The Tea Party Coalition: Some Racial Resentment, Lots of Economic Resentment

Abstract

In this paper I argue that the main impulse underlying the tea party movement is a conviction that activist government unfairly rewards the undeserving at the expense of the productive leading them to demand limited government. I say main impulse because racial resentment and other illiberal attitudes also contribute to tea party involvement. But illiberal motives do not play the dominant role that much of the leading research suggests. When tests are properly conducted, preference for limited government is the strongest and most consistent predictor of tea party support. Further I show the tea party is a heterogeneous coalition, consisting of three distinct groups. I find the largest of these subgroups has a strongly libertarian flavor and scarcely a whiff of racial animus. Social conservatives comprise another significant group, with strong preferences for limited government and moral traditionalism, and some racially conservative attitudes. Racial conservatives are a substantial subgroup too, but my analysis shows that they are no less motivated by the issue of limited government than others in the movement. These groups are different from one another but came together in the same movement largely because they shared a belief that the federal government had violated basic fairness in its response to difficult economic times.

Full text

Electronic copy available at: http://ssrn.com/abstract=2727727
1
The Tea Party Coalition:
Some Racial Resentment, Lots of Economic Resentment
Emily E. Ekins*
September 29, 2015
Draft
Abstract
In this paper I argue that the main impulse underlying the tea party movement
is a conviction that activist government unfairly rewards the undeserving at the expense
of the productive leading them to demand limited government. I say main impulse
because racial resentment and other illiberal attitudes also contribute to tea party
involvement. But illiberal motives do not play the dominant role that much of the
leading research suggests. When tests are properly conducted, preference for limited
government is the strongest and most consistent predictor of tea party support. Further
I show the tea party is a heterogeneous coalition, consisting of three distinct groups. I
find the largest of these subgroups has a strongly libertarian flavor and scarcely a whiff
of racial animus. Social conservatives comprise another significant group, with strong
preferences for limited government and moral traditionalism, and some racially
conservative attitudes. Racial conservatives are a substantial subgroup too, but my
analysis shows that they are no less motivated by the issue of limited government than
others in the movement. These groups are different from one another but came together
in the same movement largely because they shared a belief that the federal government
had violated basic fairness in its response to difficult economic times.
* Emily Ekins Ph.D. is a Research Fellow at the Cato Institute. I thank my dissertation advisor John
Zaller for his guidance of this project and my dissertation committee members Jonathan Haidt, Lynn
Vavreck, and Kathleen Bawn for their invaluable advice. I also want to thank Jeffrey Lewis for his
methodological guidance. Any deficiencies are my own.

Electronic copy available at: http://ssrn.com/abstract=2727727
2
An Emergent Movement
Few anticipated the conservative backlash that burst into American political life
in the early 2000s and came to be known as the tea party movement. The movement
emerged during one of the most serious economic downturns in American history. After
a period of credit expansion and economic boom, the housing bubble popped followed by
a severe devaluation of capital bringing down with it home prices and the highly over-
leveraged investment banks. In the fall of 2008, some of the nation’s largest financial
institutions failed or were on the verge of failing, the economy shed hundreds of
thousands of jobs each month, and home values plummeted.
In the span of only six months, Congress passed two significant spending bills in
efforts to mitigate the crisis totaling over $1.5 trillion. The first, the Emergency
Economic Stabilization Act secured $700 billion in Troubled Asset Relief (TARP)
dollars or the “bailouts” for Wall Street banks and automakers. The American Recovery
and Reinvestment Act (ARRA), also known as the stimulus, authorized roughly $800
billion toward infrastructure, social services, domestic investments, and tax cuts.
Also during this timeframe Democrats swept the 2008 election further
consolidating their power in both Congressional chambers and the executive branch,
leaving the Republican Party in shambles. In addition, the country witnessed the
historic election of President Barack Obama, the first African-American president.
It was within this context in February 2009 that CNBC contributor Rick Santelli
standing on the floor of the Chicago mercantile exchange took to the show Squawk Box
to lambast President Obama’s stimulus funds directed toward homeowners facing
foreclosure. In short, he proclaimed such actions as fundamentally unfair as he declared
that the “government is promoting bad behavior!” and questioned whether Americans
should “subsidize the losers’ mortgages” or “reward people who can carry the water

3
instead of drink the water.” In an unusual fashion, stock floor traders shouted out in
affirmation with one exclaiming: “how about we all stop paying our mortgage? It’s a
moral hazard!” Santelli proceeded to invite “capitalists” to a Chicago “tea party” to
dump some derivative securities into Lake Michigan.
Something in Santelli’s outburst struck a nerve. Within a week, thousands of
angry conservative and libertarian activists in early 2009 began holding protests under
the banner of “tea party” with explicit opposition to increased government spending, the
growing national debt, and the expansion of government power in the aftermath of the
financial crisis. Activists called for a return to constitutional principles, reining in of
federal power, retrenchment of government spending and the national debt all for the
sake of protecting individual liberty. At tea party protest and rallies, activists brought
homemade signs conveying their belief that activist government was unfair: “Big
Government= Less Freedom,” “Stop The Spending-Give Us Liberty Not Debt,” “Stop
Punishing Success; Stop Rewarding Failure,” “Don’t Spread My Wealth, Spread My
Work Ethic,” or “Free Markets Not Freeloaders” (Ekins 2010, Zernike 2010, Loc. 735).
Surveys confirmed the widespread emphasis on fiscal issues and limited
government: a 2010 CBS/New York Times survey found that 96 percent of tea party
sympathizers favored a “smaller government, providing fewer services” compared to 42
percent of non-supporters, and 81 percent of sympathizers wanted the government to
prioritize deficit reduction over spending money to create jobs, compared to 36 percent
of non-supporters (CBS News/New York Times 2010).
Although tea party members insist their true motivation is their desire for
limited government, leading scholarship on the tea party views it through a prism of
racial resentment and out-group hostility (Parker and Barreto 2013, Skocpol and
Williamson 2012, Perrin et al. 2011, Abramowitz 2012, Knowles et al. 2013).

4
In this paper I review this scholarship and subject it to empirical scrutiny with
three main results. The first result is that existing scholarship is correct in its claim that
racial attitudes correlate with support for the tea party movement. However, the second
empirical result is that, when tests of the kind used in existing studies are properly
conducted, variables measuring preferences for limited government have about twice the
statistical impact of variables measuring racial attitudes. The third empirical result,
based on latent class analysis, is that tea party supporters consist of three distinct
clusters of individuals.
The first cluster, which leans libertarian, comprises 41 percent of the movement
and highly endorses limited government but takes moderate to liberal positions on social
issues and race. The second cluster, 36 percent of all tea partiers, leans socially
conservative and is also strongly defined by limited government preferences and harbors
some racially conservative attitudes. The third group of racial conservatives, 24 percent
of all tea partiers, also shares strong preferences for limited government but also takes
racially conservative positions. While these three groups differ in their attitudes on
moral traditionalism and race, they share a strong preference for limited government. I
then supplement these findings with qualitative interviews with 53 tea party activists
across the country, including early organizers of the movement. In sum, I argue that the
tea party movement is a heterogeneous coalition, in which its prevailing impulse is
largely a sincere demand for limited government.
What Tea Partiers Say They Are About
Qualitative interviews I conducted with 53 tea party activists across the country
revealed their central explicit complaint was that government was getting too expansive
and their top priority was to reduce the size and scope of government.1 One method to
1 States included California, Washington, Utah, Texas, Kentucky, Florida, New Jersey, New York,
Pennsylvania, Virginia, North Carolina, Ohio, and Minnesota. See Appendix A for additional information.

5
quantitatively test the salience of this overarching complaint is to systematically
measure signs at a major tea party protest rally. While signs certainly do not tell us
everything that protestors care about, it tells us how they prioritized using a limited
amount of time and space to express themselves.
I conducted such an analysis of tea party signs at the 9/12 Tea Party Capitol
Hill rally in Washington, D.C. in September 2010. Walking in a systematic fashion
along each row from the back of the tea party rally to the front (about 1pm-4pm), I
took pictures of every visible sign.2 The sampling procedure of protest signs produced
234 readable pictures of signs. I coded the signs’ content, identifying sixteen unique
categories. Signs could be assigned multiple topics, and signs with even subtle
connotations were assigned the relevant category. (See Table 1).
The analysis revealed that more than half the signs (56%) themed a limited
government ethos, either in the form of mentioning the proper role of government or
problems with socialism (36%), concrete economic policies like cutting spending and
taxes (33%), or individual liberty (19%).
Forty percent expressed populist anger, including 28 percent that specifically
mentioned ousting elected officials in the upcoming election, and 12 percent mentioned
the elite versus the people. Twenty-seven percent expressed anger toward or disapproval
of President Obama. Twelve percent of the signs had patriotic or nationalistic themes,
For additional methodological details, see Chapter 4, (p. 116) in Ekins, Emily. 2015. Tea Party Fairness:
How the Idea of Proportional Justice Explains the Right-Wing Populism of the Obama Era, Doctoral
Dissertation: Retrieved from ProQuest Dissertations and Theses.
2 The sampling procedure could not prevent sign-holding protestors from moving from one place to the
other, or ensure that some protestors did not leave during the sampling procedure, which means that
some signs would be missed. Also, some signs were placed on the ground and under bags as the protestors
occasionally sat down to rest. Bias could be introduced if there were systematic differences among the
types of protestors who may have left early, especially if they were grouped together in the audience.
Also, understandably, older protestors would be more likely to become tired and place their signs on the
ground, perhaps under bags, to rest. However, I did not find a high number of signs placed on the ground
out of view.

