Partisan Polarization, Social Identity, andDeliberative Democracy in the United States

Ryan StricklerUniversity of South Carolina

strickpr@email.sc.edu

Presented at the 2016 Midwest Political Science Association Annual MeetingApril 7th, 2016

AbstractAs of late, democratic theory, research, and practice have taken a “deliberative turn,” advancing an ideal of political discourse that focuses on inclusion, consensus building, and mutual respect when faced with disagreement. How realistic is this ideal, though, in an era marked by partisan polarization and gridlock? The paper here addresses this question, examining how the psychology of partisanship impacts attitudes towards political deliberation in the public. Drawing from social identity theory, the project illustrates the separate, yet interrelated “social” and “ideological” aspects of partisan identification. Through a set of survey experiments, it then examines how these aspects work to affect one’s attitudes towards discussion characteristics deliberative theorists prioritize. Results show that partisans exhibit less open-minded, good faith consideration towards out-party political arguments, vis-à-vis in-party and neutral arguments. This effect is particularly pronounced for partisans who have a stronger social attachment to their party. However, partisans are not less likely to value reciprocity when evaluating their representatives. These results point to the impact of modern partisan polarization as well as the possibility of a more deliberative democracy in the United States.

2

“When citizens and accountable officials disagree, and also recognize that they are seeking deliberative agreement, they remain willing to argue with one another with the aim of achieving provisionally justifiable policies that they can recognize as such”

- Gutmann and Thompson, Democracy and Disagreement, 1996

“There’s nothing short of Trump shooting my daughter in the street and my grandchildren — there is nothing and nobody that’s going to dissuade me from voting for Trump,”

- Lola Butler, Donald Trump supporter from Mandeville, LA (quoted in Barbaro, Parker, and Martin 2016)

During the run-up to the 2012 Presidential Election, the radio show This American Life

produced an episode entitled “Red States Blue State,” where they declare that “everyone knows

that politics is now so divided in our country that not only do the sides disagree on the solutions

to the country’s problems, they don’t even agree on what the problems are.” In one particularly

evocative scene from the show, interviewer Lisa Pollack profiles Frank Mills and Ron Sexton.

The two men were close friends who repeatedly discussed political issues on the phone. This is

until, as Pollack describes on-air, “Ron urged Frank to support a Republican candidate for

Congress. Frank balked. Didn't Ron know he supported Democrats?

Frank Mills: And he (Ron) said, ‘Who did you vote for for president?’ And I said, ‘I voted

for Obama.’

Lisa Pollak: Apparently this had not come up before.

Frank Mills: And then he said, ‘You must be a Socialist.’

Lisa Pollak: He said this seriously or jokingly?

Frank Mills: No, seriously. ‘You must be a Socialist.’ And I said, ‘How can you make that

assumption?’ He says, ‘Well, you voted for Obama. He's a socialist, and therefore you are.’

And then I took it as if he had called me a dirty name. And we got into an argument. And

then after a while, he said, ‘Well, I'm writing you off my list, Frank. Don't ever talk to me

again. You're no longer a friend of mine,’ is how that conversation ended,” (Glass 2012).

3

The story of Frank and Ron may be common in America, and, for many, it may be

troubling. With the rise of partisan polarization debated by academic and popular commentators

alike (Abramowitz 2010; Haidt and Hetherington 2012), many point to a link between divergent

political views and an incivility and breakdown of political discourse (“Civility in America

2013”). Dating back to the ancient Greeks, political theorists have extoled the virtues of

“deliberative” democracy, predicated on the ability of the public (or at least their representatives)

to see divergent perspectives as legitimate and to be open to opinion change (Rawls 2005;

Gutmann and Thompson 1996; Dryzek 2010). The closed-mindedness and lack of consideration

evinced by Frank and Ron calls this possibility into question.

It is also perhaps puzzling, as many public opinion scholars demonstrate that, while

conservatives and liberals may have increasingly “sorted” into their “correct” party based on

ideology, the public writ large is as moderate as it has been in the past half century (Fiorina

2011). The acrimony may be real, but it may be relatively untethered to ideological or issue

position conflict; in other words, it is comprised of “a nation that agrees on many things but is

bitterly divided nonetheless,” (Mason 2015, 128). To help explain this puzzle, a body of recent

research in political psychology (Iyengar, Sood, and Lelkes 2012; Iyengar and Westwood 2014;

Mason 2015; Huddy, Mason, and AarøE 2015) has drawn a distinction between the ideological

and social identity bases of partisanship to suggest that it is not Democrats’ and Republicans’

political views, but social identities, that are polarizing.

Drawing on the concept of “social identity” polarization, this project explores how the

ideological and social identity bases of partisan attachment impact attitudes towards deliberation

in the public. In particular, the project focuses on the concept of reciprocity, broadly defined as

open-minded, good faith consideration and reason giving towards those with whom one morally

4

disagrees. The survey experiments employed here demonstrate that partisanship does impact the

amount of reciprocity one evinces towards disagreement. This effect is particularly pronounced

for partisans who have a stronger social attachment to their party (in other words, being a

Democrat or Republican is stronger part of “who they are”). This being said, partisanship has

little impact on whether one wants their representative to evince deliberative attitudes. These

results point to the impact of modern partisan polarization as well as the possibility of a more

deliberative democracy in the United States.

The paper proceeds as follows. First, after a discussion of deliberative democracy and

the concept of reciprocity, it probes the relationship between deliberation and partisanship as

well as how extant literature underexplores this relationship. Next, the paper conceptualizes

partisan identification as containing both a “social identity” and “ideological” component, and it

spells out a theoretical framework for how these distinct, yet interrelated components impact

partisans’ attitudes towards reciprocity. From there, the survey experiment approach as well as

the operationalization of key concepts are explained. Then, both descriptive statistics and

experimental results are presented. Finally, the paper concludes by placing the results in the

context of debates on partisan polarization and democratic theory.

Deliberative Democracy

In the 21st century, democracy is generally seen as a good thing. However, like other

valence terms such as “freedom” or “justice,” “democracy” is a universal good with many and at

times conflicting interpretations. Many of those that study democracy primarily focus on the

electoral connection (Schumpeter 2003; Mayhew 1974 [2004]), or the quality of representation

from those elected (Pitkin 1967). Still others focus on the protection of minority rights (Mill

1859), the importance of group competition (Dahl 1956), or the role of social justice in

5

democracy (Rawls 1999). Increasingly, however, democratic theory has taken what Dryzek calls

a “deliberative turn” (2010, 3), placing public discourse and decision making processes at the

forefront of political thought, empirical political inquiry, and real-world institutional

development. Deliberative democratic theorists do not focus on specific governing institutions

or election procedures, and they moreover do not argue that deliberation should replace an

electoral connection. They do argue, though, that the quality of political communication that

connects citizens both with each other and the ruling class matters. Dryzek lays out a complete,

succinct definition of the core elements of deliberative democracy:

“A system can be said to possess deliberative capacity to the degree it has structures to

accommodate deliberation that is authentic, inclusive, and consequential. To be authentic,

deliberation ought to be able to induce reflection upon preferences in noncoercive fashion…

and involve communicating in terms that those who do not share one’s point of view can find

meaningful and accept…To be inclusive, deliberation requires the opportunity and ability of

all affected actors (or their representatives) to participate. To be consequential, deliberation

must somehow make a difference when it comes to determining or influencing collective

outcomes,” (2010, 10).

Thus, at its core, deliberative democracy is about reason giving (Thompson 2008);

decisions are not made by fiat, but with collective, interactive discussion, and with justification

towards those disagree. Normative theorists vary in their requirements for the quality of reasons

given, the role (if any) of power or bargaining, the proper site for this discussion, as well as the

place for consensus as a discursive goal (Bachtiger et al. 2010). In a seminal work, though,

Gutmann and Thompson (1996) argue that a key to ideal deliberation is a norm of reciprocity.

While this basic moral concept broadly means that “we should return good for good, in

6

proportion to what we receive,” (Becker 1986, 4), Gutmann and Thompson apply it to the realm

of democratic politics by stating that “when citizens deliberate, they seek out agreement on

substantive moral principles that can be justified on the basis of mutually acceptable reasons,”

(1996, 55). Recognizing the collective nature of democratic decision making, political reasoning

is given in terms that those that disagree could, at least in theory, accept. Consensus is sought,

but in recognizing the reality of moral disagreement (particularly in contemporary American

politics – see Jacoby 2014), reciprocity only requires the honest effort to forge moral agreement,

and mutual respect for competing legitimate viewpoints when consensus fails.

