryo, 2010). However, the past several decades have shown intense
backlash against how different entities have created and implemented such policies, to the point that
affirmative action policies have been challenged multiple times in the judicial system and several
making their way up to the Supreme Court of the United States (SCOTUS) (Perry, 2013; Ziegler, 2018).
This backlash is often characterized as an emotional response to ”unfair” policies, frequently gets the
name ”reverse racism”, and drives opposition to affirmative action overall (Federico and Sidanius, 2002;
Hudson, 1995; Hughes, 1997).
3.2 Measures of Racism
The ANES employs two batteries of questions to measure racism. Disgust with racial minorities—which
is generally rooted in beliefs about race-based biological superiority/inferiority—and fear of the loss
of status (should the lives of minorities improve) was the driving force behind “old-fashioned” racism
and policies of exclusion (Banks and Valentino, 2012; Tesler, 2012). These authors also found that now,
anger is the emotional trigger that increases opposition to racially redistributive policies among white
racial conservatives. Moreover, their study noted that the impact of fear and disgust on political opinion
or participation was minimal compared to the impact of anger. As Tesler (2012) notes, other scholars’
findings on old-fashioned racism indicates that “old-fashioned racist” oppose any policies that work to
improve the lives of African-Americans, simply because they are black and are inferior to whites.
Racial resentment (also called “symbolic racism”), on the other hand, is a prejudiced view of minority
groups—particularly African-Americans or Blacks—that views those groups socioeconomic, political,
academic, and professional failures as a result of lacking true American values and work ethics, and
32
their successes as undeserved due to perceived unfair advantages (Knuckey and Kim, 2015; Sears,
2014) (Sears, 1988; Knuckey and Kim, 2015). Much research has shown that racial resentment and
old-fashioned racism have a massive impact on political behavior (Ford et al., 2010; Huddy and Feldman,
2010; Kam and Kinder, 2012; Kinder and Sears, 1981; Knuckey, 2005, 2011; Rabinowitz et al., 2009; ?).
Separating the two concepts can be somewhat difficult. At face value, ANES questions on racial
resentment generally focus on respondents’ opinions around the work ethic, morality, discipline, fairness
of government assistance, and other forms of American norm and value violations on the part of the
stated demographic group; questions on old-fashioned racism focused on derogatory statements and
desire for distance between races. Some scholars have attempted to distinguish the two based on
whether the question targets African-Americans as a group (racial resentment), or whether the question
asks about the qualities of individual African-Americans (old-fashioned racism) (Virtanen and Huddy,
2006).
3.3 Affirmative Action
Affirmative Action in the United States has a nearly fifty-year history. The term ”affirmative action” was
first used by presidents John F. Kennedy (1961) and Lyndon B. Johnson (1965) in an attempt to prevent
employment discrimination at the federal level. This later extended to educational settings, including
(and especially) institutions of higher education. As previously mentioned, many people—particularly
whites—were vehemently against the use of affirmative action, and saw themselves as victims when
they were unable to gain admission or employment.
There is a long line of cases that grapple with the legal and socioeconomic implications of affirmative
action, and particularly the role of the courts in determining the limits that should be imposed on
affirmative action policies. Over time, SCOTUS evaluated affirmative action policies such as the use
of certain levels of education (e.g. Griggs v. Duke Power Company (1971)), using a quota system for
particular racial-ethnic groups (e.g. Regents of the University of California v. Bakke (1978)), using race
as a general factor in admissions (e.g. Grutter v. Bollinger (2003) and Gratz v. Bollinger (2003)), and
33
the scope of the use of race and other demographic identities in university admissions (e.g. Fisher v.
University of Texas (2013, 2016)) (Perry, 2013; Rabold, 2016; Salazar Arryo, 2010; Ziegler, 2018).
Affirmative Action policies challenged in court (such as the ones listed above) are evaluated under
”strict scrunity”—a judicial tool triggered when there is a violation of rights of one or more individuals
whose identities are at least one of the 14th Amendment delineated protected classes—that requires
a policy have a compelling interest (incredibly important to society/the government) and be narrowly
tailored (cannot be any more specific to achieve the policy’s goal). As it now stands, race may be used
as a factor in university admissions (and, by extent, perhaps employment opportunities) as improving
racial-ethnic diversity is a ”compelling interest”. However, in order for the policy to be ”narrowly
tailored”, race must not be a primary factor, it must not be given a specific amount of weight (e.g. points),
and it can only be used when all other evaluative tools have been exhausted.
Several scholars contest the outcomes of these cases. Given that the disadvantages racial minorities—
especially blacks—face continue to face as a remnant of slavery and Jim Crow, some argue institutions
of higher education should have more free reign to implement race-conscious policies (Brooks, 2005;
Greenberg, 2002). This is particularly important as wealth and education gaps compound across
generations, and there is thus a need to help blacks “catch up”; affirmative action is needed to assist
in creating true equality and fairness because, in the words of President Lyndon B. Johnson, “it is not
enough just to open the gates of opportunity” (Adeniji, 2015).
3.4 Critical Race Theory
One important analytical tool in understanding the role of race in policy making—and its backlash—is
Critical Race Theory (CRT). CRT is an interdisciplinary analytical framework used to examine and
expose policies, practices, customs, and laws that continue to privilege dominant groups with power,
wealth, and resources. CRT has made the theoretical arguments for the following, and form the
backbone of the framework Delgado and Stefancic (2001); Gillborn (2007); Guinier (2004); Gonzales
Rose (2017); Vaught and Castagno (2008):
34
1. Race and racism are socially constructed, flexible categories that are central, endemic, normalized,
and permanent parts of society;
2. White privilege and White supremacy—as well as their related benefits and ideologies such as ob-
jectivity, meritocracy, liberalism, and color-blindness—are constructions of power that perpetuate
the existence of race and racism and that must be challenged and interrupted when they occur;
3. Analyses of race and racism must be historically and socially contextualized, and must support
the lived experiences of those who suffer from racism and dominant ideologies while fighting for
social justice; and finally,
4. CRT operates on the principle that racism is not an individual, psychopathological or social
“disorder” as dominant society would want us to believe, but it is a pervasive, sociocultural,
institutional, and systemic dynamic that serves to regulate the distribution of resources along
racial lines.
When critical race theorists speak of White Supremacy, they are asserting that White Supremacy
is the force that perpetuates—and the ideology that legitimizes—the complete ownership of power
and resources by whites over all other races (Ansley, 1997; Harris, 1993). These ideas are, indeed,
quite normalized, though rarely vocalized by dominant society, though are acutely felt by subordinated
populations (Gillborn, 2007). However, studying, analyzing, and critiquing whiteness is not an “assault”
on white people, but on the socially constructed and reinforced power of white identifications and
interests (Gillborn, 2007; Gonzales Rose, 2017).
In Whiteness as Property, a seminal work by Harris (1993), she explains how white people have a
property interest in their whiteness. The general definition of property (established via common law and
encoded in numerous statues) is that something is considered “propert” when a person or entity may
use the power of the state to enforce their exclusive use or control over it via their “ownership”. These
characteristics may be broken down into four elements:
1. There must be a “thing”, whether tangible or intangible
35
2. There must be a relationship between this “thing” and a person or entity
3. That person or entity must have, or expect, exclusive use and/or control over the “thing”
4. The exclusive use or control must be backed by law and the (violent) power of the state.
Part of what makes Harris’ theory so powerful is the underlying understanding that for property to
provide an advantage, others must not have it nor may they have access to it. This is what makes a
certain form of property so special and exclusive. The sharing of, or expanding access to, this property
thus removes the specialness and exclusivity.
According to Harris, the property interest in whiteness comes about because historically, Whiteness
is associated with intelligence, a strong work ethic, innocence, beauty, and being deserving of power
and resources to have an enjoyable life; other groups are seen as unintelligent, lazy, criminal, ugly, and
therefore undeserving of power and resources or the ability to enjoy life (hence, racial resentment).
Because of the property interest in these traits and their resulting advantages (that have been codified
by law), we thus have a property interest in whiteness. Whiteness becomes the default, moral, good, and
the “ultimate consolation prize” when all else goes wrong (such as when socioeconomic status drops or
identity is threatened—hence how many insults towards white reference “failing” to uphold whiteness,
e.g. hillbilly, redneck, trailer trash, etc.). Using the definition of property from above, Harris establishes
whiteness as a form of property thus so:
1. The thing is being white or being a member of the white “race”, which includes all of its positive
associations
2. Relationship to a person being their racial identity and their literal white skin
3. Exclusive use and control over being white and the positive characteristics, exemplified through
negative stereotypes, discrimination, and segregation
4. Backed by law, historically done so through institutions like slavery & Jim Crow and contem-
porarily done so through residential, transportation, economic, and other forms of segregation and
discrimination.
36
Harris then argues that this impacts white political behavior through a backlash against numerous
policies aimed at providing services to low income individuals (who are socially framed as being black)
and other marginalized groups. Most notably, she highlights the (angry) response to affirmative action
by looking at the history of litigation on the topic, arguing that the root of the anger rests in a racist
view of power and privilege within a system that prioritizes white interests (Brooks, 2005; Greenberg,
2002). The focus on racism first and self-interest second has been supported by other scholarship as
well, including on other topics such as busing (Kuklinski et al., 2007; Mcconahay, 1982).
However, neither racism nor self-interest (or group interest) alone account for the opposition to
affirmative action programs; rather, they work in tandem to provide the most thorough explanation
(Gottemoller, 2011). The effect of this theory is to demonstrate that the emotional responses to
affirmative action derive, in part, from the belief that giving African-Americans this “advantage” or “leg
up” unfairly removes the advantage or “right” from whites (Brooks, 2005; Harris, 1993). The American
ethos is to judge individuals by their “merits”, and using race as a factor undermines this American
tradition; this ignores the compounding effect of generations of whites marginalizing and excluding
racial minorities, denying them the chance to access resources to build the “merit” required to advance
in society (Greenberg, 2002; Rabinowitz et al., 2009).
3.5 Conceptualizing Feelings
The goal of this paper is to compare feeling thermometers to old-fashioned racism and racial resentment
in predicting the level of white support for affirmative action policies, specifically towards blacks.
However, the term “feeling” must be conceptualized and defined. Unfortunately, there is no consensus
of how to define “feeling” within political science, psychology, and many other fields (?). The four terms
used when discussing this kind of biopsychological response (BPR) to stimuli are—in alphabetical
order—affect, emotion, feeling, and mood. But what, exactly, do these terms mean? What is their
relationship to one another?
37
Across the literature, affect is often seen as higher in the hierarchy of the four BPRs (?Gross, 2010).
Many scholars assert that affect provides the backdrop or lens through which mood, emotion, and
feeling are interpreted and experienced. For example, affect can be considered quite abstract and is the
body’s way to prepare itself to respond to any circumstance by qualifying and intensifying an experience
(Shouse, 2005). However, this “intensifying” can also be considered slightly lower and more directing
the display or manifestation of emotion (Ketai, 1975).
Emotions are almost always at the bottom of the hierarchy, considered of shorter duration or
“episodic” in nature which is the actual display of what an individual is experiencing (???Shouse,
2005; Trivedi, 2017). This is mediated by affect and mood, understood as a subjective experience.
Emotion, overall, is claimed to be an interpretation of “feeling” and the terms are occasionally used
interchangeably. Feelings are not frequently discussed as their own distinct concept. Rather, the
other terms are used to explain how a feeling is expressed or understood. Not only are they used
interchangeably, but sometimes “feeling” is completely replaced by emotion (??Zadra and Clore, 2011).
Mood, like affect, is a term that is generally used to indicate a frame of mind that impacts BPRs
(??Gross, 2010; Ketai, 1975; ?; Shouse, 2005; Trivedi, 2017; Zadra and Clore, 2011). Trivedi (2017)
argues that moods are pervasive, but unintentional and non-targeted dispositional states which impact
“all other conscious mental events”. Similarly, Pathank et. al. (2011) asserts that moods are not
necessarily tied to specific objects or events, though Beedie et. al. found that many people believe that
mood can be altered by an emotion.
Given the wide variety of conceptualizations of—and relationships between—affect, emotion, feeling,
and mood, it is important to set out my own theoretical framework for how to understand what a
“feeling” is and what its relationship is to other measures of individuals’ BPRs. For this paper, I will be
using the following definitions:
1. Affect: The long term and general mental backdrop that mediates an individual’s BPRs to
surrounding stimuli
38
2. Mood: Influenced by affect, a somewhat shorter mental framework that mediates an individual’s
more immediate BPRs to environmental stimuli
3. Emotion: Influenced by mood, the immediate biochemical “knee-jerk” reaction to surrounding
stimuli before it can be consciously or cognitively processed
4. Feeling: The cognitive interpretation, understanding, articulation, and expression of an individ-
ual’s emotion
These distinctions are important to understand what, exactly, a feeling thermometer is measuring.
Nowhere in the 2004, 2008, 2012, or 2016 ANES surveys does it ask about a respondent’s mood. The only
context in which the survey asks about an individual’s emotion is when asking what emotion they feel
when they see the United States flag and whether or not they would describe themselves as “emotionally
stable”. Therefore, it is possible that respondents are treating the feeling thermometer as representing
their “emotion”, which is directly influenced by mood (and thus, indirectly influenced by affect). An
individual’s emotions and mood are also likely to be influenced by their partisanship and their political
ideologies, which are stable and lasting parts of an individual’s identity and understanding of the world
around them (Beaumont, 2011; Federico and Sidanius, 2002; Gottemoller, 2011; Wilcox et al., 1989).
There is a debate about the relationship between party ID, partisanship, racial resentment, and old-
fashioned racism in support for race-conscious policies and intervention. Some research highlights the
importance of political ideology and partisanship, finding that it is more predictive of racial resentment
and old-fashioned racism, and in turn, of attitudes towards race-conscious policies than measures of
racism(Link and Oldendick, 2006; Virtanen and Huddy, 2006). However, many more scholars have
found that levels of racism are far more predictive of support (Federico and Sidanius, 2002; Gottemoller,
2011; Huddy and Feldman, 2010; Hughes, 1997; Pierce, 2013; Wilcox et al., 1989; Wilson, 2007).
39
3.6 Feeling Thermometers
Feeling thermometers are used to gauge the respondent’s “feeling” towards a demographic group,
individual, policy, political party, ideology, and many more topics. However, although the tool has been
used for decades, is has been subject to wide criticism for its accuracy in measuring attitudes (Cohen,
2018). For example, scholars have used feeling thermometers as an evaluative tool to approximate
emotion and affect, substituting the term “feeling” for these other BPRs (Gottemoller, 2011). Another
issue with feeling thermometers rests in whether or not the 101-point scale has any utility. For example,
Alwin (2007) reviewed past research on optimal category size for survey questions that generally
supported the idea that the more categories, the more informative the responses. However, feeling
thermometers end up being a 9-point scale, and Alwin notes that “rarely do more than 3% to 5% of
respondents give responses outside of the 9-point scale” (Alwin, 2007).
Additionally, the method of gathering feeling thermometer responses alters responses. Face-to-face
respondents overall tend to provide warmer responses, while those taking surveys on the internet give
cooler but more reliable responses Liu and Wang (2015). Another key aspect that may be overlooked
is that some respondents are more likely to provide warmer or cooler responses in general. Moreover,
some respondents use the entire 101-point scale for their answers, while others might only use a small
range (Wilcox et al., 1989). Wilcox reviews past literature and finds that although many different tools
have been proposed to adjust the varying responses, they tend to have similar results but ultimately
does not provide significantly more information than not adjusting the responses in the first place.
As mentioned above, the exact measure of feeling thermometers is unclear. Some research suggests
that it is a measure of affect (Mcconahay, 1982; Sears, 2014) while others seem to combine feeling and
emotion into one concept or use them interchangeably (Shikano and K¨ appner, 2015; Stangor et al.,
2011; ?). Due to the lack of agreement of the different terms, there is no one right answer to what a
feeling thermometer truly is, if it is important, and if so, why. This paper intends to examine one such
potential use of feeling thermometers.
40
3.7 Methodology & Hypotheses
Given the theories explained above, I propose the following hypotheses:
H1 As cold feelings increase, whites’ opposition to affirmative action will slightly increase
H2 As racial resentment increases, whites’ opposition to affirmative action will moderately increase
H3 As old-fashioned racism increases, whites’ opposition to affirmative action will greatly increase
I will use data from four different years of the American National Election Studies (ANES): 2004,
2008, 2012, and 2016. I will use the following independent variables:
• Age (a continuous variable)
• Race (six categories, each binary with a 1 being for the selected category)
• Gender (binary, with 1 being female)
• Political Ideology (a 7 point scale from conservative to liberal)
• Party Identification (a 7-point scale from strong republican to strong democrat)
• Education (set in ranges, each level receiving one integer for its rank; a higher rank means a
higher level of education)
• Income (set in ranges, each level receiving one integer for its rank; a higher rank means a higher
income)
• Old-fashioned racism (set to an index of 0 to 1)
– In 2004, the variables were as follows:
1. Where would you rate BLACKS on [a 7-point scale ranking a group as hard working or
lazy]?
2. Where would you rate BLACKS on [a 7-point scale ranking a group as intelligent or
unintelligent]?
41
– In 2008, the variables were as follows:
1. Where would you rate BLACKS in general on [a 7-point scale ranking a group as hard
working or lazy]?
2. Where would you rate BLACKS in general on [a 7-point scale ranking a group as intelli-
gent or unintelligent]?
– In 2012, the variables were as follows:
1. Where would you rate BLACKS in general on [a 7-point scale ranking a group as hard
working or lazy]?
2. Where would you rate BLACKS in general on [a 7-point scale ranking a group as intelli-
gent or unintelligent]?
– In 2016, the variables were as follows:
1. Where would you rate BLACKS in general on [a 7-point scale ranking a group as hard
working or lazy]?
2. Where would you rate BLACKS in general on [a 7-point scale ranking a group as peaceful
or violent]?
• Racial Resentment/Symbolic Racism (set to an index of 0 to 1)
– In 2004, the variables were as follows:
1. Blacks should work their way up like other groups
2. History [generations of slavery and discrimination] makes [it] more difficult for blacks to
succeed
3. Blacks should try harder to succeed
4. Blacks [have] gotten less than they deserve
– In 2008, the variables were as follows:
1. Blacks should try harder to get ahead
42
2. Blacks [have] gotten less than they deserve
3. Generations of slavery and discrimination make it more difficult for blacks to get out of
poverty
4. Blacks should work their way up without any special favors
– In 2012, the variables were as follows:
1. Blacks should try harder to get ahead
2. Blacks [have] gotten less than they deserve
3. Generations of slavery and discrimination make it more difficult for blacks to get ahead
4. Blacks should work their way up without any special favors
– In 2016, the variables were as follows:
1. Blacks must try harder to get ahead
2. Blacks have gotten less than they deserve
3. Generations of slavery and discrimination make it more difficult for blacks to get out of
poverty
4. Blacks should work their way up without any special favors
• Feeling Thermometer towards African Americans/Blacks
The dependent variable will be approval of Affirmative Action policies. All questions were re-
configured such that higher values/scale placements meant higher approval of affirmative action
policies. ANES surveys ask about affirmative action in employment; thus, it will not exactly match
up with Harris’s theoretical explanation which focused on affirmative action in higher education.
Nevertheless, this is the closest approximation given the limits of the survey data.
• In 2004, the variables were as follows:
1. Are you FOR or AGAINST preferential hiring and promotion of blacks?
2. Should [the] government see to fair employment for blacks?
43
3. Where would you place yourself on [a 7-point scale ranking whether the government should
provide assistance to blacks]?
• In 2008, the variables were as follows:
1. Do you favor preferential hiring/promotion for blacks? (Originally followed by two more
binary questions on ”how strongly” for yes or no, combined to make a four-point scale)
2. Where would you place yourself on [a 7-point scale ranking whether the government should
provide assistance to blacks]?
• In 2012, the variables were as follows:
1. How much do you favor preferential hiring and promotion of blacks?
2. What is your opinion about the government ensuring fair jobs for blacks?
3. Where would you place yourself on [a 7-point scale ranking whether the government should
provide assistance to blacks]?
• In 2016, the variables were as follows:
1. How much do you favor preferential hiring and promotion of blacks? [4-point scale]
2. Where would you place yourself on [a 7-point scale ranking whether the government should
provide assistance to blacks]?
Although the questions for old-fashioned racism and racial resentment were consistent over all four
surveys (except for the change from asking about intelligence to asking about violence in 2016), the
questions on affirmative action varied between two and three questions. Due to this variation, the
dependent variable of support for affirmative action will be indexed for the overall approval level of
affirmative action policies as a whole. Thus, in order to approximate each year onto the same scale,
I will take the sum of the total values of the questions and then divide that sum by the number of
questions. While this is an imperfect solution, this provides the most accurate reading possible of the
overall approval rating of affirmative action policies without eliminating the third question. I will use
44
OLS regression to analyze the predictive power of old-fashioned racism, racial resentment, and feeling
thermometer scores for approval of affirmative action policies.
I also expect political ideology and party identification to play a large role in predicting opposition
to affirmative action policies, given that political ideology often correlates with beliefs normatively
considered to be racist, oppressive, or discriminatory (Ford et al., 2010; Fowler, 2016; Gottemoller, 2011;
Huddy and Feldman, 2010; Huddy et al., 2015; Rabinowitz et al., 2009; Sears, 2014; Sidanius et al.,
1996).
3.8 Results & Discussion
The regression tables in this section demonstrate measured effects and their statistical significance
on support for affirmative action, with higher values of feeling thermometer, racial resentment, and
old-fashioned racism scores meaning higher levels of cold feelings, resentment, and disgust towards
Blacks. As Table 3.1 demonstrates, racial resentment had no statistically significant predictive value
in support for affirmative action, while old-fashioned racism was the opposite in that it was highly
statistically significant across all years. Feeling thermometers were statistically significant only in 2012
and 2016, and their effect sizes were similar with the effect sizes of old-fashioned racism. With respect
to all of the independent variables at large, old-fashioned racism was the only variable to consistently
correlate with changes in levels of support for affirmative action.
Table 3.2 shows the differences in R-squared values as an additional check to the primary results.
Here, we see that across the board, feeling thermometer and racial resentment scores had almost exactly
the same amount of explanatory power as the other. However, old-fashioned racism scores soundly
trounced the other two measures; in 2004, the old-fashioned racism scores explained 42% percentage
points more of the obtained affirmative action support results than the demographic variables alone,
more than double what either the feeling thermometer or racial resentment scores explained. Of
particular note is the high level of explanatory power of all three variables of interest for support for
affirmative action in 2016, an election many consider to have been extraordinarily racially charged.
45
Looking down the line of statistically significant variables in 2016, we see that one is less likely to
support affirmative action if one is an older, more educated, more conservative, more republican-
identifying white male who has cooler feelings and feelings of disgust towards Blacks. While all reached
the same level of statistical significance, feeling thermometers and old-fashioned racism had vastly
larger effect sizes to predict whites’ levels of support for affirmative action.
Table 3.1: Results
Dependent variable:
Support for Affirmative Action
2004 2008 2012 2016
Age 0:004 0:002 0:0003 0:004