6
while less than ten percent respectively of protestors’ signs explicitly referenced health
care (9%), the US Constitution (8%), cultural or religious conservatism (5%), ethnic
“othering” (5%), anti-unauthorized immigration (3%), the mainstream media (1%),
Obama’s birthplace or “birtherism” (1%), and generational theft (<1%).
Table 1
Tea Party Protest Signs by Category at the 9/12 2010
Tea Party March on Washington D.C.
Sign Category
%
Limited Government Ethos..
56%
Role of Government
36%
Spending, Taxes, Deficits
33%
Liberty/ Freedom
19%
Populist……………………………
40%
Ousting Politicians
28%
People versus Ruling Elite
12%
Anger Toward Obama……….
27%
Patriotic/Nationalistic……….
12%
Healthcare………………………..
9%
Constitution………………….....
8%
Social Conservatism……….....
5%
Ethnic Undertones…………….
5%
Anti-illegal Immigration…….
3%
Anti-mainstream media……..
1%
“Birther”…………………………..
1%
Generational theft……………..
0.4%
Ambiguous……………………….
6%
Source: 9/12 Tea Party March on Washington, D.C., September 12, 2010; Method:
Author took pictures of every visible sign from 1pm to 4pm on the day of the event, and
walked throughout the entire crowd in a systematic fashion, going row by row during this
period, starting from the back moving to the front. Percentages add up to more than
100% since signs could be tagged with more than one category.
Overall, the central thrust of the signs confirmed the principal explicit complaint
tea partiers make: that the movement is about limited government in general, in the
form of cutting spending, taxes, or deficits, or more philosophically in terms of
promoting liberty. The signs’ second most common complaint reflects the populist
nature of the conservative uprising: these activists want to “throw the bums out” and

7
feel resentful toward elites who they believe want to tell them want to do. Thirdly, the
movement places Obama in the position of the opposition’s symbolic standard-bearer.
However, just because it’s on a sign doesn’t make it so. If latent racial anxieties
were at the root of tea party angst, we would expect most tea partiers to know better
than to put it on a sign. Indeed, further empirical work is needed to determine if
demand for limited government or racial anxieties is the movement’s prevailing impulse.
More Racialized Politics?
Over the past several decades, a number of important studies of public opinion
have stressed the importance of race and racial resentment on voting behavior and
political attitudes, especially attitudes toward government action. Moreover, attitudes
toward government functions that have little to do with race are found to be
“racialized,” such that they are impacted by underlying hostility toward African-
Americans and people of color more generally (Sears and Kinder 1971, Sears, Hensler,
and Speer 1979, Kinder and Sears 1981, Kinder and Sanders 1996, Sears et al. 1997,
Gilens 1999, Sears and Henry 2003, Hutchings and Valentino 2004, also see Kinder and
Kam 2010, Kluegel and Smith 1986). The election of President Barack Obama in 2008
led to new studies documenting the continuing impact of race on political attitudes
(Tesler 2012, Knowles, Lowery, and Schaumberg 2010).
Consequently, since the first widely known tea party protests erupted shortly
after the inauguration of the first African-American president, much of the leading work
on the tea party has viewed it within the literature on racialized politics, particularly
Parker and Barreto (2013) and to some extent Skocpol, Williamson, and Coggin (2012,
2011), Perrin et al. (2011), Abramowitz (2012), and Knowles et al. (2013).3 In the same
3 Arceneaux and Micholson (2012) find racial attitudes failed to significantly impact support for
government aid among tea party supporters in controlled experiments and thus such attitudes do not
appear to primarily undergird their opposition to government action. Knowles et al. (2013) find no

8
vein, researchers have also compared the tea party to social movements like the Ku
Klux Klan of the 1920s, Father Charles Coughlin and his followers (Parker and Barreto
2013, Berlet 2012) or the John Birch Society (Parker and Barreto 2013, Skocpol and
Williamson 2012).
Economic Resentment, Racial Resentment, or Both?
These scholars largely agree a fear of change mobilized the tea party, but contend
that the change it fears is rooted in some form of demographic change and out-group
hostility (Parker and Barreto 2013, Perrin et al. 2011, Skocpol and Williamson 2012).
Nevertheless, the literature varies in the extent to which researchers argue out-group
hostility drives tea partiers’ fears. For instance, Parker and Barreto argue that the
election of President Obama, the first non-white president, signifies subversive change—
a “plot to undermine dominance” of white Americans, in tea partiers’ view (p. 45).
Perrin et al (2011) and Skocpol and Williamson (2012) make a softer, more nuanced,
argument that nativism and racial stereotypes color tea partiers’ fears and give a strong
indication that Obama’s election played a role in stimulating these concerns.
Perrin et al (2011) finds that tea partiers are more likely than non-tea partiers to
be very concerned about “changes taking place in American society these days” leading
the authors to conclude they are similar to earlier right-wing movements concerned with
“status defense” (p. 4). The authors go on to say that this deontological insecurity (fear
of change) is blended with nativism, the fear that newcomers will change the system
and their society.
Interestingly however, the very quotes Perrin et al use to demonstrate tea
partiers’ fear of change perhaps better reflects a grievance with activist government
than it does status defense. For example, Perrin et al cite one tea party survey
association between racial prejudice and tea party identification when controlling for other variables in
their latent growth model, but that prejudice is mediated by belief in national decline and libertarianism.

9
respondent who explained, “we don’t want the big government that’s taking over
everything we worked so hard for…the government’s becoming too powerful…we want
to take back what our Constitution said” which in their view is limited government (p.
7). Instead of identifying this as a specific complaint of government being unfair, the
authors explain it as a categorical fear of change: “Such emotional statements are rooted
in feelings of ontological insecurity and are consistent with expressions of earlier right
wing movements that were based on status defense” (p. 7).
However, some researchers find that the changes tea partiers feared were not
necessarily or exclusively about losing social prestige. For instance, in their in-person
interviews with tea partiers, Skocpol and Williamson find tea partiers were distressed
over perceived changes to the rewards system in America: “...for members of the Tea
Party, it felt as though the fundamental rules about the American Dream had changed.
Working hard no longer meant getting ahead” (p. 30). These concerns are predictable
given the 2008 financial crisis that punished everyone regardless of merit. What
distinguished tea partiers was where they laid blame: “government efforts at
redistribution,” to soften the impact of the economic downturn, “skewed the rewards and
costs that should rightly [have been] apportioned by the free market,” explain Skocpol
and Williamson (p. 31). In other words, tea partiers viewed government action as the
culprit changing the rewards system in America, thereby unfairly rewarding the
“freeloaders” at the expense of the “hard working taxpayers.”
Nevertheless, Skocpol and Williamson further argue that tea partiers’ distinction
between “hard working taxpayers” and “freeloaders” has “ethnic, nativist, and
generational undertones” (p. 74) They conclude this because the authors believe tea
partiers made a “big government” exception for themselves in the form of Social Security
and Medicare. As Skocpol wrote for Politico, “[tea partiers] stand for nothing clear
besides deep generational and racial anxieties” (Skocpol 2010). If Skocpol and

10
Williamson are correct that tea partiers’ support large government programs for their
ingroup but not for other groups, this undermines supporters’ claims that the movement
is about limited government and not about race.
Skocpol and Williamson base their argument on the belief that tea partiers are
not really opposed to large government programs and services that “benefit people like
them,” they are opposed to programs going to others outside of their “social orbit” (p.
60).4 They cite an April 2010 CBS/New York Times survey which found 62 percent of
tea party supporters believe Social Security and Medicare are “worth the costs…for
taxpayers” (p. 60). Further the report from their interviews that:
“not a single grassroots tea party supporter we encountered argued for privatization of
Social Security or Medicare along the lines being pushed by ultra free market politicians
like Representative Paul Ryan (R-WI) and advocacy groups like FreedomWorks and
Americans for Prosperity…They are sure that Obama’s Affordable Care Act is
unconstitutional but elide this standard for their own entitlements” (p. 60-61).
[emphasis added]
They reason that since tea partiers oppose cutting entitlements, but favor cutting
other programs, tea partiers’ aversion to government spending is conveniently self-
serving, perhaps shaped by racial anxieties, rather than grounded in some sort of moral
orientation toward politics.
Their interpretation fits within the literature on racialization, in which some
individuals are more likely to view people in their race/ethnic group as producers and
thus deserving of government services and individuals in other race/ethnic groups as
“takers” disqualifying their desert for government services.
However, there are several shortcomings with Skocpol and Williamson’s
assessment. First, Skocpol and Williamson based their conclusion on an incomplete
understanding for how tea partiers comprehend Social Security and Medicare
entitlement programs. Contrary to the authors’ assertion, tea partiers are in fact
4 Skocpol in Politico also explains, “Rank-and-file tea partiers are not even opposed to the core of federal
spending for defense, Medicare, and Social Security” (Skocpol 2010).