Beyond Gutmann and Thompson’s work, reciprocity is developed in the work of

Habermas(Examiner.com 2010) (cite), Rawls (2005), Dryzeck (2010), Bächtiger et al. (2010),

and others. Drawing on these works and others, the full conceptualization of reciprocity for this

project includes:

Tolerance of viewpoints one disagrees with (as a necessary, but not sufficient, aspect of

reciprocity)

Acknowledgment of the legitimacy of other viewpoints

A willingness to consider, in good faith, countervailing arguments

Giving reasons/justification in political debate that all, even those that disagree, can see

as legitimate

Having the goal of consensus (while maintaining one’s moral convictions; in other

words, no ‘compromise for compromise’s sake’)

Often, scholarship on deliberation focuses on discursive norms for elite politics. A

reason for that is, for deliberative democracy to matter, it cannot be just “idle talk;” deliberative

norms have to be, as Dryzek argues, “consequential” in government decision making. Many

7

empirical studies on deliberation, thus, either examine decision making practices of legislative

bodies (Bachtiger et al. 2007) or specific “empowered” public bodies such as juries (Sunstein

2002) or advisory boards (Warren and Gastil 2015). This being said, most scholars argue that, in

an ideal democracy, deliberative norms are imbued through the whole citizenry. Dryzek, for

example, argues that a key part of a deliberative system is a vibrant, inclusive “public space,

ideally hosting free-ranging and wide ranging communication,” which plays a consequential role

in influencing the “empowered space” of policymaking (2010, 11). The “public space” consists

of all manner of formal and information political participation, including policy forums,

traditional and social media, and face to face political talk. If it is unfeasible for the public writ

large to deliberative in policymaking decisions, engaged “public space” political discussion can

nonetheless work to ensure that all groups and vantage points are represented in policymaking

processes (Dryzek and Niemeyer 2008).

In addition to the mere inclusion of the public discussion in decision making, normative

calls for reciprocity have a place in guiding public, not just elite, rhetoric. This is because, for

one, the public and the political elite inhabit the same social and cultural context, and widespread

open-mindnessess or intransigence in public political culture can impact future elites’ attitudes

towards dialogue and decision making. The “rhetorical cue” elites can take from the public is

further enhanced by the formal dependence of elites on the public’s support via the electoral

connection (Mayhew 2004) as well as the demonstrated responsiveness of elites to public

opinion (Erikson, MacKuen, and Stimson 2002). If representatives engage in “pork barrel”

spending or adopt certain policy positions to demonstrate responsiveness to their constituents

(Mayhew 2004), it is reasonable to suspect they may adopt their constituents’ preferred rhetorical

style or attitude toward ideological difference, as well. Further, despite presumptions to the

8

contrary, the public does engage in political discussion more than commonly presumed (Cook,

Delli Carpini, and Jacobs 2007). If deliberative discussion norms can lead, as theorists posit, to

better informed, higher quality opinions and decisions for elite policymakers, they can certainly

do so for the majority of the public that engages in political talk as well. Finally, as a critical

ideal, a deliberative democracy that takes both the word “deliberative” and the word

“democracy” seriously requires widespread, and reciprocal, public political participation. While

there is a clear tension between fostering both public participation and ideal standards of

deliberation (Mutz 2006), it is up to deliberative theorists and empirical researchers to explore

how institutional reform and changing cultural context can help move us closer to this ideal.

Deliberation and Partisanship

Many commentators have argued for the ideal of reciprocity in public discussion,

positing benefits that include a greater normative and empirical legitimacy, a more informed

citizenry, better and more consistent attitude formation, higher tolerance for diverse viewpoints,

greater engagement and social capital building, and higher quality, more consensual decisions

(see Mendelberg 2002, 153-4). Many commentators, in turn, have lamented the loss of

reciprocial political discourse, and these benefits that accrue, given the rise of partisan

polarization in the 21st century. Echoing Washington’s warning that “the spirit of party…

agitates the community with ill-founded jealousies and false alarms” (“Washington’s

Farewell Address” 1796 [2008]), contemporary pundits claim that polarization “pervades our

political system from top to bottom” (Galston 2014), and partisan factions “neither trust nor

understand each other” (Dionne 2013). A prime example of this consternation comes from Jon

Stewart; concluding his 2010 “Rally to Restore Sanity and/or Fear,” he encapsulates this (media-

perpetuated, he argues) concern by asking, in the age of (media-driven, he argues) polarization:

9

“Why would we work together? Why would you reach across the aisle to a pumpkin assed

forehead eyeball monster? If the picture of us were true, of course, our inability to solve

problems would actually be quite sane and reasonable. Why would you work with Marxists

actively subverting our Constitution or racists and homophobes who see no one’s humanity

but their own?” (quoted in Examiner.com 2010)

Despite the normative popular concern, the academy has not fully explored connection

between partisan polarization and ideals of deliberative democracy. For example, political

scientists have paid increasing attention to partisan polarization; scholars, however, tend to study

polarization as an ideological construct, examining whether Democrats and Republicans are

increasingly comprising ideologically homogenous, ideologically distinct camps. From this

literature, there is a clear consensus that political elites are more polarized today than they have

been in the past century(McCarthy, Poole, and Rosenthal 2006; Theriault 2008). With this

observation, some scholars have argued that the public is now similarly “polarized,” harboring

homogenous, distinct worldviews and issue positions (Abramowitz 2010; Abramowitz and

Saunders 2008). The majority consensus, though, is that while partisans in the public may be

better “sorted” (Levendusky 2009), they are just as ideologically “unconstrained” as Phillip

Converse (1964) and his colleagues surmised in the 1960’s (Fiorina 2011; Fiorina 2009; Hill and

Tausanovitch 2015).

The literature examining ideological mass polarization (or the lack thereof) provides

vital, yet incomplete, insight into the impact of 21st century partisanship on the quality of

American democracy. This literature is primarily motivated by a normative concern over the

quality of representation in America, and the electoral connection between representatives and

the represented. Given the clear ideological rift between parties in government, if the public is

10

not similarly split ideologically (as Fiorina 2009 argues), there is a clear breakdown in this

machinery. If, on the other hand, the public is ideologically polarized (as Abramowitz 2010

argues, the electoral connection is then healthy. While this line of inquiry is important and not

misplaced, a sole focus on the electoral connection is incomplete because it presumes a

“minimalist” or “aggregate” vision of democracy, where public involvement in the democratic

process is limited to an occasional “competitive struggle for the people’s vote,” (Schumpeter

1942 [2003]). Election procedure and outcomes take precedence over the processes of political

discussion and opinion formation, as well as input and justification of policymaking processes

between elections. Moreover, the concerns voiced by deliberative theorists as well as popular

commentators like Stewart – of incivility, aggression, and a lack of “reciprocity” in elite and lay

political discussion – do not factor in.

If polarization studies often miss the concerns of deliberative theorists, empirical work in

deliberative democracy (Fishkin 1995; Barabas 2004; Mansbridge 1983; Gastil 2000) often

glosses over the potential impact of partisanship on deliberation that our founders worried about.

Most of this literature takes deliberation as an independent variable, assessing its effects on

dependent variables such as trust, tolerance, consensus, etc. They also operationalize

“deliberation” holistically, arguing that the package of conditions theorists prioritize (reciprocity,

reason-giving, non-domination, accurate information, etc.) create the effects they study. There is

certainly value in taking this approach; many argue that, as deliberation is a synergistic,

communal process, breaking the characteristics of deliberation down to their component parts

means studying something that is not deliberation (Thompson 2008).

This being said, the “holistic” studies of deliberation are, at best, incomplete. For one,

they suffer from issues of endogeneity. Is the open-mindedness, opinion change, greater

11

understanding, etc., created by the deliberative process, or is it driven by the type of people that

choose to participate in the extensive research process? What’s more, Ryfe is correct in positing

that "researchers have been less interested in deliberation itself than in measuring its effects."

(2005, 54). In other words, what specifically produces the salutary benefits that studies of

deliberation find? Is it the access to accurate, expert information? The encouragement of open-

mindedness? Social contact with persons of a different opinion? By setting up a holistic

deliberative process and not subjecting each of these components of the process to scrutiny,

Mutz is correct that deliberation becomes a “moving target,” which “insulate(s) the theory from

falsification” (Mutz 2008).

Thus, studying the specific components of deliberation in isolation, with an eye towards

internal validity, is a necessary addition (but not replacement) to the extant “holistic” research

that case based and forum based studies provide. Indeed, her work (2006) as well as other work

(eg. Jackman and Sniderman 2006; Druckman and Nelson 2003; Hibbing and Theiss-Morse

2002) takes such an approach, using survey experiments and other experimental techniques to

examine isolated characteristics of deliberation such as interpersonal network diversity or

exposure to counterargument. However, like the more “holistic” research, this experimental

work, with few exceptions, takes characteristics of the deliberative setting as independent

variables, examining if they produce the salutary effects theorists contend they do. Less research

from this subfield takes the deliberative setting as the dependent variable, exploring its

preconditions. This is unfortunate, as Thompson argues that “the aspect of deliberation about

which empirical inquiry has potentially the most to say is the set of conditions that are necessary

for, or at least contributory toward, good deliberation,” (2008, 509). He posits that more

research into the cultural and institutional preconditions that lead to higher-quality deliberation

12

can help to explain the mixed results empirical inquiry into deliberation often finds. To wit,

despite the extended popular and academic debate over the rise of partisan polarization, little

work has examined the role of parties in fostering or inhibiting deliberation. However, recent

work examining partisan social identity can provide a useful tool with which this project can do

just that.