(0:006) (0:003) (0:001) (0:001)
Gender 0:017 0:178

0:004 0:006
(0:183) (0:083) (0:019) (0:036)
Education 0:010 0:001 0:001 0:016

(0:019) (0:007) (0:001) (0:002)
Income 0:109 0:049 0:020

0:015
(0:124) (0:041) (0:009) (0:018)
Liberal-Conservative Placement 0:013 0:014 0:067

(0:033) (0:008) (0:016)
Party ID 0:086 0:038 0:066 0:083

(0:191) (0:023) (0:042) (0:012)
Feeling Thermometer 0:315 0:297 0:184

0:517

(0:568) (0:220) (0:053) (0:097)
Racial Resentment 0:112 0:023 0:011 0:033
(0:104) (0:029) (0:009) (0:017)
Old-Fashioned Racism 0:507

0:386

0:106

0:536

(0:117) (0:066) (0:013) (0:020)
Constant 5:390

5:033

3:409

6:101

(0:693) (0:332) (0:063) (0:110)
Observations 59 395 3,518 1,870
R
2
0.467 0.149 0.036 0.606
Adjusted R
2
0.382 0.130 0.034 0.604
Note:

p<0.05;

p<0.01;

p<0.001
46
Figure 3.1: Support for Affirmative Action
47
Table 3.2: R-Squared Marginal Tests
2004 2008 2012 2016
Baseline 0.03 0.09 0.01 0.39
Feeling Thermometer 0.21 0.10 0.02 0.44
Racial Resentment 0.21 0.10 0.01 0.43
Old-Fashioned Racism 0.45 0.15 0.03 0.59
Feeling Thermometer
Baseline Difference 0.18 0.01 0.01 0.05
Racial Resentment
Baseline Difference 0.18 0.01 0.00 0.05
Old-Fashioned Racism
Baseline Difference 0.42 0.05 0.02 0.20
I ran three additional models to capture how each measure of racism fared against only one other
measure. In Table 3.3, we see that feeling thermometers hold no statistically significant predictive
power in either 2004 or 2008, but are extremely significant in 2012 and 2016. Racial resentment, on the
other hand, swept the board at having extremely high levels of statistical significance. However, both
measures not only had signs going the same direction, but their effect sizes were also on par with one
another even when feeling thermometers had no statistically significant value. This suggests that—as
one might expect—that higher levels of racism lead to lower levels of support for affirmative action
regardless of its predictive power compared to other variables. When comparing feeling thermometers
to old-fashioned racism, Table 3.4 actually shows that feeling thermometers had statistically significant
levels of predictive power across all four years (in contrast to old-fashioned racism’s two years) and in
the years where both variables were statistically significant, feeling thermometers had ten times the
level of effect. Examining racial resentment and old-fashioned racism together shows that old-fashioned
racism was far more predictive for level of support for affirmative action than racial resentment. As
Table 3.5 demonstrates, old-fashioned racism was statistically significant across all four years, and
48
Figure 3.2: Paired Significance Relationship
in the one year where both variables were significant, old-fashioned racism’s effect size was ten times
larger. These paired results create a sort of “rock-paper-scissors” relationship as shown in Figure 3.2.
These results lead to mixed support for the original hypotheses. As Table 3.1 demonstrates, old-
fashioned racism is the clear front-runner of predicting support of affirmative action when looking at
all three variables of interest simultaneously. In this same model, racial resentment had no predictive
power in any year, while feeling thermometers had predictive power in both 2012 and 2016. As a
thought experiment, even if both feeling thermometers and racial resentment had been statistically
significant across all four years, feeling thermometers still held more predictive power than racial
resentment. This means we cannot reject the null hypothesis that racial resentment towards Blacks
has no predictive power of white support for affirmative action policies when analyzed in tandem with
feeling thermometers and old-fashioned racism.
When old-fashioned racism is not at play, however, racial resentment does not hold much of a
greater effect than feeling thermometers in predicting white support of affirmative action policies (as
seen by Table 3.3). This rejects the hypothesis that racial resentment has a larger influence than
feeling thermometers in predicting support for affirmative action policies. Table 3.5 shows that while
old-fashioned racism still holds larger predictive power than racial resentment—thus maintaining H1
that old-fashioned racism has a larger effect than racial resentment—with a much larger effect, this is
in sharp contrast when comparing feeling thermometers with old-fashioned racism.
49
Table 3.3: Results for Feeling Thermometer & Racial Resentment
Dependent variable:
Support for Affirmative Action
2004 2008 2012 2016
Age 0:003 0:002 0:0003 0:004

(0:006) (0:003) (0:001) (0:001)
Gender 0:005 0:172

0:005 0:003
(0:180) (0:083) (0:019) (0:036)
Education 0:009 0:001 0:001 0:016

(0:018) (0:007) (0:001) (0:002)
Income 0:098 0:051 0:020

0:015
(0:122) (0:041) (0:009) (0:018)
Liberal-Conservative Placement 0:012 0:014 0:065

(0:033) (0:008) (0:016)
Party ID 0:064 0:039 0:064 0:084

(0:186) (0:023) (0:042) (0:012)
Feeling Thermometer 0:522 0:246 0:154

0:591

(0:530) (0:211) (0:047) (0:090)
Racial Resentment 0:542

0:375

0:103

0:543

(0:111) (0:065) (0:013) (0:020)
Constant 5:146

5:065

3:431

6:007

(0:648) (0:330) (0:060) (0:098)
Observations 60 395 3,526 1,879
R
2
0.456 0.148 0.035 0.604
Adjusted R
2
0.382 0.130 0.033 0.602
Note:

p<0.05;

p<0.01;

p<0.001
50
Table 3.4: Results for Feeling Thermometer & Old-Fashioned Racism
Dependent variable:
Support for Affirmative Action
2004 2008 2012 2016
Age 0:003 0:003 0:0002 0:007

(0:007) (0:002) (0:001) (0:001)
Gender 0:129 0:061 0:004 0:048
(0:211) (0:062) (0:019) (0:042)
Education 0:019 0:003 0:001 0:015