11
supportive of reforming entitlements if they can get back what they put in—consistent
with a more limited government approach—and are often more supportive of reform
than those not supportive of the tea party movement.
Social Security and Medicare are extraordinarily complicated programs, the
details of which are lost on most Americans. Further complicating the public’s
understanding is that these programs are not funded by their own tax dollars in some
sort of “lock box.” Moreover, some individuals will get back less than the real amount
they contributed plus interest while others will get back more. When overly simplistic
survey questions find the public opposes cutting entitlements, it’s unclear if the public
opposes losing redistributive benefits they could receive from the programs, or if they
are afraid of losing the dollars they could have otherwise saved in a retirement account.
To further investigate how tea partiers viewed these programs, I requested that
Reason Foundation place several questions on Social Security and Medicare on the
Reason-Rupe national telephone survey. The survey revealed that, contrary to Skocpol
and Williamson’s argument, tea party supporters were willing to support cutting
entitlement benefits if they were guaranteed to get back what they contributed into the
system, and were more likely than non-tea party supporters to favor changing
entitlement programs.
For instance, although a majority (62%) of tea party supporters thought
entitlements were worth the costs to taxpayers, a majority (74%) were also willing to
reduce their own Social Security benefits if they were guaranteed to receive benefits
equal to the amount they and their employer contributed into the system (Reason
Foundation/Princeton 2012). (See Table 2). At first a slim majority (51%) of tea
partiers would favor reducing their own Medicare benefits to help balance the federal
budget, while a majority (54%) of non-tea partiers would instead oppose. However, if
respondents were guaranteed to receive what they contributed toward the system (this

12
question did not mention employer contributions) then 67 percent of tea partiers would
support, compared to 57 percent of non-tea partiers (Reason Foundation/NSON 2011).
Furthermore, in contrast to Skocpol and Williamson’s qualitative findings, a
September 2012 national survey, which included 300 tea party sympathizers, found this
group far more supportive of voucher-like entitlement reform methods than non-tea
party supporters. Indeed, fully two-thirds of tea partiers favored a reform that would
give “seniors a credit to purchase the health insurance plan of their choice” even if it
might result in “seniors paying more money out-of-pocket for their own health care.” In
contrast, only 40 percent of non-tea partiers agreed (Reason Foundation/Princeton
2012).
Similarly in an August 2011 survey, which also included about 300 tea party
sympathizers, nearly three fourths of sympathizers agreed that “people like me should be
primarily responsible for saving enough money to meet basic expenses in retirement”
compared to 55 percent of non-supporters. Similarly six in 10 tea party supporters also
felt that “people like me should be primarily responsible for saving enough money to
purchase health insurance in retirement.” However, among non-supporters numbers were
flipped with nearly 6 in 10 instead agreeing that “people like me should primarily expect
help from the government to pay for health insurance in retirement” (Reason
Foundation/NSON 2011). Tea party supporters were also roughly 25 points more likely
to favor allowing workers to opt-out of Social Security (71 to 47 percent) and Medicare
(74 to 48 percent) than non-supporters.

13
Table 2
Beliefs about Entitlements and Entitlement Reform by Tea Party Support

14
These data indicate that tea partiers were more likely to take a more fiscally
conservative position on not only taxes, spending, and social welfare programs, but also
on entitlements as well. This does not imply tea partiers want to privatize or abolish
old-age retirement programs, but it weakens’ Skocpol and Williamson’s underlying
assumption that tea partiers selectively favor programs for people like themselves, but
not for other groups.
This leads to the next shortcoming of their analysis. Without further empirical
research, it remains unclear the extent to which racial anxieties color tea partiers’
perceptions of deservingness and attitudes toward government. This is important to
ascertain because if racial anxiety is a prevailing feature of tea party support, this
undermines members’ claims that the movement is about limited government, rather
than about race.5
Arceneaux and Micholson (2012) conduct such an empirical investigation finding
little evidence that race drives tea party opposition to government aid. Using an
experiment, the authors test if tea partiers are more likely to support increases in
unemployment benefits and college aid if such a request were paired with a Caucasian
face, compared to a Latino or African-American face. They find little statistical support
for the claim that racial animus drove tea party opposition to program expansion.
Interestingly, they find that non-conservative tea party supporters were more likely to
support unemployment aid when paired with Latino and African-American faces than
white faces (p < .05). However, even though not statistically significant, they also find
5 Furthermore, qualitative research is particularly sensitive to salient news events. For instance, Skocpol,
Williamson, and Coggin (2011) report that tea partiers voiced disproportionate concern over immigration;
however the authors conducted a bulk of their qualitative interviews in April-May 2010 the same
timeframe during which Arizona passed SB 1070, a controversial immigration law that made immigration
a nationally salient issue in the news cycle at that time. In my interviews conducted between 2010-2014
only about half of the activists brought up the issue of immigration on their own, but all brought up
issues of limited government.

15
that conservative tea partiers rated government assistance to the unemployed 4 points
higher when attributed to a white face compared to a nonwhite face (p-value .07). The
authors conclude that, “racial animus does not appear to be the primary force behind
[tea partiers’] opposition to government aid” but that it may “color…judgments.”
Interestingly Arceneaux and Micholson (2012) find that racial attitudes influence
support for aid only among conservatives who were not tea party supporters: they were
the only group who substantially and significantly offered greater aid when paired with
a Caucasian face than an African-American face by about 12 points (p < .05). If racial
anxieties were a primary motivator for tea party support, why didn’t these
conservatives join? Nevertheless, Arceneaux and Micholson do not discuss these
implications or fully address the extent to which racial anxieties may define support for
the tea party movement itself.
To date, the most systematic and quantitative study of sources for tea party
support is Christopher Parker and Matt Barreto’s Change They Can’t Believe In: The
Tea Party and Reactionary Politics in America.
Parker and Barreto take a stronger position than Skocpol and Williamson, and
Perrin et al, with their central thesis being that the election of Barack Obama
threatened a substantial number of Americans, who turned to the tea party as an
expression of their perceived loss of power and status in their own country:
“We believe that President Obama, by virtue of his position as president, and the fact that he’s
the first nonwhite person to hold the office of president, represents to some an assault upon a
specific ethnocultural conception of American identity and everything for which it stands…” (p.
35).
“We argued that if members of out-groups hold positions of power, members of the in-group are
likely to perceive the powerful out-group members’ goals to be antagonistic to those of the less
powerful in-group. Based on the power and authority the president wields, it’s no surprise that
people who support the tea party, most of whom are white, feel a loss of control” (p. 97).

16
As evidence for this view, they use original survey data they collected in 13
states, selected disproportionately from competitive states. They run a regression model
predicting tea party support, finding that preferences for limited government, racial
resentment (modern racism) and social dominance (old-fashioned racism) predict tea
party support. In the most careful statement of their findings in Table 2.8 on page 96
they report that a movement across the full range of their Limited Government scale
increases support for the tea party by 9 percentage points, while the corresponding
increase for their Racial Resentment scale is 14 percentage points and Social Dominance
is 15 percentage points.6 The measures were standardized to a 0-1 range, min to max.
This analysis has some shortcomings. One is that the Limited Government scale
consists of three dichotomous items, while the Racial Resentment scale consist of four
items, each having a range of 1 to 5 and the Social Dominance scale consists of six
items, also ranging from 1 to 5. Thus the latter measures are likely to be more reliable,
and more likely to generate minimum and maximum values that are more extreme than
those of the Limited Government measure. Conversion of both scales to a 0-1 metric
will not change the likelihood that scores of 0 and 1 will be more extreme positions on
the racial resentment scale than on the limited government scale. For both reasons –
more reliable measurement and greater extremity of min and max values – the Racial
Resentment and Social Dominance scales are likely to have an unmerited advantage in
competing for effect size with the Limited Government scale.
Another shortcoming of the analysis is that it does not determine whether the
effect of racial resentment or social dominance is statistically significantly larger than
the effect of belief in limited government.7 Yet the book refers 81 times to Racial
6 Full regression results found in Table A2.2 on p. 290
7 Table A2.2 on page 290 shows the limited government variable in an ordered logistic regression for tea
party support has a coefficient of .84 (.315) while the racial resentment variable has a coefficient of -.25
(.095) and Social Dominance .19 (.062). The racial resentment variable is negative because these items

17
Resentment as a cause of support for the tea party, but only 18 times to the impact of
Limited Government. The strong impression is that the tea party is mostly the product
of racial resentment, but this is by no means demonstrated by the book’s analyses.
An additional problem is their measure of old-fashioned racism, Social
Dominance. Researchers have found agreement with statements like “inferior groups
should stay in their place” predicts old-fashioned racism and aggressive intergroup
interactions (Ho et al. 2011). However, there are some statistical problems with using
these measures. First, few people endorse such views. For instance the 2004 National
Politics Survey finds only 13 percent of Americans agree that “inferior groups should
stay in their place,” including 11 percent of Republicans, 13 percent of independents,
and 15 percent of Democrats (Jackson et al. 2004).
These small percentages lead to the next problem: regressions using these
measures with skewed distributions have results driven by extreme cases. This may be
the reason Parker and Barreto supplemented the standard Social Dominance battery
with a separate battery that some view as related but also a separate construct that
measures egalitarian preferences. This egalitarianism battery has been shown to predict
“conservatism and opposition to redistributive social policies” (Ho et al. 2011). It is
problematic to conflate one battery than has been shown to predict racism and zero-sum
competition (social dominance) with another battery that predicts conservatism and
opposition to redistributive social policies, suggesting they indicate the same latent
preference for social dominance. This is analogous to surveying beliefs about abortion
and gay marriage and using those as a proxy for limited government preferences. They
may be related, but they are by no means the same thing. Consequently, it is
were coded in reverse (see page 271). The authors say they reversed coded racial resentment due to
“survey question order.” It is unclear why they did not recode the results when running the regression.