The Dual Bases of Partisan Identity

Fiorina, Abramowitz, and other polarization scholars tend to work with an “ideological”

concept of polarization, where partisan affiliation is (or is not) linked to ideological and/policy

preference. In a recent paper, however, Iyengar, Sood, and Lelkes (2012) argue that negative

affect towards an “out-group” party, not ideological difference, is the most important dimension

of partisan polarization. They draw on social identity theory (SIT) to explain how, as elite

polarization and the modern media environment have made partisanship an increasingly salient

“social identity” in American culture, dislike towards, and stereotyping of, out-partisans has

increased in survey responses; this phenomenon is not evident for other groups. They also show

that ideology is only, at best, a very weak predictor of “affective” partisan polarization, which

suggest that the salience of the “partisan” social grouping is primary, and it is not caused by a

prior ideological or issue-based difference.

The study of “affective” or “social” partisan polarization, may be relatively novel, but it

ties into a canon of research that suggests partisanship is a long-term, psychological attachment

(Campbell et al. 1960; Green, Palmquist, and Schickler 2002). It also resonates with

psychological research on attitude formation and group identification. Social identity theory has

long posited that group affiliations (such as attachment to one’s party) play a key role in an

individual’s identity development, and even the most arbitrary or “minimal” group affiliations

13

can drive attitude formation and behavior (Brewer 1979; Tajfel et al. 1971; Turner 1987). Much

depends on context; group categorizations become salient to an individual depending on their

cognitive accessibility, their perceived value to the individual, and relevant environmental

stimuli (Yzerbyt and Demoulin 2010, 1029; Turner 1987, Ch. 6). With salient group

identifications, though, individuals develop a host of behaviors, including:

Engaging in “self-stereotyping” by adopting in-group attitudes and behavioral norms

(Terry and Hogg 1996; Suhay 2015)

Exaggerating in-group similarity and out-group difference (Brown 2000; Ellemers,

Spears, and Doosje 1997)

Favoring the in-group and prejudice towards the out-group (Brewer 1999; Brown et al.

2001)

Positive ingroup emotions (such as pride) (Suhay 2015), and negative outgroup emotions

(such as anger) (Mackie, Devos, and Smith 2000).

While Iyengar et al. (2012), as well as subsequent studies similarly leveraging social

identity theory (Iyengar and Westwood 2014; Huddy, Mason, and AarøE 2015; Mason 2015),

are right to highlight the social and affective dimensions of mass partisan polarization, it would

be wrong to suggest that ideology is irrelevant. If affect is all that matters, it would be difficult

to explain how individuals are increasingly “sorting” into the “correct” party ideologically

(Levendusky 2009). Lodge and Taber, moreover, argue that humans are “motivated reasoners,”

facing implicit pressure to bring their ideological worldview in line with their affective feelings

of partisan support (Lodge and Taber 2013; see also Kunda 1990). In short, with regards to

partisan polarization, it is important to look at social identity and ideology, not one or the other.

The figure to the side presents a proposed schematic for individual-level partisan polarization,

14

which consists of two dimensions of one’s “distance” from his or her out-party. On the x-axis is

“ideological” polarization, or how close one’s policy views are from the median individual in his

or her in-party. The right side of the axis represents one who is ideologically similar from his or

her in-party, the left side represents one who is ideologically dissimilar. On the y-axis is “social”

polarization, or the level of group identity-based attachment one registers for his or her in-party.

The top of the line represents a strong partisan social identity, and the bottom of the line

represents party being a weak part of one’s identity.

Figure One Here

Theoretical Outline

Drawing on social identity theory and the distinction between the social and ideological

bases of partisan attachment, this project seeks to explore how partisan polarization impacts the

potential for a more deliberative democracy. The general expectation is that partisan social

identity salience is negative related to deliberative attitudes in the public, focusing on the key

attitude of reciprocity. This connects to the insight from SIT that, for salient group identities,

one finds out-group bias and exaggeration of difference from the in-group (why engage in good

faith discussion with someone who’s opinion you hold in low regard?). It also connects to recent

research on partisan psychology that suggests that, far from open-mindedness, one’s partisanship

conditions the quality and hue of information that he or she receives from their environment

(Iyengar and Hahn 2009; Ramsay et al. 2010), as well as how that information is processed and

interpreted (Gaines et al. 2007; Lodge and Taber 2013). The project here extends the theoretical

purview of SIT and partisan psychology research by positing a negative connection to

deliberative democratic ideals.

15

I expect separate effects for both the “social identity” and “ideological” dimensions of

one’s partisan identification. As discussed above, salient social identities have a strong impact

on behavior; what’s more, salient in-group identities produce anger against the out-group

(Mackie, Devos, and Smith 2000), and anger, as an emotion rooted in a negative affective

response, is a primary driver of cognition and attitude formation (Zajonc 1980; Ellsworth and

Scherer 2003). Thus, I expect partisan social identity salience to have the strongest negative

effect on reciprocity. However, since “self-stereotyping” is also an indicator of social identity

salience (Mackie 1986; Mason 2015), I expect one’s “ideological” partisanship (as an indicator

of adopting prototypical in-group norms) to be a secondary contributor to this negative

relationship to reciprocity.

Data and Method

This project assesses the impact of the ideological and social identity components of

partisanship on deliberation through an online survey experiment1 of 1,619 respondents during

the summer of 2015. The survey was developed and administered through Qualtrics web-based

survey software, and respondents were recruited through Amazon’s mTurk human intelligence

task service. Each recruited respondent was paid $0.90 to complete the survey2. For this study, I

am interested only in the impact of treatment on partisan respondents – those identifying or

“leaning” toward either the Democratic or Republican party based on the traditional, ANES party

identification scale. This subset includes 1,350 respondents3.

1 As there is strong debate as to whether mTurk experiments are externally valid (Berinsky, Huber, and Lenz 2012; Buhrmester, Kwang, and Gosling 2011; Krupnikov and Levine 2014), key results from this study will be replicated through a representative telephone sample (800 respondents) of the South Carolina population in the spring of 2016.2 The full survey instrument can be found at http://ryanstrickler.weebly.com 3 I include both “outright” partisans and those that “lean” toward one party because it is well established that “leaners” behave much like those that explicitly identify with a party (Keith et al. 1992). The analysis, however, holds up whether one includes or excludes “leaners”

16

In one set of experiments embedded in the survey, there is a manipulation of the framing,

or the relevant considerations the respondent receives, before answering a series of questions on

political attitudes. “Issue” or “emphasis” framing effects experiments like this have been well

developed in studies on partisanship and political communication, notably by Druckman

(Druckman and Nelson 2003; Druckman 2001; Druckman, Peterson, and Slothuus 2013). There

is a long lineage of political communication research on framing; a typical example is framing a

question on whether a hate group has the right to conduct a rally as a “free speech” versus a

“public safety” issue, and measuring the difference in aggregate opinion (Chong and Druckman

2007, 105-6). In this experiment, three frames will be employed (assigned through simple

random assignment): a “neutral” frame, a “partisan” frame, and a “partisan polarization” frame. I

expect that, for Democrat or Republican respondents, introducing “partisanship” and “partisan

polarization” as a consideration one receives (or does not receive) in a frame will affect the

salience of partisanship to them as a social identity. This in turn (per social identity theory) will

have a negative effect on reciprocal attitudes.

The second set of experiments consists of “vignette” questions (see Mutz 2011, Ch. 4 for

a discussion of vignette experiments). The first two vignettes ask the respondent about a

political issue, and then introduce a counterargument to the respondent’s position4. This

approach draws from Jackman and Sniderman (2006); the key manipulation, though, is that the

counterargument will be attributed to either a Democrat, a Republican, or a person without a

party identifier (the respondent is assigned through simple random assignment to one of these

three groups). Respondents will then be asked a series of questions about the argument; the

hypothesis is that partisans will be more likely to respond with an attitude of reciprocity to an in-

4 These counterarguments were developed from recent public political stances on these issues (Gearen 2015; Hook 2015) (Gearen 2015; Hook 2015), as well as a recent Pew Research Center survey (2014).

17

party counterargument, and less likely to respond with reciprocity to an out-party

counterargument. The final vignette question gauges the respondents’ support for candidates

that exhibit ‘deliberative’ behaviors. It is a two-by-two vignette (vignettes assigned through

simple random assignment), presenting a scenario which manipulates whether an in-party

candidate (either a Democratic or Republican) is working with both parties to pass legislation, or

worked in opposition to the other party to pass legislation. I expect partisans to be less likely to

support in-party candidates that work across the aisle, and more likely to support candidates that

work in opposition to the opposing party. All framing and vignette treatments can be found in

the Appendix.