(0:022) (0:006) (0:001) (0:003)
Income 0:039 0:028 0:007 0:072

(0:142) (0:029) (0:009) (0:021)
Liberal-Conservative Placement 0:063

0:003 0:190

(0:024) (0:008) (0:018)
Party ID 0:129 0:081

0:144

0:153

(0:221) (0:017) (0:041) (0:014)
Feeling Thermometer 1:254

0:393

0:253

1:001

(0:612) (0:169) (0:052) (0:112)
Old-Fashioned Racism 0:236

0:023 0:001 0:132

(0:117) (0:022) (0:009) (0:020)
Constant 4:253

4:271

3:182

5:566

(0:746) (0:176) (0:057) (0:126)
Observations 59 795 3,541 1,875
R
2
0.266 0.101 0.019 0.457
Adjusted R
2
0.165 0.092 0.017 0.455
Note:

p<0.05;

p<0.01;

p<0.001
51
Table 3.5: Results for Racial Resentment & Old-Fashioned Racism
Dependent variable:
Support for Affirmative Action
2004 2008 2012 2016
Age 0:004 0:002 0:0003 0:004

(0:006) (0:003) (0:001) (0:001)
Gender 0:009 0:191

0:001 0:022
(0:181) (0:082) (0:019) (0:036)
Education 0:011 0:001 0:001 0:016

(0:018) (0:007) (0:001) (0:002)
Income 0:111 0:056 0:021

0:019
(0:123) (0:040) (0:009) (0:018)
Liberal-Conservative Placement 0:008 0:016 0:064

(0:033) (0:008) (0:016)
Party ID 0:084 0:041 0:084

0:087

(0:189) (0:023) (0:042) (0:012)
Racial Resentment 0:131 0:006 0:003 0:068

(0:097) (0:024) (0:008) (0:016)
Old-Fashioned Racism 0:532

0:389

0:113

0:557

(0:107) (0:065) (0:013) (0:020)
Constant 5:453

5:053

3:424

6:154

(0:679) (0:327) (0:063) (0:110)
Observations 59 401 3,533 1,871
R
2
0.464 0.146 0.032 0.600
Adjusted R
2
0.391 0.129 0.030 0.599
Note:

p<0.05;

p<0.01;