18
questionable that this egalitarianism construct is properly measuring if tea partiers
believe Obama becoming president symbolizes “subversion.”
Hypotheses and Method
Further analyses are needed in order to determine if tea partiers’ explicit
claims—concerns about limited government—or out-group hostility are primarily
driving their support. I will test these competing explanations in a series of head-to-
head tests of variables measuring both factors. However, these tests will take care to
ensure both variables are measured with comparable reliability and will utilize tests to
compare the relative explanatory power in predicting tea party support.
Data Analysis I
First, I have selected the CBS/New York Times (CBS/NYT) April 2010 dataset
to test these hypotheses for several reasons. First, the CBS/NYT survey was one of the
earliest surveys conducted of the tea party movement, being fielded in April 2010,
capturing earlier joiners. Second, in part because of its earlier fielding, it asks about the
issues most salient when the tea party emerged, for instance about bailouts, the
stimulus, racial resentment, where Obama was born, about prioritizing economic or
social issues, etc. Third, and perhaps most importantly, the survey includes an
oversample of tea party supporters (775) that allows for greater precision in predicting
tea party support.
The survey includes questions that can measure the variables of interest.
(Question wording can be found in Appendix B). I measure preferences for limited
government and racial anxiety using one variable each, coded 0-1; the variables were
selected to be as similar in variance as possible.8 I also run an additional regression
8 The CBS/NYT question used to measure racial anxiety arguably maps onto the standard battery of
racial resentment questions used in the literature, “Has too much or tool little been made of the problems

19
model including a measure of unauthorized immigration anxiety, since Skocpol and
Williamson found this issue featured prominently in their interviews.
Data Results I
Table 3 shows the results of the logit regression model predicting tea party
support. I run a separate model with each primary variable of interest and the standard
demographic variables, and then combine the primary variables of interest into one
regression model.
Similar to Parker and Barreto’s results, the model finds both preferences for
limited government and racial anxiety are statistically significant predictors, even when
included in the same model. The model finds tea partiers are significantly more likely
to be male, Republican and independent, evangelical, have more education, be older,
and middle income. These demographics fit what one would expect of a middle class,
conservative social movement.
Table 3 Model 4 adds a measure for unauthorized immigration anxiety finding
this to also significantly predict tea party support, without significantly altering the
significance or coefficients for limited government or racial anxiety.
The full regression model offers an opportunity for a “contest” between the two
primary explanations of tea party support—preferences for limited government or racial
anxiety. The results reveal that preference for limited government has about twice the
impact of racial attitudes. A linear-combinations-of-estimates test indicates that the
coefficient for limited government is a statistically more influential predictor of tea
party support than racial attitudes (p <0.01). 9
facing black people, or is it about right?” However, to differentiate this question from the standard four, I
labeled it “racial anxiety” rather than racial resentment.
9 The linear combinations of estimates test reveals the odds ratio between limited government and racial
anxiety is 1.47 in Model 6 and 1.39 in Model 7.

20
Table 3
Predicting Tea Party Support:
Regressions on Measures of Preference for Limited Government, Racial Attitudes, and
Immigration Anxiety (CBS/NYT April 2010 Survey)

21
Limited government has about three times the impact of unauthorized immigration
anxiety, and a linear-combinations-of-estimates test also finds the coefficient for limited
government is a statistically more influential predictor (p <0.001).10
Predicted probability plots with associated rug plots (see Figure 1) shows how
the probability of being a tea party supporter increases as one moves from the min to
max values (0 to 1) for the limited government variable and racial attitudes variable
respectively. The slopes in the predicted probability plots visually demonstrate that tea
party support is more strongly related to preferences for limited government than racial
attitudes.
Figure 1 Predicted Probability of Tea Party Support: Limited Govt and Racial
Attitudes
(CBS/NYT April 2010 Survey)
Note: Figure displays predicted probability plots moving from min to max values for limited government
preferences and racial attitudes in predicting tea party support. Random noise was added to the rug plots to
enhance ability to observe frequency of responses. Without random noise, the responses would be on top of
each other. For this reason, the rug plots are slightly shifted to the right. Rug plots show distributions of tea
party supporters’ responses.
10 The linear combinations of estimates test reveals the odds ratio between limited government and
immigration anxiety is 2.17 in Model 5 and 2.14 in Model 7.
||| || || || ||| | || || ||| || || || | || | |||| || | ||| ||| |||| || | | || | || ||| |||| || | ||| | ||| || || || |||| || ||| || | || || | |||| ||| ||| | || | |||| | || || ||| |||| || || || || || || || || || || |||| ||| || ||| || || | || | |||| || | ||| | || | || || || ||| | | || ||| ||| ||| ||| || || | ||| | | || || || | ||| | | ||| ||| || ||| || | || ||| | ||| ||| || |||| || |||| ||| |||| | || || ||| | ||| || | ||| | || | | ||| || | ||| ||| ||||||||| |||| |||| ||||| || ||| ||| || | ||| ||| || |||| | |||| | | | || ||| || ||| || ||||| | || | |||| ||| || | |||| | ||||| || || | || | ||| || ||| ||| |||| | | || | || ||| || |||| || ||||| | |||| || |||| | || || | |||| ||| | || | || | | || | || | | || | ||| ||| ||| ||| || |||| | ||| | ||||| | || | || | ||| || | |||| || || |||| ||| ||| || | || || ||||| || || || | | ||| | || | ||| || | || | || |||| | || ||||| || || ||| ||| ||| | || || || || || | | || ||||| || || | || ||| | |||| | ||| | |||| || ||| || || | || | | | || | | || || || || ||| ||| || | | || ||| | ||| || ||| || | ||| || ||| ||| || || || || ||| || | || |||| || || | || || || ||| | | || | | || || || ||| || | || ||| ||| || |||| | || | || ||| ||| |||| | |||| || ||| | | |||| || | ||| | | ||| | || || || | || | ||| ||| || ||| || |||| ||| | || |||| || |||| | ||| ||| || || ||||| || ||| || | | || ||| | ||| || ||| ||| ||| || || ||| || | ||| | ||||| | || || || |||| || ||| || || | ||||| || ||| |||| | ||| | ||| || ||| || ||||| || || || || | |||| ||| | ||| || | ||| || || || ||| |||| || ||| | ||| |||||| || | || || || || || || ||| |||| || ||| || || ||| |||| || | ||| ||| || ||||| | ||| | | |||| ||| ||||| ||||| || ||| || |||||| | || || || ||||| | |||| | | |||| |
0.05 .1 .15 .2
Pr(Tea Party Supporter)
0.2 .4 .6 .8 11.2
(X-Axis Move Across Variables' Min to Max)
Limited Govt Preference Racial Anxiety
*Rug Plots show distributions of tea party responses
Predicting Tea Party Support
Limited Govt Preference v Racial Anxiety

22
The rug plots display the distributions of responses among tea party supporters.
The rug plots indicates that almost all tea party supporters (96%) indicated a high
preference for limited government (blue plot). However, a little more than half (55%) of
tea party supporters indicated a racially conservative position, while the remaining were
at the national median or below (45%) (red plot).
Calculating marginal effects shows that, while holding other variables at their
observed values, moving from low to high support for limited government increases the
probability of being a tea party supporter 29 percent. In contrast, moving from low to
high racial anxiety increases the probability of being a tea party supporter by 16
percent, almost half the effect of limited government (see Figure 2).
Figure 2 Probability of Supporting the Tea Party, Marginal Effects
(CBS/NYT April 2010 Survey)
Note: Predicted probabilities of significant predictors of tea party support. All
predictors shown are significant at the p <0.05 level.

23
Data Analysis II
In this section I replicate this model using the American National Election
Studies 2012 Evaluations of Government and Society Survey (ANES12). Although it
lacks the tea party oversample of the CBS/NYT survey, and is conducted 3 years after
the tea party mobilized, there are several benefits to using the data. First, the dataset
includes the four standard racial resentment questions most often used in the academic
literature and four questions about the size of government or its economic scope. The
advantage of having four survey items each is that research has shown that averaging
multiple survey questions reduces measurement error of policy preferences
(Ansolabehere, Rodden, and James M. Snyder 2008). Additionally, when averaging the
four items into scale, they can be standardized in such a way to reduce the influence of
extreme cases.
If using the standard battery of racial resentment questions, and controlling for
extreme cases, we once again find limited government a significantly more influential
predictor of tea party support, this would provide additional compelling evidence that
the tea party is primarily about the size of government.
In this dataset I average responses across four items for both limited government
preferences and racial resentment respectively, taking care they are measured with
comparable reliability and comparable variance.11 (Question wordings found in
Appendix B). I then transform both variables such that moving from 0 to 1 indicates
moving across the interquartile range, reducing the influence of outliers. (This also will
reduce the size of the coefficients relative to coding 0-1 from min to max.)
11 It is important to ensure both the variables measuring limited government preference and racial
resentment have comparable reliability and comparable variance to prevent one variable from having an
unmerited advantage in competing for effect size. The 4 racial resentment questions range from 1-5, 2 of
the limited government questions range from 1-3 and another two range from 1-7. Combining these
questions into respective scales creates a limited government measure with a standard deviation of .68,
and a racial resentment measure with a standard deviation of .75.