On Operationalizing Partisan Social Identity and Ideological Partisanship

Beyond looking writ large at how partisans respond to variations in party frames and

vignettes, I am interested in the CATE for different partisan subgroups, based on measures of

partisan social identity (PSI) and ideological partisanship. For the latter, the survey gives the

respondent 11 declarative statements concerning an array of policy issues, allowing for a five-

point response ranging from “Strongly Agree” to “Strongly Disagree.” These items draw

directly from ANES Time Series surveys. Six of the items have been in every iteration of the

ANES since 1982; using these items will allow me in the future to attest to the import of the

result of this project, given changes in ideological partisanship over time. An additional five

policy items in this survey are drawn from the 2012 ANES Time Series survey. An exploratory

factor analysis (EFA) using every policy question in the 2012 ANES was conducted, and five

items that loaded clearly on either the first or second factors were included. The 11 policy items

range in topic from social welfare to offshore drilling to gay marriage. They are all, however,

18

policies that have a clear “liberal vs. conservative” dimension. For more information on these

items and the results of the EFA, see Appendix.

Following a technique employed by Carsey and Layman (2002) as well as Iyengar, Sood,

and Lelkes (2012), the 11 policy items were then used to create ideological partisanship scores

for each respondent using one-dimension confirmatory factor analysis (CFA) in the following

way. After conducting the CFA and producing scores, the mean score for “true independents”

(based on the traditional ANES party identification measure) was set to zero (again following

Carsey and Layman 2002). Then, the measure was folded so respondents increasingly more

liberal or more conservative than true independents would have increasingly high scores.

Finally, the measure was re-centered, with zero being the new median of the final ideological

partisanship measure for all partisan respondents. Thus, scores above zero on ideological

partisanship indicate that one is more ideologically consistent with one’s party than the median

respondent, and scores below zero indicate one is less ideological consistent with one’s party.

Partisan Social Identity (PSI) is operationalized by the survey instrument in two ways;

the first is based on the following three questions:

When I talk about Democrats (Republicans), I usually say “we” rather than “they”

When someone criticizes Democrats (Republicans), it feels like a personal insult

I don’t have much in common with most Democrats (Republicans)

The respondent received either the “Democrat” or “Republican” version of these

questions based on the traditional ANES party ID measure - whether they identified or “leaned”

towards one of the parties. For each item, the respondent has a five-point response choice

ranging from “Strongly Agree” to “Strongly Disagree.” These three items, like with the

ideological partisanship score, were used to create a folded index score with confirmatory factor

19

analysis, with the median set at zero for all partisan respondents. Thus, for the PSI measure, a

score above zero indicates an above the median social identity attachment to one’s party, and a

score below zero indicates a lower than the median social identity attachment.

Both the PSI and ideological partisanship measures will be used to divide partisan

respondents into subgroups, in order to assess differences in treatment effects across groups.

Figure Two and Table One indicate how respondents are distributed across the PSI and

ideological partisanship measures, as well as what proportion of respondents fall into each of

these subgroups. As one can see, partisans are allayed across the range of social and ideological

attachment. Given the prevalence of partisan ideological sorting (Levendusky 2009), one may

have expected fewer “High PSI, Low Ideology” and “Low PSI, High Ideology” respondent.

This may, though, highlight the line of research suggesting that, even for partisans, the public

does not have ideologically consistent or polarized attitudes (Converse 1964; Fiorina 2011).

Figure Two Here

Table One Here

Descriptive Statistics

Given established concerns about the unrepresentativeness of survey respondents

recruited through mTurk (Krupnikov and Levine 2014), it is not surprising to find that the

respondents for this study are not a microcosm of the country. Table Two indicates that the

1,619 survey respondents are whiter, have more education, are more liberal, and prefer the

Democratic Party more than the population writ large. While these respondents may be more

representative than a convenience sample of college students (Buhrmester, Kwang, and Gosling

2011), and the major focus with this experimental approach is internal validity, external validity

20

is still a legitimate concern. As such, the key results from this study will be replicated with a

representative telephone sample of South Carolina residents in the spring of 2016.

Table Two Here

Even though non-representativeness in the sample compels us to focus primarily on the

experimental results, some non-experimental statistics concerning attitudes toward deliberation

are worth noting. Table Three provides results from partisan respondents for a series of ‘forced

choice’ questions. As one can see, partisans scoring above the median on the social identity

measure (“high PSI” respondents) are more likely than those scoring below the median (“low

PSI” respondents) to believe that political discussion can lead to finding common ground.

However, high PSI respondents are also more likely to see issue stances as clearly right or

wrong, as opposed to believe that intelligent people can disagree. This seems to suggest that

those with a strong social attachment to their party are more likely to feel that consensus is

possible, but are also more likely to demand consensus on their terms.

Table Three Here

There were no significant differences between “high” and “low” PSI respondents for the

two forced choice questions from the survey asking about attitudes towards representatives’

willingness to compromise. This, though, belies different dispositions towards compromise

across parties, as suggested by Figure Three. Democrats in the sample are much more likely

than Republicans to want a representative to work toward compromise, and they are much more

likely to tolerate a change in mind from their representative. This may be, in part, due to

differences in values underlying the parties, but it also may be due to the politics of the day. For

Republicans in 2015, representative “compromise” might simply imply the Republican Congress

working with the Obama administration (and thus forgoing policy priorities). Likewise, for

21

Democrats in 2015, “compromise” might simply mean moving parts of a Democratic agenda

through Congress. The desire for compromise in one’s representative, in other words, may well

reverse with a Republican president and a Democratic Congress.

Figure Three Here

Experimental Results

As stated earlier, the survey included two sets of experiments – a “framing” experiment

and a “vignette” experiment. While the “framing” portion of the survey produced some

interesting non-experimental statistics with the forced choice questions discussed above, there

were no significant results from experimentally manipulating the frame one received. This is, in

part, likely do to the weak treatment that a frame on an online survey screen (which one can

easily click through) provides. The concern over processing of the frame is redoubled by the

nature of mTurk respondents. Often “mTurkers” take surveys as a source of supplemental

income; while in this survey they have a remarkably high per question response rate (no question

with a response rate less than 99%), they are nonetheless motivated to move through a question-

less screen quickly. In addition, social desirability bias might be muting any treatment effect in

the questions following the frame respondents’ receive. Many of these questions ask about the

respondents general disposition to argument and disagreement; as such, there may be an inherent

desire to “appear deliberative” that washes out any treatment effect.

As Table Four shows, however, significant treatment effects were found with the

“vignette” questions; for partisan respondents, whether one finds a counterargument to their

opinion reasonable or worth considering hinges on whether the argument is coming from an

inparty, neutral, or outparty interlocutor. This effect holds for both opinion on immigration and

opinion on civil liberties issues. What’s more, comparing both the “in-party” and “out-party”

22

experimental groups to the “neutral” group, one finds that treatment effect is mostly driven by a

move away from the outparty, as opposed to a move toward the inparty. For three out of the

four questions, respondents receiving an outparty counterargument are significantly less likely to

see the claim as reasonable or worth considering. In the other question, which asks whether the

immigration counterargument is worth considering, the outparty treatment effect, while not

statistically significant, is in the same, theoretically expected direction.

Table Four Here

Furthermore, a subgroup analysis shows that the results in Table Four are driven

primarily by partisans scoring above the median on the partisan social identity indicator (“high

PSI” respondents). On the questions asking whether the counterargument is “worth

considering,” Figures Four and Five give the conditional average treatment effect for “high PSI”

and “low PSI” respondents. Here, for simplicity, the responses by group are summarized with an

average ranging from 0 (implying more consideration) to 2 (implying less consideration)5. The

figures show that “high PSI” respondents are significantly likely to give more consideration to an

inparty counterargument and less consideration to an outparty counterargument. This is true for

both issues examined. Partisans scoring lower on the PSI measure, however, are not equally

moved by the vignette treatment, and do not exhibit a statistically significant difference when

given either an inparty or an outparty counterargument. This evidence is suggestive of the link

between social identity and deliberation. For partisan respondents, as one’s party becomes a

more important part of their social identity, one is less likely to approach political disagreement

in the “good faith,” reciprocial way that Gutmann and Thompson (1996) extol.

Figures Four and Five Here

5 The results hold up when examining responses across each response category. The results are also consistent with the results for questions asking whether the counterargument is “reasonable”

23

What about the ideological dimension of partisanship? While it was hypothesized that

greater ideological adherence to one’s party would have a similar, yet secondary, negative effect

on reciprocity, the results presented in Figures Six and Seven provide a more nuanced picture.

These figures again give results for the questions on whether the counterargument is “worth

considering”; here, they focus only on respondents scoring above the median on the PSI

indicator, and they compare subgroups of these respondents based on whether they scored above

or below the median on the ideological partisanship measure. For the immigration question,

respondents scoring above the median on the ideological partisanship measure are significantly

more likely to see an inparty argument as worth considering, and less likely to see an outparty

argument as worth considering. This effect does not hold up for those scoring low on the

ideological partisanship measure. This accords with the theory outlined earlier that ideological

partisanship is an indicator of social identity salience (exemplifying in-group norm adoption),

thus helping to drive an unwillingness to deliberate with the outparty.

Figures Six and Seven Here

However, the effects of the vignettes across these subgroups are reversed when looking at

the question on civil liberties. Here, the subgroup of those scoring below the median on the

ideological partisanship measure are the ones responding to treatment, and there are no

significant effects looking at the subgroup of those above the median ideological partisanship

score. One possible explanation for this is that the question concerning NSA data collection and

civil liberties is less familiar to respondents scoring lower on the ideological partisanship score.