p<0.001
52
Overall, this means that disgust—an emotion—is generally seen as the strongest factor at play
when told to consider affirmative action policies. Given the decades of affirmative action being publicly
discussed and included in both news and entertainment media, it is unlikely that the survey is the
first time an individual has heard about affirmative action policies. Therefore, the broader topic
of ones opinions towards Blacks can be considered a general “mood” when prompted to consider
affirmative reaction. In this case that would be represented by racial resentment, which would always
be lurking in the background. The specific feeling towards Blacks—theoretically captured by the feeling
thermometer—would be the self-interpretation of an individual’s level of disgust. If one feels disgust
towards a group, then it follows that they would not feel particularly warmly towards the group.
However, the election of Obama as President arguably flipped the racial narratives upside down by
showing the world the success of a black man in America (though news media tended to overlook the
fact that in actuality, Obama is multiracial). On the other hand, his success then had the undesirable
result of making it seem like Blacks had no barriers to achieve the American Dream, and thus did not
need any additional assistance. Therefore, any blacks who could not improve themselves—in other
words, to the public eye that would be most blacks due to blacks disproportionately being low-income,
incarcerated, etc.—had individual-level/group level defects (old-fashioned racism) and/or needed to put
in the effort since they had been helped enough and there was no longer a need for government support
or any kind of affirmative action (racial resentment). Going question by question, we see the following:
• Old-Fashioned Racism
1. Where would you rate BLACKS in general on [a 7-point scale ranking a group as hard
working or lazy]? (Obama worked hard and became president; therefore, other blacks were
lazy for not working hard enough to achieve similar levels of success)
2. Where would you rate BLACKS in general on [a 7-point scale ranking a group as peaceful
or violent]? (Obama was not seen as a violent person, but blacks are disproportionately
incarcerated; therefore, blacks must generally be violent)
53
3. Where would you rate BLACKS in general on [a 7-point scale ranking a group as intelli-
gent or unintelligent]? (Obama must have been intelligent to graduate from Harvard Law
School; therefore, since blacks generally don’t have as high levels of education, they must be
unintelligent)
• Racial Resentment
1. Blacks must try harder to get ahead (Obama tried hard and got ahead; therefore, other blacks
should be able to try hard and achieve success)
2. Blacks have gotten less than they deserve (Obama became president, the highest position in
the United States; therefore, blacks now have gotten the same opportunities as whites)
3. Generations of slavery and discrimination make it more difficult for blacks to get out of
poverty (Obama was able to succeed; therefore, the negative impact on blacks of slavery and
discrimination is over)
4. Blacks should work their way up without any special favors (Obama was able to achieve
success; therefore, the need for special favors is over)
Of course, many of these justifications contain several assumptions. For example, one such assump-
tion would be that Obama did not benefit from affirmative action. Another would be assuming his
success is the rule, rather than being the exception. Finally, that Obama did not benefit from his mother
(Ann Dunham) being white and thus not barred from accessing many resources generally unavailable
to blacks. These examples do not create an exhaustive list, as there are a myriad of other possible
assumptions being made about Obama’s success. Nevertheless, this could explain the distinction we see
in 2004 (before Obama’s election) from subsequent years.
This leaves questions about how the feeling thermometer may be impacted by the respondents’
affect, mood, or emotions at the time of taking the survey. Specifically, what is the impact of the overall
political climate, which may be heightened or emphasized during an election year? What about an
individual’s emotions throughout the course of the survey? Or any emotions immediately dredged up
54
when asked these racially-charged questions? While these issues theoretically impact all responses,
it is especially important when attempting to directly measure respondents’ feelings. Even though
some assumptions could be made with regards to a narrower affect, mood, emotion, or feeling and their
interactions with one another, this does not provide the complete picture. Without any questions on the
survey attempting to ascertain the respondent’s current affect, mood, or emotional state, the feeling
thermometer scores itself should be taken with caution.
3.9 Conclusion
This paper sought to compare the utility of feeling thermometers in predicting support for affirmative
action when compared to the utility of measures of old-fashioned racism and racial resentment. While
all three tools were generally statistically significant across all four years of ANES surveys, the impact
of each measure was dramatically different and changed depending on which other measures were being
tested at the same time. However, when feeling thermometers are present—and regardless of their
level of statistical significance—they do not tell us any information that could not be ascertained from
the other two measures. Moreover, the efficacy of feeling thermometers to provide accurate readings
becomes questionable without considering measuring the respondents’ current affective state, mood,
and emotional state.
Future studies could look at feeling thermometers to predict support for other types of policies
or candidates. The statistical significance and predictive power of party identification and political
ideology when compared to the other three measures varied, thus making it a prime candidate for
further comparison. Finally, more research could be done to evaluate why these three tools may have
varied so much between presidential election years.
55
Chapter 4
The Case(s) For Advocating for, and Supporting, LGB Rights
Introduction
Given that non-voting political participation in the United States is relatively low, what encourages the
participation of those individuals who do get involved in the political process? What helps to create
cultural shifts and changes in public opinion to support a previously marginalized group? What helps
to explain the rapid change in public opinion regarding the rights of lesbian, gay, and bisexual (LGB)
populations?
Using data from multiple years of surveys and incorporating theories from across disciplines
(including political science, sociology, media studies, communications, law, and more), this paper explore
the questions above to more precisely connect various takes on explaining political participation and
support for LGB civil and political rights. Part I will provide a broad overview of the literature around
political participation and creating support for particular policies and ideologies. Part II will provide a
truncated explanation of different strategies that have been used to achieve LGB(T) rights over the
past several decades, focusing on the use of litigation. Part III will explain the analytical methodology
employed, while Part IV will discuss the results. Part V will mention limitations to the findings of the
article, followed by concluding remarks.
56
4.1 Political Participation in Review
Political participation and civic engagement are considered to be cornerstones of a healthy and func-
tioning democratic society, and democracies should not only value all voices but work to ensure that
all said voices receive their due representation (Acemoglu et al., 2005; Berinsky and Lenz, 2011). One
frequently studied issue in the social sciences is what, exactly, predicts or motivates individuals to
participate. The topic has been researched from numerous different angles, and the below sections
will briefly review some of the various takes that are most pertinent to this paper. Section A will focus
on studies on demographic identity. Section B will look at the role of media consumption. And finally,
Section C will address political polarization and emotion.
4.1.1 Demographic Identity
Several scholars have investigated the link between an individual’s level of education and their level
of political participation, and found a positive correlation between them (with a slight drop-off for
individuals with PhDs or their field’s equivalent highest degree) (Almond and Verba, 1989; Jackson,
1995; Glaeser et al., 2007; Persson et al., 2016). Other scholars have examined demographic factors
such as race, ethnicity, and religion (Bany, 2012; Cunningham and Regan, 2012; Read, 2015), as well
as various occupational backgrounds (Wright-Phillips, 2015; Soon and Kluver, 2014), all finding that
a strong sense of shared identity or behavior (including type of jobs or shared hobbies, particularly
those that are tied to ones community or culture) positively correlates to increased engagement with,
and mobilization around, political issues. One of the most salient findings in this line of research is
that of linked fate, which is the idea that one’s own social, economic, and/or political fate or destiny is
inextricably linked to the fate or destiny of other members of one’s racial group (Dawson, 1994). Linked
fate often, but not always, correlates with individual political political behavior and policy preferences
(Dawson, 1994; Masuoka, 2006; Rim, 2009; Sanchez and Medeiros, 2016; Simien, 2005). Though the
concept of linked fate was coined by Dawson with African-Americans in mind, the logic can be—and has
been—expanded to other demographic groups as well.
57
Perhaps most important for this paper is the ability to predict the political participation of those
who identify as LGB. Swank and Fahs had strong and thorough studies on the topic, and they note that
gays and lesbians who routinely spoke with other gays and lesbians ended up being more politically
active overall; being more active within LGB communities lead to a larger likelihood of voting for liberal
political candidates; and being contacted by the Democratic party specifically increased the likelihood
of voting for a lesbian candidate (Swank and Fahs, 2013). However, they find that two of the largest
motivators for political participation are self/community identification and framing (e.g. whether or not
one is “out”, “closeted”, “silenced”, etc.) and surviving hate crimes or other forms of homophobic bullying
(Swank and Fahs, 2013). Both motivators also relate to prior theories on the importance of demographic
identity and a sense of linked fate in inspiring greater levels of political participation..
4.1.2 Media
Another line of research examines the relationship between media consumption and political par-
ticipation, with much scholarship finding a positive effect of the amount of political news media
consumed—particularly partisan news—and the viewer’s political participation (Bachmann et al., 2010);
(Bachmann and De Z ´ u ˜ niga, 2013); (Tien-Tsung and Lee, 2005; Wojcieszak et al., 2016). For example,
self-identified liberals who consumed news media they considered hostile to their political beliefs
correlated with increased political participation, while self-identified conservatives who considered news
media hostile to their political beliefs actually decreased political participation (Feldman et al., 2017).
The primary exception uncovered was that among weak conservatives and moderates, hostile media
perceptions had no relationship with activism (Feldman et al., 2017). Relatedly, another study observed
in an overall sample (not separated by political party or ideology) that consumption of pro-attitudinal
news increased political participation, while counter-attitudinal and balanced news sources did not
have an effect on political participation (Wojcieszak et al., 2016); In this case, pro-attitudinal news could
still describe ideas or policies hostile to the consumer’s political views, though such information would
be filtered through an ideologically-similar lens..
58
In contrast, Suzanna Dilliplane found partisan political news consumption had no effect on the
“strongly habitual nature of voting” (Dilliplane, 2011). This falls in line with Anthony Downs’s assertion
that voting is largely seen as a duty, and therefore the chance of a voter turning out is primarily affected
by the voter’s subjective determinations of how likely it is that their vote will matter, the cost of voting
(e.g. temporal, financial, etc.), and their sense of civic duty (Downs, 1957). The formula for determining
the likelihood of voter turnout (V) is V = pBC +D, where p is the probability that a voter’s vote
will matter; B is the perceived benefit of the voter’s chosen candidate/policy; C is the cost of voting;
andD is the voter’s level of civic duty. From this, note thatB would need to be extraordinarily high in
order to offsetp, whose value will generally be infinitesimally small, but incredibly influential. C and
D are left to mitigate or enhance one another and/or the result ofpB.. Although Dilliplane found no
relationship between partisan news and voter turnout, she did find that consuming partisan news had a
positive correlation with when a voter made their substantive electoral decisions (Dilliplane, 2011). Vote
decisions were made earlier (measured as time before election day) if the voter consumed like-minded
news, while decisions were made later if the voter consumed conflicting or opposing news. However, her
findings illustrated that partisan news consumption was strongly correlated with political campaign
involvement; once again, like-minded news increased involvement while conflicting news decreased
involvement (Dilliplane, 2011)..
4.1.3 Political Polarization and Emotion
While partisan media is not a new phenomenon, the level of political polarization within the United
States has recently been given increased attention by scholars and society at large. Scholars are
particularly interested in the perceived and/or actual growing divide between political parties, political
ideologies, individual politicians, and the American public at large (Baldassarri, 2011; Dixit and Weibull,
2007; Fiorina et al., 2010; Carmines et al., 2012; Westfall et al., 2015). Of particular concern and
relevance to this paper is the (perhaps, in some ways, predictable) impact of party identification and
ideology on participation and support for gay rights. Increased polarization between parties—that is,
59
tendency for political views to become more extreme rather than moderate—will impact the degree
of partisanship seen in political news (as the political news being reported will be inherently more
partisan) which in turn will affect the political participation of the news consumer (Dilliplane, 2011).
Unsurprisingly, emotions also play a large role in political mobilization and mediating other political
interactions such as news consumption or interpersonal debates (Namkoong et al., 2012; Weber, 2013;
Woods et al., 2012). In general, “negative” emotions—such as anger, fear, threat, and anxiety—mobilize
individuals to a greater extent than “positive” emotions—such as hope and reassurance—and in general
play a central role in politics (Huddy et al., 2015, 2016; Ost, 2004; Valentino et al., 2011; Weber,
2013). For example, disgust with racial minorities— rooted in beliefs about race-based biological
superiority/inferiority—and fear of the loss of status (should the lives of minorities improve) was the
driving force behind “old-fashioned” racism and policies of exclusion (Banks and Valentino, 2012). Now,
anger is the emotional trigger that increases opposition to racially redistributive policies among white
racial conservatives (Banks and Valentino, 2012). This study also found that the impact of fear and
disgust on political opinion or activism was minimal compared to that of anger.. Similar to how fear,
anger, and disgust prompts backlash against racial minorities, these same emotions reflect homophobic
reactions to decades of social progress made for the benefit of lesbians, gays, and bisexuals (LGB). To
be clear, there are multiple sexual orientations and identities outside of “heterosexual” than the three
noted here (for example, transgender is often included for the acronym LGBT). However, the focus of
this paper is on same-sex marriage and thus will focus on sexual orientations that allow for same-sex
marriage. Additionally, the majority of the cases and surveys cited in this paper use this language, so
mirroring such language allows for consistency. Therefore, further use of LGB and LGBT is not meant
to be exclusionary or divisive; rather, it is reflective of the scope of this paper.
This paper draws upon many of the theories described in refsubsec:media3above to explore and
better understand what potential factors correlate with support for gay rights and what factors correlate
with different levels of political participation.
60
4.2 Tactics for Achieving LGB Rights
LGB individuals have had an uphill battle to gain and/or assert their civil, political, and/or human
rights, with increasing visibility over the past 60 years. There is a rich and diverse history of various
political and community organizations mobilizing to protest homophobic policies, lobby for support from
politicians, and provide resources to other LGB individuals and their families. Section A will provide a
very brief overview of activist strategies that have been used, while Section B will focus on litigation as
a strategy.
4.2.1 LGB Activism
Various forms of political participation generally have the goal of securing civil, political, and sometimes
human rights for a particular group, and the LGB population is no exception. Although LGB activism
in the United States preceded the Stonewall Riots by decades, Poindexter’s research echoes a myriad
of other historians who assert that the eruption of political upheaval Stonewall represented was in
part due to the extreme oppression experienced by lesbian, gay, bisexual, and transgender individuals
(especially those of color), and also was a continuation of the prior decade’s political movements and
countercultural philosophies that “supported equality, freedom, choice, and peace” by “highlighting civil
rights, equal opportunity, and fair treatment” (Poindexter, 1997). Furthermore, Poindexter, highlighting
prior research, noted that the ability of LGB organizations to bring issues of homophobic discrimination
to the American public and, perhaps more importantly, to mobilize LGB communities at large was
due—at least in part—to them having access to certain resources (including money, communication, and
labor, among others), experienced social activists, media outlets, and less visible sources of mobilization
such as health clinics (Poindexter, 1997). This led Poindexter to conclude that, rather than the birth of
a movement, Stonewall was one defining feature of the LGBT rights movement, akin to the March on
Washington during the 1960s period of the (Black) Civil Rights Movement.
Human and Mertus focus on the different framing tactics used by various LGBT advocacy groups
to advance gay rights. On the international level, gay rights were (and still are) often framed as
61
human rights (such rights were formalized into the Yogyakarta Principles in 2006 after a meeting of
international human rights groups in Yogyakarta, Indonesia), which advocates tended to agree made
sense on a pragmatic level (Human and Mertus, 2007). This framing was less accepted at the US level
since “policymakers, the general public, and even many social change advocates still view[ed] human
rights as something that applies [sic] not at home, but in some distant land” and thus would not be
successful as a tactic (Human and Mertus, 2007). Part of the struggle among advocacy groups centered
on whether strategies aimed at assimilation or confrontation were most appropriate, with groups like
the AIDS Coalition to Unleash Power (ACT UP) preferring more militant actions and legal advocacy
groups focusing on issues around equality and integration, such as adoption and marriage (Human and
Mertus, 2007). Many militant groups framed themselves as somewhat revolutionary, and sought to
upend the social institutions that imposed specific sex- and gender-based roles.
Human and Mertus argue that Bowers v. Hardwick provides an illustrative example of these
seemingly strategically ideologically opposed groups uniting to protest the outcome of the case–which
upheld anti-sodomy laws–and the government’s poor handling of the AIDS crisis (Human and Mertus,
2007). Despite activist leaders from multiple groups declaring the protests—inherently a disruptive
strategy—as “relatively successful in focusing national attention on the plight and demands of American
gays and lesbians”, many groups retreated away from using disruptive such tactics towards the end of
the 1980s and, in the 1990s, channeled their energy towards reducing discrimination and increasing
access to mainstream resources, which in turn mobilized LGBT people to engage in normative forms of
political participation such as voting, lobbying, and supporting LGBT-friendly candidates (Human and
Mertus, 2007).
4.2.2 Litigation Strategy
Litigation gained traction as a popular strategy in the 1990s, with countless complaints filed in local,
state, and federal courts in order to gain civil and political rights for LGBT citizens (Human and Mertus,
2007). These approaches to collective legal mobilization were not always seen as focusing on law as an
62
exogenous tool for social change or consciousness raising; instead, those approaches seen as a culturally
relevant and systemic tool to communicate socially embedded knowledge around values such as equality
and civil rights from within the very social structures activists were seeking to change (Morrill et al.,
2015).
Despite very real socioeconomic barriers to litigation, Nejaime explains that litigation is still an
attractive option for groups who are disadvantaged through the normative political processes of lobbying
politicians and relying on public support to elect legislators willing to act on their behalf (Nejaime,
2012). This is because “courts generally have an obligation to hear and consider a group’s grievance,
even when lawmakers do not provide a forum”, thus making them far more attractive (Nejaime, 2012).
The independence of courts is challenged by several scholars such as Lewis, who highlights that the
judicial branch of state governments are rarely independent due to judges facing elections to keep their
jobs and thus biases decisions towards those approved by the electorate (Lewis et al., 2014). Though the
courts can still advance the cause of a group when society is still hostile towards the advancement of its
rights, public opinion may be swayed by court rulings which are then influenced by public opinion, thus
creating a cycle moving in unknown directions.
Lewis notes that one role of the courts is to protect and advance minority rights in direct democracy
states which have a greater ability to pass legislation favored by the majority (Lewis et al., 2014). Using
variables such as level of the court, public opinion on a particular issue, justice ideology, and whether
there is a retention election, Lewis finds that the likelihood of a justice in a direct democracy state
casting a vote in favor of gay rights is, on average, reduced by 16% (Lewis et al., 2014). However, state
supreme courts were generally more willing to support LGBT rights claims on state constitutional
grounds in comparison to the federal courts on federal constitutional grounds (Zschirnt, 2016).