24
Data Results II
Table 4 shows the results of the logit regression using the ANES12 data
predicting tea party support. As before, I run a separate model with each primary
variable of interest and standard demographic variables, and then combine the primary
variables of interest into one regression model. In this analysis, the coefficients represent
moving from the 25th to the 75th percentile (the interquartile range) for both the limited
government and racial resentment scales.
Like the CBSNYT model, the ANES12 model finds both preferences for limited
government and racial resentment are statistically significant predictors of tea party
support. (See Table 4). Being Republican or independent remain significant predictors,
and in several models tea partiers are significantly less likely to be in the highest income
group.
In the contest between preference for limited government or racial resentment,
similar to the CBSNYT model, the former has nearly twice the impact of the latter.
Moreover, a linear-combinations-of-estimates (lincom) test indicates that the coefficient
for limited government is a statistically more influential predictor of tea party support
than racial anxiety (p <0.01).12 Limited government has about 3 times the impact of
unauthorized immigration anxiety, and a lincom test also finds the coefficient for limited
government is a statistically more influential predictor (p <0.01).13
Predicted probability plots of the interquartile range (See Figure 3) again show a
steeper slope for limited government preferences than racial resentment, demonstrating
the strong relationship. Calculating marginal effects shows that when holding the other
variables at their observed values, moving across the limited
12 The linear combinations of estimates test reveals the odds ratio between limited government and racial
resentment is 1.75 in Model 6 (p < .05) and 2.02 in Model 7 (p < .001).
13 The linear combinations of estimates test reveals the odds ratio between limited government and
immigration anxiety is 2.59 in Model 6 (p < .001) and 2.43 in Model 7 (p < .001).

25
Table 4
Predicting Tea Party Support:
Regressions on Measures of Preference for Limited Government, Racial
Attitudes, and Immigration Anxiety (ANES EGSS 2 Survey)
government variable’s interquartile range increases the probability of being a tea party
supporter by 16 percent. In contrast, moving across the racial resentment measure’s
interquartile range increases the probability by 9 percent, almost half the effect of
limited government.

26
Figure 3 Predicted Probability of Tea Party Support:
Limited Govt and Racial Attitudes
(ANES 2012 EGSS 2)
Note: Figure displays predicted probability plots moving from 25th-75th percentiles for
limited government preferences and racial resentment in predicting tea party support.
Figure 4 Probability of Supporting the Tea Party, Marginal Effects
(ANES 2012 EGSS 2)
Note: Predicted probabilities of significant predictors of tea party support.
All predictors shown are significant at the p < 0.05 level. Figure displays
predicted probabilities moving from the 25th-75th percentiles for each
variable.
.15 .2 .25 .3 .35
Pr(Tea Party Support)
0.2 .4 .6 .8 1
Interquartile Range (25th-75th Percentiles)
Limited Govt Preference Racial Resentment
Predicting Tea Party Support (Across IQR)
Limited Govt Preference vs Racial Resentment

27
These results indicate that even when muting the effect of min and max values
by examining the interquartile ranges of these variables, desire for small government is
a significantly more influential variable than racial resentment.
In sum, even when using different datasets, and different questions within these
datasets to operationalize the two variables of interest, the ratio of limited government
to racial anxiety remains almost 2 to 1. The results from these two investigations
provide compelling evidence that the tea party is more about limited government than
racial resentment.
The Tea Party as a Coalition
Initially using these standard logit regression models are advantageous because
most of the academic literature has been thinking about the tea party as a homogenous
block. Using improved measures in head-to-head contests demonstrate that even if the
tea party is homogenous, limited government preferences are the prevailing impulse,
ahead of racial anxiety.
However, what this standard logit model (and the models used in Parker and
Barreto (2013) and (Abramowitz 2012)) ignore, is the possibility that the tea party
movement is comprised of dissimilar types of people. Standard logit regression models
assume that every tea party supporter places the same weight on all the variables.
However, the results cannot distinguish whether the tea party is comprised of a
homogenous group of conservatives in which everyone harbors some resentment but are
more strongly motivated by limited government preferences, or if the tea party is a
coalition of those primarily motivated by limited government preferences, others
primarily motivated by racial resentment, and some motivated by a mix of both. Under
either scenario, the aforementioned regression models wouldn’t distinguish between the
two.

28
There are several reasons to think that different people in the tea party place
different weights on various considerations. First, both my activist interviews and
supplementary activist surveys revealed that supporters held dramatically different
positions on privacy, civil liberties, race, immigration, gay rights, but all were very
fiscally conservative. Second, running the standard logit model predicting tea party
support among subgroups in the CBSNYT dataset reveals that when offered the ability
to do so, tea partiers do place substantially different weights on limited government and
racial anxiety. (See Table 5).
Interestingly, among tea partiers with high levels of economic information
(reported both knowing a lot about the Federal Reserve and the federal budget deficit,
25%), racial resentment drops out as a significant predictor, while preference for limited
government remains significant and larger than the coefficient in the standard model.
Among those with less than high economic information, both competing variables are
significant, but limited government remains significantly more influential.
Dividing tea partiers into social conservatives (58%) and social moderates (42%)
in the sample demonstrates that the latter are significantly more predicted by their
support for limited government than racial anxiety by a factor of almost 3 to 1. 14
However, among social conservatives, even though limited government has a larger
coefficient, tests fail to reject the null that’s its significantly larger than racial anxiety.
In another model, libertarian-leaners are also not significantly predicted by
unauthorized immigration anxiety, while social conservatives are.
14 Socially moderate tea party supporters were also significantly more likely than socially conservative tea
partiers to know Obama was born in the United States, to say Roe v Wade was a good thing, and to
prioritize economic issues over social issues.

29
Table 5 Predicting Tea Party Subgroups:
Regressions on Measures of Preference for Limited Government and Racial Attitudes
(CBS/NYT April 2010 Survey)

30
Third, Those who know “a lot” about the tea party (56%) place significantly
more weight on limited government than racial anxiety, compared to those who are less
engaged in the movement (44%). This indicates that more energized supporters are
better predicted by their concerns about the size of government than racial anxiety they
may harbor.
The following additional groups placed significantly greater weight on limited
government than racial anxiety: non-southerners, non-evangelicals, non-birthers, and
those with average or above average education levels. Conversely, for southern,
evangelical, less educated tea partiers, tests failed to reject the hypothesis that
supporters placed equal weight on both variables.
These results imply that the core tea party supporters are more motivated by the
size of government than racial anxiety, but dissimilar types of people joined the
movement. While nearly all are strong fiscal conservatives, they vary in the extent to
which racial attitudes inform their political judgments.
Data Analysis III
As a more incisive test of whether the tea party following consists of different
kinds of people, I will run a latent class regression model for polytomous outcome
variables (poLCA) (See Blaydes and Linzer (2008)) to identify clusters of like-minded
tea partiers. Latent class models define response clusters according to an unobserved, or
latent, categorical variable, and assign respondents’ probabilities of class membership in
each class. The average posterior probabilities indicate the predicted share of
respondents within each latent class.
Consequently, this analysis can distinguish whether the tea party is largely
homogenous, or if it contains different groups who place different weights on competing
considerations. It will identify what people place more weight upon limited government

31
concerns or racial concerns or a combination of both. I can then examine these groups,
who they are, where they are from, their differences and similarities.
To identify the clusters, I include both economic and race related questions:
limited government, taxing the wealthy, the problems facing African-Americans,
Obama’s treatment of Caucasians and African-Americans, and Obama’s birthplace. I
include several manifest variables on social issues as well. (See Appendix C for question
wording). As covariates to the latent class regression, I include partisanship, gender,
income, education, age, region, and evangelicalism. I run the poLCA model only on the
tea party subset sample of 735 respondents to identify clusters within the tea party
movement.15
Data Results III
Based on an assessment of minimizing both the Akaike Information Criterion
(AIC) and Bayesian Information Criterion (BIC), I determine that three clusters is the
appropriate number of classes (See Figure 5). Doing so reveals a cluster representing 41
percent of tea party supporters who are libertarian leaning and racially moderate and
very economically conservative. Another cluster making up 36 percent of tea party
supporters are socially conservative, also very economically conservative, and in between
being racially moderate and conservative. The third group are racial conservatives,
comprising 24 percent of the tea party movement. A key finding is that all three tea
party clusters are strong economic conservatives, but vary substantially in their
positions on social issues and race. It should be noted that the group denoted as
libertarian-leaning, or libertarian, are not necessarily Libertarian Party identifiers, but
rather take economically conservative positions and moderate to liberal positions on
social and cultural issues.
15 The CBS/NYT April 2010 survey included an oversample of 775 tea party supporters. However, in the
poLCA analysis I excluded 40 observations with missing data.

32
Figure 5:
The Tea Party Coalition
Tea Party Clusters Discovered Through Latent Class Analysis
Demographic Profiles
Table 6 provides basic demographic information on the three types of tea party
supporters.
Libertarian Leaning Tea Partiers
The libertarian group has significantly more education, higher incomes, are
middle aged, male, and are less likely to attend church regularly or be evangelical. This
group also feels less comfortable within the Republican Party with half identifying as
independent, almost twice as many as the other two tea party groups. These tea
partiers are also much less likely to identify as staunch conservatives and are less
favorable toward President George W. Bush compared to the other groups. Libertarian-
leaning tea partiers are solidly favorable toward former Fox News host Glenn Beck and
former Alaska Gov. Sarah Palin, but considerably less so than the other groups. Only a
third of libertarian leaners think Palin would make an effective president.