There is a well established relationship between political knowledge and ideological consistency

(Zaller 1992), and if it is true that less “ideological” partisans are not familiar with the issue, they

may be looking to the party of the interlocutor as a cue for their disposition to the argument.

24

This would be consistent with a long line of research on partisan heuristics used by low-

knowledge citizens (Lupia 1994; Lau and Redlawsk 2001; Darmofal 2005). For the immigration

question, though, it may be that the partisan positions on this issue are already clear to

respondents regardless of ideology. If that is the case, “cueing” effects would be limited, and the

social identity effects of in-group norm adoption would predominate. This all being said, as

there are not questions of political knowledge on this survey, all one can say for now is that the

effect of ideological partisanship on attitudes of reciprocity varies by context.

Finally, how does partisanship impact the willingness to support reciprocity evinced by

an elected representative? Looking at the 2 x 2 vignette constructed to examine this question, the

results suggest that one’s support for a Congressperson’s effort to promote domestic energy

production hinges on both the party of the Congressperson and whether he or she is working with

both parties. Given the same information concerning efforts to promote domestic energy

production, partisan respondents are more willing to support these efforts if they come from an

inparty rather than an outparty representative. Interestingly, regardless of party, they are also

more likely to support these efforts if the representative “worked with both parties to increase

domestic energy production, as opposed to simply “worked to increase domestic energy

production.” As Figure Eight shows, these effects are similar when looking at both “high PSI”

and “low PSI” partisan subgroups. The only difference of note is that “high PSI” respondents

are even less likely than “low PSI” respondents to support a Congressperson if it is an outparty

Congressperson not “working with both parties.” In total, though, the results suggest that, while

the effect of party on support for a Congressperson are strong, partisans (even those with strong

social attachments to their party) value when either a Democrat or Republican Congressperson

works across the aisle on an issue.

25

Figure Eight Here

Discussion and Conclusion

Does partisanship, and in particular the social identities of partisans, impact the

possibility of a more deliberative democracy? The survey experiments discussed here suggest

that they do; when faced with an argument running counter to their opinion, partisans take into

account whether the argument comes from an inparty or outparty member. In particular,

partisans are significantly less like to evince deliberative or “reciprocal” attitudes when faced

with an outparty, as opposed to neutral or inparty, counterargument. The effects on reciprocity

are particularly pronounced for the subset of partisans for whom party is an important part of

their social identity (in other words, being a Democrat or Republican is an important part of

“who they are”). That being said, though, the experimental results show that all partisans, even

those with strong partisan social identities, prefer to see a representative of either party work

across the aisle on legislation than simply work alone. This suggests that the social identity

dimension of partisanship drives anti-deliberative attitudes with regards to interpersonal

discussion, but it does not drive a desire to see representatives adopt these same attitudes.

Finally, the ideological dimension of party attachment has an effect on reciprocal attitudes, but it

is nuanced and hinges on the specific issue being considered.

While the experimental approach maximizes the internal validity of the findings, a lack of

external validity is a potential limitation to this study. One way external validity is potentially

limited is through the lack of representativeness of the sample; Table Two shows that the mTurk

respondents are whiter, more educated, more liberal, and more Democratic than the United States

population as a whole. Another concern with external validity is with the use of a survey

approach more broadly. For example, it can be questioned as to whether receiving a

26

counterargument statement through an online survey screen adequate simulates encountering

“real world” political argument. Also, it can be questioned whether a small set of survey

questions can adequate tap into nuanced, multifaceted concepts such as “reciprocity,”

“ideological partisanship,” and/or “partisan social identity.” While some of these limitations are

perhaps intractable (and indeed necessary in order to maintain the internal validity of the

findings), in the near future, key results from this study will be replicated in a representative

telephone survey of South Carolina residents. In addition, future iterations of this project will

include more observational data analysis from national sources such as the ANES and CCES, as

well as more qualitative analysis of public deliberation. These additions, hopefully, will lessen

concerns of external validity.

Limitations aside, the results nonetheless help us to understand the impact of partisan

polarization in the 21st century. Despite hand-wringing in the popular media suggesting that

mass polarization is tearing the public apart, the extant academic literature on partisan

polarization traditionally argues that the phenomenon is overblown (Fiorina 2011; Hill and

Tausanovitch 2015). More recent research from a social identity perspective, however, posits

that the public is splitting along partisan lines, but it is an social and affective, not ideological,

split (Iyengar, Sood, and Lelkes 2012; Mason 2015). The results here point to the impact this

partisan social polarization is having on political discourse. While it seems to have little impact

on the desire to hold representatives accountable to standards of reciprocity, in everyday

interpersonal talk, it makes good faith discussion and efforts at forging consensus across party

lines more difficult. To the extent that the ideal of deliberative democracy demands a

deliberative public, not just deliberative elites (Dryzek 2010), the results show that our current

era of polarization makes this ideal more difficult to obtain.

27

The goal of this study is empirical – to examine how partisanship impacts the attainment

of ideals of deliberative democracy, not to examine those ideals as such. Nonetheless, this study

can contribute to the normative assessment of modern democracy in the United States. This

assessment hinges on the yardstick one is using; if one hews to a “procedural” or “minimal” view

of democracy, the results here are, in a way, positive. Partisan social polarization does not

impact the likelihood of one supporting a representative’s efforts in bipartisan legislating; ceretis

paribus, bipartisan legislating is what partisans - even “socially polarized” partisans - want to

see. Yes, gridlock in Washington is still driven by elite polarization, party activist interests, and

institutional rules, but there is no basis for the electoral connection to be a motivating factor.

If, however, one takes the deliberative ideals of Dryzeck, Gutmann and Thompson, and

others seriously, the results here are more troubling. If the ideal of deliberative democracy

requires an engaged public adhering to norms of reciprocity, mass social polarization is driving

us away from this ideal. Perhaps, as critics suggests, “deliberative democracy” is unrealistic

(Shapiro 1999), exclusionary (Fish 1999), or irreconcilable with ideals of participatory

democracy (Mutz 2006). Rebuffing these critics and defending deliberation is beyond the scope

of this paper. However, a cursory look at the 2016 Presidential primary suggests that we should

at least consider what our standards of political discourse should be, and when those standards

are transgressed. It is a primary season marked by misogyny, calls for outgroup

exclusion/vilification, misinformation and ad hominem as a rhetorical norm, violence, and little

in the way of substantive debate. These are not characteristics of a healthy democracy; to move

in a more productive direction, it may be worth it to consider the ideals laid out by deliberative

theory, as well as how factors such as partisanship impact the likelihood of adopting these ideals.

28

Tables

Table One: Distribution of Partisan Respondents across Subgroups of Partisan IndicatorsPartisan Social Identity (PSI) score

Ideological Partisanship High PSI Low PSILow Ideological Partisanship 22.76% 27.99%High Ideological Partisanship 27.31% 21.94%n = 1350

Table Two: Comparison of mTurk Sample to US PopulationSample US Population*

% Non-Hispanic White 75.86% 62.77%% Female 46.75% 50.81%% College Educated (18+) 50.60% 26.74%% Identifying or Leaning Democrat

59.58% 46.27%

% Liberal or Strongly Liberal 39.41% 13.05%*US Population data comes from the 2014 American Community Survey as well as the 2012 ANES Time Series Study

Table Three: Partisan Respondents’ Attitudes Towards Deliberation, by Partisan Social Identity (PSI)

Low PSI High PSIOn important issues facing America today…It is important for politicians to work towards compromise 80.87% 80.27%It is important for politicians to stand firm 19.13% 19.73%

When people discussion political issues of the day… Low PSI High PSIThey tend to find common ground 20.99%*** 27.57%***It tends to highlight their differences 79.01%*** 72.43%***

When I vote for a politician… Low PSI High PSIIt is important for him or her to stick to their platform 55.09% 56.93%It is fine for him or her to have a change in mind 44.91% 43.07%

On most political issues… Low PSI High PSIThere is a clear right and wrong 20.78%*** 29.45%***It is reasonable for intelligent people to be on opposing sides 79.22%*** 70.55%***

n = 1338; **significant at p < 0.01

29

Table Four: Vignette Treatment Effect for All Partisans and “Leaners”

Is the immigration counterargument reasonable?Vignette Very

reasonableSomewhat Not at all

In-Party 10.13%* 60.79%* 29.07%*Neutral 7.71% 56.39% 35.90%Out-Party 9.55% 51.14% 39.32%

Is the immigration counterargument worth considering?Vignette A good deal Somewhat Not at allIn-Party 14.54% 64.10% 21.37%Neutral 11.45% 63.88% 24.67%Out-Party 13.18%** 55.68%** 31.14%**

Is the civil liberties counterargument reasonable?Vignette Very

reasonableSomewhat Not at all

In-Party 12.85% 56.53% 30.62%Neutral 11.90% 56.98% 31.12%Out-Party 6.74%*** 55.51%*** 37.75%***

Is the civil liberties counterargument worth considering?Vignette A good deal Somewhat Not at allIn-Party 18.42% 55.67% 25.91%Neutral 18.54% 52.40% 29.06%Out-Party 11.43%*** 54.30%*** 33.86%***

n = 1350; ***significant at p < 0.01, **significant at p < 0.05, *significant at p < 0.1Significance is measured against the “Neutral” subgroup

30

Figures

Figure One: Dimesons of Partisan Identity

Figure Two

31

Figure Three

Figure Four

Figure Five

32

Figure Six

Figure Seven

33

Figure Eight

34

Works Cited

Abramowitz, A. I. 2010. “Transformation and Polarization: The 2008 Presidential Election and the New American Electorate.” Electoral Studies 29 (4): 594–603.