When it came to litigation, many groups were selective in which cases they supported or which cases
they directly led. For example, in the last 20 years of the 20th century, Lambda Legal was involved in
27% of 286 state-court appellate cases dealing with LGBT rights, though rather than litigating all of the
case themselves, they primarily did so through amicus curiae briefs (Kane and Elliott, 2014). Lambda
63
directly litigated Romer v. Evans in 1996, and in 1998, Gay and Lesbian Advocates and Defenders
(GLAD) chose to litigate Bragdon v. Abbot (Human and Mertus, 2007). Despite not being in the forefront
of the case, Lambda was able to persuade state courts in Montana, Tennessee, Kentucky, and Georgia
to strike down anti-sodomy laws; later, in 2003, the Supreme Court of the United States (SCOTUS)
declared all anti-sodomy statutes unconstitutional via Lawrence v. Texas, a case that was selectively
and directly led by Lambda. (Human and Mertus, 2007).
Litigation was but one strategy selectively used by LGB rights organizations, but it raises the
question of how or when the choice is made to use litigation as a tool to gain rights, change public
opinion, and garner more sociopolitical support for LGB populations rather than using other existing
structural tools such as elections and lobbying. There is a tension within scholarship as to whether or
not organizations get involved based on the likelihood of success or whether it is able to create enough
publicity that will increase public support, donations, and media presence (Kane and Elliott, 2014;
Nejaime, 2012).
On the one hand, we can synthesize what scholars have argued to reach the understanding that in
many cases, the way that judges vote tends to depend upon some combination of whether or not judges
in that jurisdiction are voted into their seats, public opinion on the issue, and the ideological makeup of
the court (Lewis et al., 2014; Zschirnt, 2016). On the other hand, however, the choice of which cases to
litigate may be more of a function of which cases get far enough in the appeals process, rather than
which cases to an organization should be involved with at the very start of the proceedings.
For example, Kane notes that Lambda Legal is less likely to participate in custody cases than sodomy
law cases(Kane and Elliott, 2014). This supports Leachman’s assertion that—after compiling data
on over 1100 cases pursued by GLAD, Lambda, and the National Center for Lesbian Rights (NCLR)
and tracking their media presence from 1996 to 2006—cases about same-sex relations will create
more media attention and, in line with some of the theories mentioned above, will increase political
participation (Stoutenborough et al., 2006; Leachman, 2016b). Setting aside issues of cost (which is a
very real barrier), almost anyone can start the process of litigation (as the court system is theoretically
64
open to all), and even that single act could be enough to generate media and publicity which can once
again move along the cycle of altering public opinion and political activism. Note that Lambda, the
American Civil Liberties Union, and NCLR all decided not to initiate Baehr v. Miike. There may have
been strategic reasons for this choice such as fear of trying to push the needle of LGBT rights too far
(Leachman, 2016a).
4.3 Methodology
This part will discuss how I examined the relationship between demographic group, court cases and the
surrounding media coverage of cases, support for gay rights, and political participation. The part shall
be divided into two sections. The first section will explain the variables I used to define and measure
political participation and support for gay rights. The second section will explain the cases I highlighted
as potentially having impacted political participation and support for gay rights.
4.3.1 Measurement
I used data from four different years of the American National Election Studies (ANES): 2004, 2008,
2012, and 2016. ANES surveys have been conducted since 1948 to capture data from a random sample
of the electorate on topics such as demographics, political beliefs, political knowledge, and support for
particular politicians or policies. Data collection starts approximately one year before US presidential
election years and concludes a few weeks after the presidential election itself. ANES surveys contain
approximately 2000 questions and have sample sizes from 1,000 to 6,000 individuals, meeting the needs
to have a sufficient sample size.
The individual variables that I used consisted of standard demographic information, which is all
self-reported. Age is a continuous variable (truncated to the nearest integer) beginning at 17. The
ANES has six racial categories: white, Black, Native American, Asian/Asian-American Pacific Islander,
non-white Hispanic/Latino, and other. The “other” category was selected for respondents who were
multi-racial or who did not feel that the other five categories accurately described them. Unfortunately,
65
due to the small sample size of the “other” category across all four surveys being used, the “other”
category was not used in the analyses of this paper. Each respondent received a 1 in the category for
which they indicated their race, and a 0 for each category they did not. Consistent with the survey
options, gender was similarly a single binary variable, where “male” is set to 0 and “female” is set to 1
(the default setting for most social science statistical analyses); no other gender options are presented.
Income and education are both ranked and ordered factors, but shall be treated as continuous data
for analytical purposes. Rather than selecting a specific number of months or years of education, or
rather than specifying their exact income, respondents select predetermined ranges in which they fall.
As an example, one range of income is $15;000$17;499, which will have its own single integer value
(in this case, 5). As an individual’s income bracket increases, so does its integer value (its rank). For this
paper, I also included the question of whether a person considers themselves heterosexual, homosexual,
or bisexual (this is the phrasing of the question). Because the sample sizes of the homosexual and
(especially) the bisexual individual categories are quite small, for the purposes of this paper I combined
into a single response. While this is problematic in that it could reinforce stigmas and stereotypes facing
bisexuals (e.g. “bisexuality is a stepping stone to being gay”), for this paper, bisexual and homosexual
respondents both face discrimination based on their sexual orientation and the fact that they do not
identify as “heterosexual”, and—by definition—are vastly more likely than heterosexuals to be in
same-sex romantic or sexual relationships. Therefore, identifying as heterosexual was set to 0 and
identifying as LGB was set to 1.
The dependent variables are support for LGB rights and political participation. In order to measure
political participation, I used a battery of five questions that have been consistently used across
large social science research to determine political participation. Each survey uses a different time
measurement for consideration. In 2004 and 2016, the survey asks if the respondent had done one of
the above actions in the past 12-months/year. In 2008, the survey asks if the respondent has ever done
one of the above actions, with no mention of a time range. In 2012, the survey asks the respondent to
consider the past four years. These inconsistencies mean that the findings should be taken with caution;
66
however, these surveys are also considered the best ones to study US elections and politics due to their
routine nature, but they are all the ANES offers. The questions ask whether or not the respondent has:
• Participated in a political protest
• Contacted your representative
• Signed an internet petition
• Signed a paper petition
• Donated to a social or political organization
These questions are binary in nature, with a 0 indicating that a person has not done a particular
action and a 1 indicating that they have. To control for scaling for overall participation, I created a
participation index in which a respondent’s score is totaled and divided by 5 to create a 0 to 1 range.
Voting is excluded from this analysis, as much literature demonstrates that voting to is best understood
as a form of civic duty rather than other forms of political engagement engagement and the decision
whether or not to vote has not been found to have been influenced by political media consumption
or substantive political beliefs (as opposed to, for example, party identification). Note that the 2004
ANES did not ask about signing a petition in 2004, nor did the 2016 ANES specify someone signing
either a paper petition or an internet petition (it simply asked about signing “a petition”). Given the
ubiquitousness of the internet in 2016, for the purposes of this study I chose to interpret “petition” as
“internet petition”.
A similar method was used to determine approval of LGB rights. I used a battery of four questions
that ask about individual’s support for the right for gays to:
• Be protected from job discrimination
• Serve in the military
• Adopt children
67
• Become legally married
Because each individual question varies in their response options within each year (e.g., some are
binary while others provide a varying number of other options), I will not be able to create the same
scale-controlled index as with political participation. However, as it is still important to get a sense
of overall levels of support, an additive score was created with higher values always indicating more
support for LGB rights. Therefore, for both political participation and support for LGB rights, higher
numbers always indicate higher levels of political participation and higher levels of support for LGB
rights.
4.3.2 Case Selection
Cases were selected based on their media coverage from 2000 to 2016. This is due to the fact that
the ANES questions vary in how long in the past a respondent has politically participated. However,
to be inclusive of any impact cases may have had at the end of one year, I will include cases that
received media attention in 2000 as well. Moreover, cases often receive publicity well before they are
actually argued in front of the Supreme Court; therefore, in order to capture the impact of media
leading up to the Supreme Court hearing, the timeline was extended. Appellate cases are often the most
important cases as their decisions generally become the leading cases on the issue and establish critical
precedents(Kane and Elliott, 2014). However, while the cases mentioned will focus on state supreme
court, federal circuit court, and SCOTUS cases. Other cases will be included should they have received
national publicity. Note that although the battery of questions for support of LGB rights includes issues
beyond same-sex marriage, the media around those types cases are tremendously overshadowed by
cases around same-sex relations (e.g. intercourse). Therefore, the poll questions and case selections will
be based around the issue of same-sex relations.
Although the ANES does not ask specifically about particular court cases, past research has shown
that court decisions—especially SCOTUS decisions—can have an impact on public opinion, particularly
in the area of gay civil rights. For example, Bowers v. Hardwick was a case that, after a decision had
68
landed, was directly tied to a 12-percentage point drop in public support for the legality of same-sex
relations (Stoutenborough et al., 2006). In contrast, Dale v. Boy Scouts of America and Romer v. Evans
did not receive the same amount of publicity, and thus public opinions on the legality of same-sex
relations did not shift. Prior research also asserts that once SCOTUS has issued a landmark ruling on
a particular issue—for example, laws criminalizing sodomy—subsequent rulings are less likely to cause
a shift in public opinion. Again, even though Romer was a landmark decision, since neither it nor Dale
dealt specifically with same-sex relations, no shifts in public opinion were seen (Stoutenborough et al.,
2006).
The decision in Lawrence v. Texas, on the other hand, was associated with an 8-percentage point
drop in public opinion on the legality of same-sex relations, which the authors explain is due to the
“specifics of the decision, the high media coverage, the content of that coverage, and the political context”
(Stoutenborough et al., 2006). This demonstrates the power the courts have in helping to shape national
public opinion, as even a single judicial decision—regardless of whether it advances or sets back a
cause—can significantly alter the process of social progress (Nejaime, 2012). The complete list of cases
around which I will use public opinion data are as follows (appearing as Case Name (Decision date and
court)):
1. Baehr v. Miike (December 9, 1999 by the Supreme Court of Hawaii)
• Facts: Three same-sex couples sued the Hawaii Department of Health for denying them
marriage licenses on the basis of them being two individuals of the same sex, thus violating
the state’s Equal Protection Clause.
• Legal Question: Whether marriage was a fundamental right that extended to same-sex
couples.
• Ruling: The 1996 ruling stated that the exclusion of same-sex couples from obtaining
marriage licenses constituted discrimination, which the state would have to justify with a
compelling interest. The case was remanded for said fact finding; by 1998, however, a state
amendment had been passed that allowed the state to define “marriage” as opposite-sex
69
couples. This removed the legal foundation of the plaintiff, and the Supreme Court of Hawaii
entered a judgment for the defendant.
2. Baker v. Vermont (December 20, 1999 by the Vermont Supreme Court)
• Facts: Three same-sex couples sued their local Vermont jurisdictions for denying them
marriage licenses on the basis of them being same sex couples, thus violating Vermont’s
constitution.
• Legal Question: Whether the denial of marriage benefits violated the state constitution.
• Ruling: The denial of marriage benefits violated the state constitution, but denying marriage
licenses did not violate the state constitution.
3. Lawrence v. Texas (June 26, 2003 by the Supreme Court of the United States)
• Facts: Two men were arrested for being caught engaging in same-sex intercourse which
violated Texas’s anti-sodomy law.
• Legal Question: Whether the anti-sodomy law violated the Constitution.
• Ruling: All individuals—including gay individuals—had a right to engage in private relations
(excluding sexual relations); that social and/or moral animus of gay “lifestyles” was not a
justification for the anti-sodomy laws; and moral and sexual beliefs and behaviors were
protected under the Constitution.
4. Goodridge v. Department of Public Health (November 18, 2003 by the Massachusetts Supreme
Judicial Court)
• Facts: The Massachusetts Department of Public Health denied marriage licenses to seven
same-sex couples.
• Legal Question: Whether the denial of marriage licenses to same-sex couples violated the
state constitution.
70
• Ruling: Denying same-sex couples the legal benefits of marriage on the basis of being
same-sex couples was a violation of the state constitution.
5. Lewis v. Harris (October 25, 2006 by the Jew Jersey Supreme Court)
• Facts: Seven same-sex couples sued the New Jersey Department of Human Services for
denying them marriage licenses.
• Legal Question: Whether the denial of marriage licenses to same-sex couples violated the
state constitution.
• Ruling: The denial of marriage licenses—and the accompanying legal benefits—violated
same-sex couples’ rights to equal protection under the state constitution.
6. In re Marriage Cases (May 15, 2008 by the Supreme Court of California)
• Facts: In 2004, the city and county of San Francisco began allowing same-sex couples to file
for, and obtain, marriage licenses. This led to anti-marriage equality advocacy groups to file
suit, requesting the lower court to prohibit the distribution of marriage licenses to same-sex
couples. The lower court refused to issue the stay, but the groups appealed to the Supreme
Court of California who then issued the stay. San Francisco then brought a constitutional
challenge against the marriage laws that barred marriage for same-sex couples.
• Legal Question: Whether the denial of marriage licenses to same-sex couples violated the
state constitution.
• Ruling: The denial of marriage licenses—and the accompanying legal benefits—violated
same-sex couples’ rights to equal protection under the state constitution (this was temporarily
invalidated by the passing of Proposition 8 which banned same-sex marriages, though the
proposition itself was overturned in 2010 (though on a substative issues separate from the
constitutional right to marriage for same-sex couples) and eventually led to Hollingsworth v.
Perry).
71
7. Hollingsworth v. Perry (June 26, 2013 by the Supreme Court of the United States)
• Facts: Following the In re Marriage Cases, Proposition 8—which banned same-sex marriages—
was challenged by several groups. Eventually, the case was appealed to SCOTUS. Both
Governor Arnold Schwarzenegger and succeeding Governor Jerry Brown refused to defend
Prop 8, as did new state attorney general Kamala Harris. Thus, the sponsoring organization
requested to file as replacement defendants in lieu of the state government, which SCOTUS
allowed.
• Legal Question: In addition to the legal question of whether Prop 8 violated constitutional
rights, the defendants also had to argue that they had standing to be defendants in the case.
• Ruling: Ignoring the question of the constitutionality of same-sex marriage, SCOTUS ruled
that the defendants lacked Article III standing to file suit. This had the effect of allowing the
lowest court’s ruling—which overturned Prop 8—to stand as the final ruling.
8. United States v. Windsor (June 26, 2013 by the Supreme Court of the United States)
• Facts: A lesbian couple living in New York were married in Canada. One of the women died
and left her estate to her spouse, who then attempted to claim the estate tax exemption.
Although New York state recognized their marriage, the federal government did not due to the
Defense of Marriage Act (DOMA), barring her from obtaining the exemption. Lower courts
ruled that DOMA was unconstitutional, and the Department of Justice denied to defend
DOMA. They did, however, file an appeal to let another group (Bipartisan Legal Advisory
Group) intervene and conduct the legal fight.
• Legal Question: In addition to the legal question of whether the Defense of Marriage Act
(DOMA) violated the Constitutional right of equal protection, the defendants also had to
argue that they had standing to be defendants in the case and whether the DOJ’s refusal to
directly appeal—in other words, agree with—the ruling meant that there was no actual legal
issue to decide.
72
• Ruling: DOMA defining marriage as between one man and one woman violated the Consti-
tution.
9. Obergefell v. Hodges (June 26, 2015 by the Supreme Court of the United States)
• Facts: Multiple court cases combined, all of them dealing with some states refusing to
recognize same-sex marriages of couples living in their state when the marriage license was
obtained from other states.
• Legal Question: Whether the Fourteenth Amendment require states to recognize marriages
from other states, and whether the Fourteenth Amendment requires states provide marriage
licenses to same-sex couples who wish to marry.
• Ruling: All states must provide marriage licenses to same-sex couples who seek to marry
within their state, and must recognize same-sex marriages conducted in other states.
10. Masterpiece Cakeshop v. Colorado Civil Rights Commission (June 4, 2018 by the Supreme Court
of the United States)
• Facts: A bakery refused to provide a wedding cake for the wedding of a same-sex couple. The
couple brought the issue to the Colorado Civil Rights Commission, who deemed the baker
had violated state’s anti-discrimination law. The baker argued that forcing him to provide
the cake violated his First Amendment rights, and the case wound up in the Supreme Court.
• Legal Question: “Whether applying Colorado’s public accommodations law to compel
Phillips to create expression that violates his sincerely held religious beliefs about mar-
riage violates the Free Speech or Free Exercise Clauses of the First Amendment.”
• Ruling: SCOTUS ruled that the Colorado Civil Rights Commission was not religiously
neutral when it investigated the couple’s claim, avoiding the issue of the conflict between
anti-discrimination laws and First Amendment Rights.
73
Polls rarely refer to the cases by name or note the passage of a case. Rather, the polls ask respondents
about their views on same-sex marriage before and after a case decision lands (roughly one month
before and one month after). Additionally, it was not possible to use the same source for all of the cases,
so all data is somewhat approximate. Finally, though the names of the courts may differ, if a state is
named that court is the highest level court in that state. While massive public attention to a case may
show the success of choosing to litigate a case more generally, whether a case so succeeds depends on
the media narratives associated with the issue. This paper will situate itself in Leachman’s work, which
found a direct relationship between the controversial nature of a case, the media coverage around it,
and politically charged public reaction to it (Leachman, 2016b). Taking it a step further, I posit the
following cycle best explains this relationship:
Case Impact Cycle
Figure 4.1: The relationship between controversial court cases, the media, and social reactions
For example, between 1996 and 2006, 40% of cases brought by GLAD, Lambda, and NCLR that
were covered by the media were marriage cases, despite marriage cases accounting for roughly 10%
of all cases litigated by those organizations within that time frame(Leachman, 2016a). With this in
mind, highlighted cases centered on marriage or same-sex relations and all had significant, as there is
little theoretical indication that other cases will move the needle on public opinion (and thus, political
participation).
4.3.3 Poll Data
All poll data was gathered from the Polling the Nation database. The below information will state the
full question, the date of the question’s release, and the entity that conducted the poll or survey. These
74
polls will be used to roughly examine before-and-after the LGB same-sex relations cases delineated
in the prior section. Because of the irregularity in when these polls were conducted, I avoided using
polls from 18 months before or after the court published their decision to ensure the most contextually
relevant information on public opinion. Additionally, viewer audiences differ between different news
media outlets, which means that some polls may be somewhat biased. Finally, the exact wording of the
polls were rarely the same across surveys, and the entities conducting the survey also rarely conducted
both the before and after surveys. However, polls from different sources are still cited in news media,
and thus I contend that they are still useful approximate measures of public opinion.
4.3.3.1 Baehrv. Miike andBakerv. Vermont
While these two cases were decided in 1999, the most recent survey I was able to find on the question of
same-sex marriage was from 1996. Even though it lined up with the 1996 Baker v. Miike, I excluded
that question from the “before” as such opinions may have changed in the subsequent 3 years.
Before:
• Thinking about what might be done to protect the rights of gays and lesbians - do you think there