33
Table 6 Tea Party Clusters:
Demographics, Political Identification, Tea Party Engagement
(CBS/NYT April 2010 Survey)
Socially Conservative Tea Partiers
Social conservatives are middle class, middle-aged evangelicals, and staunch
conservative Republicans who attend church weekly. They are solidly favorable of the
Republican Party and George W. Bush, and are also overwhelmingly positive toward

34
Glenn Beck and Sarah Palin. Nevertheless, this group doesn’t think Palin should be the
president. In comparison to libertarian leaners, these tea partiers have less education,
are more middle income, more female, but roughly the same age and similarly likely to
be employed.
Racially Conservative Tea Partiers
Racial conservatives are considerably older, retired, male, and from the South. In
fact, fully half are over 65 years old, and a majority (56%) are from the South. This
group is similar to social conservatives in terms of education, partisan identification,
and favorability toward the Republican Party, George W. Bush, and Sarah Palin.
However, racial conservatives are the only tea party group that thinks Sarah Palin
would be an effective president (60%). This group differs from social conservatives on
religiosity, is significantly less evangelical and less likely to attend church weekly but is
not as secular as libertarian leaners. This seems to translate into this group being in
between social conservatives and libertarians on ideology.
When it comes to tea party participation, no one group seems to have the
advantage. Nearly equal numbers (slightly more than half) say there is a tea party
group active in their community, about a fifth have participated in a tea party group
either by attending a rally or donating money, and a third have visited tea party
websites. These similar participation rates are particularly notable given the fairly
heterogeneous demographic profiles across these groups’ ages, incomes, education,
religiosity, and region.
Political Beliefs
Taking a closer look at these tea party groups’ political beliefs (see Table 7),
reveals they are all very fiscally conservative, but differ across partisan loyalties, social
issues, and race.

35
All three tea party groups equally favor smaller government offering fewer
services (roughly 96%) and overwhelmingly oppose raising taxes on high-income
households (roughly 85%). Moreover, 7 to 8 in 10 still favor smaller government even if
it requires cuts to defense, education, Social Security and Medicare. However, when it
comes to priorities, 95 percent of libertarian tea partiers prioritize economics over social
issues, compared to 57 percent of social conservatives and 80 percent of racial
conservatives.
Table 7
Political Beliefs Across Tea Party Clusters and Non-Tea Party Groups
(CBS/NYT April 2010 Survey)

36
In comparison to their non-tea party Republican counterparts, tea partiers are
substantially more fiscally conservative. Compared to Republicans who do not support
the movement, tea partiers are about 30 points more likely to favor smaller government,
and oppose raising taxes on upper income households, and about 20 points more likely
to continue favoring small government if it requires difficult cuts (significant at p <
.001) and to believe that social services fosters dependency (significant at p <.01).
Libertarian tea partiers are also about 20 points more likely to prioritize economics over
social issues compared to regular Republicans (p < .001), but socially conservative tea
partiers are about 20 points less likely (p < .001). Racial conservatives are similar to
regular Republicans. Less surprisingly, independents and Democrats are far less likely to
select fiscally conservative positions.
Major differences emerge among tea party groups on the issue of race and moral
traditionalism. On race, racial conservatives (75%) are nearly twice as likely as
libertarian tea partiers (41%) and nearly 20 points more likely than social conservatives
(58%) to say “too much” has been made of the problems facing African-Americans.
Racial conservatives were more than 10 times as likely as both libertarian leaners and
social conservatives to say that President Obama favors black Americans over white
Americans (84% to 8%). Only 15 percent of libertarian tea partiers thought President
Obama was born outside of the United States, compared with 59 percent of racial
conservatives and 31 percent of social conservatives.
Interestingly, on issues of President Obama and race, libertarian tea partiers take
more liberal positions than non-tea party independents and Republicans, or are about
the same. For instance, only 8 percent of libertarian tea partiers said they believed
Obama favored black Americans over white Americans, compared to nearly three times
that among Republicans who do not support the tea party movement, with this
difference being statistically significant (p < .001), and were statistically similar to

37
independents. In addition, libertarian tea partiers were more than 20 points more likely
than regular Republicans (significant p < .001) and 12 points more likely than
independents (significant p < .05) to believe President Obama was born in the United
States. Furthermore, a comparison of means tests finds that libertarian tea partiers are
equally likely as non-tea party Republicans to say “too much” has been made of the
problems facing African-Americans, but significantly more likely than independents (p
< .001).
Similar to libertarian tea partiers, socially conservative tea partiers are
significantly less likely than non-tea party Republicans to say that the president favors
African-Americans over Caucasians (p < .001), and are statistically similar to
independents. However, they are less likely to believe President Obama was born in the
United States than regular Republicans (p < .01) and independents (p < .001), but
equally likely to go so far to say he was born abroad. While not as high as racial
conservatives, social conservatives are considerably more likely than non tea party
Republicans and independents to say “too much” has been made of the problems facing
African-Americans (significant, p < .001).
Given the heterogeneous attitudes on race and Obama, it is particularly
interesting that tea partiers are still nearly universally unfavorable toward him. Fully
90-99 percent of all three groups say they “disapprove” of Obama’s job performance, and
nearly equal numbers say he’s moving the country toward socialism and expanded the
role of government too much. Rather than abstract trepidation, socialism connotes a
very specific outcome across tea party groups. Interestingly, despite differential levels of
education, the three tea party groups correctly defined socialism in roughly equal
proportion. Tea party supporters were the only political group in the CBSNYT survey
that could accurately define socialism using their own words in response to an open-
ended survey question. While nearly 6 in 10 tea partiers could accurately define

38
socialism, only 40 percent of non-tea party Republicans, 19 percent of independents, and
16 percent of Democrats could do the same. Overwhelming numbers also disbelieve the
president shares their values; however, libertarian-leaners are less likely to believe they
have disparate values (68% versus ~90%). These data indicate that tea partiers may
reach similar unfavorable conclusions about President Obama but perhaps for different
reasons.
On traditionalism, 77 percent of socially conservative tea partiers say there
should be no legal recognition for same-sex couples and 74 percent think abortion should
be illegal, In comparison, only 8 percent of libertarian tea partiers agree. Instead 62
percent of libertarian tea partiers say that Roe v Wade was a “good thing” compared to
8 percent of social conservatives. On each of these social issues, racial conservatives are
in between (50 and 23 percent respectively).
Again, libertarian tea partiers take more liberal positions on social issues
compared to some non-tea party groups. For instance, non-tea party Republicans are
nearly 5 times as likely to oppose legal recognition of same-sex couples compared to
libertarian tea partiers, 38 to 8 percent respectively, (significant at the p < .001). Non-
tea party independents and Democrats are about 3 times as likely to oppose as
libertarian tea partiers, (significant at p < .001). Similarly on abortion, non-tea party
Republicans are about 4 times as likely to say abortion should be illegal and
independents are about 3 times as likely, compared to libertarian tea partiers (both
significant at p < .001). Libertarian tea partiers are statistically as likely as Democrats
to say abortion should be illegal. Six in 10 libertarian tea partiers say Roe v. Wade was
a good thing, statistically similar with non-tea party independents and Democrats. On
immigration, libertarian tea partiers are 12 points less likely than non-tea party
Republicans to support decreasing immigration levels (p < .05) and statistically similar
to non-tea party independents and Democrats. Socially conservative tea partiers are

39
statistically similar with non-tea party Republicans and independents on lowering
immigration levels.
Table 8 reports some additional information on how the three subgroups of tea
partiers differ from the rest of the population. Each column of the table reports results
for a different subgroup. In the first column, for example, the dependent variable is
membership in the tea party libertarian subgroup versus everyone else in the CBSNYT
sample. Results from logit models show that preference for limited government is a large
and significant predictor of membership in each of the three tea party subgroups.
However, racial anxiety loses its statistical significance in predicting tea party
libertarians and is considerably less influential among social conservatives. Among the
latter group, limited government has roughly 2.5 times the effect of racial anxiety, and a
lincom test reveals it's a significantly more influential predictor (p < .01).16 For the
racially conservative quarter of tea partiers, both racial anxiety and preference for
limited government are equally significant predictors of this group.
The model verifies that libertarian tea partiers are significantly more likely to be
male, independent or Republican, be older, have more education and have higher
household incomes, and are less evangelical. Socially conservative tea partiers are
significantly more Republican (but not independent), evangelical, and middle income.
Being older and Republican are significant predictors of being racially conservative.
In sum, these three tea party groups are extraordinarily similar in their fiscal
conservatism and preference for small government—the very issues they say they are
about. However, they differ substantially across race and moral traditionalism—issues
scholars have emphasized as defining the movement. The presence of these three groups
in the tea party indicates that while race and traditionalism matter, and that racial
16 The linear combinations of estimates test reveals the odds ratio between limited government and racial
attitudes is 7.82.