Abramowitz, A. I., and K. L. Saunders. 2008. “Is Polarization a Myth?” Journal of Politics 70 (2): 542–55.

Bachtiger, Andre, Simon Niemeyer, Michael Neblo, Marco R. Steenbergen, and Jurg Steiner. 2010. “Disentangling Diversity in Deliberative Democracy: Competing Theories, Their Blind Spots and Complementarities.” Journal of Political Philosophy 18 (1): 32–63.

Bachtiger, Andre, Markus Sporndli, Marco R. Steenbergen, and Jurg Steiner. 2007. “Deliberation in Legislatures: Antecedents and Outcomes.” In Deliberation, Participation, and Democracy: Can the People Govern?, edited by Shawn W. Rosenberg. New York: Palgrave Macmillan.

Barabas, Jason. 2004. “How Deliberation Affects Policy Opinions.” American Political Science Review 98: 678–701.

Barbaro, Michael, Ashley Parker, and Jonathan Martin. 2016. “Rank and File Republicans Tell Party Elites: We’re Sticking With Donald Trump.” New York Times, March 4, Web edition. http://www.nytimes.com/2016/03/05/us/politics/donald-trump-republican-party.html.

Bargh, John A., Mark Chen, and Lara Burrows. 1996. “Automaticity of Social Behavior: Direct Effects of Trait Construct and Stereotype Activation on Action.” Journal of Personality and Social Psychology 71 (2): 230.

Becker, Lawrence. 1986. Reciprocity. Chicago: University of Chicago Press.

Berinsky, A. J., G. A. Huber, and G. S. Lenz. 2012. “Evaluating Online Labor Markets for Experimental Research: Amazon.com’s Mechanical Turk.” Political Analysis 20 (3): 351–68.

Bishop, Bill. 2008. The Big Sort: Why the Clustering of Like-Minded Americans Is Tearing Us Apart. Boston: Mariner Books.

Branscombe, N. R., R. Spears, N. Ellemers, and B. Doosje. 2002. “Intragroup and Intergroup Evaluation Effects on Group Behavior.” Personality and Social Psychology Bulletin 28 (6): 744–53.

Brewer, Marilynn B. 1979. “Ingroup Bias in the Minimal Intergroup Situation: A Cognitive-Motivational Analysis.” Psychological Bulletin 86: 307–24.

———. 1999. “The Psychology of Prejudice: Ingroup Love or Outgroup Hate?” Journal of Social Issues 55 (3): 429–44.

35

Brown, Rupert. 2000. “Social Identity Theory: Past Achievements, Current Problems and Future Challenges.” European Journal of Social Psychology 30: 745–78.

Brown, Rupert, Pam Maras, Barbara Masser, James Vivian, and Miles Hewstone. 2001. “Life on the Ocean Wave: Testing Some Intergroup Hypotheses in a Naturalistic Setting.” Group Processes & Intergroup Relations 4 (2): 81–97.

Buhrmester, M., T. Kwang, and S. D. Gosling. 2011. “Amazon’s Mechanical Turk: A New Source of Inexpensive, Yet High-Quality, Data?” Perspectives on Psychological Science 6 (1): 3–5.

Campbell, Angus, Philip E. Converse, Warren E. Miller, and Donald E. Stokes. 1960. The American Voter. New York: John Wiley & Sons, Inc.

Carsey, Thomas M., and Geoffrey C. Layman. 2002. “Party Polarization and ‘Conflict Extension’ in the American Electorate.” American Journal of Political Science 46 (4): 786–802.

———. 2006. “Changing Sides or Changing Minds? Party Identification and Policy Preferences in the American Electorate.” American Journal of Political Science 50 (2): 464–77.

Chong, Dennis, and James N. Druckman. 2007. “Framing Theory.” Annual Review of Political Science 10 (1): 103–26.

Converse, Philip E. 1964. “The Nature of Belief Systems in Mass Publics.” In Ideology and Discontent, edited by David E. Apter, 206–61. New York: The Free Press.

Cook, Fay Lomax, Michael X. Delli Carpini, and Lawrence R. Jacobs. 2007. “Who Deliberatives? Discursive Participation in America.” In Deliberation, Participation, and Democracy: Can the People Govern?, edited by Shawn W. Rosenberg, 25–44. New York: Palgrave Macmillan.

Dahl, Robert. 1956. A Preface to Democratic Theory. Chicago: University of Chicago Press.

Darmofal, David. 2005. “Elite Cues and Citizen Disagreement with Expert Opinion.” Political Research Quarterly 58 (3): 381–95.

Dionne, E.J. 2013. “In Church and Country, a Crisis of Governance.” Washington Post, March 3, Web Edition edition. http://www.washingtonpost.com/opinions/ej-dionne-in-catholicism-and-politics-a-crisis-of-governance/2013/03/03/b5c1a1f4-82c7-11e2-a350-49866afab584_story.html.

Druckman, James N. 2001. “On the Limits of Framing Effects: Who Can Frame?” Journal of Politics 63: 1041–66.

Druckman, James N., and Kjersten R. Nelson. 2003. “Framing and Deliberation: How Citizens’ Conversations Limit Elite Influence.” American Journal of Political Science 47 (4): 729–45.

36

Druckman, James N., Erik Peterson, and Rune Slothuus. 2013. “How Elite Partisan Polarization Affects Public Opinion Formation.” American Political Science Review 107 (01): 57–79.

Dryzek, John S. 2010. Foundations and Frontiers of Deliberative Governance. Oxford: Oxford University Press.

Dryzek, John S., and Simon Niemeyer. 2008. “Discursive Representation.” American Political Science Review 102 (04): 481.

Ellemers, Naomi, Russell Spears, and Bertjan Doosje. 1997. “Sticking Together or Falling Apart: In-Group Identification as a Psychological Determinant of Group Commitment versus Individual Mobility.” Journal of Personality and Social Psychology 72 (3): 617.

Ellsworth, P. C., and K. R. Scherer. 2003. “Apprasial Processes in Emotion.” In Handbook of Affective Sciences, edited by R. J. Dalton, K. R. Scherer, and H. R. Goldsmith, 572–95. Oxford: Oxford University Press.

Erikson, Robert S., Michael B. MacKuen, and James A. Stimson. 2002. The Macro Polity. Cambridge: Cambridge University Press.

Examiner.com. 2010. “Rally to Restore Sanity - Jon Stewart’s Closing Speech (full Text),” October 30. http://www.examiner.com/article/rally-to-restore-sanity-jon-stewart-s-closing-speech-full-text.

Fiorina, Morris P. 2009. Disconnect: The Breakdown of Representation in American Politics. Oklahoma: University of Oklahoma Press.

———. 2011. Culture War? The Myth of a Polarized America. 3rd ed. New York: Pearson Longman.

Fishkin, James S. 1995. The Voice of the People. New Haven, CT: Yale University Press.

Fish, Stanley. 1999. “Mutual Respect as a Device of Exclusion.” In Deliberative Politics: Essays on Democracy and Disagreement, edited by S. Macedo, 88–102. New York: OUP.

Gaines, Brian J., James H. Kuklinski, Paul J. Quirk, Buddy Peyton, and Jay Verkuilen. 2007. “Same Facts, Different Interpretations: Partisan Motivation and Opinion on Iraq.” Journal of Politics 69 (4): 957–74.

Galston, William A. 2014. “Americans Are as Polarized as Washington.” Wall Street Journal, June 3. http://online.wsj.com/articles/william-a-galston-americans-are-as-polarized-as-washington-1401837373.

Gastil, John. 2000. By Popular Demand. Berkeley, CA: University of California Press.

Gearen, Anne. 2015. “Clinton Speaks Forcefully for Path to Citizenship for Illegal Immigrants.” Washington Post, May 5, Web edition. http://www.washingtonpost.com/politics/clinton-

37

speaks-forcefully-for-path-to-citizenship-for-illegal-immigrants/2015/05/05/21b18f20-f330-11e4-b2f3-af5479e6bbdd_story.html.

Green, Donald P., Bradley Palmquist, and Eric Schickler. 2002. Partisan Hearts and Minds: Political Parties and the Social Identities of Voters. New Haven, CT: Yale University Press.

Gutmann, Amy, and Dennis Thompson. 1996. Democracy and Disagreement. Cambridge, MA: Harvard University Press.