should or should not be legally-sanctioned gay marriages? (Newsweek on March 11, 2000)
– 34% should
– 57% should not
– 9% don’t know
• Thinking now about gay marriage, that is allowing a couple who are gay or lesbian to marry one
another legally, do you support or oppose gay marriage? Do you strongly support or oppose, or only
somewhat support or oppose? (Los Angeles Times on June 16, 2000)
– 22% support strongly
– 12% support somewhat
– 9% oppose somewhat
75
– 49% oppose strongly
– 8% don’t know
After: Do you feel that long-term gay couples should be permitted to marry, given the option of some
legal status short of marriage, or that their union should have no legal status at all? (O’Neill Associates
on October 8, 2001)
• 22% permit to marry
• 19% other legal status
• 49% no status
• 11% don’t know
4.3.3.2 Lawrencev. Texas andGoodridgev. DepartmentofPublicHealth
Before: Do you feel that long-term gay couples should be permitted to marry, given the option of some
legal status short of marriage, or that their union should have no legal status at all? (O’Neill Associates
on October 8, 2001; while this case is just outside of the 18-month window, I was unable to find any other
polls between this one and when the case came down specifically on same-sex relationship approval.
• 22% permit to marry
• 19% other legal status
• 49% no status
• 11% don’t know
After:
• Do you favor or oppose allowing gay and lesbian couples to enter into same-sex marriages? (NBC
News/Wall Street Journal on July 26, 2003)
– 32% favor
76
– 51% oppose
– 4% depends
– 13% don’t know
• Would you favor or oppose a law that would ban gay marriage, requiring that marriage should be
between a man and a woman? (Associated Press on August 19, 2003)
– 52% favor
– 41% oppose
– 7% don’t know or refused to answer
• Today/ yesterday, the Massachusetts Supreme Judicial Court ruled in favor of gay marriage in
that state. Do you favor or oppose same-sex marriage? (Fox Broadcasting Company on November
21, 2003)
– 25% favor
– 66% oppose
– 9% not sure
4.3.3.3 Lewisv. Harris
Before: There has been a proposal to change the US Constitution to ban gay marriage. Do you think
amending the Constitution to ban gay marriage is a good idea, or a bad idea? (Pew Research Center on
June 27, 2006)
• 33% favor
• 33% oppose/good idea
• 19% oppose/bad idea
• 3% don’t know/refused
77
• 12% oppose, don’t know
After: Do you believe gays and lesbians should be allowed to get legally married, allowed a legal part-
nership similar to but not called marriage, or no legal recognition given to gay and lesbian relationships?
(Fox Broadcasting Company on November 6, 2006)
• 30% legally married
• 30% legal partnership
• 32% no legal recognition
• 7% don’t know
4.3.3.4 InreMarriageCases
Before: Would you favor or oppose a law that would ban gay marriage, requiring that marriage should
be between a man and a woman? (Associated Press/AOL on January 22, 2007)
• 51% favor
• 45% oppose
• 4% not sure
After: Do you strongly favor , favor, oppose, or strongly oppose allowing gay and lesbian couples to
marry legally? (Pew Research Center on May 29, 2008)
• 15% strongly favor
• 23% favor
• 20% oppose
• 29% strongly oppose
• 13% don’t know
78
4.3.3.5 Hollingsworthv. Perry andUnitedStatesv. Windsor
Before: Do you strongly favor, favor, oppose, or strongly oppose allowing gay and lesbian couples to
marry legally? (textupPew Internet & American Life Project on April 25, 2013)
• 26% strongly favor
• 22% favor
• 18% oppose
• 25% strongly oppose
• 9% don’t know or refused to answer
After: Overall, do you approve or disapprove of the US Supreme Court’s allowing gay marriage in
California? Do you feel that way strongly, or somewhat? (AB C News/ Washington Post on July 3, 2013)
• 34% approve strongly
• 17$ approve somewhat
• 11% disapprove somewhat
• 34% disapprove strongly
4.3.3.6 Obergefellv. Hodges
Before: The US Supreme court could decide that gays have a constitutional right to marry which
would have the effect of legalizing gay marriage throughout the country. Would you favor or oppose the
Supreme Court taking this action? (NBC News/Wall Street Journal Poll on June 25, 2015)
• 45% favor strongly
• 12% favor not strongly
• 8% oppose not strongly
79
• 29% oppose strongly
After: Do you support or oppose the US Supreme Court ruling legalizing gay marriage? Do you feel
that way strongly or somewhat? (ABC News/Washington Post on July 22, 2015)
• 37% support strongly
• 15% support somewhat
• 35% oppose strongly
• 4% no opinion
4.3.3.7 Masterpiece Cakeshop v. Colorado Civil Rights Commission
Before: Would you support or oppose a constitutional amendment allowing states to ban gay marriage?
(Economist/YouGov on May 11, 2016)
• 35% support
• 47% oppose
• 17% no opinion
After: Would you support or oppose a constitutional amendment allowing states to ban gay marriage?
(Economist/YouGov on July 5, 2016)
• 34% support
• 51% oppose
• 15% no opinion
4.3.4 Hypotheses
The goal of this project is to look at polling data from both before and after each case decisions is handed
down. I begin with the assumption (as discussed above) that the choice for advocacy organizations
80
selecting these cases was due to the belief that they would generate substantial media attention which
would then move the needle of public opinion. Therefore, my hypotheses are as follows:
• Polls about support for overall gay rights will be most dramatically impacted by the media around,
and the resulting decisions of, SCOTUS cases dealing with intimate same-sex relationships (rather
than with LGB identity more generally), leading to a sustained increase in the subsequent ANES.
• Support for specific LGB rights will be most impacted by the similarity of the right implicated
in the case and the right measured by the ANES. Since the cases all revolve around same-sex
relations, I therefore expect the largest differences in public opinion to be on same-sex marriage,
while other opinions on the other rights will see a smaller difference.
• Increased support for LGB rights will correlate with increased rates of political participation.
One expected caveat to my hypotheses is that as support for gay rights increases over time (from
ANES to ANES), the correlation of support to political participation will similarly decrease as support
for LGB rights becomes more normalized. However, I still expect an immediate shift in public opinion
in response to the media around a particular case as well as the subsequent SCOTUS decision of each
case.
4.4 Results
Multiple cases were on judicial dockets or decided between 2000 and the 2008 elections, including civil
union and marriage cases in California, New Jersey, Connecticut, and the Supreme Court landmark case
Lawrence v. Texas. However, between 2008 and 2012, the cases docketed included the much contested,
much publicized cases in the Hollingsworth v. Perry related litigation over California’s Proposition 8
(which limited marriage between a man and a woman) and United States v. Windsor and associated
cases challenging Section 3 of the Defense of Marriage Act (which barred same-sex married couples
from being considered “spouses” under federal law). Two major cases receiving publicity between 2012
81
and 2016 were the Obergefell v. Hodges and Masterpiece Cakeshop v. Colorado Civil Rights Commission
lines of litigation (both of which were initiated in 2012).
The models presented below—all with the level of significance set atp = 0:05—illustrate whether a
particular, isolated variable was found to increase support for gay rights or political participation. Each
model shows data from each of the four years (2004, 2008, 2012, and 2016) unless a particular line of
question or option was not available at the time of the survey (e.g., the question of whether gays should
be allowed to serve in the military was not present in the 2016 ANES). The full list of independent
variable names as used in the models are as follows:
• Gender = gender
• Income = income bracket
• Education = level of education
• Age = age in whole years
• Liberal-Conservative Scale = self-placement on a 1 (strong liberal) to 7 (strong conservative))
liberal-conservative scale, with 4 being moderate (whole numbers only)
• PID = self-placement on a 1 (strong democrat) to 7 (strong republican) party identification scale,
with 4 being independent (whole numbers only)
• Gay = Heterosexual or Homo/Bisexual identity (unfortunately, this demographic question was not
included in the 2004 ANES)
• AAPI = Asian/Asian-American Pacific Islander
• Black = Black or African-American
• White = White or Caucasian
• Native American/Alaskan = Native American/Alaskan
• Hispanic = Non-white Hispanic
82
The results will be split into two sections. The first section shows the findings for predicting political
participation, while the second section presents findings for predicting support for gay rights. Both
sections lead with the model of the participation/gay rights index, followed by models for each individual
component of those indices. Given the large number of independent and dependent variables being
observed, any independent variables that are not expressly mentioned as being correlated with the
dependent variables “Overall Political Participation” or “Overall Support for Gay Rights” can be assumed
to have no statistically significant correlation. For individual component dependent variables, I will
focus the two independent variable of “gay identity” and “support for gay rights”, though the lack of
mentioning other variables does not indicate lack of significance. As an additional note, the role of
confounding variables will be discussed in Part 4.5.
A quick way to interpret the models—particularly for the variables I will not discuss—is to first see
which side of “0” the dot for a given year is, which indicates a positive or negative effect. Second, if
the dot itself or the lines on either side of the dot cross 0, then the effect is considered not statistically
significant. Finally, longer lines mean larger standard errors.
4.4.1 Predicting Political Participation
In what seems to go against most existing literature, Figure 4.2 shows that education was negatively
correlated in all years but 2012 (which had a positive correlation). Party ID was positively correlated
on in 2004, though the statistical significance is quite weak. In 2004, identifying as white or Hispanic
were positively correlated with participation, while in 2012, identifying as AAPI or Hispanic were
negatively correlated with participation. Finally, identifying as gay or bisexual was positively correlated
with participation in 2012, while there was a negative correlation in 2016. The change in political
participation in 2012 and 2016 among those who identify as gay or bisexual seems odd for a number of
reasons.
First, Obergefell, Hollingsworth, and Windsor were all decided after the 2012 elections. Both
Hollingsworth and Windsor—whose decisions landed in 2013—dealt with longstanding issues and had
83
Overall Political Participation
Figure 4.2: Predicting respondents’ overall political participation
84
received a lot of media attention. Hollingsworth in particular had a long, public, yet narrow history due
to the process by which California Proposition 8 passed in 2008 and the many responses and internal
litigation that ensued. DOMA had also been a contentious topic since its passing, and after years of
“mini-DOMAs” being struck down, additional interest and media coverage may have mobilized activists
into action (Leachman, 2016a).Given the amount of publicity and protest that surrounded those cases,
why did participation drop between those two years?
One simple answer is based on the time of data collection, as ANES surveys generally begin
interviews in the September before the presidential election. The decisions for Hollingsworth and
Windsor came down on June 26, 2013, which—along with Obergefell whose decision came down exactly
two years later—would have been outside the “past 12-month” time frame of the question asked in the
2016 survey. The only remaining case receiving public attention within a year of the 2016 election was
the then upcoming Masterpiece Cakeshop litigation, whose decision would not come down until June 4,
2018.
It may be true that there being fewer cases around which to mobilize or there being a potential
survey shortcoming could singularly or jointly lead to gay identity being less predictive of participation.
However, another perspective is that the act of politically participating around issues of gay rights no
longer depends on actually identifying as gay, lesbian, or bisexual. This mirrors the general trend of
increasing public approval for gay rights—illuminated further in Section 4.4.2—which itself is part
of a cycle of changing social values, media coverage, political battles, and generational shifts. Taken
together, this cycle and our understanding of shared identity positively correlating with increased
political participation suggest that others may have joined in to fight for rights, rather than gays and
lesbians no longer participating (Stoutenborough et al., 2006).
Across each tested component form of political participation—donating to a political organization
(Figure 4.3), contacting an elected official (Figure 4.4), signing a paper or internet petition (Figure
?? and Figure 4.6 respectively), or joining a protest/demonstration (Figure 4.7)—gay identity was
not significantly correlated in 2008, positively correlated in 2012, and negatively correlated in 2016.
85
Hollingsworth and Windsor were being litigated throughout 2012 (with final arguments heard December
of 2012), and given the media attention it garnered, one could expect a heightened sense of urgency
entering into the feedback loop surrounding the cases.
Supporting gay rights overall, however, had no significant effect in any year for any form of par-
ticipation except in 2004, where there was a positive correlation with contacting an elected official.
This provides some support for the above theories given that Lawrence was decided in 2003, therefore
increasing the likelihood of more individuals contacting their representatives about supporting gay
rights (and using that mindset to help them decide how they will vote in the upcoming election).
4.4.2 Predicting Support for Gay Rights
Figure 4.8 shows that gender had the strongest positive correlation in 2008, then decreased in strength
until 2016; given the rise in overall support for gay rights and prevalence of gay rights cases between
those years, this difference highlights how the difference in support of LGB rights between men and
women shrank. Education was negatively correlated in 2004—supporting the across-the-board drop
in public support for gay rights after Lawrence—then jumped to being strongly positively correlated
in 2008 and decreasing in strength of effect through 2016. Party ID took the most interesting path.
Strangely, in 2004, increased republican identification was positively correlated with support for gay
rights albeit only barely reaching a level of statistical significance.
From 2008 on, though, increased republican party ID was negatively correlated with support for
gay rights, with its zenith in 2012 (following the trend above in being the year when two of the most
important cases were argued). The same pattern holds true on the liberal-conservative scale. Identifying
as gay, however, most strongly predicted support for gay rights in 2008, then decreased in positive
correlation through 2016. The salience of identifying as LGB as a correlate to supporting gay rights
provides some support for applying linked fate to LGB populations (though in these surveys, the concept
was not directly measured).
86
Figure 4.3: Predicting respondents donating to a social or political organization
87
Figure 4.4: Predicting respondents contacting an elected representative or official
88
Figure 4.5: Predicting respondents signing a paper petition
89
Figure 4.6: Predicting respondents signing an internet petition
90
Figure 4.7: Predicting respondents being involved in a protest
91
Overall Support for Gay Rights
Figure 4.8: Predicting respondents’ overall support for gay rights
A trend can be observed when examining each individual component of gay rights—adoption
(Figure 4.9), military service (Figure 4.10), protection from job discrimination (Figure 4.11), and same
sex marriage (Figure 4.12)—in which the salience of identifying as LGB correlating with support for a
particular right decreased over from 2008 to 2016 as overall approval for gay rights rose. Placement
on the liberal-conservative scale—moreso than party ID—was salient for all four rights as well, with
its correlation generally revealing that the less conservative a respondent was, the more likely the
respondent supported a particular right.
4.5 Limitations and Further Research
A standard caveat for this line of research would be the presence of confounding variables, such as other
world events, the excitement of a particular election (including at the local and state levels), types of
entertainment media being presented, and more. Entertainment media played an incredibly important
role, as “normalizing” shows (which is controversial in its own right) such as Will Grace helped to
shift public opinion, increasing activism (particularly among allies), and the creation of LGBT-friendly
legislation (Sobel, 2015). Political participation does not occur in a vacuum, but as this paper suggests,
92
Figure 4.9: Predicting respondents’ support for LGB individuals adopting childing
93
Figure 4.10: Predicting respondents’ support for LGB individuals serving in the military
94
Figure 4.11: Predicting respondents’ support for protecting LGB individuals from job discrimination
95
Figure 4.12: Predicting respondents’ support for same sex marriage
96
the participation of particular demographic groups shifts depending on the importance of the issue to
that population.
One key aspect that was not explored via ANES data was individual media consumption. There are
dozens of questions in ANES surveys asking about different forms of media consumed, specific networks
or brands consumed, frequency of consumption, and more that could be investigated and compared to
the revelations from this paper. These details may more strongly connect the theories of the role of
media consumption specifically with LGB populations. Of course, with the rise of social media, even that
information might not accurately capture the impact of media consumption, as most of that research is
based in more traditional media engagement (e.g. television, newspapers, radio broadcasts, etc.).
Finally, this paper was limited by the (often subtle) changes in the battery of questions used to track
participation and support of gay rights over time. Future research could attempt to corroborate these
findings by using other datasets, such as the General Social Survey (with the understanding that the
goals of the surveys and time points of data collection may be markedly different). However, even those
batteries of questions may have changed and may not be parallel to the ones in these surveys.
4.6 Conclusion
This paper was exploratory in nature, rather than testing specific hypotheses. Many trends in political
participation and support for gay rights mirrored those that were expected from other research given
the timing of court cases and the media publicity surrounding said cases. The findings here provide
additional legitimacy to those theories, but in a more contemporary and demographic-specific setting. I
emphasize the importance of the role of media surrounding litigation as an effective strategy to push
the needle on public opinion, therefore increasing the likelihood of being able to gain additional rights.
Additionally, activists should tap into the shared identity of the demographic group facing discrimi-
nation (while, as Sobel argues, still making room to value and display the diversity within the group) in
order to increase visible support. With more concrete knowledge of the role a multifaceted approach
using litigation and media can play in encouraging political participation, advocates and activists (for
97
any and all sides of an issue) should more strongly consider emphasizing this strategy in future fights
for social justice.
98
Chapter 5
Conclusion and ongoing work
The overarching goal of this dissertation was to examine the impact public opinion might have on
the rates of political participation. Independently, both public opinion and political participation can
strongly impact future governmental policy decisions. Moreover, both can influence who is elected to be
a representative and the choices elected officials make when it comes to drafting, supporting, voting for,
or enacting different pieces of legislation. Knowing and understanding the public’s opinion and stance
on an issue can increase awareness of the topic, alert officials to the general position of different sets of
constituents, and even drive research into the topic. And, whether participation is through voting for
officials or policies (including, in some jurisdictions, judges) or the vocalization of an individual’s support
for an issue, a representative government is responsible for responding to those interests. Additionally,
both public opinion and political participation can influence media coverage of the topic (which, as an
aside, often means additional polling and measures of public opinion), feeding into the cycle of adding
pressure onto the government to act.
For example, while political efficacy does have an impact on political participation, even major
shifts in electoral policies that prioritized campaign funding from large and wealthy individuals or
entities—such as those resulting from Citizens United and other cases—may not strongly correlate with
long-term levels of political efficacy. While there could be enough other variables—such as political
knowledge—that impact political efficacy (or, alternatively, counteract the effect of a different variable),
it could also be that this situation is an example of an issue that was never tied to an individual’s sense
99
of political efficacy in the first place. This would mean that changes on this issue are unlikely to affect
one’s level of participation.
This finding seemingly contradicts the impact of Supreme Court (SCOTUS) cases on same-sex
relations on public opinion and thus, political participation. However, the topic of sexual orientation and
gender identity (SOGI) seems to strike a much more personal chord with individuals and arguably may
be understood as having a greater impact on someone’s personal life. Whereas campaign finance may be
a structural issue, SOGI is much more commonly seen as a moral issue, thus eliciting stronger emotional
and psychological responses. This difference could explain why there was evidence for shifts in both
public opinion and political participation after SCOTUS decided cases around same-sex relations.
Whereas emotional and psychological responses may have influenced levels of support for LGB
policies rooted in morality, feelings towards other demographic groups also impacts public opinion for
policies perceived to be unfair towards ones own group. Similar to support for LGB policies, white
support for affirmative action was strongly correlated with other measures of racism towards Blacks,
with the strongest predictor resulting from the most intense emotional and psychological reaction
out of the three: disgust. This correlation reifies what other scholars have found re: the impact of
emotion on opinion and participation, but these specific findings call into question which of the three
measures—old-fashioned racism, racial resentment, and feeling thermometers towards Blacks—is most
useful. Additionally, what can feeling thermometers predict? Could they also be used to predict political