40
conservatives and social conservatives were drawn to the tea party, the prevailing
impulse of the movement is what they say: it’s primarily about limited government.
TABLE 8
Predicting Tea Party Cluster Membership:
Regressions on Measures of Limited Government Preferences and Racial Attitudes
(CBS/NYT April 2010 Survey)
Furthermore, these results indicate the tea party is comprised of libertarian
leaners who take moderate positions on race and social issues, a group of older, largely
Southern, racial conservatives who also care about limited government, and social

41
conservatives who are primarily concerned about limited government but also harbor
some racial resentments.
Discussion
These results somewhat reflect groupings Lipset and Raab (1970) identified
nearly 50 years previous when they wrote “there are at least three different kinds of
right-wing strains” (Lipset and Raab 1970, p. 472). The first comprised economic
conservatives who were higher income, more educated and did not hold prejudicial views
towards African-Americans, similar to the libertarian tea party group. The second group
comprised “radical rightists” were also very economically conservative, but less educated,
working class, more religious, with a strong desire to preserve or bring back the past
with which they identified, similar to the tea party social conservatives. The authors
labeled the third group the pejorative term “rednecks” who were less economically
conservative, but shared demographic characteristics with the radical rightists, and were
“susceptible to status preservatism, whenever it becomes or is made salient and when it
jibes with their economic position” and that the backlash target is “typically ethnic or
racial in nature” (Lipset and Raab 1970, p. 474). This third group seems most similar
to, although not the same as, the third group of tea party racial conservatives.17
There are also useful parallels between the modern tea party movement and the
1890s populists. For instance, few scholars deny that some 1890s populists engaged in
anti-Semitic conspiratorial rhetoric or revealed prejudice (see Johnston 2007, Collins
1989, Hofstadter 1955, Ostler 1995, Handlin 1951, Pollack 1962, Woodward 1959).18
17 Lipset and Raab further report that radical rightists and economic conservatives were united in their
support of Republican Barry Goldwater’s presidential candidacy and the John Birch Society; however,
they were divided in the support for George Wallace’s third party bid in 1968 (Lipset and Raab 1970,
pgs. 475-476).
18 In Age of Reform Richard Hofstadter (1955) caused controversy by proposing there were strains of
anti-Semitism in the 1890s populist movement: “In the books that have been written about the Populist
movement, only passing mention has been made of its significant provincialism; little has been said of its

42
However, as the debate over Richard Hofstadter’s Age of Reform reveals,19 historians
largely agree that such illiberal strains were not central to the 1890s populist core.
Instead, they argue that populists’ economic grievances were the most important motive
of their mobilization and thus largely define scholarly understanding of the movement
(Woodward 1959, Nugent 1963, Johnston 2007, Ostler 1995, see Hicks 1931, Goodwyn
1976).20 This is similar to my argument of the contemporary tea party, which is that
illiberal attitudes are present in some parts of the movement, but they are not the
relations with nativism and nationalism; nothing has been said of its tincture of anti-Semitism”
(Hofstadter 1955, Loc. 990, also see Handlin 1951). However, it was Hofstadter’s bolder argument which
instigated much debate: “It is not too much to say that the Greenback-Populist tradition activated most
of what we have of modern popular anti-Semitism in the United States” (Hofstadter 1955, Loc. 1231-
1263). However, other historians remained unconvinced by Hofstadter’s argument, evidenced by Lawrence
Goodwyn’s (1991) recent declaration that “the world of populism constructed by Hofstadter now
languishes in ruin” (cited in: Ostler 1995, p. 2, see Woodward 1959, Pollack 1960, 1962, Nugent 1963,
Goodwyn 1976). However, few historians today deny that there were strains of anti-Semitism found
among some parts of the 1890s populist movement (see Johnston 2007, Collins 1989, Hofstadter 1955,
Ostler 1995, Handlin 1951, Pollack 1962, Woodward 1959). Robert Johnston (2007) points out that
“without a doubt many Populists did engage in a search for “Shylock” bankers…and most of these writers
were not innocent of the cultural work of the term “Shylock” (p. 132). (The term “shylock” is considered
an anti-Semitic slur). Rhetoric often made references to conspiracies involving Jews versus the “toilers,” or
like populist orator Mary E. Lease accused President Grover Cleveland as being the “the agent of Jewish
bankers and British gold” (Higham 1957, p. 132, Johnston 2007). In populist leader Tom Watson’s 1892
campaign book he wrote: “Did [Jefferson] dream that in 100 years or less his party would be prostituted
to the vilest purposes of monopoly, that redeyed Jewish millionaires would be chiefs of that Party, and
that the liberty and prosperity of the country would be…constantly and corruptly sacrificed to Plutocratic
greed in the name of Jeffersonian Democracy?” (Watson 1892/1975 cited in: Kazin 1995, Loc. 252).
Woodward (1959) pointed out that among some in the movement there was a “tendency to turn cranky,
illiberal, and sour…to take off after race phobias, religious hatreds, and witch hunts” (p. 70). Nevertheless,
the prevailing scholarly view is that illiberal attitudes did not define the movement or the primary
grievances of the movement adherents.
19 Richard Hofstadter was perplexed by the backlash against his proposition that late 19th century
populism had strains of anti-Semitism. Robert Collins (1989) writes that some historians viewed
Hofstadter’s research as reducing the Populists to “a horde of xenophobic, anti-Semitic, delusional cranks”
(p. 152, see Pollack 1960, Pollack 1962). In a letter to historian C. Vann Woodward in 1963 Hofstadter
wrote “if I had known what an imbecile fuss would be raised about my having mentioned the occasional
anti-Semitic rhetoric among the Populists, I would either have dropped it as not worth the trouble or else
spent even more time than I did in clarifying what I was saying” (cited in Johnston 2007, p. 130).
20 For instance, C. Vann Woodward (1959) contended “Whatever concern the farmers may have had for
their status was overwhelmed by desperate and immediate economic anxieties….while their legislative
program may have been often naive and inadequate, it was almost obsessively economic” (p. 63).

43
dominant force behind tea party members’ grievances and mobilization. Just as
economic grievances are viewed as the dominant force defining the 1890s populists, so to
do limited government preferences define the tea party movement of the early 21st
Century.
Conclusion
This article has evaluated competing claims that the tea party is primarily
motivated by racial attitudes or a desire for limited government. This has been done
using standard logit regression models that assume homogeneity of the movement,
latent class models allowing for heterogeneity, and then specialized models predicting
membership in the classes identified in the LCA model. Opinion data starting from 2010
onwards show that preference for limited government is a significantly more influential
predictor of tea party support than racial anxiety.
The latent class analysis reveals three distinct clusters of tea partiers, one being
libertarian-leaning tea partiers who are racially moderate and very fiscally conservative,
another being socially and very fiscally conservative who also hold some racially
conservative attitudes, and then a third group of racial conservatives who are also
fiscally conservative.
In light of this heterogeneity in the composition of the tea party movement, these
results further imply that scholars should use care in making statements that apply to
all members. In particular, my results imply that it is not quite accurate to make
statements of the form, “tea party membership is partly predicted by racial resentment,”
because this statement implies that all tea party supporters are to some degree
predicted by racial resentment. Rather, based on my results, one should say that some
tea party members are predicted by racial resentment and some are not.
The tea party is surely not the only social movement with a heterogeneous
composition. Latent class analysis should therefore probably be used more often than it

44
is in the study of public opinion. When, for various practical reasons, scholars cannot
test for the sort of heterogeneity uncovered by latent class analysis, they should use
caution in statements they make on the basis of simple regression analysis.
In sum, these results demonstrate that the tea party movement attracted many
different kinds of people, including secularized fiscal conservatives, social conservatives,
and some racially anxious people as well. While these groups vary substantially in their
attitudes on race and moral traditionalism, what all three groups share in common is an
explicit desire for small government. In sum, race is some but not all of the tea party
story. Instead, fiscal conservatism and a preference for limited government is the
movement’s prevailing impulse.

45
APPENDIX A
Activist Information
Between 2010 and 2014 53 interviews were conducted with tea party activists in California, Florida,
Georgia, Kentucky, Minnesota, New Jersey, New York, North Carolina, Ohio, Pennsylvania, Tennessee,
Texas, Utah, Virginia, Washington, and the District of Columbia.

46
APPENDIX B
CBS/New York Times April 2010 National Survey
Tea Party Issues
• Q22. If you had to choose, would you rather have a smaller government providing fewer services, or a bigger
government providing more services?
Racial Resentment
• Q72. In recent years, do you think too much has been made of the problems facing black people, too little
has been made, or is it about right?
Immigration Anxiety
• Q65. What about illegal immigration, how serious a problem do you think the issue of illegal immigration is
for the country right now — very serious, somewhat serious, not too serious, or not at all serious?
Tea Party Libertarian-Leaners/Social Moderates and Social Conservatives
Tea party supporters were divided into two groups, reflecting the divide I found on the ground among tea party
activists, and the entrance poll I conducted. While most of the tea party is extremely fiscally conservative, one part is
fairly socially moderate while the other part is socially conservative. Using the method developed in Boaz and Kirby
(2006), I categorize tea party libertarian-leaners as those who are socially moderate to liberal and tea party social
conservatives as those who are socially conservative.
In this dataset I use a question about same-sex marriage and another on abortion. Tea party social conservatives say
there should be no legal recognition for same-sex couples and abortion should not be permitted. Tea party libertarian
leaners favor civil unions or same sex marriage and say abortion should be generally available or available with limits.
• Which comes closest to your view? Gay couples should be allowed to legally marry, or gay couples should be
allowed to form civil unions but not legally marry, or there should be no legal recognition of a gay couple's
relationship?
• Which of these comes closest to your view? Abortion should be generally available to those who want it,
abortion should be available but under stricter limits than it is now, abortion should not be permitted?

47
APPENDIX B (Cont.)
ANES EGSS 2 2012 National Survey
Tea Party Fiscal Issues
• c2_zh_ Do you think that the government should provide more services than it does now, fewer services than it
does now, or about the same number of services as it does now?
• c2_zr1 Some people feel that the government in Washington should see to it that every person has a job and a
good standard of living. Suppose these people are at one end of a scale, at point 1. Others think the government
should just let each person get ahead on their own. Suppose these people are at the other end, at point 7. And of
course, some other people have opinions somewhere in between, at points 2, 3, 4, 5, or 6. Where would you place
yourself on this scale, or haven't you thought much about this?
• c2_w7 As you may know, about 10 years ago the federal government cut income taxes for Americans at all
income levels through 2010. In December 2010 the tax cuts were extended until 2012. Do you think the tax cuts
should have been extended for all income levels, extended only for incomes under $250,000, or not extended at
all?
• c2_zk3 Do you favor, oppose, or neither favor nor oppose the government trying to make this income difference
(income inequality) smaller?
Scale Reliability Coefficient: .676; Standard Deviation .68
Racial Resentment
• c2_zgg1 Irish, Italians, Jewish and many other minorities overcame prejudice and worked their way up. Blacks
should do the same without any special favors.
• c2_zgg2 Generations of slavery and discrimination have created conditions that make it difficult for blacks to
work their way out of the lower class.
• c2_zgg3 Over the past few years, blacks have gotten less than they deserve.
• c2_zgg4 It’s really a matter of some people not trying hard enough;; if blacks would only try harder they could
be just as well off as whites.
Scale Reliability Coefficient: .795; Standard Deviation .75
Unauthorized Immigration Anxiety
• c2_zb1 Which comes closest to your view about what government policy should be toward unauthorized
immigrants now living in the United States? Should the government
o Make all unauthorized immigrants felons and send them back to their home country.
o Have a guest worker program that allows unauthorized immigrants to remain in the United States in
order to work, but only for a limited amount of time.
o Allow unauthorized immigrants to remain in the United States and eventually qualify for U. S.
citizenship, but only if they meet certain requirements like paying back taxes and fines, learning
English, and passing background checks.
o Allow unauthorized immigrants to remain in the United States and eventually qualify for U. S.
citizenship, without penalties.

48
APPENDIX C
CBS/New York Times April 2010 National Survey
Manifest Variables Included in Latent Class Regression Model
• If you had to choose, would you rather have a smaller government providing fewer services, or a bigger
government providing more services?
• Do you think it is a good idea or a bad idea to raise income taxes on households that make more than
$250,000 a year in order to help provide health insurance for people who do not already have it?
• In recent years, do you think too much has been made of the problems facing black people, too little has
been made, or is it about right?
• According to the Constitution, American Presidents must be "natural born citizens." Some people say
Barack Obama was not born in the United States, but was born in another country. Do you think Barack
Obama was born in the United States, or do you think he was born in another country?
• In general, do you think the policies of the Obama administration favor whites over blacks, favor blacks over
whites, or do they treat both groups the same?
• Which comes closest to your view? Gay couples should be allowed to legally marry, or gay couples should be
allowed to form civil unions but not legally marry, or there should be no legal recognition of a gay couple's
relationship?
• Which of these comes closest to your view? Abortion should be generally available to those who want it,
abortion should be available but under stricter limits than it is now, abortion should not be permitted?

49
References
Abramowitz, Alan I. 2012. "Chapter 8: Grand Old Tea Party: Partisan Polarization and
the Rise of the Tea Party Movement." In Steep: The Precipitous Rise of the Tea
Party, edited by Larence Rosenthal and Christine Trost, 195-211. Berkeley:
University of California Press.
Ansolabehere, Stephen, Jonathan Rodden, and Jr. James M. Snyder. 2008. "The
Strength of Issues: Using Multiple Measures to Gauge Preference Stability,
Ideological Constraint, and Issue Voting." American Political Science Review
102 (2):215-232. doi: 10.1017/S0003055408080210.
Arceneaux, Kevin, and Stephen P. Micholson. 2012. "Who Wants to Have a Tea Party?
The Who, What and Why of the Tea Party Movement." PS: Political Science
and Politics 45 (4):700-710.
Berlet, Chip. 2012. "Chapter 2: Reframing Populist Resentments in the Tea Party
Movement." In Steep: The Precipitous Rise of the Tea Party, edited by Larence
Rosenthal and Christine Trost. Berkeley: University of California Press.
Blaydes, Lisa, and Drew A. Linzer. 2008. "The Political Economy of Women's Support
for Fundamentalist Islam." World Politics 60:576-609.
Boaz, David, and David Kirby. 2006. The Libertarian Vote. In Policy Analysis.
Washington DC: Cato Institute.
CBS News/New York Times. 2010. CBS/New York TImes: Government Tea Party
Movement Poll. New York: CBS/New York Times.
Ekins, Emily. 2010. Los Angeles.
Ekins, Emily. 2015. Tea Party Fairness: How the Idea of Proportional Justice Explains
the Right-Wing Populism of the Obama Era, Doctoral Dissertation: Retrieved
from ProQuest Dissertations and Theses.
Gilens, Martin. 1999. Why Americans Hate Welfare: Race, Media, and the Politics of
Antipoverty Policy. Chicago: University of Chicago Press.
Ho, Arnold K., Jim Sidanius, Felicia Pratto, Shana Levin, Lotte Thomsen, Nour Kteily,
and Jennifer Sheehy-Skeffington. 2011. "Social Dominance Orientation: Revisitng

50
the Structure and Function of a Variable Predicting Social and Political
Attitudes." Personality and Social Psychology Bulletin XX (X):1-24.
Hutchings, Vincent L., and Nicholas A. Valentino. 2004. "The Centrality of Race in
American Politics." Annual review of Political Science 7:383-408.
Jackson, James, Vincent Hutchings, Ronald Brown, and cara Wong. 2004. National
Politics Study 2004. edited by Inter-university Consortium for Political and
Social Research. Ann Arbor, MI.
Kinder, Donald R., and Cindy D. Kam. 2010. Us Against Them: Ethnocentric
Foundations of American Opinion Chicago: University of Chicago Press.
Kinder, Donald R., and Lynn M. Sanders. 1996. Divided by Color: Racial Politics and
Democratic Ideals, American Politics and Political Economy Series. Chicago:
University of Chicago Press.
Kinder, Donald R., and David 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.
Kluegel, James R., and Eliot R. Smith. 1986. Beliefs About Inequality. New York:
Aldine De Gruyter.
Knowles, Eric D., Brian S. Lowery, and Rebecca L. Schaumberg. 2010. "Racial
prejudice predicts opposition to Obama and his health care reform plan."
Journal of Experimental Social Psychology 46:420-423.
Knowles, Eric D., Brian S. Lowery, Elizabeth P. Shulman, and Rebecca L. Schaumberg.
2013. "Race, Ideology, and the Tea Party: A Longitudinal Study." PLoS ONE 8
(6):1-11.
Lipset, Seymour Martin, and Earl Raab. 1970. The Politics of Unreason: Right-Wing
Extremism in America, 1790-1977. 2nd ed. Chicago: University of Chicago Press.
Maddox, William S., and Stuart A. Lilie. 1984. Beyond Liberal and Conservative:
Reassessing the Political Spectrum. Washington DC: Cato Institute.
Parker, Christopher S., and Matt A. Barreto. 2013. Change They Can't Believe In: The
Tea Party and Reactionary Politics in America. Princeton: Princeton University
Press.

51
Perrin, Andrew, Steven J. Tepper, Neal Caren, and Sally Morris. 2011. "Cultures of the
Tea Party." Contexts 10 (74).
Reason Foundation/NSON. 2011. "Reason-Rupe/NSON August 2011 National
Telephone Poll." Reason Foundation/NSON.
Reason Foundation/Princeton. 2012. "Reason-Rupe September 2012 National Telephone
Poll." Reason Foundation Accessed August 1.
Sears, David O., and P. J. Henry. 2003. "The origins of symbolic racism." Journal of
Personality and Social Psychology 85 (2):259-275. doi:
papers3://publication/doi/10.1037/0022-3514.85.2.259.
Sears, David O., Carl P. Hensler, and Leslie K. Speer. 1979. "Whites' Opposition to
"Busing": Self-Interest or Symbolic Politics?" The American Political Science
Review 73 (2):369-384. doi: 10.2307/1954885.
Sears, David O., and Donald R. Kinder. 1971. "Racial Tensions and Voting in Los
Angeles." In Los Angeles: Viability and Prospects for Metropolitan Leadership,
edited by W.Z Hirsch, 51-88. New York: Praeger.
Sears, David O., Colette Van Laar, Mary Carrillo, and Rick Kosterman. 1997. "Is It
Really Racism? The Origins of White Americans' Opposition to Race-Targeted
Policies." Public Opinion Quarterly 61 (1):16-53.
Skocpol, Theda. 2010. "Politico Arena: GOP/Tea party linkage savvy move or sign of
desperation?". Politico, Last Modified September 20, 2010 Accessed August 5,.
http://www.politico.com/arena/perm/Theda_Skocpol_E8BD76F7-7169-44FC-B779-
3993550FC1E6.html.
Skocpol, Theda, and Vanessa Williamson. 2012. The Tea Party and the Remaking of
Republican Conservatism. New York: Oxford University Press.
Skocpol, Theda, Vanessa Williamson, and John Coggin. 2011. "The Tea Party and the
Remaking of Republican Conservatism." Perspectives on Politics 9 (1):25-43.
Tesler, Michael. 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.
Zernike, Kate. 2010. Boiling Mad: Inside Tea Party America. New York: Times Books
Henry Holt and Company.