———. 2012. The Spirit of Compromise. Princeton, NJ: Princeton University Press.

Hibbing, R., and Elizabeth Theiss-Morse. 2002. Stealth Democracy: American’s Belief About How Government Should Work. Cambridge: Cambridge University Press.

Hill, Seth J., and Chris Tausanovitch. 2015. “A Disconnect in Representation? Comparison of Trends in Congressional and Public Polarization.” The Journal of Politics 77 (4): 1058–75.

Hook, Janet. 2015. “Immigration Puzzle Confounds Republican 2016 Field,” May 8, Web edition. http://www.wsj.com/articles/immigration-puzzle-confounds-republican-2016-field-1431077405.

Huddy, Leonie, Lilliana Mason, and Lene AarøE. 2015. “Expressive Partisanship: Campaign Involvement, Political Emotion, and Partisan Identity.” American Political Science Review 109 (01): 1–17.

Iyengar, Shanto, and Kyu S Hahn. 2009. “Red Media, Blue Media: Evidence of Ideological Selectivity in Media Use.” Journal of Communication 59 (1): 19–39.

Iyengar, Shanto, G. Sood, and Y. Lelkes. 2012. “Affect, Not Ideology: A Social Identity Perspective on Polarization.” Public Opinion Quarterly 76 (3): 405–31.

Iyengar, Shanto, and Sean J. Westwood. 2014. “Fear and Loathing across Party Lines: New Evidence on Group Polarization.” American Journal of Political Science, December, n/a – n/a.

Jackman, Simon, and Paul M. Sniderman. 2006. “The Limits of Deliberative Discussion: A Model of Everyday Political Arguments.” Journal of Politics 68 (2): 272–83.

Jacoby, William G. 2014. “Is There a Culture War? Conflicting Value Structures in American Public Opinion.” American Political Science Review 108 (04): 754–71.

Keith, Bruce E., David Magleby, Candice J. Nelson, Elizabeth Orr, Mark C. Westlye, and Raymond E. Wolfinger. 1992. The Myth of the Independent Voter. Berkeley, CA: University of California Press.

38

Krupnikov, Yanna, and Adam Smith Levine. 2014. “Cross-Sample Comparisons and External Validity.” Journal of Experimental Political Science 1: 59–80.

Kunda, Ziva. 1990. “The Case for Motivated Reasoning.” Psychological Bulletin 108 (3): 480–98.

Lau, Richard R., and David P. Redlawsk. 2001. “Advantages and Disadvantages of Cognitive Heuristics in Political Decision Making.” American Journal of Political Science 45 (4): 951–71.

Levendusky, Matthew. 2009. The Partisan Sort: How Liberals Became Democrats and Conservatives Became Republicans. Chicago: University of Chicago Press.

Lodge, Milton, and Charles S. Taber. 2013. The Rationalizing Voter. Cambridge: Cambridge University Press.

Lupia, Arthur. 1994. “Shortcuts Versus Encyclopedias: Information and Voting Behavior in California Insurance Reform Elections.” American Political Science Review 88: 63–76.

Mackie, Diane M. 1986. “Social Identification Effects in Group Polarization.” Journal of Personality and Social Psychology 50 (4): 720.

Mackie, Diane M., Thierry Devos, and Eliot R. Smith. 2000. “Intergroup Emotions: Explaining Offensive Action Tendencies in an Intergroup Context.” Journal of Personality and Social Psychology 79 (4): 602–16.

Madison, James. 2003. “The Federalist No. 10.” In The Democracy Sourcebook, edited by Robert Dahl, Ian Shapiro, and Jose Antonio Cheibub, 118–22. Cambridge, MA: The MIT Press.

Mansbridge, Jane. 1983. Beyond Adversary Democracy. Chicago: University of Chicago Press.

Mason, Lilliana. 2015. “I Disrespectfully Agree.” American Journal of Political Science 59: 128–45.

Mayhew, David R. 2004. Congress: The Electoral Connection. 2nd ed. New Haven, CT: Yale University Press.

McCarthy, Nolan, Keith T. Poole, and Howard Rosenthal. 2006. Polarized America: The Dance of Ideology and Riches. Cambridge, MA: MIT Press.

Mendelberg, Tali. 2002. “The Deliberative Citizen: Theory and Evidence.” In Research in Micropolitics: Political Decisionmaking, Deliberation and Participation, edited by Michael X. Delli Carpini, Leonie Huddy, and Robert Shapiro, 151–93. Greenwich, CT: JAI Press.

Miller, J. M., and Jon A. Krosnick. 1996. “News Media Impact on the Ingredients of Presidential Evaluations: A Program of Research on the Priming Hypothesis.” In Political Persuasion

39

and Attitude Change, edited by Diana C. Mutz, Paul M. Sniderman, and R. A. Brody, 79–100. Ann Arbor, MI: University of Michigan Press.

Mill, John Stuart. 1859. On Liberty. Edited by Elizabeth Rapaport. Indianapolis, IN: Hackett Publishing Company.

Moghaddam, Fathali. 2016. The Psychology of Democracy. Washington D.C.: American Psychological Association.

Mutz, Diana C. 2006. Hearing the Other Side: Deliberative versus Participatory Democracy. Cambridge: Cambridge University Press.

———. 2008. “Is Deliberative Democracy a Falsifiable Theory?” Annual Review of Political Science 11: 521–38.

———. 2011. Population-Based Survey Experiments. Princeton: Princeton University Press.

Pitkin, H. F. 1967. The Concept of Representation. Berkeley, CA: University of California Press.

“Political Polarization in the American Public.” 2014. Pew Research Center. http://www.people-press.org/files/2014/06/6-12-2014-Political-Polarization-Release.pdf.

Ramsay, Clay, Steven Kull, Evan Lewis, and Stefan Subias. 2010. “Misinformation and the 2010 Election: A Study of the US Electorate.” University of Maryland: World Public Opinion. http://drum.lib.umd.edu/handle/1903/11375.

Rawls, John. 1999. A Theory of Justice. Revised. Cambridge, MA: Harvard University Press.

———. 2005. Political Liberalism. New York: Columbia University Press.

Ryan, Timothy J. 2014. “Reconsidering Moral Issues in Politics.” The Journal of Politics 76 (02): 380–97.

Ryfe, David M. 2005. “Does Deliberative Democracy Work?” Annual Review of Political Science 8 (1): 49–71.

Schumpeter, Joseph. 2003. “Capitalism, Socialism, and Democracy.” In The Democracy Sourcebook, edited by Robert Dahl, Ian Shapiro, and Jose Antonio Cheibub, 5–11. Cambridge, MA: The MIT Press.

Shapiro, Ian. 1999. “Enough about Deliberation: Politics Is about Interests and Power.” In Deliberative Politics: Essays on Democracy and Disagreement, edited by S. Macedo, 28–38. New York: OUP.

Sherman, S. J., Diane M. Mackie, and D. M. Driscoll. 1990. “Priming and the Differential Use of Dimensions in Evaluation.” Personality and Social Psychology Bulletin 16: 405–18.

Stoker, Laura, and M. Kent Jennings. 2008. “Of Time and the Development of Partisan Polarization.” American Journal of Political Science 52 (3): 619–35.

40

Suhay, Elizabeth. 2015. “Explaining Group Influence.” Political Behavior 37: 221–51.

Sunstein, Cass. 2002. “The Law of Group Polarization.” Journal of Political Philosophy 10 (2): 175–95.

Tajfel, Henri, M. G. Billig, R. P. Bundy, and Claude Flament. 1971. “Social Categorization and Intergroup Behavior.” European Journal of Social Psychology 1 (2): 149–78.

Terry, D. J., and Michael A. Hogg. 1996. “Group Norms and the Attitude-Behavior Relationship: A Role for Group Identification.” Personality and Social Psychology Bulletin 22: 776–93.

Theriault, Sean M. 2008. Party Polarization in Congress. Cambridge: Cambridge University Press.

Thompson, Dennis F. 2008. “Deliberative Democratic Theory and Empirical Political Science.” Annual Review of Political Science 11 (1): 497–520. doi:10.1146/annurev.polisci.11.081306.070555.

Turner, John C. 1987. Rediscovering the Social Group: A Self-Categorization Theory. Oxford: Basil Blackwell.

Warren, Mark E., and John Gastil. 2015. “Can Deliberative Minipublics Address the Cognitive Challenges of Democratic Citizenship?” The Journal of Politics 77 (2): 562–74.

“Washington’s Farewell Address.” 2008. Yale University Law School. http://avalon.law.yale.edu/18th_century/washing.asp.

Yzerbyt, Vincent, and Stephanie Demoulin. 2010. “Intergroup Relations.” In Handbook of Social Psychology, edited by Susan T. Fiske, Daniel T. Gilbert, and Gardner Lindzey, 2:1024–83. Hoboken, NJ: Wiley.

Zajonc, Robert B. 1980. “Feeling and Thinking: Preferences Need No Inferences.” American Psychologist 35 (2): 151.