efficacy or support for other types of policies?
The three papers presented here highlight the relationships between shifts in public opinion and
political participation around heated issues within US politics. All three issues have been litigated, and
SCOTUS has decided cases on these topics multiple times. They continue to receive media coverage
and play a role in politics today (though potentially at varying levels), and it is likely that other major
issues—such as gun control and reproductive rights—may similarly see changes in public opinion and
political participation due to SCOTUS decisions, biases, and other psychoemotional reactions.
100
These studies are limited to the ANES data that was collected over 4 separate presidential elections,
and the 2016 election cycle was perhaps one of the most animated, contentious, and dramatic presiden-
tial elections in the past several decades; many of the issues raised have continued to be hotly debated
throughout President Trump’s term. Including data from the 2020 election cycle may help to determine
whether 2016 might be treated as more of an anomaly or if it is representative of the direction in which
US politics is heading. Additionally, this would allow for the investigation of a shift in the constitutional
approaches of the Supreme Court, as it is possible that the types of rulings may impact public opinion
and political participation. Future research could add in additional variables around political knowledge,
religion, and expand the use of different bias measurements to different demographic groups to attempt
to hone in on other relationships that influence the opinion and participation of the American public.
After all, those are some of the foundational aspects of a democracy, and if democracy is to survive,
knowledge of the beliefs and wills of the public will be paramount to our country’s success.
101
Bibliography
Abraham, J. R. (2010). Saving Buckley: Creating a Stable Campaign Finance Framework. Law
Review 110(4), 1078–1122.
Acemoglu, D., S. Johnson, J. A. Robinson, and P. Yared (2005). From education to democracy? American
Economic Review 95(2), 44–49.
Achilov, D. (2016). When actions speak louder than words: examining collective political protests in
Central Asia. Democratization 23(4), 699–722.
Adeniji, L. (2015). Affirmative Action and the Struggle Against Systematic Appression: How Race-Based
Admissions are Not a Threat to America. UMKC Law Review De Jure 4(4), 1–17.
Alexander, M. C. (2011). Citizens United and Equality Forgotten. New York University Review of Law &
Social Change 35, 499–527.
Almond, G. A. and S. Verba (1989). The civic culture : political attitudes and democracy in 5 nations.
Sage Publications.
Alwin, D. F. (2007). Feeling Thermometers Versus 7-Point Scales. Sociological Methods & Research 25(3),
318–340.
Ansley, F. L. (1997). White supremacy (And what we should do about it). In R. Delgado and J. Stefancic
(Eds.), Critical white studies: Looking behind the mirror, Chapter 98, pp. 592–595. Temple University
Press.
Asal, V., R. Legault, O. Szekely, and J. Wilkenfeld (2013). Gender ideologies and forms of contentious
mobilization in the Middle East. Journal of Peace Research 50(3), 305–318.
Austin v. Mich. Chamber of Commerce (1990). 494 u.s. 652.
Bachmann, I. and H. G. De Z ´ u ˜ niga (2013). News platform preference as a predictor of political and civic
participation. Convergence 19(4), 496–512.
Bachmann, I., K. Kaufhold, S. C. Lewis, and H. G. D. Z ´ u ˜ niga (2010). News Platform Preference
: Advancing the Effects of Age and Media Consumption on Political Participation. International
Journal 5(1), 34–47.
Baldassarri, D. (2011). Partisan Joiners: Associational Membership and Political Polarization in
America (1974–2004). Social Science Quarterly 92(3), 631–655.
Banks, A. J. and N. A. Valentino (2012). Emotional Substrates of White Racial Attitudes. American
Journal of Political Science 56(2), 286–297.
Bany, J. A. (2012). Civic Engagement, Assimilation, and Identity among Latinos in Southern California.
Ph. D. thesis, University of California Irvine.
Beaumont, E. (2011). Promoting political agency, addressing political inequality: A multilevel model of
internal political efficacy. Journal of Politics 73(1), 216–231.
102
Berinsky, A. J. and G. S. Lenz (2011). Education and Political Participation: Exploring the Causal Link.
Political Behavior 33(3), 357–373.
Bodnaruk Jazayeri, K. (2016). Identity-based political inequality and protest: The dynamic relationship
between political power and protest in the Middle East and North Africa. Conflict Management and
Peace Science 33(4), 400–422.
Bopp, J., R. Elf, and A. Y. Milanovich (2015). Contribtion Limits after McCutcheon v. FEC. Valparaiso
University Law Review 49(Winter), 361–396.
Briffault, R. (2011). Corporations, Corruption, and Complexity: Campaign Finance after Citizens
United. Cornell Law Review 20(3), 643.
Brooks, R. L. (2005). American Democracy and Higher Education for Black Americans: The Lingering-
Effects Theory. Journal of Law and Social Challenges 7, 1–80.
Bros, C. and V. K. Borooah (2013). Confidence in public bodies, and electoral participation in India.
European Journal of Development Research 25(4), 557–583.
Buckley v. Valeo (1976). 424 U.S. 1.
”Burroughs, e. a. v. U. S.
Carmines, E. G., M. J. Ensley, and M. W. Wagner (2012). Who Fits the Left-Right Divide? Partisan
Polarization in the American Electorate. American Behavioral Scientist 56(12), 1631–1653.
Ceka, B. (2013). The Perils of Political Competition: Explaining Participation and Trust in Political
Parties in Eastern Europe. Comparative Political Studies 46(12), 1610–1635.
Citizens United v. Fed. Election Com’n (2010). 558 u.s. 310.
Cohen, J. (2018). From Antisemitism to Philosemitism? Trends in American Attitudes toward Jews
from 1964 to 2016. Religions 9(4), 107.
Coleman, J. J. and P. F. Manna (1998). Congressional Campaign Spending and the Quality of Democracy.
THE JOURNAL OF POLITICS 62(3), 53.
Corrigall-Brown, C. and R. Wilkes (2014). Media exposure and the engaged citizen: How the media
shape political participation. Social Science Journal 51(3), 408–421.
Cunningham, G. B. and M. R. Regan (2012). Political activism, racial identity and the commercial
endorsement of athletes. International Review for the Sociology of Sport 47(6), 657–669.
Dawson, M. E. (1994). Behind the mule: Race and class in African-American politics. Princeton
University Press.
Delgado, R. and J. Stefancic (2001). Critical Race Theory: An Introduction (First ed.). New York: New
York University Press.
Denver, D. and G. Hands (1990). Does Studying Politics Make a Difference? The Political Knowledge,
Attitudes And Perceptions Of School Students. British Journal of Political Science 20(2), 263–279.
Desposato, S. and B. Norrander (2009). The Gender Gap in Latin America: Contextual and individual
influences on gender and political participation. British Journal of Political Science 39(1), 141–162.
Dilliplane, S. (2011). All the news you want to hear: The impact of partisan news exposure on political
participation. Public Opinion Quarterly 75(2), 287–316.
Dixit, A. K. and J. W. Weibull (2007). Political polarization. Proceedings of the National Academy of
Sciences 104(18), 7351–7356.
103
Downs, A. (1957). An Economic Theory of Democracy. New York: Harper.
Esenberg, R. M. (2010). The lonely death of public campaign financing. Harvard Journal of Law and
Public Policy 33(1), 283–332.
Eveland, W. P. and M. H. Hively (2009). Political discussion frequency, network size, and ”heterogeneity”
of discussion as predictors of political knowledge and participation. Journal of Communication 59(2),
205–224.
Fed. Election Comm’n v. Colo. Republican Fed. Campaign Comm. (2001). 533 U.S. 431.
Fed. Election Comm’n v. Mass. Citizens for Life, Inc. (1986). 479 u.s. 238.
Fed. Election Comm’n v. Nat’l Right to Work Comm. (1982). 459 u.s. 197.
Federico, C. M. and J. Sidanius (2002). Racism, ideology, and affirmative action revisited: The an-
tecedents and consequences of ”principled objections” to affirmative action. Journal of Personality
and Social Psychology 82(4), 488–501.
Feldman, L., P. S. Hart, A. Leiserowitz, E. Maibach, and C. Roser-Renouf (2017). Do Hostile Media
Perceptions Lead to Action? The Role of Hostile Media Perceptions, Political Efficacy, and Ideology in
Predicting Climate Change Activism. Communication Research 44(8), 1099–1124.
Fevurly, G. (2010). Davis v. Federal Election Commission: A Further Step towards Campaign Finance
Deregulation and the Preservation of the Millionaires. University of Colorado Law Review 81(2),
627–676.
Fiorina, M., S. Abrams, and J. Pope (2010). Culture War? The Myth of a Polarized America (2nd ed.).
New York: Pearson, Longman.
Ford, P. K., A. Maxwell, and T. Shields (2010). What’s the Matter with Arkansas? Symbolic Racism and
2008 Presidential Candidate Support. Presidential Studies Quarterly 40(2), 286–302.
Fowler, M. (2016). Race still matters: Political ideology and the south. PS - Political Science and
Politics 49(2), 215–220.
Gil de Z ´ u ˜ niga, H., T. Diehl, and A. Ard´ evol-Abreu (2017, jul). Internal, External, and Government Polit-
ical Efficacy: Effects on News Use, Discussion, and Political Participation. Journal of Broadcasting &
Electronic Media 61(3), 574–596.
Gillborn, B. D. (2007). Education policy as an act of white supremacy: Whiteness, critical race theory
and education reform. Journal of Education Policy 20(4), 485–505.
Glaeser, E. L., G. A. Ponzetto, and A. Shleifer (2007). Why does democracy need education? Journal of
Economic Growth 12(2), 77–99.
Gonzales Rose, J. B. (2017). Toward a critical race theory of evidence.
Gottemoller, P. G. (2011). White Americans’ Affect Toward African Americans: Predictive Power on
Political Behavior and Measurement Problems. Ph. D. thesis, Southern Illinois University Carbondale.
Grasso, M. T. (2014). Age, period and cohort analysis in a comparative context: Political generations
and political participation repertoires in Western Europe. Electoral Studies 33, 63–76.
Greenberg, J. (2002). Affirmative Action in Higher Education: Confronting the Condition and Theory.
Boston College Law Review 43(3), 521–621.
Gross, J. J. (2010). The future’s so bright, i gotta wear shades. Emotion Review 2(3), 212–216.
104
Guasti, P. (2016). Development of citizen participation in Central and Eastern Europe after the EU
enlargement and economic crises. Communist and Post-Communist Studies 49(3), 219–231.
Guinier, L. (2004). From Racial Liberalism to Racial Literacy: Brown v. Board of Education and the
Interest-Divergence Dilemma. Journal of American History 91(1), 92.
Harris, C. I. (1993). Whiteness as Property. HARVARD LAW REVIEW 106(8), 1707–1791.
Hasen, R. L. (2011). Citizens united and the illusion of coherence. Michigan Law Review 109(4),
581–623.
Hirschmann, D. (1991). Women and political participation in Africa: Broadening the scope of research.
World Development 19(12), 1679–1694.
Huddy, L. and S. Feldman (2010). On Assessing the Political Effects of Racial Prejudice.
Huddy, L., S. Feldman, C. Taber, G. Lahav, L. Huddy, and S. Brook (2016). Threat , Anxiety , and
Support of Antiterrorism. American Journal of Political Science 49(3), 593–608.
Huddy, L., L. Mason, and L. Aarøe (2015). Expressive Partisanship: Campaign Involvement, Political
Emotion, and Partisan Identity. American Political Science Review 109(1), 1–17.
Hudson, J. B. (1995). Simple justice: Affirmative action and american racism in historical perspective.
Black Scholar 25(3), 16–23.
Hughes, M. (1997). Symbolic Racism, Old-fashioned Racism, and Whites’ Opposition to Affirmative
Action. In S. A. Tulch and J. K. Martin (Eds.), Racial Attitudes in the 1990s: Continuity and Change,
pp. 45–75. Westport: Praeger.
Human, J. M. and J. Mertus (2007). The Rejection of Human Rights Framings: The Case of LGBT
Advocacy in the US HUMAN RIGHTS QUARTERLY The Rejection of Human Rights Framings: The
Case of LGBT Advocacy in the US. Article) Human Rights Quarterly 29(4), 1036–1064.
Ikeda, K., T. Kobayashi, and M. Hoshimoto (2008). Does political participation make a difference? The
relationship between political choice, civic engagement and political efficacy. Electoral Studies 27(1),
77–88.
Jackson, R. A. (1995). Clarifying the relationship between education and turnout. American Politics
Research 23(3), 279–299.
Kam, C. D. and D. R. Kinder (2012). Ethnocentrism as a Short-Term Force in the 2008 American
Presidential Election. American Journal of Political Science 56(2), 326–340.
Kane, M. D. and T. A. Elliott (2014). Turning to the Courts: A Quantitative Analysis of the Gay and
Lesbian Movement’s Use of Legal Mobilization. Sociological Focus 47, 219–237.
Kentmen-C ¸ in, C ¸ . (2015). Participation in social protests: comparing Turkey with EU patterns. Journal
of Southeast European and Black Sea 15(3), 223–237.
Ketai, R. (1975). Affect, Mood, Emotion, and Feeling : Semantic Considerations. American Journal of
Psychiatry 132(11), 1215–1217.
Kinder, D. R. and D. O. Sears (1981). Prejudice and politics: Symbolic racism versus racial threats to
the good life. Journal of Personality and Social Psychology 40(3), 414–431.
Knuckey, J. (2005). Racial resentment and the changing partisanship of southern whites. Party
Politics 11(1), 5–28.
105
Knuckey, J. (2011). Racial Resentment and Vote Choice in the 2008 U.S. Presidential Election. Politics
and Policy 39(4), 559–582.
Knuckey, J. and M. Kim (2015). Racial Resentment, Old-Fashioned Racism, and the Vote Choice of South-
ern and Nonsouthern Whites in the 2012 U.S. Presidential Election. Social Science Quarterly 96(4),
905–922.
Kuklinski, J. H., P. M. Sniderman, K. Knight, T. Piazza, P. E. Tetlock, G. R. Lawrence, and B. Mellers
(2007). Racial Prejudice and Attitudes Toward Affirmative Action. American Journal of Political
Science 41(2), 402.
Leachman, G. M. (2016a). Institutional Essentialism: Mechanisms of Intersectional Subordination
within the LGBT Movement. Wisconsin Law Review 152, 655–682.
Leachman, G. M. (2016b). Media, Marriage, and the Construction of the LGBT Legal Agenda. Rutgers
University Law Review 69, 691.
Levinson, J. A. (2013). The Original Sin of Campaign Finance Law: Why Buckley v. Valeo is Wrong.
University of Richmond Law Review 47(3), 881–938.
Lewis, D. C., F. S. Wood, and M. L. Jacobsmeier (2014). Public opinion and judicial behavior in direct
democracy systems: Gay rights in the american states. State Politics and Policy Quarterly 14(4),
367–388.
Link, M. W. and R. W. Oldendick (2006). Social Construction and White Attitudes toward Equal
Opportunity and Multiculturalism. The Journal of Politics 58(1), 149–168.
Liu, M. and Y. Wang (2015). Data collection mode effect on feeling thermometer questions: A comparison
of face-to-face and Web surveys. Computers in Human Behavior 48, 212–218.
Masuoka, N. (2006). Among Asian Americans and Latinos Together They Become One : Examining the
Predictors of Panethnic Group Consciousness Among Asian Americans. Social Science Quarterly 87(5),
993–1011.
Mcconahay, J. B. (1982). Self-Interest versus Racial Attitudes as Correlates of Anti-Busing Attitudes in
Louisville: Is It The Buses or the Blacks? The Journal of Politics 44, 692–720.
McConnell v. Fed. Election Comm’n (2003). 540 u.s. 93.
Milyo, J. (2012). Do State Campaign Finance Reforms Increase Trust and Confidence in State Govern-
ment? SSRN Electronic Journal.
Milyo, J. D. and D. M. Primo (2017). Public Attitudes and Campaign Finance: Report Prepared for the
Campaign Finance Task Force. Technical report.
Morrill, C., M. Feddersen, and S. Rushin (2015, jan). Mobilization of Law. Elsevier.
Namkoong, K., T. K. Fung, and D. A. Scheufele (2012). The Politics of Emotion: News Media Atten-
tion, Emotional Responses, and Participation During the 2004 U.S. Presidential Election. Mass
Communication and Society 15(1), 25–45.
Nejaime, D. (2012). THE LEGAL MOBILIZATION DILEMMA. Emory Law Journal 61(4), 663–736.
Nixon v. Shrink Mo. Gov’t PAC (2000). 528 u.s. 377.
Ost, D. (2004). Politics as the Mobilization of Anger. European Journal of Social Theory 7(2), 229–244.
Overton, S. (2004). The Donor Class: Campaign Finance, Democracy, and Participation. University of
Pennsylvania Law Review 153(1), 73–118.
106
Perry, B. A. (2013). Bakke to the future: affirmative action and the US Supreme Court. Politics, Groups
and Identities 1(4), 602–607.
Persson, M., K. O. Lindgren, and S. Oskarsson (2016). How does education affect adolescents’ political
development? Economics of Education Review 53, 182–193.
Pierce, J. L. (2013). White racism, social class, and the backlash against affirmative action. Sociology
Compass 7(11), 914–926.
Pinkleton, B. E., E. W. Austin, Y. Zhou, J. F. Willoughby, and M. Reiser (2012). Perceptions of news
media, external efficacy, and public affairs apathy in political decision making and disaffection.
Journalism and Mass Communication Quarterly 89(1), 23–39.
Poindexter, C. C. (1997, nov). Sociopolitical Antecedents to Stonewall: Analysis of the Origins of the
Gay Rights Movement in the United States. Social Work 42(6), 607–615.
Primo, D. M. and J. Milyo (2006, mar). Campaign Finance Laws and Political Efficacy: Evidence from
the States. Election Law Journal: Rules, Politics, and Policy 5(1), 23–39.
Rabinowitz, J. L., D. O. Sears, J. Sidanius, and J. A. Krosnick (2009). Why do white americans oppose
race-targeted policies? Clarifying the impact of symbolic racism. Political Psychology 30(5), 805–828.
Rabold, R. N. (2016). The Smug Assumption of Reverse Discrimination: Abigail Fisher and Fisher v.
University of Texas at Austin. Kentucky Law Journal 105, 705.
Read, J. G. (2015). Gender, religious identity, and civic engagement among Arab muslims in the United
States. Sociology of Religion: A Quarterly Review 76(1), 30–48.
Rim, K. H. (2009). Racial context effects and the political participation of asian americans. American
Politics Research 37(4), 569–592.
Salazar Arryo, L. A. (2010). TAILORING THE NARROW TAILORING REQUIREMENT IN THE
SUPREME COURT’S AFFIRMATIVE ACTION CASES. Cleveland State Law Review 58(3), 649–684.
Sanchez, G. R. and J. Medeiros (2016). Linked Fate and Latino Attitudes Regarding Health-Care Reform
Policy*. Social Science Quarterly 97(3), 525–539.
Sears, D. (2014). Is It Really Racism ? the Origins of White Americans ’ Opposition To Race-Targeted
Policies Carrillo. The Public Opinion Quarterly 61(1), 16–53.
Shaykhutdinov, R. and D. Achilov (2014). Islam, Islamism, and collective action in Central Asia.
Trames 18(4), 383–405.
Shikano, S. and K. K¨ appner (2015). Exploiting Feeling Thermometer Scores: A simultaneous Estimation
of Spatial Political Constellations and Valence in party Competition.
Shouse, E. (2005). Feeling, Emotion, Affect. M/C Journal 8(6).
Sidanius, J., F. Pratto, and L. Bobo (1996). Racism, Conservatism, Affirmative Action, and Intellectual
Sophistication: A Matter of Principled Conservatism or Group Dominance? Journal of Personality
and Social Psychology 70(3), 476–490.
Simien, E. M. (2005). Race, gender, and linked fate. Journal of Black Studies 35(5), 529–550.
Sobel, S. (2015). CULTURE SHIFTING AT WARP SPEED: HOW THE LAW, PUBLIC ENGAGEMENT,
AND WILL & GRACE LED TO SOCIAL CHANGE FOR LGBT PEOPLE - ProQuest. St. John’s Law
Review 89(1), 143–194.
107
Soon, C. and R. Kluver (2014). Uniting political bloggers in diversity: Collective identity and web
activism. Journal of Computer-Mediated Communication 19(3), 500–515.
Spencer, D. M. and A. K. Wood (2012). Citizens United, States Divided: An Empirical Analysis of
Independent Political Spending.
Stangor, C., L. A. Sullivan, and T. E. Ford (2011). Affective and Cognitive Determinants of Prejudice.
Social Cognition 9(4), 359–380.
Stoutenborough, J. W., D. P. Haider-Markel, and M. D. Allen (2006). Reassessing the impact of Supreme
Court decisions on public opinion: Gay civil rights cases. Political Research Quarterly 59(3), 419–433.
Swank, E. and B. Fahs (2013, jul). Predicting electoral activism among gays and lesbians in the United
States. Journal of Applied Social Psychology 43(7), 1382–1393.
Takhar, S. (2011). The construction of political agency: South Asian women and political activism.
Community Development Journal 46(3), 341–350.
Tambe, E. B. (2016). Who Votes in East Asia? European Journal of East Asian Studies 15(2), 149–173.
Tambe, E. B. (2017). Electoral participation in African democracies: the impact of individual and
contextual factors. Commonwealth and Comparative Politics 55(2), 119–140.
Teachout, Z. (2009). The Anti-Corruption Principle. Cornell Law Rev 94, 341.
Tesler, M. (2012). The Spillover of Racialization into Health Care: How President Obama Polarized
Public Opinion by Racial Attitudes and Race. American Journal of Political Science 56(3), 690–704.
Tien-Tsung, L. and T. T. Lee (2005). MEDIA EFFECTS ON POLITICAL DISENGAGEMENT RE-
VISITED : A MULTIPLE-MEDIA APPRO. Journalism and Mass Communication Quarterly 82(2),
416–433.
Trivedi, S. (2017). Emotions, Moods, and Feelings. In Imagination, Music, and the Emotions: A
Philosophical Study, Chapter 1, pp. 9–28. Albany: SUNY Press.
United States Declaration of Independence (1776).
Valentino, N. A., T. Brader, E. W. Groenendyk, K. Gregorowicz, and V. L. Hutchings (2011). Election
night’s alright for fighting: The role of emotions in political participation. Journal of Politics 73(1),
156–170.
Valenzuela, S., A. Arriagada, and A. Scherman (2012). The Social Media Basis of Youth Protest Behavior:
The Case of Chile. Journal of Communication 62(2), 299–314.
Valenzuela, S., N. M. Somma, A. Scherman, and A. Arriagada (2016). Social media in Latin America:
Deepening or bridging gaps in protest participation? Online Information Review 40(5), 695–711.
Vaught, S. E. and A. E. Castagno (2008). ”I don’t think I’m a racist”: Critical Race Theory, teacher
attitudes, and structural racism. Race Ethnicity and Education 11(2), 95–113.
Virtanen, S. V. and L. Huddy (2006). Old-Fashioned Racism and New Forms of Racial Prejudice. The
Journal of Politics 60(2), 311–332.
Volokh, E. (2006). Freedom of Speech, Permissible Tailoring and Transcending Strict Scrutiny. Univer-
sity of Pennsylvania Law Review 144(6), 2417.
Weber, C. (2013). Emotions, Campaigns, and Political Participation. Political Research Quarterly 66(2),
414–428.
108
Westfall, J., L. Van Boven, J. R. Chambers, and C. M. Judd (2015). Perceiving Political Polarization in
the United States: Party Identity Strength and Attitude Extremity Exacerbate the Perceived Partisan
Divide. Perspectives on Psychological Science 10(2), 145–158.
Wilcox, C., L. Sigelman, and E. Cook (1989). Some Like It Hot: Individual Differences in Responses to
Group Feeling Thermometers. Public Opinion Quarterly 53(2), 246.
Wilson, T. C. (2007). Whites’ Opposition to Affirmative Action: Rejection of Group-based Preferences as
well as Rejection of Blacks. Social Forces 85(1), 111–120.
Wojcieszak, M., B. Bimber, L. Feldman, and N. J. Stroud (2016). Partisan News and Political Participa-
tion: Exploring Mediated Relationships. Political Communication 33(2), 241–260.
Woods, M., J. Anderson, S. Guilbert, and S. Watkin (2012). ‘The country(side) is angry’: emotion and
explanation in protest mobilization. Social & Cultural Geography 13(6), 567–585.
Wright-Phillips, M. (2015, dec). Identity, Agency, and Emotion: Political Activism Among Anti-War
Military Veterans. Dissertation.
Zadra, J. R. and G. L. Clore (2011). Emotion and perception: the role of affective information. Wiley
interdisciplinary reviews. Cognitive science 2(6), 676–685.
Ziegler, M. (2018). WHAT IS RACE?: THE NEW CONSTITUTIONAL POLITICS OF AFFIRMATIVE
ACTION. Connecticut Law Review 50, 279.
Zschirnt, S. (2016). Gay Rights, the New Judicial Federalism, and State Supreme Courts: Disentangling
the Effects of Ideology and Judicial Independence. Justice System Journal 37(4), 348–366.
109