Zaller, John R. 1992. The Nature and Origins of Mass Opinion. Cambridge: Cambridge University Press.

41

Appendix A –Factor Analysis for Issue Questions

Below are the results of a principal component factor analysis of all pre-election policy questions asked in the 2012 American National Election Study (means were inserted for missing values). Results for only the 11 items used in this study are presented (each of these items is accompanied by a 5-point response set, ranging from “Strongly Agree” to “Strongly Disagree”). All items load highly on either the first dimension (which focuses on domestic policies that typically track along liberal/conservative lines) or the second (which focus on social issues more specifically). They are also well rounded, representing a myriad of policy domains. From these 11 policy questions, the ideological partisanship score was constructed for each respondent.

Question 1st Factor

2nd Factor

ANES Questions Asked Every Year Since 1982The government should provide fewer services even in areas such as infrastructure and education in order to reduce spending.

-.701 -.085

All people’s medical expenses should be paid by individuals through private insurance plans, not the government

.702 .143

The government should see to it that every person has a job and a good standard of living.

.705 .001

The government should spend much less money for national defense than it currently does

.410 .172

The government should make every effort to improve the social and economic position of African-Americans.

.610 -.020

By law, abortion should never be permitted -.136 -.683Liberal-Conservative Self-Placement .613 .399

Question Drawn from 2012 ANESThe government should make it more difficult for people to buy a gun .461 .018The government should allow more offshore drilling for oil and natural gas in U.S. waters

-.494 -.079

The government should repeal the Affordable Care Act (the health care reform law passed in 2010).

.668 .228

In order to protect the environment and create jobs, the government needs to regulate business

.624 .221

Gay and lesbian couples should be allowed to legally marry. .196 .819

VARIANCE EXPLAINED BY FACTOR 18.9% 7.7%

42

Appendix B – Framing and Vignette Treatments and Questions (full questionnaire is available at http://ryanstrickler.weebly.com)

Framing Treatments – Respondents were randomly assigned one of these conditions, where a screen would display the text. Once respondents clicked “next” on the screen, they answered a series of 11 questions concerning their attitude toward reciprocity, tolerance, and political disagreement.

Control Condition: Next, there will be a series of statements concerning how issues are discussed and decided on in political life.

Partisan Condition: Next, there will be a series of statements concerning how issues are discussed and decided on by Democrats and Republicans in political life.

Partisan Polarization Condition: Next, there will be a series of statements concerning how issues are discussed and decided on by Democrats and Republicans in political life. This is an important topic, as research shows Democrats and Republicans in Washington are more divided in opinion on the issues now than any time in the past century.

43

Issue Vignette Treatments – Below are the prompts, questions, counterarguments, and survey logic used for the “immigration” and “civil liberties” experimental vignettes discussed in this paper.

Immigration Vignette

Next, I would like to ask you for your opinion on some more current issues. Imagine a scale from 0 to 10, where 0 stands for “strongly disagree,” 10 stands for “strongly agree” and 5 stands “neither agree nor disagree.” Where would you place yourself, from 0 to 10, in response to the following statement?

Unauthorized immigrants in the United States should have a path towards qualifying for US citizenship. (Q116)

(SCALE)

The branch the respondent goes down from here for the next two questions depends on his or her answer to the question above. If the respondent chooses 0-4, proceed to branch #1 below. If the respondent chooses 6-10, choose branch #2 below. If the respondent chooses #5, randomize the branch he or she follows

Branch #1 (Q117, Q120-133)Others have a different perspective. Some (Democrats/Republicans/no identifier) have argued that many unauthorized immigrants are hard-working and contribute to the economy and their communities. They suggest that denying these individuals an opportunity for citizenship would be unfair.

What do you think of this argument? Is it a very reasonable, somewhat reasonable, or not at all reasonable opinion to have on the issue?

A. VERY REASONABLEB. SOMEWHAT REASONABLEC. NOT AT ALL REASONABLE

Do you think this argument is worth considering a good deal, somewhat, or not at all in forming your own view on the issue?

A. A GOOD DEAL WORTH CONSIDERINGB. SOMEWHAT WORTH CONSIDERINGC. NOT AT ALL WORTH CONSIDERING

Now, I would like for your opinion again. Imagine a scale from 0 to 10, where 0 stands for “strongly disagree,” 10 stands for “strongly agree” and 5 stands “neither agree nor disagree.” Where would you place yourself, from 0 to 10, in response to the following statement?

Unauthorized immigrants in the United States should have a path towards qualifying for US citizenship (SCALE)

44

Branch #2 (Q134-148)Others have a different perspective. Some (Democrats/Republicans/no identifier) have argued that unauthorized immigrants, if given a path towards citizenship, would compete with American workers for jobs. Offering a path towards citizenship would thus be unfair to workers who are in the country legally.

What do you think of this argument? Is it a very reasonable, somewhat reasonable, or not at all reasonable opinion to have on the issue?

A. VERY REASONABLEB. SOMEWHAT REASONABLEC. NOT AT ALL REASONABLE

Do you think this argument is worth considering a good deal, somewhat, or not at all in forming your own view on the issue?

A. A GOOD DEAL WORTH CONSIDERINGB. SOMEWHAT WORTH CONSIDERINGC. NOT AT ALL WORTH CONSIDERING

Now, I would like for your opinion again. Imagine a scale from 0 to 10, where 0 stands for “strongly disagree,” 10 stands for “strongly agree” and 5 stands “neither agree nor disagree.” Where would you place yourself, from 0 to 10, in response to the following statement?

Unauthorized immigrants in the United States should have a path towards qualifying for US citizenship (SCALE)

{NEW PAGE}

45

Civil Liberties Vignette

Again, imagine a scale from 0 to 10, where 0 stands for “strongly disagree,” 10 stands for “strongly agree” and 5 stands “neither agree nor disagree.” Where would you place yourself, from 0 to 10, in response to the following statement?

The government’s collection of telephone and internet data is necessary as part of its anti-terrorism efforts.

(SCALE)

The branch the respondent goes down from here for the next two questions depends on his or her answer to the question above. If the respondent chooses 0-4, proceed to branch #1 below. If the respondent chooses 6-10, choose branch #2 below. If the respondent chooses #5, randomize the branch he or she follows

Branch #1Others have a different perspective. Some (Democrats/Republicans/no identifier) have argued that government collection of telephone and internet data has made our country safer, and it is needed to stop the next terrorist attack.

What do you think of this argument? Is it a very reasonable, somewhat reasonable, or not at all reasonable opinion to have on the issue?

A. VERY REASONABLEB. SOMEWHAT REASONABLEC. NOT AT ALL REASONABLE

Do you think this argument is worth considering a good deal, somewhat, or not at all in forming your own view on the issue?

A. A GOOD DEAL WORTH CONSIDERINGB. SOMEWHAT WORTH CONSIDERINGC. NOT AT ALL WORTH CONSIDERING

Now, I would like for your opinion again. Imagine a scale from 0 to 10, where 0 stands for “strongly disagree,” 10 stands for “strongly agree” and 5 stands “neither agree nor disagree.” Where would you place yourself, from 0 to 10, in response to the following statement?

The government’s collection of telephone and internet data is necessary as part of its anti-terrorism efforts. (SCALE)

{NEW PAGE}

46

Branch #2 Others have a different perspective. Some (Democrats/Republicans/no identifier) have argued that government collection of telephone and internet data violates the liberties that our founding fathers gave us and is unconstitutional.

What do you think of this argument? Is it a very reasonable, somewhat reasonable, or not at all reasonable opinion to have on the issue?

D. VERY REASONABLEE. SOMEWHAT REASONABLEF. NOT AT ALL REASONABLE

Do you think this argument is worth considering a good deal, somewhat, or not at all in forming your own view on the issue?

D. A GOOD DEAL WORTH CONSIDERINGE. SOMEWHAT WORTH CONSIDERINGF. NOT AT ALL WORTH CONSIDERING

Now, I would like for your opinion again. Imagine a scale from 0 to 10, where 0 stands for “strongly disagree,” 10 stands for “strongly agree” and 5 stands “neither agree nor disagree.” Where would you place yourself, from 0 to 10, in response to the following statement?

The government’s collection of telephone and internet data is necessary as part of its anti-terrorism efforts (SCALE)

{NEW PAGE}

47

Representative Support Vignette – Each respondent received the following question, where both the partisanship of the Congressperson as well as whether they worked with both parties was randomly assigned. It is thus a 2 x 2 vignette design, which each respondent receiving one of four possible treatments:

During the course of the 2016 congressional campaign, you hear a speech from a (Democratic/Republican) congressman in your state running for re-election. The speech focuses on energy policy, and it touts how the congressman (“worked with both parties to increase domestic energy production”/”worked to increase domestic energy production”). Upon hearing this, how would it affect your likelihood of supporting for this congressman?

A. I WOULD BE MORE LIKELY TO SUPPORT THE CONGRESSMANB. I WOULD BE NEITHER MORE OR LESS LIKELY TO SUPPORT THE

CONGRESSMANC. I WOULD BE LESS LIKELY TO SUPPORT THE CONGRESSMAN