Abstract (if available)

Abstract Political participation and public opinion help to drive electoral and policy decisions. Within these three articles, I examine whether political efﬁcacy decreased after the passing of Citizens United

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Information, public opinion, and international relations

PDF

The political psychology of scarcity in American society: a study of culture, discourse, and public opinion

PDF

Neither purse nor sword: strategic minority opinions and judicial policy-making

PDF

Subverting state violence through the art of hood politics: an exploratory study of Black and Latinx students' critical consciousness and political efficacy

PDF

Descriptive representation in a multi-racial America

PDF

"Hum so the devil doesn't hear you": rural Black Southern political existences as the otherwise

PDF

Politics is something we do together: identity and institutions in U.S. elections

PDF

Public opinion and international affairs: a multi-method approach to foreign policy attitudes

PDF

Changing political attitudes and behavior in a diverse America: incorporating individual and contextual determinants

PDF

Challenges from below: the origins of status competitions in world politics

PDF

Political decision-making in an uncertain world

PDF

Egalitarianism begins at home and in cohorts: egalitarianism in United States public opinion, 1952-2000

PDF

Political identity and political participation: China's post-80's generation

PDF

The changing state of tolerance and tolerance in changing states

PDF

The role of merits briefs in the generation of Supreme Court decisional language

PDF

Forced march to modernity: State-imposed cultural change and regime stability in 20th-century east Asia

PDF

Identity salience and political violence: centering social identity in the study of intrastate conflict

PDF

Legitimizing self-determination: advancing the sovereignty of separatist movements

PDF

Do no harm: an evaluation study of providing culturally competent services to African American women who are victims of domestic violence

PDF

Running for office? Three different models in U.S. political campaigns & elections

Asset Metadata

Creator Coombs, Kristopher Jr. (author)

Core Title Political participation, public opinion, and the law

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Political Science and International Relations

Publication Date 08/11/2020

Defense Date 08/11/2020

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag Affirmative Action,campaign finance,Civil rights,critical race theory,critical theory,donating,external efficacy,feeling thermometer,gay and lesbian rights,gay marriage,Gay rights,internal efficacy,Law,LGBT,LGBTQ,OAI-PMH Harvest,old fashioned racism,political efficacy,political opinion,Political participation,public opinion,Public Policy,Race,racial resentment,Racism,Same sex marriage,Supreme Court,voting

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Junn, Jane (committee chair), Armour, Jody (committee member), Lo, James (committee member)

Creator Email kris.coombs.2017@lawmail.usc.edu,kris.coombs@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-364526

Unique identifier UC11666118

Identifier etd-CoombsKris-8926.pdf (filename),usctheses-c89-364526 (legacy record id)

Legacy Identifier etd-CoombsKris-8926.pdf

Dmrecord 364526

Document Type Dissertation

Rights Coombs, Kristopher Jr.

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA