La eleccion de la pandemia: analyzing Latino political behavior during the 2020 election

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Request accessible transcript

Transcript (if available)

Content

La Eleccion de la Pandemia
Analyzing Latino Political Behavior During the 2020 Election

By

Jarred R Cuellar

A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
POLITICAL SCIENCE & INTERNATIONAL RELATIONS

August 2022

Copyright 2022 Jarred R Cuellar
ii

Dedication

I dedicate this dissertation to my wife, Ailyn Manjarrez Cuellar, who was with me since
day one of my doctoral journey and has supported me even when I could not do so for myself. I
equally dedicate this to my mother, Michelle Cuellar, without whom I would never have thought
it possible to attain a PhD.

iii

Acknowledgements

I would, first, like to acknowledge and thank Dr. Louis DeSipio, Dr. Christian Grose, Dr.
James Lo, and Dr. Darren Ruddell, who all agreed to serve on my committee and helped me
overcome my academic short comings.
I would like to send a very special thanks to my advisor, Dr. Ange-Marie Hancock
Alfaro, who guided and mentored me throughout my years at USC. It was thanks to her that I not
only overcame imposter syndrome, but also grew into a confident academic. You always will be
my “academic Mom”.

iv

TABLE OF CONTENTS
Dedication ....................................................................................................................................... ii
Acknowledgements ........................................................................................................................ iii
List of Figures ..................................................................................................................................v
List of Tables & Models ................................................................................................................ vi
Abstract ......................................................................................................................................... vii
Chapter 1: Introduction ....................................................................................................................1
Chapter 2: Latino Evangelical Vote Choice in the 2020 Presidential Election ...............................7
Literature Review .....................................................................................................................9
Theory .....................................................................................................................................16
Data .........................................................................................................................................17
Variables & Measurement ......................................................................................................19
Methodology ...........................................................................................................................23
Findings & Implications .........................................................................................................30
Conclusion ..............................................................................................................................32
Chapter 3: What Causes Latinos to Believe in Conspiracy Theories? ..........................................34
Literature Review ...................................................................................................................36
Theory .....................................................................................................................................43
Data .........................................................................................................................................45
Variables & Measurement ......................................................................................................46
Methodology ...........................................................................................................................50
Discussion ...............................................................................................................................56
Conclusion ..............................................................................................................................57
Chapter 4: Questions Matter: Polling the Latino Electorate in 2020 .............................................59
Literature Review ...................................................................................................................62
Survey Weighting ...................................................................................................................65
Analyzing Measures of Vote Choice ......................................................................................66
Breaking Down the Latino Vote .............................................................................................68
Discussion ...............................................................................................................................69
Bibliography ..................................................................................................................................71

v

List of Figures

Figure 1: Evangelical Latino Vote Theory Model .........................................................................18
Figure 2: Covid Impact by Race/Ethnicity ....................................................................................43
Figure 3: What causes Latinos to Believe in Conspiracy Theories ...............................................44

vi

List of Tables & Models

Table 1: 2020 Presidential Vote .....................................................................................................23
Table 2:2020 Latino Vote ..............................................................................................................23
Table 3: If the Election Were Held Today (Late Oct/Early Nov ..................................................66
Table 4: What Percentage of Your Social Circle Will Vote For (Mean) .......................................68
Table 5: Latino Vote 2020 US Presidential Election .....................................................................68
Model 1: Regression Using Latino as a Variable ..........................................................................24
Model 2: Effects of Being Evangelical on Latino Vote .................................................................26
Model 3: Update to Include Interactions .......................................................................................27
Model 4: Looking at the General Population (Without Interaction) ..............................................28
Model 5: Looking at the General Population (With Interactions) .................................................29
Model 6: Belief in Hydroxychloraquine ........................................................................................51
Model 7: Covid is No Worse than Cold/Flu ..................................................................................52
Model 8: Masking Wearing Protects Against Covid .....................................................................53
Model 9:Mail-in-Voting Causes Widespread Voter Fraud ............................................................54
Model 10: Deep-State Tried to Remove Trump ............................................................................55

vii

Abstract
While the study of American politics has long studied the effects of religion on political
behavior, there is a dearth of knowledge on how religion affects Latino political behavior, given
the long-held belief that the Latino community is a religious monolith, with Latinos being
overwhelming Catholic in faith. This has caused a gap in knowledge in which we know
relatively little about how religious denomination may impact voter behavior. In this dissertation,
I will go onto show that Evangelical Latinos were significantly more likely to vote for Trump.
While those under 45 do not follow this trend, my findings highlight the necessity to stop
omitting religious denomination from further Latino politics studies.
Similar to the study of Evangelical Latino voting behavior, there is not much known
about Latinos and what drives their belief in conspiracy theories. While Latinos have been found
to be more susceptible to conspiratorial thinking, we are still unsure what causes Latinos this
phenomenon. 2020 saw the emergence of numerous conspiracy theories related to Covid-19 and
the 2020 election. With this said, I use my second article (Chapter 3) to show that college
educated Latinos, and those who are associated with the Democratic Party are less likely to
subscribe to contemporary conspiracy theories. In contrast, I show that Latinos with an external
locus of control are more likely to believe in these Covid related conspiracy theories.
Recently we have seen much of the public lose faith in pre-election polls due to incorrect
forecasts. In this article I show that inaccurate survey weighting may not actually be to blame,
like many pollsters believe. Rather, in this new era of hyper-partisanship it appears that many
Latino respondents do not feel comfortable disclosing their true intentions. I use the experimental
“social circle” question from the USC Poll to show that it may be time to ask new questions that
more accurately forecast, not only election results, but also the Latino vote.
1

Chapter 1:
Introduction

2

In March of 2020 much of the United States was forced to shelter-in-place due to Covid-
19 virus. This emergence of this new virus resulted in a global pandemic, the likes of which the
US has not faced since the Spanish Flu Pandemic of 1918 (Center of Disease Control and
Prevention 2018). To make matters even more interesting, the US was in the middle of a
Presidential election year, with the primaries taking place when the pandemic was declared. The
uniqueness of this election cycle cannot be overstated, as Covid response turned into one of the
top wedge issues that year (Hillygus & Shields 2009; Calvo & Ventura 2021). Furthermore,
Covid divided much of the population, with many Republicans arguing against is severity, and
holding stances against measures meant to stop the spread of the virus (Bierwiaczonek 2020).
The divide on Covid response led to many having inaccurate information on how to stop or treat
the virus (Fadilah et al. 2020), especially Latinos.
The US Latino population faced a disproportionate amount of hardship due to Covid-19,
with Latinos not just catching the virus at higher rates, but also having an increased chance of
dying from Covid. This comes as a result of several things- the first being misinformation, as
many Latinos in the US have incorrect information on the US healthcare system. Other reasons
being multigenerational housing and a greater chance of being an essential worker (Rodriguez-
Diaz et al. 2020; Hendrix 2021; Sena & Weber 2021). However, a key reason for this was the
fact that President Donald Trump continuously downplayed the severity of Covid, and he even
helped spread conspiracy theories regarding the virus (Dyer 2020a; Uscinski et al. 2020).
Ironically enough, Latino support for Trump had been on the rise since taking office in
2017 (USC Poll 2020), with many baffled as to why. Latino politics scholars have consistently
informed the world that US Latinos are far from a homogenic voting bloc (Beltran 2010), an
example being the comparison of the 2000 and 2008 elections. George W. Bush, a Republican,
3

received over 40% of the Latino Vote in 2000/2004 (Varela 2020), while Barack Obama, a
Democrat, received roughly 70% of the Latino vote in 2008. However, in the case of Bush, this
support can be explained by his history as governor of Texas and being vocal about his support
for the Latino community during his campaign (Nuno 2007). This was not the case with Trump,
who had a history of anti-Latino rhetoric and policy positions (Corral & Leal 2020). This causes
one to wonder what might have led to this increase in support.
When asked why he thought there was increased support for Trump amongst Latinos,
President Obama argued that the explanation was simple- many Latinos are Evangelical, and
these individuals turned out on Election Day (Scott 2020). Many dismissed Obama’s claims as
an oversimplification, despite a vast body of literature showing that Evangelicals, in general, are
more likely to vote for Republican candidates (Thompson 1986; Rozzell & Wilcox 1996;
Patrikios 2013; Wald & Calhoun-Brown; Whitehead et al. 2018). Given these prior findings, I
found it premature to dismiss President Obama’s claims without empirical evidence. In the
following chapter I do just this, by using data from the 2020 USC Dornsife/LA Times
Presidential Poll to show that increased Latino support for Trump was indeed tied to the growing
Evangelical population (Cuellar 2022). Regression analysis will show that Latino Evangelicals
were more likely to vote for Trump in 2020, but this may not be the case with younger
individuals.
However, the election itself was not the only issue to arise in 2020. Another key issue
was how to conduct the election itself, with many states adopting vote-by-mail policies that
allowed individuals to vote, but not put themselves at risk for Covid. Given his low poll numbers
(fivethirtyeight.com 2020), Trump used these new policies to argue that the election would be
marred by voter fraud and therefore the results would not be legitimate. Despite political
4

scientists and other election experts attempts to show that widespread voter fraud was nowhere
near probable (Minnite 2011; Smith 2017; Edelson et al. 2017; Holman & Lay 2019), many
believed Trump’s conspiracy theory.
Furthermore, the belief in conspiracy theories, in general, was on the rise (Dyer 2020;
Uscinski et al. 2020) due to Covid/Covid response. These theories ranged from the belief in
unproven medicine to treat Covid all the way to the idea that the “Deep State” was trying to
remove Trump from office. Many researchers have shown that the key catalyst for these
phenomena was partisan affiliation, with those who affiliate themselves with the Republican
party being more susceptible to these conspiracy theories. What has not been analyzed in depth is
the relationship between US Latinos and these contemporary conspiracy theories.
Prior studies on conspiracy theories have found that there are several groups of
individuals who may be prone to conspiratorial thinking, with one of these being Latinos
(Goertzel 1994; Abalakina-Papp et al. 1999). Along with Black Americans, Latinos were found
to have a higher chance of believing in conspiracy theories that specifically paint the United
States in a negative light (Parsons et al. 1998; Miller et al. 2016). This makes sense considering
the long history of brutality and prejudice that Latinos in the US have faced (Shaw et al. 2018).
Such brutalities include lynchings in the Southwest, repatriation (Balderrama & Rodriguez 2008;
Shaw et al. 2020), and the 2006 Sensenbrenner Bill (Barreto et al. 2010). It is also
understandable when one considers that much of the mistreatment towards racial and ethnic
minorities, despite being a reality, was relegated to conspiracy theory status by the general public
(King 1992; Gorbie-Smith 1999; Reverby 2009).
Further analyzation showed that race and ethnicity may not be the key to understanding
conspiratorial thinking, as it was also shown that low-income White conservatives also subscribe
5

to conspiracy theories at higher rates (Oliver & Wood 2014). This led many to argue that while
Black and Latino Americans are more likely to believe in conspiracy theories at higher rates, the
key driver of this is having an external locus of control (Abalamkina-Papp et al. 1999). In other
words, those who feel that external forces are responsible for one’s successes/failures are more
likely to believe in conspiracy theories.
This theory is highly plausible and applicable to the Latino community, given the
aforementioned injustices that Latino Americans have faced, which has resulted in many these
individuals having an external locus of control. However, it is unknown if this holds with
conspiracies that spread during 2020, given the high levels of widespread fear that are brought
about by a novel virus that medical professionals do not know how to treat. The partisan nature
of contemporary conspiracy theories may also lead to more nuanced results. Therefore, I use
Chapter 3 to uncover what factors drove Latinos to believe in conspiracy theories regarding the
pandemic and the 2020 election. I use several OLS regressions to show that Latinos Democrats
are significantly less likely to believe in conspiracy theories, while Latinos with an external locus
of control tend to be more susceptible.
I previously mentioned that Trump propagated the fear of widespread voter fraud due to
his standings in the polls. While it was true that many polls (fivethirtyeight.com 2020) predicted
a blowout win for Biden, this was not the case when it came to the actual results. Biden did win
the presidency, but not with the wide margin that pollsters predicted. This was reminiscent of
2016 when most polls predicted a Clinton victory, only to be wrong on Election Day. This led
many pollsters to believe that either their survey weighting was wrong, or that traditional polling
questions (Flannely et al. 2000; Wang et al. 2015; Gelman et al. 2016) no longer garner accurate
results.
6

To combat the latter, pollsters at the USC Dornsife/LA Times Poll began to ask questions
that do not ask who the respondent is voting for, but rather who their social circle will vote for.
When analyzing the post-mortem results, we
1
found that the social circle question more
accurately predicted the margin of victory than traditional polling questions. However, the results
of the traditional questions also became more accurate when weighted according to the actual
vote, which also led to the idea that the weights may have simply been off.
When looking at Latinos, the USC Poll was fairly accurate at predicting the group’s vote
choice by using the social circle question, while the traditional question was not as accurate. I
use the final chapter of this dissertation to discuss best practices in Latino polling, where I show
weighting may not always necessarily be the issue, but rather asking questions such as the social
circle one may allow for more accurate poll results. The reason being that many may not feel
comfortable, even under anonymous circumstances, stating for whom they intend to cast their
vote. Rather, a question that acts as a proxy-such as the social circle question, may be a better
predictor of election results.

1
I was part of the USC Dornsife/LA Times Poll research team, working as a research assistant. I am able to
therefore use these data in my dissertation.
7

Chapter 2:
Latino Evangelical Vote Choice in the 2020 Presidential Election

8

It is well known that many Latinos tend to have strong religious ties that shape their
identity. While this may be common knowledge, we in the field of political science have rarely
looked at the effect church/religion has on Latino political behavior. Now may be the optimal
time to study this effect, especially given recent claims by former President Barack Obama
(2020) that Latino support for Donald Trump had increased due to the growing number of
Latinos who religiously identify as Evangelicals. While the existing body of literature on religion
and political behavior, amongst Latinos, is slim, studies on Latinos and religion have shown that
it can play a role in several areas including partisanship and voter behavior (Kelly & Kelly
2005), with some findings showing that Evangelicals of color tend to be more conservative
(Wong 2018). A study by Correa & Leal (2001) also showed that churches, in the Latino
community, can act as civic associations and in turn have a positive effect on Latino voter
turnout. When drawing upon other REP findings we can see a link between religiosity/church
attendance and political action (Harris 1993) that is caused by the marginalization of
underrepresented Americans. In turn, minority churches, especially those considered “political”
(Calhoun-Brown 1996), have been found to act as a catalyst for civic engagement.
While these studies provide a peak into how religion affects Latino political behavior,
little is actually known on how identifying as an evangelical affects behavior. We have seen
studies show that, historically, Evangelicals hold more conservative political beliefs (Brint &
Arbutyn 2010) when compared to Catholics (Newman 2009; Campbell et. al. 2010; Wong 2018),
however recent studies have shown that this may no longer be the case. Political scientists have
found that Catholic voters are increasingly becoming more conservative (Newman 2009), with
vote choice becoming more evenly dispersed amongst Democrats and Republican candidates.
However, these studies (Converse et al. 1961; Converse1966; Gray et al. 2006; Newman 2009
9

Rozell 2018) have also shown that Evangelicals continue to support conservative/Republican
candidates at high levels (Patrikios 2013; Schwadel 2017; Margolis 2020). What is not so clear is
whether this is also true amongst the growing number of Latino evangelicals, as these individuals
belong to both a traditionally Democratic and traditionally Republican voting blocs.
In this study I will utilize data from the 2020 USC Dornsife Presidential Poll to show that
President Obama was fairly accurate when he stated Evangelical Latinos were indeed more
likely to vote for Trump. However, there is a level of nuance to it. I will show that while Latino
Evangelicals were overall more likely to vote for Trump, Evangelical Latinos under 45 are not
(Chaturvedi 2014; Wong 2018). I will go on to discuss why this is not a surprising finding and
what it means for Latino politics going further.
Literature Review:
History of Religion and Politics
Academics who study politics have been long been fascinated by the impact of religion
(Converse 1966; Thompson 1986; Gray et al. 2006; Wald & Calhoun-Brown 2014) since the
historic election of John F Kennedy (Converse 1961). During this time researchers were
intrigued by the possible effect Kennedy’s Catholicism had on the 1960 presidential vote
breakdown (Converse 1966). At first glance Kennedy appeared to have over-preformed with
Catholic voters by garnering 80% of the Catholic vote share in 1960, compared to 50% for
Stevenson in 1956. The Protestant vote stayed roughly the same in these two elections, leading
many to believe that Kennedy’s religious ID was to thank for his outstanding performance with
Catholics. However, after deeper analyzation Converse found that this was not the case. With
Eisenhower’s extreme popularity as a WWII hero, he managed to siphon off a large amount of
traditionally Democratic voters (Converse 1966; Hyman & Sheatsley 1953), which included a
10

significant amount of Catholic and Protestant Democrats. Kennedy merely gained back those
Catholics who defected to Eisenhower in 1956. However, Kennedy’s Catholicism did cost him
Protestant votes (Converse 1966; Warwick 1996), as his religious identity acted as a deterrent for
traditionally Democratic voting Protestants who voted for Eisenhower due to his war hero status.
These findings give way to the notion that religion can have significant effects on voter behavior,
but perhaps not in the way we might expect.
Besides being able to help explain candidate vote choice, religion can also be a driving
factor in ideology formation (Morgan & Meier 1980), as it can act as a guide for policy positions.
This is especially true for minorities, and it can be illustrated throughout history with Black
churches using the pulpit to not only advocate for civil rights (Morgan & Meier 1980; Calhoun-
Brown 2000; McDaniel 2009), but also as a mobilizing tool for Black political participation. This
can be seen across the Christian diaspora. This is seen with both the Catholic Church, as it has
traditionally led the fight against abortion rights (Fleishman 2000; Feree et al. 2002); and the
Protestant church, which historically fought for morality-based legislation that includes
outlawing gambling and pornography (Lienesch 1982; Rozell & Wilcox 1996).
It has been argued that that religion may play a bigger role in proposition and referendum
voting than it does in presidential elections (Hutcheson & Taylor 1973; Fairbanks 1977; Morgan
& Meier 1980), given findings that show protestants to be significantly more likely to support
morality-based propositions and referendums such as alcohol and gambling prohibition. These
findings go on to show spurious relationship with candidate voting. However, these are
predominantly White samples, and the results are much more nuanced when including
racial/ethnic minorities. REP research has found Black churches, and to a lesser degree Latino
churches, act as mobilizing tools to advance the interests of the group as a whole (Calhoun-
11

Brown 1996; Barreto et al. 2006; Wald & Calhoun-Brown 2014). With this said, it is safe to
assume that if churches influence policy positions and ideology, they will naturally influence
vote choice.
Contemporary politics and the influence of religion
When trying to untangle the role of religion in contemporary US politics, researchers
have studied what is known as the “religion gap” (Olson & Green 2006; Smidt et al. 2010). This
term stems from the 2004 election when it was discovered that voters who attended church
regularly had a 2-1 propensity to vote for Republican candidates, while those who did not
attended church regularly had a 2-1 propensity to vote for Democrat candidates. Candidate
religiosity was also shown to also play a role, especially when a candidate’s religious affiliation
aligns with partisan ideology. An example being George W. Bush, who garnered a large share of
the evangelical vote because much of the GOP platform (i.e. anti-abortion, anti-gay marriage,
etc.) aligned with Evangelical ideals. Thus, showing that the policy stances may mean more than
the candidate’s religiosity (Marti 2019; Margolis 2020).
This was exemplified in 2016, when we saw significant Evangelical support for Trump,
as a large amount of evangelicals viewed Donald Trump as the best choice given his policy
stances rather than personal religiosity, which could be argued was nonexistent (Gorski 2017).
These voters rationalized their support for Donald Trump by arguing that he was the best chance
they had at achieving evangelical oriented policy. While most originally saw Trump as the
“lesser of two evils”, a significant number went on to see him as their ideal candidate (Gorski
2017; Whitehead et al. 2018). These individuals also being shown to ascribe to White Christian
nationalist policies, so much so that they believed that voting for Trump was a symbol of their
faith and their way of preserving American Christian values (Whitehead et al. 2018). However, it
12

remained unclear if this belief was only prevalent amongst White Evangelicals or Evangelicals
of all groups.
As previously mentioned, the evangelical population is growing amongst non-whites with
approximately 25% of Evangelicals coming from racial/ethnic minority groups. When it comes
to these groups, support for Donald Trump in 2016 was significantly less enthusiastic (Gorski
2017). Gorski (2017) and Whitehead et al. (2018) mention that the lack of minority support is
why the term “White Christian nationalism” was used frequently when discussing radical Trump
supporters, and can be further demonstrated with most ethnic/racial minorities opting to vote for
Clinton in 2016. With this said, despite the insights these studies provide on the role of religion,
they should be taken with a grain of salt when it comes to fully understanding the role it plays
within the US electorate. Why is this so? While they do explain that minorities, while
increasingly evangelical, do not behave the same as their White evangelical counterparts, not
much detail backs this argument.
Most studies, rather, only provide descriptive statistics to simply to show that most
evangelicals supported Trump, apart from minorities. They do not parse out the potential
differences amongst these Evangelical minority groups (see Wong 2018, as she does separate by
race/ethnicity). They go on to explain why White evangelicals supported Trump, but not why or
which minority evangelicals did not. By grouping all minority groups together (Gorski 2017;
Whitehead 2018), we remain in the dark about how Latino Evangelicals voted. This is in part
due to the fact that it can be difficult to explain minority vote choice, especially Latino vote
choice (Hero 1992; Jones-Correa & Leal 1996; Manforti & Sanchez 2010). Regardless, we are
still unsure if President Obama’s premise that a significant chunk of the Latino electorate
supported Trump due to their evangelism is accurate.
13

Latinos and political behavior
Before answering this question, I must provide a brief discussion the history of Latino
political behavior. This topic truly took off in the late 1990s and early 2000s with an emerging
collective of political scientists conducting studies first on Latino voters (De la Garza & DeSipio
1992; Pachon & DeSipio 1992; Arivizu & Garcia 1996). As more Latino politics scholars
entered the academy, the study of Latino voting behavior really began to take off. Early findings
showed that Latino voter behavior is much more nuanced when compared to Black and White
voters, with Latinos being best mobilized by canvassing and coethnic candidates (Barreto 2007;
Michaelson 2003, 2005, 2006). This essentially shows that Latinos participate at higher numbers
when made to feel as though they are an integral part of the US polity (Sinclair et al. 2013).
This theory, that inclusion leads to greater participation, has shown to extend beyond
voting as it can also be seen in Barreto et al.’s (2010) article on the 2006 immigration protests. In
this situation the Catholic Church, along with several other organizations, made a point to
mobilize Latinos from all over the US. The efforts made by the Catholic Church, Spanish media,
and Latino dominant public schools again created the sense of inclusion that led Latinos to
believe that they were an important part of the US polity. This in turn, led to enormous amounts
of Latino participation during the 2006 immigration protests. What remains unclear is whether
the Catholic Church was a powerful mobilizer because of its role as a religious institution or if it
is because it is considered a reliable Latino ally. What is clear is that like in Black community,
Latino churches can and do act as political mobilizers, however whether this carries over to the
growing Latino evangelical population is unclear.

14

Latinos, Religion, and voting behavior
On this note, those who have analyzed the role of religion within the Latino community,
have found that there continues to be a significant division in the belief systems of traditionally
Catholic Latinos and the ever-growing Evangelical Latino population (Ellison et al. 2011 &
Kelly and Kelly 2005). Ellison et al. (2011) found that Latinos, are more likely to see church as
an informational source, which translates into these churches having influence on the political
behavior of their congregations. Interestingly, Evangelical Latinos seem to be influenced more
by church attendance than their catholic counterparts (Kenski 1995; Lockerbie 2013), as studies
have found that not only do Protestant/evangelical Latinos differ from Catholics, but they also
have a much greater tendency to hold more conservative beliefs if they actively attend church
services (Ellison et al. 2011 & Bartowski et al. 2012; Chaturvedi 2014; Jones-Correa et al. 2018).
These beliefs include opposition to gay marriage and pre-marital cohabitation, as well as pro-life
abortion stances. While it is not exactly clear if denomination is what leads
2
to these differences
of belief, it has been shown that, unlike Whites, there seems to be a distinguishable difference
when comparing the beliefs of Catholic Latinos to their Evangelical counterparts.
Whether this difference translates into vote choice is much more complex. As it has been
shown that Latino evangelicals do hold more conservative viewpoints on specific social issues
(Chaturvedi 2014; Jones-Correa et al. 2018), such as same-sex marriage. This is seen in
Chaturvedi’s findings which show that identifying as Evangelical makes Latinos significantly
more likely to hold negative attitudes toward same-sex marriage. This finding is less true for
younger individuals and 2
nd
generation and beyond citizens. Jones-Correa (2018) helps give

2
It is unclear if one becomes conservative due to their religious affiliation or if one self-selects into a denomination
based on an already conservative ideology
15

backing to this notion by showing generational status to be a key driver in party affiliation, with
1st generation citizens identifying with the Democratic Party at a 2:1 margin and growing to a
3:1 margin with later generations. It is therefore plausible to assume that these findings may
potentially translate over to help explain vote choice.
When looking directly at the political attitudes of Evangelical Latinos, Janelle Wong
(2018) found that this group is much more conservative than non-Evangelical Latinos, as well as
Evangelical Black Americans. She went on to show that while Evangelical Latinos may be less
conservative than White Evangelicals, this group still voted for Republicans and held
conservative positions at substantial rates (Wong 2018).
Explaining the Latino vote
While some Latinos will choose the GOP candidate, most will still vote for the
Democratic ticket. This should come as no surprise given, unlike White voters, who tend to
follow the Michigan School of voting model (Converse et al. 1960), Black and Latino voters
follow a more Dawsonian model of voting (Dawson 1994). This can be seen when analyzing
Latino voting through a certain lens, which explains that Latinos vote for the Democratic Party
due to their progressive stances on Latino issues (de la Garza and Cortina 2007; Barreto 2007;
Barreto et al. 2010).
However, it must be kept in mind that while most Latinos do utilize a Dawsonian (1994)
approach, it is to a significantly lesser degree than the Black community. This is easily seen
when looking at the difference in Black voters when compared to Latino voters, as over 90% of
Black voters chose Joe Biden while only about 65% of Latinos did the same (UCLA Latino
Policy and Politics Initiative 2021). So, while most Latinos do follow the Dawsonian model, a
significant minority do not. Who is this minority?
16

As mentioned, it is plausible to assume that a growing Evangelical population may help
explain what drives certain Latinos to vote for the GOP. In the next sections of this article, I will
lay out my theory and findings which show that one Evangelical identity is indeed one of the key
predictors of Latino support for Trump. However, I will also go on to show that, like recent
findings from Chaturvedi (2014) show, this is entirely dependent on age group.
Theory
As previously mentioned, Latinos have a very unique relationship with religion, most
notably with the Catholic Church (Barreto et al. 2009). The Evangelical church, however, has
also had a significant impact on Latino public opinion, especially when it comes to social issues
(Chaturvedi 2014; Wong 2018; Sherman 2017); and as seen in the literature review, the power to
influence social policy positions can and does translate into vote choice. Furthermore, in the case
of the 2006 immigration protests, the Catholic Church was forced to prime ethnicity and frame
HR 4437 as a Latino issue in order to have the impact that it did. Whereas simply identifying as
an Evangelical was enough to lead to negative attitudes toward same sex marriage (Chaturvedi et
al. 2014; Wong 2018).
With this said, I argue that the application of a Dawsonian lens to the study of Latino
behavior, is appropriate in many cases, however this is not the case here. Rather, this study must
be rooted in Chaturvedi (2014) and Wong’s (2018) findings, given they best illustrate why one
must take into account Latino Evangelism. Showing that it can ultimately produce more
conservative viewpoints. I argue that these viewpoints, in turn, will cause Latino Evangelicals to
be more likely to vote for Republican Donald Trump. By utilizing Chaturvedi’s framework, I
will show that Latino evangelicals had an increased likelihood to vote for Trump, as he was seen
as the candidate who was more representative of their policy positions. With these positions
17

being much more conservative when compared to other Latinos (Wong 2018). Also, in line with
Chaturvedi’s theory, I argue that the predictive power of identifying as a Latino evangelical will
not hold for those under 45 years of age.
Ultimately, I posit a theory than can be seen in figure 1. This figure shows that most
Latinos will be more inclined to vote for the Democratic candidate as they represent the interests
of Latinos better than the Republican candidate. However, it also shows that Latino Evangelicals
will be more likely to vote for the GOP candidate because they better represent the interests of
Evangelicals. Lastly, one can see that my latter argument is dependent on age, with Latino
Evangelical identity playing no significant role for those under 45 years of age.
Hypotheses:
H1: Latinos were significantly more likely to cast their vote for Joe Biden in the 2020
Presidential Election
H2: Evangelical Latinos were significantly more likely to cast their vote for Donald Trump in the
2020 Presidential Election
H3: Evangelical Latinos under 45 will not be significantly more likely to cast their vote for
Donald Trump in the 2020 Presidential Election
Data
To conduct this study, I will be using data from the 2020 USC Dornsife Presidential Poll.
This poll was done through the USC Dornsife Center for the Political Future, and it was used as
the official poll of the Los Angeles Times during the 2020 presidential election. It is important
to note that this election survey was also part of a larger study called the Understanding America
Study/Survey (UAS), which focused on Covid-19 and other nonpolitical issues.

18

Latino Voters
Prioritizing
Latino Salient
Issues
Identify as
Evangelical
Prioritize
Evangelical
Salient
Issues
Increase
Likelihood to
vote for
Democratic
candidate
Increase
Likelihood to
vote for
Republican
candidate
Figure 1:
Evangelical Latino
voting theory

19

Participants were gathered by inviting roughly 8,000 eligible voters, who were already
active members of the USC Understanding America Study, to participate in an ongoing survey
on the 2020 election. Roughly 96% of those invited agreed to participate every other week on an
assigned day leading to approximately 6,400 respondents in each survey wave, with roughly
500-800 Latinos per sample.
Like many national election polls, this survey was fielded in waves from mid-August to
mid-November, however for this study, I will only be using the post-election poll to garner the
most accurate results regarding vote choice. The post-election poll was an extensive post-mortem
survey meant to understand actual vote choice and the reasons behind said vote choice. I argue
that the polling data is highly appropriate for this study for several reasons. The first being that
this survey has a significant number of Latinos, which allows for generalizable results. This is
key given not many surveys are able to garner a large enough sample size to make generalizable
claims about the US Latino population (Barreto et al. 2018; Frasure-Yokley et al. 2020). Second,
to further correct for any sampling error or generalizability issues, this survey has been weighted
accordingly
3
, to allow for even greater degree of certainty. Lastly, this survey asks specific
questions regarding the influence of religion on political attitudes. This allows me to analyze
how influential being Evangelical truly is when it comes to forming presidential vote choice.
Variables & Measurement
Presidential vote choice
As I mentioned above, I am utilizing a post-election survey which asks respondents
which candidate they cast their vote for in the 2020 Presidential Election. Originally, this
question included all candidates listed on the ballot which included a choice for Joe Biden,

3
Please see article 3 on best practices regarding weighting election data
20

Donald Trump, Howie Hawkins, Joe Jorgensen, and “other”. However, for the sake of this study
I recoded this question to only include those who chose either Donald Trump (coded as “0”) and
Joe Biden (coded as “1”), while dropping the other observations. I did so for three main reasons.
The first being methodological, as less than 2% of Latinos voted for these candidates in total so
including these individuals did not seem necessary. The second being that this study is meant to
understand whether Latino evangelicals were more likely to vote for the GOP candidate,
meaning that there is no need to include third party candidates. The last being that Duverger’s
Law (Riker 1982; Palfrey 1988; Fujiwara 2011) shows us that there is rarely a need to include
third party candidates outside of certain studies. With this in mind, I utilize the recoded vote
choice variable as the dependent variable in several regressions that will follow in the methods
section.
Evangelical
Like many political surveys, the USC Dornsife Presidential Election Poll asked several
questions regarding demographics, two of which happened to be about religious affiliation. One
asked respondents to report their religious affiliation and listed two different choices for
Christianity- Catholic and Protestant (USC Dornsife Presidential Poll 2020). The other asked if
respondents identified as “Evangelical or Born Again”, which has been coded as a binary (0/1 for
no/yes).
I argue that the latter question will be more appropriate for this study, as I am not
focusing on Protestant Latinos as a whole, but specifically on those who identify as Evangelical.
Furthermore, the binary measurement used for the Evangelical question allows one to clearly
observe the relationship between identifying as Evangelical and vote choice (Chaturvedi 2014).
Party Affiliation
21

While the focus of this study is to analyze the role that identifying as Evangelical plays in
presidential vote choice, it is always important to account for party identification for several
reasons. Converse et al. (1960) showed us that one of the strongest predictors of vote choice is
party affiliation. This has shown to hold up over time, and research has even shown that it is not
only party registration status that can predict vote choice, but party affiliation (Barnes et al.
1988; Petrocik 2009; Gerber et al. 2010). This is different than registration status, as it asks
respondents to report the party that they feel closest to, rather than asking which party they
belong to (Greene 1999; Petrocik 2010). It is for these reasons that I include party affiliation in
my logit models, and I expect to be rather strong predictor. It is measured as a binary with those
who reported an affiliation with the GOP coded as 0, and those who affiliate with the Democratic
Party being coded as 1. Unlike Wong’s (2018) prior findings, I expect party to be rather
impactful.
Immigration Status
When it comes to Latino politics it is important to account for immigration status, as
roughly 80% of Latinos are either first, second, or third generation immigrants (USC Dornsife
Poll 2020). This common bond amongst Latinos is also a determinant of vote choice, as it has
been demonstrated by Jones-Correa et al. (2018) that Latinos who are first generation immigrants
significantly differ from other generations when it comes to voting, with there being only a 2-1
likelihood of voting for the Democratic Party. In turn, 2
nd
generation and beyond are more likely
to vote for Democrats at a 3-1 margin. Therefore, I have included a 4-point measure of
immigration status, which begins at 1 for first generation immigrants and ends at 4 for fourth
generation and beyond/non-immigrant. This is ideal as it allows me to account for the unique
experiences of 2
nd
and 3
rd
generation immigrants, whereas a binary measurement would not
22

accurately capture the differences between 2
nd
generation Latino immigrants and those whose
ancestors resided in the southwest when it was ceded to the US via the Treaty of Guadalupe
Hidalgo (Castillo 1992).
Control Variables- Education, Age, and Gender
Aside from my primary independent variable, I have included several control variables,
one of which is education. It is key to include education as it has been found to be a major
predictor in voting behavior, with education being shown to make individuals more politically
sophisticated which translates into a greater likelihood to vote (Luskin 1990; Sondheimer &
Green 2010; Dun & Jesse 2020). While little relationship has been shown between education and
vote choice it is best to include to avoid omitted variable bias (Clarke 2005; Gschwend &
Schimmelfennig 2007). Education is measured on a 14-point scale where respondents select the
highest level of education they have attained, with the lowest value being 5
th
/6
th
grade
completion and the highest being doctoral degree.
The other demographic control variables that I have included in my models are age and
gender. These are not new to the study of political science (Pollock 2003), as these variables are
consistently used as a means to ensure that the primary independent variable, in this case
identifying as Evangelical, truly holds explanatory power over the dependent variable. When it
comes to age, I am interested in looking at the difference between two key groups- those who are
45 and under, and those over 45. I argue that by using a binary that is coded as 0 for 45 and
under, and 1 for over 45 will allow us to see the potential divide between generations more
clearly. Gender is also coded as a binary with women coded 0, and men coded as 1.
4

4
The author recognizes the issues that come with coding gender as a binary, but for the sake of this study it would
not be feasible to do otherwise until a better measure is used in presidential polling
23

Methodology
Using these aforementioned variables, I have created several binary logistic regression
models in STATA to analyze how identifying as Evangelical may impact vote choice. Before I
get to these models, I have provided a breakdown of Latino vote choice in the 2020 election. As
one can see, roughly 65% of Latino voters cast their ballot for Joe Biden, while about 32% voted
for Donald Trump. This shows a rather large majority of Latinos supported Biden, while a
sizable minority went for Trump. To understand the impact of being Latino on vote choice, I
have gone on to provide a model that includes all respondents who voted in the 2020 presidential
election, and it includes Latino as a variable to show that being Latino was a significant predictor
of casting a vote for Joe Biden. One can see that this is the only model that excludes the
Evangelical variable, as this initial regression is solely meant to show that on the surface, simply
being Latino will cause one to be significantly more likely to vote for Biden.

Table 1

Table 2
2020 PRESIDENTIAL VOTE
JOE BIDEN 51%
DONALD TRUMP 47%
2020 LATINO VOTE
JOE BIDEN ~ 65%
DONALD TRUMP ~ 32%
24

5 Regression showing Latino as an independent variable and vote choice as dependent variable, includes all voters

Regression Showing Latino as a Variable
5

Coefficient Standard Error t P > t
Evangelical -.739 .253 -2.92 0.003
Education .234 .061 3.84 0.000
Party Affiliation 7.206 .287 25.12 0.000
Immigrant
Generation -.136 .105 -1.30 0.194
Latino 1.004 .368 2.73 0.006
Constant -5.635 .771 -7.31 0.000
Model 1
25

Next, I have created several binary logistic regressions which only include Latino
respondents and are weighted for accuracy and generalizability. Model 2 uses the mentioned
variables to uncover whether Evangelicals were more likely to vote for Trump. As one can see,
those who identify as evangelical were significantly more likely to vote for Donald Trump, while
Democrats and those under 45 were significantly more likely to vote for Joe Biden.
Unsurprisingly, party affiliation yields the most explanatory power with a beta coefficient of
11.8, while identifying as Evangelical came in second with a beta coefficient of -4.1. Higher
educated individuals were also more likely to vote for Biden, but with at a smaller magnitude
with a beta coefficient of 0.5.
Interaction terms and models
While Model 2 does show that Evangelicals are significantly more likely to vote for
Donald Trump, I wanted to be sure of this connection. Furthermore, I wanted to see if this holds
for Latinos under the age of 45. To test this, I have created a third model (Model 3) that uses
interaction terms to assess whether the interaction of Evangelical ID and age yields significant
results. While one might ask why I did not simply subset the data to those under 45, I argue that
this is not feasible, as the interaction term more accurately captures how age may mitigate the
impact of identifying as Evangelical (Brambor et al. 2005). I have also included a different term
to test how the interaction of gender and Evangelical identity may play a role in Latino vote
choice (Monforti 2017; Junn & Masuoka 2020).

[SEE NEXT PAGE ]

26

Effects of Being Evangelical on Latino Vote Choice
Coefficient Standard Error t P > t
Evangelical -.831 .243 -3.41 0.001**
Age .768 .250 3.07 0.002**
Education .134 .052 2.58 0.010*
Party Affiliation 6.936 .261 26.61 0.000***
Immigrant
Generation -.0140 .122 -0.11 0.909
Gender -.002 .234 -0.01 0.995
Constant -4.754 .789 -6.03 0.000***
Model 2
27

Update to Include Interaction Terms
Coefficient Standard Error t P > t
Evangelical -4.096 1.575 2.60 0.009**
Age 4.193 1.781 2.35 0.019*
Evangelical*
Age -.895 2.372 0.38 0.706
Education .482 .209 2.31 0.021*
Party Affiliation 11.902 2.232 5.33 0.000***
Immigrant
Generation .439 .341 1.29 0.197
Gender .948 1.094 0.87 0.386
Gender*
Evangelical 1.060 2.277 0.47 0.642
Constant -12.714 3.479 3.65 0.000***
Model 3
28

Looking at the General Population (Without Interactions)
Coefficient Standard Error t P > t
Evangelical -.831 .243 -3.41 0.001**
Age .768 .250 3.07 0.002**
Education .134 .052 2.58 0.010*
Party Affiliation 6.936 .261 26.61 0.000***
Immigrant
Generation -.0140 .122 -0.11 0.909
Gender -.0012 .234 -0.01 0.995
Constant -4.754 .789 -6.03 0.000***
Model 4
29

Looking at the General Population (With Interactions)
Coefficient Standard Error t P > t
Evangelical -.585 .398 -1.47 0.142
Age .770 .316 2.44 0.015*
Evangelical*Age .037 .487 0.08 0.940
Education .134 .052 2.56 0.010*
Party Affiliation 6.944 .262 26.55 0.000***
Immigration Gen -.014 .123 -0.11 0.909
Gender .209 .306 0.68 0.495
Gender*Evangelical -.514 .475 -1.08 0.279
Constant -4.868 .804 -6.05 0.000
Model 5
30

Even with the interaction terms there is still a significant relationship between identifying
as Evangelical and vote choice, as well as party affiliation and vote choice. Interestingly, there is
not a significant relationship between vote choice and the interaction of Evangelical ID and age.
This indicates that Evangelical Latinos under 45 are not as likely to vote for Trump as those who
are over 45. It is also important to note that in this model, education is no longer significant,
which is not surprising (Nuno 2007). In the following section I will break down these results and
explain their implications for further studies that focus on Latino vote choice.
Comparing the Latino Community and General Population
What also stands out is the comparison between the Latino community and the general
public, as these models show that being Evangelical was actually more influential amongst
Latinos. This is true in both the interaction and non-interaction models, which further shows that
Latino Evangelicals are more impacted by their religious affiliation than the average American
(Barreto et al. 2010). This is a unique finding because while Latinos are known to be more
religious than other groups, this was always attributed to high levels of Catholicism.
Findings & Implications
As one can see in Model 2, aside from party affiliation, identifying as Evangelical was a
key predictor of Latino vote choice, with Latino Evangelicals being significantly more likely to
cast their vote for Trump. Further analyzation went on to show that the interaction of age and
identifying as Evangelical did not yield significant results. What the latter shows is that while
Latino Evangelicals were significantly more likely to vote for Trump, this did not hold for
Latinos evangelicals under 45. These results fall in line with Chaturvedi’s (2014) previous
findings which demonstrated that while Evangelical Latinos do hold more conservative beliefs,
31

this is not necessarily the case with younger generations. This also is not surprising, as younger
cohorts tend to be more progressive than those that preceded them (Hooghe 2004; Dalton 2015).
As the models show, the evangelical variable is statistically significant, but what also
stands out is the beta coefficient of 4.1, which shows that the magnitude of this effect is rather
large (Peterson & Brown 2005). Even when accounting for the interaction of age and
Evangelical identity, the lone “Evangelical” variable remained the second strongest predictor of
Latino vote choice, behind party affiliation. This further shows how important it is to the study of
Latino politics to account for effects caused by religion.
In this vein, these results show that political scientists should not assume that religious
homogeneity exists within the Latino community. While Latinos are commonly seen as a
monolith (Schmidt et al. 2000; Barreto et al. 2010; McKenzie & Rouse 2012), the increasing
number of Latino evangelicals means that we in the academy must now correctly account for
religious variation. It should also be evident that future studies in Latino voting behavior cannot
simply account for traditional religious denomination, rather it is key to include a standalone
question on Evangelical identity. This will allow researchers to more accurately capture the
unique behavior of Evangelicals in US politics (Chaturvedi 2014). Asking this standalone
question avoids conflating certain Protestant religions with Evangelicals.
Furthermore, these findings show that, like Wong (2018) shows, Latino Evangelicals are
more likely than other Evangelicals to support Trump. This means that not only is it time to
study the effects of religion on the Latino community, but that Gorski (2017) and Whitehead et
al. (2018) findings which show much of Trump’s Evangelical support coming from White
Christian nationalists may need to be adjusted. Rather it appears that Latino Evangelicals are
32

more likely to support Trump, but instead of voting on the basis of nationalism, Latinos voted to
attain policies that are more in line with Evangelical ideals.
Conclusion
The overall goal of this study was to test President Obama’s (2020) claim that Latino
support for Trump was driven by Evangelicals. Ultimately, I found that while the former
president was fairly accurate in his assumption, there is a degree of nuance to it. A more correct
version of this statement would exclude Evangelical Latinos under 45, who seemed to have a
spurious relationship with vote choice.
Aside from testing President Obama’s statement, I used this study to help open the door
to studying the impact of religion on Latino political behavior. Many prior studies do not take
religion into account when studying Latino politics due to the stereotype that all Latinos are
Catholic (Schmidt et al. 2000; Barreto et al. 2010; McKenzie & Rouse 2012). The notion that
Latinos are a religious monolith has caused researchers to conclude that religion will simply
yield no effect. However, the past years have seen the Latino evangelical population rise
dramatically (Ellison et al. 2011; Reyes-Barrientez 2019), which has at the same time caused a
gap in literature that fails to account for this increase. Along with a few other recent studies
(Chaturvedi 2014; Wong 2018; Sherkat 2017), this article has helped reconcile this issue by
showing Evangelical Latinos, specifically older ones, do behave rather differently than their non-
evangelical coethnics.
Lastly, it is not enough to account for religion only in the study of Latino voting
behavior, rather it must be included when studying Latino political behavior in general. With the
rise in the Latino evangelical population, researchers must consider the role it plays in voting,
33

protest participation, ideology, party ID formation, etc. This will be important going forward if
we would like to truly understand the political behavior of Latino Americans.

34

Chapter 3:
What Causes Latinos to Believe in Conspiracy Theories?

35

There were two key topics in the 2020 presidential election, the first being Covid-19,
while the second was fear of widespread election fraud. While we in the discipline of political
science are well aware that widespread voter fraud is a myth (Minnite 2011; Smith 2017;
Edelson et al. 2017; Holman & Lay 2019), then President Donald Trump did all in his power to
instill such a fear in the general public, both before and after losing the presidency to Joe Biden.
This was not the first time that Trump had cried election fraud, as was the case back in 2016
when he claimed that he only lost the popular vote due to millions of undocumented individuals
voting in California (Smith 2017; Cotrell et al. 2017). This time around he utilized the other key
topic, the Covid-19 pandemic, to cry fraud by constantly arguing that widespread vote-by-mail
was unconstitutional and would lead to millions of illegal votes (James & Clark 2020;
Pennycook & Rand 2021). Ironically enough, this was not the first Covid conspiracy theory that
Trump had endorsed (Dyer 2020a; Uscinski et al. 2020). Months prior, in late Spring 2020,
Trump began to posit the notion that Covid-19 was not as bad as medical professionals were
making it seem, and that it was not a real threat to public health. Aside from downplaying the
severity of Covid, Trump consistently appeared to question the science behind mask wearing
(Greene et al. 2022; Young et al. 2022). Despite being wrong on all of these issues
6
, many
followed Trump’s lead and adopted these conspiratorial beliefs, including many Latinos.
While studying conspiracy theories and their impact on politics is nothing new to the
discipline (Oliver & Wood 2014; Douglas et al. 2017; Radnitz et al. 2020), rarely does one come
across research analyzing the conspiracy theories from strictly from an REP standpoint (Parsons
et al. 1999; Simmons & Parsons 2005; Cortina & Rottinghaus 2022). The current literature on

6
Vote-by-mail was ruled constitutional with no evidence being found for widespread fraud (Texas v Pennsylvania
2021)
36

conspiracy theories has produced relatively limited findings which show Latino and Black
Americans are inclined to believe conspiracy theories that paint the government in a negative
light (Parsons et al. 1998; Miller et al. 2016).
7
Not much else is known about the relationship
between Latinos and CTs because studies that are focused solely on this topic are nearly
nonexistent.
I argue that now is the perfect time to change this. To do so, I ask, “what caused Latinos
to be susceptible to adopting conspiracy theories regarding Covid-19 and the 2020 Election?” In
this article, I use the 2020 USC Dornsife Election Post-election Poll to answer this question by
first showing that Latino Democrats were less likely to subscribe to conspiracy theories. I will go
on to show that Latinos with an external locus of control are significantly more likely to adopt
conspiracy theories surrounding Covid/election fraud. This article will then be followed by a
third and final analyzation that focuses on best practices in calculating the Latino vote.
Literature Review
The Adoption of Conspiratorial Thinking
To understand why individuals believe in contemporary Covid-19/Election fraud related
CTs, I must step back to explain how and why individuals adopt conspiratorial thinking in
general. One of the earlier explanations as to why this phenomenon occurs argues that this belief
stems from the psychological yearn for there to be simple answers that explain complex events
(Hofstadter 1965). Goertzel (1994) added to this theory when he found that there were several
other common factors that make one more likely believe in CTs. The first being a lack of trust in
other people, which goes hand-in-hand with the second factor- high levels of anomie, with this

7
Examples include believing the government was involved in the MLK assassination, believing that police racially
profile, and that the criminal justice system is unfair to people of color
37

term being defined as the “belief that the situation of the average person is getting worse”
(Goertzel 1994, Oliver & Wood 2014; Wood 2016; McCarthy et al. 2021; Casara et al. 2022).
These latter two arguments point the notion that individuals who feel somewhat alienated from
society are highly prone to believing in CTs. Furthermore, these individuals feel that the system
is rigged against them and yearn for simple answers to explain their situations (Hofstader 1965;
Goertzel 1994; Marchlewska et al. 2018; Smallpage et al. 2020).
As more scholars began to study CTs, the role that anomie and alienation play in their
adoption became more defined, as it was shown that not only do these feelings lead to a belief in
CT’s, but they are actually more significant than Hofstader’s (1965) original finding regarding
the yearn for simple answers to complex situations (Abalakina-Papp et al. 1999). Moreover,
results showed susceptibility to CTs comes from a lack of trust, anomia, and an external locus of
control (Abalamkina-Papp et al. 1999). These all were found to be drivers in not just subscribing
to specific CTs, but also in the likelihood to subscribe to CTs in general.
When taking political ideology into account, it becomes rather clear that ideology matters
to a significant degree, with political conservatives having a higher likelihood of believing in
CTs when compared to political progressives (Barberá et al., 2015; Jost et al., 2018; Stern et al.,
2014; van der Linden et al. 2021). The explanation as to why is two pronged, with the first being
the GOP politicization of scientific issues (Bolsen & Druckman 2018; Chinn et al. 2020; Halpern
2020; Bolsen & Thorton 2021). This is a result of scientific issues being transformed into wedge
issues during election years (Dunlap & McCright 2008; Kintisch 2013; Helmuth et al. 2016). The
second explanation being that “conservatism is often associated with stronger in-group echo
chamber effects that may more easily breed conspiratorial thinking” (Min 2021). This is why
Oliver and Wood (2014) found that conservative men and women are the most susceptible to
38

modern CTs, and why they can be described as monological. This meaning that strong in-group
echo chamber allows for conservatives to find the logic for one CT embedded within another CT.
In the following section I will discuss why this is the case and show that CTs are not simply a
conservative phenomenon.
Racial & Ethnic Minorities & conspiracy theories
Since the 1960s it has been shown that some of the highest levels of conspiratorial
thinking is amongst racial minorities, especially within the Black and Latino communities
(Hofstadder 1965; Goertzel 1994; Abalakina-Papp et al. 1999). The tendency to believe in CTs is
very high amongst these groups (Hofstadder 1965; Goertzel 1994) for several reasons. The first
being that Black and Latino Americans are more likely to have higher levels of anomie and an
external locus of control due to a history of discrimination which only helps to perpetuate these
feelings. Furthermore, Black Americans, are more inclined to believe in CTs because they have
suffered harshly at the hands of the US government with examples ranging from structural
oppression to medical experimentation, which have, in the past, been relegated to conspiracy
theory status. One example of such is being the Tuskegee Syphilis Experiments (Brandt 1978;
Gray 1998), which led to the death of over 100 Black men and countless cases of preventable
Syphilis in Black women and children (King 1992; Gorbie-Smith 1999; Reverby 2009). This
gruesome history has caused many in the Black community to have higher rates of belief in
conspiracy theories which paint the US in a negative light.
Unfortunately, there is not as much knowledge on why Latinos are likely to believe in
CTs. Most studies that do offer any insight in this area do not go beyond including “Hispanic” or
“Latino” as a variable on a regression/correlation table (Goertzel 1994; Abalakina-Papp et al.
1999). While there are numerous studies that illustrate the relationship between Black Americans
39

and CTs, studies on Latinos and political conspiracy theories are nearly nonexistent, however it
is appropriate to draw upon the literature on Black American’s and CTs to get an understanding
of the why Latinos are susceptible. I argue it is appropriate when looking at the history between
Latinos and the US
8
, as this history also includes countless acts of brutality by the dominant
Anglo population toward Latinos (Shaw et al. 2018). Numerous instances illustrate this points,
such as the US defaulting on the Treaty of Guadalupe Hidalgo, countless Latino lynchings
9

(Delgado 2009) in the Southwest, the repatriation movement (Balderrama & Rodriguez 2008),
“Operation Wetback”, and the “Sensenbrenner Bill” (Barreto et al. 2010). So, it is plausible to
assume that this mistreatment helped contribute to higher levels of CT belief amongst Latinos
that has been found since the 1990s (Goertzel 1994).
Latinos & Misinformation
Further insight on contemporary CTs is also provided by looking into misinformation
within the Latino community (Post 1990; Shrestha 2018; Austin et al. 2021). While not the same
as belief in a specific CT, these studies help show us that the Latino community is susceptible to
conspiratorial thinking. While this thinking may be caused by misinformation campaigns, it still
demonstrates the high levels of deviation from factual information.
History shows that misinformation and its relevance in the Latino community is not a
new occurrence. Rather, it dates back to the 1990s (Post 1990) when it was shown that the Latino
community in the United States had been misinformed about the price of college tuition. Many
Spanish speaking Latino homes had dramatically overestimated the cost of college tuition at the

8
For the sake of this paper “Latinos” describes individuals of Latin American descent who reside within the United
States
9
It should be noted that the term “lynching” means that there was little to no effort on the side of law
enforcement/government officials to stop or even investigate these murders. The only efforts to stop these events
came from the Mexican Government, who asked the US to step in to stop these events.
40

community college and university level, which eventually led to a downward trend in Latino
college enrollment. Fast forward 28 years to the 2018 midterm election, where we saw Latino
communities become the target of online misinformation campaigns, which tried to convince
voters that President Trump was supported by a majority of Latinos (Flores-Saviaga & Savage
2019) in an attempt to portray Trump as the better choice for Latinos.
Situations such as this have become increasingly common within the Latino community,
with misinformation being continuously spread during the 2020 election (Seitz & Weisser 2021).
In one instance Facebook ads targeted Florida and Texas voters with claims that Joe Biden was a
bonafide communist. Latinos in Florida were even targeted by ads which likened him to
Venezuela’s socialist President Nicolas Maduro (Seitz & Weitzer 2021; Busby 2022; Foreman
2022), in an attempt to sway voters of Venezuelan decent
10
. While none of this was true, as Joe
Biden was a self-professed capitalist, this was done in attempt to sway Latino voters in two
potential swing states (Seitz & Weitzer 2021). These ads significantly help spread
misinformation at high numbers within the Latino community, as Nielsen (2021) found that
Latinos are more likely than any other demographic group to consume and share political
misinformation. Cortina & Rottinghaus (2022) go onto show that this misinformation, for Texas
Latinos, gains traction the most amongst Trump supporters and those who rely on Spanish
media.
Conspiracy theories and the Covid 19 Pandemic/2020 Election
The 2020 election saw an unprecedented number of claims arguing that widespread voter
fraud was the reason Joe Biden emerged victorious (Eggers et al. 2021; Pennycook & Rand
2021). What made this situation especially unique, is this conspiracy theory was started by sitting

10
Florida is the home to the largest Venezuelan population in the US
41

president, Donald Trump, which in turned caused for upwards of 25% (USC Poll) of the
population to believe that it must be true. Trump made this argument prior to Election Day 2020
after the majority of polls predicted he would lose to Joe Biden. To combat this narrative, he
argued that widespread fraud would be the only way he could lose. While political scientists and
election officials have consistently shown that there is no evidence of such fraud (Berlinski et al.
2021; Eggers et al. 2021; Pennycook & Rand 2021), many still clung to this belief. Like prior
findings have shown, levels of belief were highest amongst White conservatives who would not
have otherwise believed in such a claim. However, this was also the case amongst Latinos who
relied on Spanish language media (Cortina & Rottinghaus 2022)
Latinos and Covid-19
It is also very important to recognize that non-election related CTs related to Covid-19
were also spreading throughout the US (Bierwiaczonek et al. 2020; Miller 2020; Romer &
Jamieson 2020a; Romer& Jamieson 2020b; Hansen et al. 2021). These theories ranged from
falsely believing in certain medications to the belief that Covid was no more severe than a cold
or flu (Jamieson et al. 2020; Prooijen et al. 2021). With this said, one must also account for these
health-related conspiracy theories. It is especially important to include these CTs due to the
disproportionate impact that the Covid-19 had on the Latino community (Podwelis et al. 2020;
Vargas & Sanchez 2020).
When compared to the general population, the Covid-19 Pandemic hit the Black and
Latino populations the hardest. Recent studies on this topic have shown that a much larger
percentage of the Latino community (Rodriguez-Diaz et al. 2020)
11
tested positive for Covid.

11
Case counts amongst the US population averaged 82 positive cases per 100,000, while the Latino community
averaged 91 cases per 100,000
42

Furthermore, it was found that just being Latino increased one’s likelihood of catching and/or
dying from Covid-19 (Podwelis et al. 2020; Vargas & Sanchez 2020; Mude et al. 2021). The
reasons for this include high density housing situations, higher rates of essential workers, and a
larger number of monolingual (non-English speaking) households.
The disproportionate impact that Covid had on Latinos is best illustrated by looking at the
non-traditional destination of North Carolina (Hendrix 2021), which saw Latinos account for
approximately 40% of all positive Covid cases, while only making up about 10% of the NC
population. Researchers are quick to point out that a majority of these cases are amongst North
Carolina born Latinos, who are fluent in English (Hendrix 2021; Sena & Weber 2021). Rather it
was found that Latinos, especially those in nontraditional destinations (Harris & Feldmeyer
2013), are much more likely to be classified as essential workers (Rodriguez-Diaz et al. 2020;
Hendrix 2021; Sena & Weber 2021). Furthermore, Latinos are more likely to live below the
poverty line which leads to many Latinos living in high density housing situations (Cohen &
Casper 2002; Mendez 2005; Hendrix 2021) and rely more on public transit. The combination of
these factors makes social distancing a challenge, the lack of which directly leads to an increased
likelihood of catching Covid-19. Latinos also have a tendency to believe that the cost of US
healthcare outweighs the care that is provided. This is a direct result of many healthcare
institutions inaction when it comes to Latino outreach (Hendrix 2021).
Numerous medical journal articles have shown that not only were Latinos more likely to
catch Covid-19, but those who had been hospitalized reported high levels of Covid
misinformation consumption (Cervantes et al. 2021; Ornelas & Ogedegbe 2021). Much of this
misinformation led many Latinos to have inaccurate knowledge regarding Covid transmissibility,
prevention, and severity (Cervantes et al. 2021; Ornelas & Ogedegbe 2021). Latinos were also
43

highly susceptible to Covid misinformation due to their limited access to health care information
(American Medical Association 2020). This lack of access to information is a result of Latinos
being the least insured racial/ethnic group in the US.

Figure 2: Covid Impact by Race/Ethnicity (CDC 2021)
Theory
When looking at past literature, several recurring themes stick out that allows me present
a theoretical framework which explains why Latinos may be more susceptible to Covid-
19/election CTs. The first being that unlike in my prior article (Cuellar 2022a), identifying as
Evangelical will not yield significant predictive power when it comes to believing in
Covid/election CTs. Rather, I argue that party affiliation (Barnes et al. 1988; Petrocik 2009;
Gerber et al. 2010; Barberá et al., 2015; Jost et al., 2018; van der Linden et al. 2021) will yield
significant results, with Latino Democrats being the less likely to subscribe to these CTs. This
stems from Trump’s politicization of Covid-19 and GOP backing of election fraud CTs, which
causes partisan affiliation to become highly influential. Furthermore, I posit that party affiliation
44

Affiliated with
Republican
Party
External
locus of
control
Increase
Likelihood to
believe in
contemporary
CTs
Latinos
Figure 3: What Causes Latinos to Believe
in Conspiracy Theories

45

is better than ideology or candidate preference when studying Latinos, given Latinos tend to hold
conservative viewpoints, but mainly identify with the Democrats (De la Garza 2007).
Unlike Cortina & Rottinghaus (2022), I theorize that Latinos with an external locus of
control will be more inclined to believe in Covid/election CTs. With Individuals with an external
locus of control being those who believe that most of their life experiences are out of their
control (Abalakina-Papp 1999; Uscinski & Parent 2014), and overall feel like they have very
little influence. This falls in line with previous findings that show individuals who feel this way
are significantly more likely to believe in CTs in general.
12

Data
To test this theory, I have used data from the 2020 USC Dornsife Presidential Election
Poll, which was carried out by the USC Dornsife Center for the Political Future from mid-
August to the days that immediately followed the Election. Like my previous article (Cuellar
2022a), this study solely relies on data from the post-election portion of the poll, which was
fielded after Election Day, as this specific survey asks the most questions on attitudes toward
CTs regarding Covid and election fraud. This survey is also ideal given it contains questions that
can appropriately measure locus of control.
In total I focus on five specific CTs, with three focusing on Covid-19 and two focusing
on voter fraud. I use these conspiracy theories as dependent variables in five separate regression
models. In the following section I will describe each CT in depth and provide a brief description
of my independent variables. For this study I use a 5 point scale to code all dependent variables

12
I want to point out that this theory does not apply to all CTs, as we have now entered a new era where factually
inaccurate CTs are endorsed by the Republican Party, and prominent political figures within its ranks. This is what
causes partisan affiliation to become the greatest predictor of Latino belief in CTs, as those who identify as
Republicans are naturally going to be more likely to endorse Covid/election conspiracy theories. See figure 1
46

with the questions asking how strongly respondents agree or disagree with certain statements
regarding CTs.. Answers range from “strongly disagree” (1) to “strongly agree” with the option
for “neither” being coded as 3.
Variables and Measurements
Hydroxychloroquine Conspiracy Theory
Due to the lack of knowledge regarding Covid-19, many individuals turned to unproven
antiviral/de-wormer medications to potentially treat the effects of Covid (Bertin et al. 2020;
Wertheimer 2022). Most recently, we have heard many tout the “efficiency” of Ivermectin, a
horse de-wormer, when it comes to treating Covid-19 (New York Times 2021). This comes
despite medical professionals, the FDA, and the CDC continuously showing that Ivermectin is in
no way effective at treating Covid (Schmith et al. 2020). Interestingly enough, this is not the first
time that individuals have turned to a drug that was never meant to combat Covid.
In the early days of the pandemic, individuals turned to the anti-malarial drug
Hydroxychloroquine, which like Ivermectin, has been found to be ineffective at treating the
Covid-19 virus (Bertin et al. 2020; Friedman 2021). However, given the notion that it could
potentially lead to the end of the pandemic, many clung onto the idea that Hydroxychloroquine
was an effective treatment. One of these individuals was President Donald Trump, who used the
bully pulpit to spread this theory, despite evidence showing it to be false. The reason I categorize
the “belief that Hydroxychloroquine is an effective treatment for Covid” as a conspiracy theory
is simple- the general public were warned of the ineffectiveness and danger of treating Covid
with this anti-malarial medication as early as May of 2020 with it being condemned as a
treatment in in Summer 2020 (Bertin et al. 2020; Friedman 2021). This means that by November
47

of 2020, it was very clear that there was no scientific evidence showing Hydroxychloroquine to
be effective.
Mask Wearing
In April of 2020 the CDC recommended that the general public wear masks or face
coverings when leaving their homes to slow the spread of Coronavirus. From the beginning the
CDC was adamant that mask wearing slowed the spread of Covid but did not necessarily protect
the wearer (Ippolito et al. 2020; Kahler & Hain 2020). Immediately after this recommendation
President Trump stated that mask wearing remained optional and that he would not be wearing
one himself (Greene et. al. 2022; Young et al. 2022). Despite these remarks, many states and
localities imposed mask mandates to slow the spread. However, due to Trump’s defiance
(DeMora et al. 2021; Kahane 2021), the issue became politicized and many failed to believe that
mask wearing was in anyway beneficial. It is for this reason that I use a five-point scale to
measure the degree to which one believes that mask wearing is effective at preventing the spread
of Covid.
13
I argue that this can be classified as a CT because the medical community very
clearly articulated the benefits of mask wearing for eight months prior to the 2020 Election.
Politicization of the topic also contributed to classifying this as a CT.
Covid is No Worse than the Flue
The final Covid conspiracy theory that I am analyzing is the belief that Covid is no worse
than the seasonal flu or common cold. This was a theory that spread due to many Covid patients
reporting they experienced flu like symptoms such as a fever, sore throat, and body aches (Jang
& Jung 2021). A high recovery rate amongst Covid patients also helped to spread this theory.

13 It is necessary to point out that this variable is specified to only ask about mask wearing being effective at
slowing the spread of Covid. I do not conflate slowing the spread with personal protection, as the question I utilize
does not ask if masks prevent Covid-19. Rather, it specifically asks if masks are effective at protecting others.

48

However, medical professionals continuously combatted this argument by constantly providing
evidence showing Covid to be very different than the flu. One key piece of evidence being that
the seasonal flu does not cause a nationwide healthcare crisis that results in a shortage of
ventilator machines and ICUs being at capacity (Alban Pascal et al. 2020; Litton et al. 2021).
Furthermore, when comparing mortality rates-an average of 12,000 flu related deaths occurred
annually between 2010 and 2019, while over 350,000 died due to Covid-19 in 2020. By
November 2020 it was very obvious that Covid was more serious than the seasonal flu. Thus,
making the belief that Covid is no worse than the flu nothing more than another conspiracy
theory.
Deep-state Tried to Remove Trump from Office
When it comes to election related CTs I opted to include one theory that fall more in line
with traditional CTs than the Covid specific ones. The first being the belief that the “Deep-state”
was behind a plot to remove Trump from Office (Amarasingam &Argentino 2020; Rosenblum &
Muirhead 2021). This stems from the first Trump impeachment that ended just before the
pandemic in February 2020, after which many conservatives argued that a cabal of liberal elites
were behind the impeachment as a way to take the presidency away from Trump. As Election
Day approached, this theory was renewed as one of the potential reasons why Trump would lose
the presidency, with many conservatives believing that a Trump loss would be due to this “Deep-
state”. While this theory was not as heavily endorsed by Trump, it was also never condemned or
corrected by the sitting president, which allowed it to gain even more traction amongst the
public. Again, this is measured on a five-point scale with the options ranging from strongly
disagree to strongly agree.

49

Vote-by-Mail Will Cause Widespread Voter Fraud
The last conspiracy theory that I will be looking at was the belief that policies making it
easier to vote-by-mail would lead to widespread voter fraud (Pennycook & Rand 2021; Eggers et
al. 2021). This CT came about because, as previously mentioned, Covid cases began to rise as
Election Day approached which caused many states to implement policies that made voting by
mail significantly easier (Bonica et al. 2021). Given pre-election polls consistently showed that
Joe Biden leading the race for President, Trump began to argue that if he were to lose the
election then it would be because wide-scale mail-in-voting lead to voter fraud (Pennycook &
Rand 2021). While this was in no way true, many voters believed Trump’s argument. Given the
complete lack of evidence and consistent findings which show the absence of widespread fraud,
it is safe to classify the belief the vote-by-mail led to election fraud as a CT.
14

Party Affiliation, Locus of Control, and other Independent Variables
Independent variables for this study include party affiliation, locus of control,
Evangelical ID, age, education level, and gender. I use traditional measures for all variables
except two-party affiliation and locus of control. When it comes to party, rather than only
including party registration, I have measured “party affiliation” as a binary variable in which
respondents were forced to choose which party they felt closest to- Democrat or Republican.
This allows me to properly measure the effect that party has on Latinos and CTs, as ideology
would not be appropriate.
To measure locus of control, I use a question which asks individuals whether they
“disagree”, “neither agree/disagree”, or “agree” with the following statement: “people like me
don’t have much say in government”. Those who chose “agree” are coded as having an external

14
Also measured using a five-point scale that ranges from “strongly disagree” to “strongly agree”
50

locus of control, as these individuals believe that, despite the ability to civically participate, they
are still powerless over what happens in government (Abalakina-Papp et al. 2002; Parsons 2010).
Methodology
Many studies on conspiracy theories use correlation analysis to find a connection
between groups and CTs (Hofstader 1965; Goertzel 1994; Oliver & Wood 2014). While
significant correlation results may show a connection between certain groups and CTs, there is
no ability to know if your variables actually influence one another. Rather than using correlation
analysis, I have run several regressions which should clearly show what drives Latinos to believe
in Covid/election CTs. As mentioned, I use data from the 2020 USC Presidential Post-election
Poll that has been weighted according to the results of the presidential election for the most
accurate results.
In total, I ran 5 OLS regressions which each look at a different CT. Each model uses a
specific CT as the dependent variable and party affiliation, Evangelical ID, anomia, education
level, gender, and age as independent variables. The first three (Model 1) look at the Covid
specific CTs while the last two focus on the election ones. This allows me to parse out the
potential differences amongst Covid and election CTs, given predictors of believing in Covid
theories may be different than those pertaining to the election
Below I have presented all the Covid related CT results as a way to show how each CT
compares to one another. The biggest thing that stands out is how much party affiliation
influences the belief in Covid-19 conspiracy theories, which is most likely a result of Trump’s
endorsement of these CTs (Uscinski et al. 2020; Kahane 2021; Greene et al. 2022). Aside from
party affiliation, age and education level seem to significantly affect the degree to which Latinos
believe in Covid-19 CT. Moreover, in line with prior findings, an external locus of control yields
51

Belief in Hydroxychloraquine
Coefficient Standard Error t P > t
Evangelical .119 .100 1.18 0.238
Age -.135 .086 -1.57 0.117
Education -.038 .020 -1.92 0.056
Party Affiliation -1.343 .097 -13.81 0.000
Immigrant
Generation .047 .044 1.08 0.283
Gender -.015 .090 -0.17 0.869
Locus of Control .159 .053 3.01 0.003
Constant 3.446 .314 10.96 0.000
Model 6
52

Covid is No Worse Than Cold/Flu
Coefficient Standard Error t P > t
Evangelical .018 .132 0.13 0.894
Age .305 .113 2.71 0.007
Education -.196 .026 -7.52 0.000
Party Affiliation -.930 .128 -7.26 0.000
Immigrant
Generation .027 .058 0.47 0.639
Gender -.553 .118 -4.68 0.000
Locus of Control .205 .069 2.96 0.003
Constant 5.025 .414 12.14 0.000
Model 7
53

Mask Wearing Protects Against Covid
Coefficient Standard Error t P > t
Evangelical .057 .103 0.56 0.577
Age -.377 .087 -4.31 0.000
Education .069 .020 3.36 0.001
Party Affiliation 1.422 .099 14.31 0.000
Immigrant
Generation .039 .045 0.86 0.389
Gender .324 .092 3.53 0.000
Locus of Control .045 .054 0.84 0.400
Constant 2.450 .321 7.63 0.000
Model 8
54

Mail-in-Voting Causes Widespread Voter Fraud
Coefficient Standard Error t P > t
Evangelical -.036 .117 -0.31 0.758
Age .364 .099 3.66 0.000
Education -.086 .023 -3.69 0.000
Party Affiliation -2.303 .128 -20.37 0.000
Immigrant
Generation .122 .051 2.40 0.017
Gender .124 .104 -1.19 0.235
Locus of Control .201 .061 3.29 0.001
Constant 4.356 .365 11.92 0.000
Model 9
55

Deep-State Tried to Remove Trump
Coefficient Standard Error t P > t
Evangelical .179 .134 1.34 0.182
Age -.019 .114 -0.16 0.870
Education -.116 .027 -4.34 0.000
Party Affiliation -1.261 .129 -9.76 0.000
Immigrant
Generation -.117 .058 -2.01 0.045
Gender -.464 .119 -3.89 0.000
Locus of Control .245 .069 3.51 0.001
Constant 4.895 .418 11.72 0.000
Model 10
56

significance in all, but the “mask wearing” model (Goertzel 1994, Oliver & Wood 2014; Wood
2016; McCarthy et al. 2021; Casara et al. 2022).
Turning to the election specific CTs we see a similar trend that was present in the Covid
models, as party affiliation, education, and locus of control all were found to also be statistically
significant. Unlike with the Covid CTs, the election models show that age is no longer a
consistent predictor, with it not reaching significance when looking at the “Deep State” theory.
However, what does seem to be a consistent predictor of both election related CTs is
immigration status, but with a positive association occurring with the vote-by-mail model and a
negative association in the other.
Discussion
When taking all of this in, a few things become clear. The first being that Latino
Democrats are the significantly less likely to endorse CTs than their Republican coethnics. This
is not surprising given most of these theories were started or endorsed by Donald Trump, who
was the leader of the Republican Party. Therefore, it is fair to say that this is a result of Trump
setting his agenda (DeMora et al. 2021). Furthermore, it is not surprising that Latino Democrats
are less likely to support these theories given the belief in CTs to be skewed towards ideological
conservatives (Barberá et al., 2015; Jost et al., 2018; Stern et al., 2014; van der Linden et al.
2021).
My theory that Latinos with an external locus of control will be more prone to CTs also
holds, as it was found that Latinos who believed they have no say in what happens in the US
government, were significantly more likely to believe in conspiracy theories. Like party
affiliation, this is shown to be true in each regression model, meaning that it can be considered a
consistent predictor of support for CTs. This is also not surprising, given prior studies have
57

shown that, in general, individuals with an external locus of control are naturally more likely to
believe in CTs than those with an internal locus of control (Abalakina-Paap et al. 1999; Oliver &
Wood 2014).
Aside from party affiliation and locus of control, these findings also show that, in most
cases, higher educated and younger Latinos tend to be less susceptible to conspiratorial thinking.
Despite, the significance of education being just outside the threshold of significance in the
hydroxychloroquine model (P=0.056), it is still safe to assume that educated Latinos are, in
general, less likely to subscribe to CTS. Age, however, is lacking significance in multiple models
meaning that we cannot make a generalized claim about the relationship between age and
susceptibility to CTs, rather we see that younger Latinos were less likely to believe in certain
CTs.
Conclusion
I conclude with this- past studies may have found that Latinos are susceptible to
conspiracy theories (Goertzel 1994; Oliver & Wood 2014), but little has been done to understand
why this is so up until now (Cortina & Rottinghaus 2022). Given the contemporary salience of
conspiracy theories, now was the opportune time to answer this question. This study has shown
that Latino Democrats, and educated Latinos tend to be less susceptible to CTs. Again, this is not
surprising given the politicization of Covid and the 2020 election by Republican President
Donald Trump (Uscinski et al. 2022). Furthermore, it is no surprise that educated Latinos are less
susceptible given educated individuals in general tend to be less susceptible to CTs (Goertzel
1994; Parsons 2010).
While it has been known that Latinos are significantly more likely to believe in CTs that
paint the US in a negative light, it was never clear what caused Latinos to be susceptible to
58

modern day conspiracy theories. This study helps fill this gap, as I have shown that, like the
general population, Latinos with an external locus of control were significantly more likely than
other Latinos to subscribe to CTs. Furthermore, this study has found that an affiliation with the
Democratic Party causes one to be significantly less likely to believe in contemporary CTs.
Taken in tandem, it becomes clear that Latino belief in contemporary conspiracy theories is
driven by partisan politics. Furthermore, these theories do not just affect partisans, but prey upon
those who feel that they yield no influence in American politics.

59

Chapter 4:
Polling Latino Voters in 2020

60

Since the behavioral revolution in the 1950s and 1960s, voting has been a hot topic
amongst political scientists (Downs 1957; Campbell et al. 1980; Dawson 1996; Beltran 2010).
To date, the Michigan School (Campbell et al. 1980), Downsian approach (Downs 1957), and the
theory of linked fate (Dawson 1996) have been some of the most well-known theoretical
products of said revolution. These theories have stuck given their generalizability as they have
shown to be applicable during wartime, periods of prosperity, and recessions. However, we in
the Academy are still in the dark when it comes to proper polling approaches to gauge potential
voting behavior, especially during contemporary presidential elections (Burden & Hillygus 2009;
Hillygus 2011; Graefe 2014; Madson & Hillygus 2020). This comes after traditional polling
techniques (USC/LA Times 2016, 2020; Valentino et al. 2017; Kennedy et al. 2018) have been
shown to be ineffective at accurately predicting true voter preference. This was recognized
during the 2016 election cycle, during which most polls significantly overestimated support for
the Democratic candidate, Hillary Clinton (fivethirtyeight.com 2016; USC Poll 2016; Valentino
et al. 2017), with 2016 polls predicting a wide margin of victory for Clinton. While Clinton did
win the popular vote, she did not do so at a margin predicted by polls. Similar issues arose in
2020 when polls predicted a blowout win for Joe Biden, only for Biden to garner roughly 51% of
the vote.
To date, polling methodology laid down by the Gallup poll has been considered the gold
standard in fielding political surveys. However, the recent events I have discussed above, show
that new types of methods may need to be adopted. Pollsters believe this occurs due to the
negative connotation that may come with voicing support for Donald Trump (Bruin et al. 2021;
Olsson et al. 2019; Olsson et al. 2021), as many have found that respondents may feel
61

embarrassed to admit that they plan on voting for Trump. The 2020 USC Dornsife Presidential
Poll, which also overestimated support for Biden, has argued that asking respondents how think
their social group will vote is a much more accurate at predicting vote choice (Galesic et al.
2018; Olsson et al. 2021; Bruin et al. 2022). What is remains unclear is if measuring perceptions
about social circle voting accurately predicts the Latino vote.
Furthermore, the importance of accurately weighting a sample cannot be understated
(Biemer & Christ 2008; Meijer 2016). This is especially true when trying to gauge the Latino
community, who tend to be more averse to pollsters and who consistently are undercounted or
mis-weighted during presidential polls (Barreto et al. 2018; Fraser-Yokely et al. 2020). With this
said, it is not only certain questions that goes into accurately assessing the Latino vote.
Relatively little research (Nuño 2007; Barreto 2007a; Barreto 2007b) discusses how to
accurately measure potential Latino vote choice, and even less attempt to reconcile inaccurate
poll results with final vote choice. In this study, I will attempt to do just this by using data from
the 2020 USC Presidential Election Poll to test whether social circle voting questions more
accurately predict Latino vote choice, as they did for aggregate vote choice (Brancaccio & Shin
2020; Ylanan et al. 2020; Bruin et al. 2021). The 2020 USC Presidential Election poll is
appropriate for this study because it is one of the only polls that continuously asked about social
circle voting. Furthermore, the post-election poll is accurately weighted (Dominguez-Villegas et
al. 2021), so I can test how the social circle voting question held up against the final results. I
will go on to show that, while not perfect, social circle voting questions do more accurately
capture how Latinos will vote on Election Day.

62

Literature Review
Polling is not a new phenomenon, but it began to garner a lot of attention when Gallup
accurately predicted Roosevelt’s 1936 victory (Hillygus 2011; Graefe 2014). With Literary
Digest incorrectly predicting a loss for Roosevelt, Gallup became the gold standard for
presidential polling. Past studies on polling methods have found that not much attention is paid
to the actual methodology that goes into presidential election polls, with specific techniques not
being widely public. The argument for why there was a lack of transparency being that such
simple math (i.e. percentages) should not require published techniques (Hillygus 2011). As time
went on, we in the field of political science began to provide academic input into the world of
polling (Abramowitz 1988; Gelman & King 1993; Voss et al. 1995; Park et al. 2004; Wang et al.
2015).
Amongst this input was the conceptualization of public opinion polls to help explain their
three key functions- election forecasting, formulating campaign strategy, and understanding
voter behavior (Hillygus 2011). The public places an emphasis on the first function-accurate
forecasting, with pollsters held to a high standard when it comes to their ability to predict the
overall winner (Abramowitz 1988; Gelman & King 1993; Hillygus 2011; Graefe et al. 2014;
Dowdle et al. 2016). However, most do not judge these polls on which accuracy as it pertains to
forecasting the popular vote. While it may sound as though these are the same thing, they are not.
As polls which correctly predict the winner, but drastically overestimate the margin of victory
are seen as superior when compared to those who predict the wrong winner but are off by only a
small margin (Hillygus 2011; Key 2020). This is not the case in the political science community,
where we place an emphasis on accurate models which predict choice and breakdown.
63

With this said, academics have attempted to come up with better measures of vote choice
which more accurately predict the vote breakdown (Beck & Stegmaier 2014; MacWilliams 2015;
Kennedy et al. 2017). This has led to academic and other election polls to go beyond the
traditional “If the election were held today, who would you cast your vote for?” question in an
attempt to more accurately predict the winner and the margin of victory (Terhanian 2017). Part
of this need goes beyond traditional types of questions can be attributed to the finding that many
respondents do not report how they will actually vote.
The latter has caused pollsters to use new measures of vote choice, with one example
being to ask respondents to report who they believed would ultimately win the election (Hillygus
2011). Graefe (2014) found this question more accurately predicted the winner and margin of
victory in elections ranging from 1932 through 2012, noting that the simplicity of the question
ultimately lead to it being the most accurate predictor of vote choice. He went on to explain that
the traditional “[i]f the election were held today…” question is an improper measure of vote
choice because it is inherently not predictive, rather it offers a “snapshot of public opinion”.
With this in mind, pollsters have not only focused their attention on asking who they
believe will emerge victorious on Election Day, but also have turned their attention to asking
questions on friends and families. Questions asking respondents how their “social circle” will
vote, was a key part of the USC Dornsife 2020 Presidential Poll, which found that asking these
questions more accurately predicted the margin of victory and vote breakdown (Hillygus 2011;
Graefe et al. 2015; Campbell et al. 2017). This was not unique to USC, as the 2020 Fox News
Presidential Poll also asked respondents questions on how their friends and family will vote.
Representatives from the USC Poll argued that the social circle questions produced more
accurate results because respondents did not accurately report their preferred vote choice. This
64

meaning that respondents, who actually voted for Trump, were less likely to say they were going
to vote for the incumbent during pre-election polls. However, these individuals were more likely
to report that their social circle was going to vote for Trump (Bruin et al. 2021; Olsson et al.
2021). Whether these social circle voting questions also accurately forecast Latino vote choice is
my focus, as this has remained unclear up to this point.
Surveying Racial and Ethnic Minorities
Political science has come to understand that many surveys have a tendency to under-
sample minorities, especially Latinos (Barreto et al. 2017). To fix this issue, many who study
race and ethnic politics have gone two routes, the first being to strive for an oversample of
minority respondents. This allows survey data to more accurately reflect the true
behavior/attitudes of racial and ethnic minority groups. Examples of such include the
Collaborative Multi-racial Post-election Survey, which uses strategic sampling efforts to over-
sample Black, Latino, and Asian Americans. This practice, while the best way to go about
attaining a representative sample, is often unfeasible when doing pre-election polling.
Another route to take is to use weighting techniques to account for the errors that may
have occurred during sampling (Biemer & Christ 2008; Meijer 2021). Weighting is a common
practice in survey methods which provides different “weights” to certain respondents to make
the sample more representative. For example, a sample that under samples Latinos and over-
counts White Americans may weight Latino responses to count as 1.3 rather than just 1, while
weighting the White respondents to .7. This process is widely used during pre-election political
polls, such as the USC Dornsife Presidential Poll (Olsson et al. 2021; Meijer 2021), of which the
main focus is to accurately forecast the election by making the sample reflect the general
population, demographically speaking.
65

Latino Social Circles
As mentioned, USC Pollsters argued that their “social circle” question (Darling 2021;
Olsson 2021) was a better method of forecasting vote choice, but they did not explain if this
carries over to the Latino community. Before I can test if this does actually carry over, I must
analyze the rationale for why this would occur. As mentioned by the USC Polling team, many
respondents did not report their true intentions when it came to how they would vote. In turn,
many respondents used the social circle question to more accurately reflect their vote choice
(Bruin et al. 2021; Olsson et al. 2021a, 2021b). It is logical to assume that this would also occur
within the Latino community, as Latino Trump supporters may have faced even greater pressure
to lie about for whom they would cast their ballot. This being due to the majority of Latinos
holding negative feelings about the sitting president due to his consistent anti-Latino rhetoric
(Sanchez et al. 2017; Gutierrez et al. 2019; Gonzalez 2019).
Survey Weighting
Before I move on to show how the traditional and social circle voting questions held up
against the actual breakdown of the Latino vote, I must explain the importance of weighting.
Weighting is a common tool in survey methods that is used for several reasons. The first reason
being to correct for unequal sampling probabilities. When it comes to the USC Poll, this base
weight accounts for unequal probabilities of being chosen according to zip code and then the
probability of a household within that zip code being chosen for the survey. This base-weight is
developed after collecting the sample, and the same weight is used for all surveys in the poll.
The second type of weight used in many election polls is post-stratification weights
(Jagers et al. 1985; Jagers 1986; Kulas et al. 2018; Lauderdale et al. 2020; Meijer 2021), which
are used to help each survey wave’s sample align with the demographics of the population. The
66

USC Poll, like most other election polls, create their post stratification weights based on the
following factors: gender, age, education, race/ethnicity, income, marital status, citizenship
status, employment status, and location (USC Poll 2020; Meijer 2021; Olsson et al. 2021).
Furthermore, these weights were based on the most recent data from the Basic Monthly Current
Population Survey (Meijer 2021). The post-election weight is a bit different as it also is weighted
according to the actual results of the 2020 election.
Analyzing Measures of Vote Choice

Table 3

Traditional polling questions
As I have mentioned, pollsters ask multiple questions to forecast elections, but most
focus on the traditional question when discussing their results (Flannely et al. 2000; Wang et al.
2015; Gelman et al. 2016). The USC Dornsife Poll acknowledges that their traditional question
did predict the correct winner but did not accurately forecast the aggregate margin of victory of
the 2020 Election. However, it is not clear if this was true across all groups, such as Latinos.
IF THE ELECTION WERE HELD TODAY (LATE OCT/EARLY NOV)
JOE BIDEN 56% (+/-2)
DONALD TRUMP 24% (+/-2)
UNDECIDED/REFUSE TO SAY 12% (+/-2)
67

When looking at the late October/early November wave of the poll, we can see that the Latino
vote was forecasted to go to Biden, with the Democrat expected to win roughly 55% of the
Latino vote. In the same poll, Trump was forecasted to win only 24% of the Latino vote, while
about 7% being undecided. This dramatically underestimated the number of Latinos who
actually voted for Trump.
Social Circle Voting Questions
The USC polling team began to ask questions about how one’s social circle will vote
during the 2016 US presidential election, and ultimately found that these questions, while not
perfect, were much more accurate at predicting election outcomes at both the national and state
level (Bruin et al. 2021; Meijer 2021; Olsson et al. 2021). Confidence in these findings grew
when they also accurately forecasted the outcome of the 2017 French presidential election and
the 2018 US midterm elections (Galesic et al. 2018). This led pollsters at USC to argue that
social circle questions may be the future of political polling, as they better account for last
minute changes in vote choice and the “dynamics of echo chambers” (Levy & Razin 2019;
Barbera 2020; USC Poll 2020. Furthermore, they argue that social circle questions also remedy
the issues that come with questions which ask respondents to predict the winner of the election,
given these new questions ask respondents to report what percentage of their social circle that
each candidate. While these questions continued to prove accurate at predicting the 2020 election
results, it is not clear if they also accurately predict the Latino vote. Below I show two tables, the
first of which shows the general results of the social circle questions, the second showing only
Latino respondents.

68

WHAT PERCENTAGE OF YOUR SOCIAL CIRCLE WILL VOTE FOR (MEAN)
JOE BIDEN 60 % (+/-2)
DONALD TRUMP 34% (+/-2)
Table 4
Breaking down the Latino Vote
When turning to the post-election poll we can see that while pre-election polls predicted
that 55% of the Latino vote would go to Joe Biden, he actually received roughly 65%. This is in
line with most findings, including UCLA Latino Policy and Politics Center (Dominguez-Villegas
2021), which first shows the accuracy of the post-election poll weights. The next thing to stick
out is the inaccurate forecast provided via the traditional voting question. This showing one of
two things, the first potential reason being that the original weighting procedure was off and did
not accurately account for the fact that the sample of Latinos was skewed in some way.

LATINO VOTE: 2020 US PRESIDENTIAL ELECTION
JOE BIDEN 65 % (+/-1)
DONALD TRUMP 35% (+/-1)
Table 5

The other reason why the traditional voting question may have been unable to properly
forecasting the Latino vote may be due to respondents inaccurately reporting who they planned
to vote for (Epstein et al. 2006; Galesic et al. 2018; Myers & Russell 2019; Olsson et al. 2021).
69

According to the USC Poll, this should be corrected by looking at the social circle question,
given it acts as a proxy that allows respondents to report their true preference. These questions
ask respondents “what percentage of your social circle will vote for Biden/Trump?”, and the
weighted mean is then compared to the actual election results. As one can see in table 4, this
question is indeed more accurate at forecasting overall vote choice, with the mean more closely
resembling the actual election outcome when compared to the traditional voting question.
Looking to table 5, we see that the social circle question is also a more accurate method
of forecasting the Latino vote. By looking solely at the sample of Latinos we see that, on
average, respondents believed that 60% of their social circle would vote for Biden, while 34%
would opt for Trump. When comparing this to the actual vote breakdown, we see that the Latino
vote is more closely reflected via the social circle question, which accurately predicted the
percentage of Latinos that voted for Trump. It also was much closer to accurately forecasting
Biden’s vote share as well, being 4-5 percentage points off rather than the traditional question’s
9. While it is not perfect, it is safe to say that like with the general public, the social circle
question does more accurately forecast the Latino vote when compared to the traditional “if the
election were held today…” question.
Discussion
My analyses have found that, like with the general public, traditional voting questions are
no longer accurate at forecasting the Latino vote. The reason for this is not last-minute changes
in voters’ preferences, rather inaccurate reporting of said preference during pre-election polling
surveys. It has been found that many Trump supporters do not report their true vote preference
either out of embarrassment or in an attempt to corrupt the poll (Darling 2020; Bruin et al. 2021;
Olsson et al. 2021a, 2021b). The former is the key reason that Latino Trump voters were also
70

less likely to state their true preference, with there being even greater pressure within the Latino
community to vote for Biden (Sanchez et al. 2017; Gutierrez et al. 2019; Gonzalez 2019). This
was somewhat remedied by the 2020 USC Dornsife Presidential Poll due to their social circle
voting question, which proved rather accurate at predicting the Latino vote in the 2020
presidential election.
Furthermore, I have shown that weighting was not necessarily the issue that caused the
USC Poll to inaccurately forecast the election results, as was the case in 2016 (USC Poll 2016).
While the post-election poll proved to have the most accurate weights, this is true of all post
election polls which are weighted to account for the actual election results. Rather, this article
shows that using a non-traditional voting question, specifically USC’s social circle question, may
be the key to more accurately predicting election outcomes across groups, given it accuracy at
forecasting both the general and Latino vote.

71

Bibliography
Abalakina-Paap, Marina, Walter G. Stephan, Traci Craig, and W. Larry Gregory. 1999. “Beliefs
in Conspiracies.” Political Psychology 20(3): 637–47.

Abramowitz, Alan I. 1988. “An Improved Model for Predicting Presidential Election
Outcomes.” PS: Political Science & Politics 21(4): 843–47.

Academic Pipleine Through Data Access, Publication, and Networking Opportunities.” PS:
Political Science & Politics 53(1): 150–51

Alban Pascal, Noudjom et al. “Pandemic Means the Whole World: COVID-19 and Global
Bioethics.” https://repository.library.georgetown.edu/handle/10822/1059419 (May 2, 2022).

Amarasingam, Amarnath, and Marc-Andre Argentino. 2020. “The QAnon Conspiracy Theory: A
Security Threat in the Making?” Combating Terrorism Center at West Point.
https://ctc.westpoint.edu/the-qanon-conspiracy-theory-a-security-threat-in-the-making/
(May 2, 2022).

Arvizu, J. R., & Garcia, F. C. (1996). Latino voting participation: Explaining and differentiating
latino voting turnout. Hispanic Journal of Behavioral Sciences, 18(2), 104-128.

Austin, Erica Weintraub, Porismita Borah, and Shawn Domgaard. 2021. “COVID-19
Disinformation and Political Engagement among Communities of Color: The Role of Media
Literacy.” Harvard Kennedy School Misinformation Review.
https://misinforeview.hks.harvard.edu/?p=6232 (April 26, 2022).

Balderrama, Francisco E., and Raymond Rodriguez. 2006. Decade of Betrayal: Mexican
Repatriation in the 1930s. UNM Press.

Barberá, Pablo et al. 2015. “Tweeting From Left to Right: Is Online Political Communication
More Than an Echo Chamber?” Psychological Science 26(10): 1531–42.

Barnes, Samuel H., M. Kent Jennings, Ronald Inglehart, and Barbara Farah. 1988. “Party
Identification and Party Closeness in Comparative Perspective.” Political Behavior 10(3):
215–31.

Barnes, Samuel H., M. Kent Jennings, Ronald Inglehart, and Barbara Farah. 1988. “Party
Identification and Party Closeness in Comparative Perspective.” Political Behavior 10(3):
215–31.

Barreto, Matt A. 2007a. “Isí Se Puede! Latino Candidates and the Mobilization of Latino
Voters.” American Political Science Review 101(3): 425–41.

———. 2007b. “The Role of Latino Candidates in Mobilizing Latino Voters: Revisiting Latino
Vote Choice.” Latino politics: identity, mobilization, and representation: 63–89.
72

Barreto, Matt A., Loren Collingwood, and Sylvia Manzano. 2010. “A New Measure of Group
Influence in Presidential Elections: Assessing Latino Influence in 2008.” Political Research
Quarterly 63(4): 908–21.

Barreto, Matt A., Lorrie Frasure-Yokley, Edward D. Vargas, and Janelle Wong. 2018. “Best
Practices in Collecting Online Data with Asian, Black, Latino, and White Respondents:
Evidence from the 2016 Collaborative Multiracial Post-Election Survey.” Politics, Groups,
and Identities 6(1): 171–80.

Barreto, Matt A., Sylvia Manzano, and Ricardo Ramírez. “Mobilization, Participation, and
Solidaridad Latino Participation in the 2006 Immigration Protest Rallies.” Urban Affairs
Review. (2009).

Beltran, Cristina. 2010. The Trouble with Unity: Latino Politics and the Creation of Identity.
OUP USA.

Berlinski, Nicolas et al. 2021. “The Effects of Unsubstantiated Claims of Voter Fraud on
Confidence in Elections.” Journal of Experimental Political Science: 1–16.

Bertin, Paul, Kenzo Nera, and Sylvain Delouvée. 2020. “Conspiracy Beliefs, Rejection of
Vaccination, and Support for Hydroxychloroquine: A Conceptual Replication-Extension in
the COVID-19 Pandemic Context.” Frontiers in Psychology 11.
https://www.frontiersin.org/article/10.3389/fpsyg.2020.565128 (April 26, 2022).

Biemer, Paul P., and Sharon L. Christ. 2008. “Weighting Survey Data.” International handbook
of survey methodology 2008: 317–41.

Bierwiaczonek, Kinga, Jonas R. Kunst, and Olivia Pich. 2020. “Belief in COVID-19 Conspiracy
Theories Reduces Social Distancing over Time.” Applied Psychology: Health and Well-
Being 12(4): 1270–85.

Bolsen, Toby, and James N. Druckman. 2018. “Validating Conspiracy Beliefs and Effectively
Communicating Scientific Consensus.” Weather, Climate, and Society 10(3): 453–58.

Bolsen, Toby, and Judd R. Thornton. 2021. “Candidate and Party Affective Polarization in U.S.
Presidential Elections: The Person-Negativity Bias?” Electoral Studies 71: 102293.

Bonica, Adam, Jacob M. Grumbach, Charlotte Hill, and Hakeem Jefferson. 2021. “All-Mail
Voting in Colorado Increases Turnout and Reduces Turnout Inequality.” Electoral Studies.

Brambor, Thomas, William Roberts Clark, and Matt Golder. 2006. “Understanding Interaction
Models: Improving Empirical Analyses.” Political Analysis 14(1): 63–82.

73

Brancaccio, David, and Daniel Shin. 2020. “USC Poll Looks at What It Got Right, Wrong with
2020 Election.” Marketplace. https://www.marketplace.org/2020/11/19/2020-election-
polling-results-usc-dornsife-daybreak-poll/ (May 2, 2022).

Brandt, Allan M. 1978. “Racism and Research: The Case of the Tuskegee Syphilis Study.” The
Hastings Center Report 8(6): 21.

Brint, Steven, and Seth Abrutyn. “Who’s Right About the Right? Comparing Competing
Explanations of the Link Between White Evangelicals and Conservative Politics in the
United States.” Journal for the Scientific Study of Religion 49, no. 2 (2010): 328–50.

Bruine de Bruin, Wändi et al. 2022. “Asking about Social Circles Improves Election Predictions
Even with Many Political Parties.” International Journal of Public Opinion Research 34(1):
edac006.

Burden, Barry C., and D. Sunshine Hillygus. 2009. “Polls and Elections: Opinion Formation,
Polarization, and Presidential Reelection.”

Busby, Robert. 2022. “Trump, Populism and the Pandemic.” In Political Marketing in the 2020
U.S. Presidential Election, Palgrave Studies in Political Marketing and Management, ed.
Jamie Gillies. Cham: Springer International Publishing, 125–41.
https://doi.org/10.1007/978-3-030-86559-7_8 (April 25, 2022).

Calhoun-Brown, Allison. 1996. “African American Churches and Political Mobilization: The
Psychological Impact of Organizational Resources.” The Journal of Politics 58(4): 935–53.

Calhoun-brown, Allison. 2007. “Upon This Rock: The Black Church, Nonviolence, and the Civil
Rights Movement”. PS: Political Science and Politics, Vol. 33, No. 2. (Jun., 2000), pp. 168-
174.

Campbell, Angus et al. 1980. The American Voter. University of Chicago Press.

Campbell, David, J. Quin Monson, and John C. Green. 2009. “Framing Faith: How Voters
Responded to Candidates’ Religions in the 2008 Presidential Campaign.”

Campbell, James E. et al. 2017. “A Recap of the 2016 Election Forecasts.” PS: Political Science
& Politics 50(2): 331–38.

Casara, Bruno Gabriel Salvador, Caterina Suitner, and Jolanda Jetten. 2022. “The Impact of
Economic Inequality on Conspiracy Beliefs.” Journal of Experimental Social Science.

Castillo, Richard Griswold del. 1992. The Treaty of Guadalupe Hidalgo: A Legacy of Conflict.
University of Oklahoma Press.

Cervantes, Lilia et al. 2021. “Experiences of Latinx Individuals Hospitalized for COVID-19: A
Qualitative Study.” JAMA Network Open 4(3): e210684.
74

Chaturvedi, Neilan S. 2014. “The next Generation? A Reexamination of Religious Influence on
Mexican-American Attitudes toward Same-Sex Marriage.” Politics, Groups, and Identities
2(4): 589–606. doi: 10.1080/21565503.2014.969740.

Chinn, Sedona, P. Sol Hart, and Stuart Soroka. 2020. “Politicization and Polarization in Climate
Change News Content, 1985-2017.” Science Communication 42(1): 112–29.

Clarke, Kevin A. 2005. “The Phantom Menace: Omitted Variable Bias in Econometric
Research.” Conflict Management and Peace Science 22(4): 341–52.

Cohen, Philip N., and Lynne M. Casper. “In Whose Home? Multigenerational Families in the
United States, 1998–2000.”

Converse, Philip E., Angus Campbell, Warren E. Miller, and Donald E. Stokes. 1961. “Stability
and Change in 1960: A Reinstating Election.” The American Political Science Review
55(2): 269–80.

Converse, Phillip. “Religion and Politics: The 1960 Elections*.” Survey Research Center:
University of Michigan, 1966.

Cortina, Jeronimo, and Brandon Rottinghaus. 2022. “Conspiratorial Thinking in the Latino
Community on the 2020 Election.” Research and Politics 9(1): 205316802210835. doi:
10.1177/20531680221083535.

Cottrell, David, Michael C. Herron, and Sean J. Westwood. 2018. “An Exploration of Donald
Trump’s Allegations of Massive Voter Fraud in the 2016 General Election.” Electoral
Studies 51: 123–42.

Dalton, Russell J. 2015. The Good Citizen: How a Younger Generation Is Reshaping American
Politics. CQ Press.

Darling, Jill. 2021. “Experiences with U.S. National Election Polling in 2016 and 2020 Using a
Probability-Based Internet Panel and Experimental Outcomes Models.” University of
Southern California. Polling Report.

Dawson, Michael C. 2020. Behind the Mule Behind the Mule: Race and Class in African-
American Politics. Princeton University Press. (May 2, 2022).

De la Garza, R. O., & DeSipio, L. (1993). “Save the baby, change the bathwater, and scrub the
tub: Latino electoral participation after seventeen years of voting rights act coverage.” Texas
Law Review, 71(7), 1479-1540.

75

De la garza, Rodolfo, and Jeronimo Cortina. 2007. “Are Latinos Republicans But Just Don’t
Know It? The Latino Vote in the 2000 and 2004 Presidential Elections.” American Politics
Research. 35: 202–23.

Delgado, Richard. 2014. “The Law of the Noose: A History of Latino Lynching.”.
DeMora, Stephanie L., Jennifer L. Merolla, Brian Newman, and Elizabeth J. Zechmeister. 2021.
“Reducing Mask Resistance among White Evangelical Christians with Value-Consistent
Messages.” Proceedings of the National Academy of Sciences 118(21): e2101723118.

Domínguez-Villegas, Rodrigo et al. 2021. “Vote Choice of Latinos in the 2020 Presidential
Election.” UCLA Latino Policy and Politics Initiative: 41.

Douglas, Karen M., Robbie M. Sutton, and Aleksandra Cichocka. 2017. “The Psychology of
Conspiracy Theories.” Current Directions in Psychological Science 26(6): 538–42.

Dowdle, Andrew J., Randall E. Adkins, Karen Sebold, and Jarred Cuellar. 2016. “Forecasting
Presidential Nominations in 2016: #WePredictedClintonANDTrump.” PS: Political Science
& Politics 49(4): 691–95.

Downs, Anthony. 1957. An Economic Theory of Democracy.

Dun, Lindsay, and Stephen Jessee. 2020. “Demographic Moderation of Spatial Voting in
Presidential Elections.” American Politics Research 48(6): 750–62.

Dunlap, Riley E., and Araon M. McCright. 2008. “A Widening Gap: Republican and Democratic
Views on Climate Change.” Environment: Science and Policy for Sustainable Development
50(5): 26–35.

Dyer, Owen. 2020. “Trump Claims Public Health Warnings on Covid-19 Are a Conspiracy
against Him.” BMJ: m941.

Edelson, Jack et al. 2017. “The Effect of Conspiratorial Thinking and Motivated Reasoning on
Belief in Election Fraud.” Political Research Quarterly 70(4): 933–46.

Eggers, Andrew C., Haritz Garro, and Justin Grimmer. 2021. “No Evidence for Systematic Voter
Fraud: A Guide to Statistical Claims about the 2020 Election.” Proceedings of the National
Academy of Sciences 118(45): e2103619118.

Ellison, Christopher G., Gabriel A. Acevedo, and Aida I. Ramos-Wada. 2011. “Religion and
Attitudes Toward Same-Sex Marriage Among U.S. Latinos: U.S. Latino Attitudes Toward
Same-Sex Marriage.” Social Science Quarterly 92(1): 35–56.

Epstein, David L., Richard H. Pildes, Rodolfo de la garza, and Sharyn O’Halloran. 2006. The
Future of the Voting Rights Act. Russell Sage Foundation.

76

Fadillah, Mariatul et al. 2020. “The Effect of Online Seminar on Public Knowledge in Facing
Misinformation Regarding the Covid 19 Pandemic in the New Normal Era.” Community
Medicine and Education Journal 1(1): 10–16. doi: 10.37275/cmej.v1i1.2.

Fairbanks, David. 1977. “Religious Forces and ‘Morality’ Policies in the American States.” The
Western Political Quarterly 30(3): 411–17.

Ferree, Myra Marx, ed. 2002. Shaping Abortion Discourse: Democracy and the Public Sphere in
Germany and the United States. Cambridge, UK ; New York: Cambridge University Press.

Flannelly, Kevin J., Laura T. Flannelly, and Malcolm S. McLeod. 2000. “Reducing Undecided
Voters and Other Sources of Error in Election Surveys.” International Journal of Market
Research 42(2): 1–6.

Fleishman, R. “The Battle Against Reproductive Rights: The Impact of the Catholic Church on
Abortion Law in Both International and Domestic Arenas.” Emory International Law
Review, 2000.

Flores-Saviaga, Claudia, and Saiph Savage. 2019. “Anti-Latinx Computational Propaganda in
the United States.” (April 25, 2022).

Frasure-Yokley, Lorrie, Janelle Wong, Edward Vargas, and Matt Barreto. 2020. “The
Collaborative Multiracial Post-Election Survey (CMPS): Building the Academic Pipeline
Through Data Access, Publication, and Networking Opportunities.” PS: Political Science &
Politics 53(1): 150–51.

Friedman, Richard. 2020. “Why Humans Are Vulnerable to Conspiracy Theories.” Psychiatric
Services. https://ps.psychiatryonline.org/doi/full/10.1176/appi.ps.202000348 (April 26,
2022a).

Fujiwara, Thomas. 2011. “A Regression Discontinuity Test of Strategic Voting and Duverger’s
Law.” Quarterly Journal of Political Science 6(3–4): 197–233.

Galesic, M. et al. 2018. “Asking about Social Circles Improves Election Predictions.” Nature
Human Behaviour 2(3): 187–93.

Gelman, Andrew, and Gary King. 1993. “Why Are American Presidential Election Campaign
Polls So Variable When Votes Are So Predictable?” British Journal of Political Science
23(4): 409–51.

Gelman, Andrew, Sharad Goel, David Rothschild, and and Wei Wang. 2016. “High-Frequency
Polling with Non-Representative Data.” In Political Communication in Real Time,
Routledge.

77

Gerber, Alan S., Gregory A. Huber, and Ebonya Washington. 2010. “Party Affiliation,
Partisanship, and Political Beliefs: A Field Experiment.” American Political Science Review
104(4): 720–44.

Goertzel, Ted. 1994. “Belief in Conspiracy Theories.” Political Psychology 15(4): 731–42.

Gonzalez, Eduardo. 2019. “Stereotypical Depictions of Latino Criminality: U.S. Latinos in the
Media during the MAGA Campaign.” Democratic Communiqué. (May 2, 2022).

Gorbie-Smith, Giselle. 1999. “The Continuing Legacy of the Tuskegee Syphilis Study:
Considerations for Clinical Investigation.” The American Journal of the Medical Sciences
317(1): 5–8.

Gorski, Philip. 2017. “Why Evangelicals Voted for Trump: A Critical Cultural Sociology.”
American Journal of Cultural Sociology 5(3): 338–54.

Graefe, Andreas, Helmut Küchenhoff, Veronika Stierle, and Bernhard Riedl. 2015. “Limitations
of Ensemble Bayesian Model Averaging for Forecasting Social Science Problems.”
International Journal of Forecasting 31(3): 943–51.

Graefe, Andreas. 2014. “Accuracy of Vote Expectation Surveys in Forecasting Elections.”
Public Opinion Quarterly 78(S1): 204–32.

Gray, Fred D. 1998. The Tuskegee Syphilis Study: The Real Story and Beyond. New South
Books.

Gray, Mark M., Paul M. Perl, and Mary E. Bendyna. 2006. “Camelot Only Comes but Once?
John F. Kerry and the Catholic Vote.” Presidential Studies Quarterly 36(2): 203–22.

Greene, Steven, Marc Hetherington, Rahsaan Maxwell, and Timothy J. Ryan. 2022. “Public
Service Announcements and Promoting Face Masks During the COVID-19 Pandemic.” PS:
Political Science & Politics 55(1): 7–12.

Greene, Steven. 1999. “Understanding Party Identification: A Social Identity Approach.”
Political Psychology 20(2): 393–403.

Gschwend, Thomas, and Frank Schimmelfennig. 2007. “Introduction: Designing Research in
Political Science — A Dialogue between Theory and Data.” In Research Design in Political
Science, eds. Thomas Gschwend and Frank Schimmelfennig. London: Palgrave Macmillan
UK, 1–18. http://link.springer.com/10.1057/9780230598881_1 (April 4, 2022).

Gutierrez, Angela, Angela X. Ocampo, Matt A. Barreto, and Gary Segura. 2019. “Somos Más:
How Racial Threat and Anger Mobilized Latino Voters in the Trump Era.” Political
Research Quarterly 72(4): 960–75.

78

Hall Jamieson, Kathleen, and Dolores Albarracín. 2020. “The Relation between Media
Consumption and Misinformation at the Outset of the SARS-CoV-2 Pandemic in the US.”
Harvard Kennedy School Misinformation Review.
https://misinforeview.hks.harvard.edu/article/the-relation-between-media-consumption-and-
misinformation-at-the-outset-of-the-sars-cov-2-pandemic-in-the-us/ (April 26, 2022).

Halpern, Lucy Wang. 2020. “The Politicization of COVID-19.” AJN: The American Journal of
Nursing 120(11): 19–20.

Hansen, Michael A. et al. 2021. “The Partisan Impact on Local Government Dissemination of
COVID-19 Information: Assessing US County Government Websites.” Canadian Journal of
Political Science 54(1): 150–62.

Harris, Casey T., and Ben Feldmeyer. 2013. “Latino Immigration and White, Black, and Latino
Violent Crime: A Comparison of Traditional and Non-Traditional Immigrant Destinations.”
Social Science Research 42(1): 202–16.

Harris, Fredrick C. 1994. “Something Within: Religion as a Mobilizer of African-American
Political Activism.” The Journal of Politics 56(1): 42–68.

Helmuth, Brian, Tarik C. Gouhier, Steven Scyphers, and Jennifer Mocarski. 2016. “Trust,
Tribalism and Tweets: Has Political Polarization Made Science a ‘Wedge Issue’?” Climate
Change Responses 3(1): 3.

Hendrix, George J. 2021. “Sidebar: Addressing North Carolina Hispanics and Latinos Affected
by COVID-19: A Grassroots Approach.” North Carolina Medical Journal 82(1): 64–65.

Hero, Rodney E. 1992. Latinos and the U.S. Political System: Two-Tiered Pluralism. Temple
University Press.

Hillygus, D. Sunshine. 2011. “The Evolution of Election Polling in the United States.” Public
Opinion Quarterly 75(5): 962–81.

Hillygus, D. Sunshine, and Todd G. Shields. 2008. The Persuadable Voter: Wedge Issues in
Presidential Campaigns. Princeton university press.

Hofstadter, Richard. 1964. “The Paranoid Style in American Politics.” Harper’s Magazine
November 1964. https://harpers.org/archive/1964/11/the-paranoid-style-in-american-
politics/ (April 25, 2022).

Holman, Mirya R., and J. Celeste Lay. 2019. “They See Dead People (Voting): Correcting
Misperceptions about Voter Fraud in the 2016 U.S. Presidential Election.” Journal of
Political Marketing 18(1–2): 31–68.

Hooghe, Marc. 2004. “Political Socialization and the Future of Politics.” Acta Politica 39(4):
331–41.
79

Hutcheson, John D., and George A. Taylor. 1973. “Religious Variables, Political System
Characteristics, and Policy Outputs in the American States.” American Journal of Political
Science 17(2): 414–21.

Hyman, Herbert H., and Paul B. Sheatsley. 1953. “The Political Appeal of President
Eisenhower.” The Public Opinion Quarterly 17(4): 443–60.

“Inclusion, Information and Intersection: The Truth about Connecting with U.S. Latinos.”
https://www.nielsen.com/us/en/insights/report/2021/inclusion-information-and-intersection
(April 25, 2022).

Ippolito, Mariachiara et al. 2020. “Medical Masks and Respirators for the Protection of
Healthcare Workers from SARS-CoV-2 and Other Viruses.” Pulmonology 26(4): 204–12.

Jagers, P., A. Oden, and L. Trulsson. 1985. “Post-Stratification and Ratio Estimation: Usages of
Auxiliary Information in Survey Sampling and Opinion Polls.” International Statistical
Review/Revue Internationale de Statistique: 221–38.

Jagers, Peter. 1986. “Post-Stratification against Bias in Sampling.” International Statistical
Review/Revue Internationale de Statistique: 159–67.

James, Toby S., and Alistair Clark. 2020. “Electoral Integrity, Voter Fraud and Voter ID in
Polling Stations: Lessons from English Local Elections.” Policy Studies 41(2–3): 190–209.

Jamieson, Kathleen H. and Dalores Albarracin. 2020. “The Relation between Media
Consumption and Misinformation at the Outset of the SARS-CoV-2 Pandemic in the US.”
The Harvard Kennedy School of Government Misinformation Review.
https://dash.harvard.edu/handle/1/42661740 (April 26, 2022).

Jang, Sou Hyun, and Gowoon Jung. 2021. “How Does COVID-19 Differ from the Flu/Cold? A
Study of Multilevel Information Seeking among Korean Immigrant Women in the U.S.”
American Journal of Health Behavior 45(4): 665–76.

Jones-correa, Michael A., and Cornell University. 2001. “Political Participation: Does Religion
Matter.” Political Research Quarterly: 751–70.

Jones-Correa, Michael, Helen B. Marrow, Dina G. Okamoto, and Linda R. Tropp. 2018.
“Immigrant Perceptions of U.S.-Born Receptivity and the Shaping of American Identity.”
RSF: The Russell Sage Foundation Journal of the Social Sciences 4(5): 47–80.

Jost, John et al. 2018. “How Social Media Facilitates Political Protest: Information, Motivation,
and Social Networks: Social Media and Political Protest.” Political Psychology 39: 85–118.

Kahane, Leo H. 2021. “Politicizing the Mask: Political, Economic and Demographic Factors
Affecting Mask Wearing Behavior in the USA.” Eastern Economic Journal 47(2): 163–83.
80

Kähler, Christian J., and Rainer Hain. 2020. “Fundamental Protective Mechanisms of Face
Masks against Droplet Infections.” Journal of Aerosol Science 148: 105617.

Kelly, Nathan J. and Morgan, Jana, "Religion and Latino Partisanship in the United States"
(2005). Political Science Publications and Other Works.
https://trace.tennessee.edu/utk_polipubs/14

Kennedy, Ryan, Stefan Wojcik, and David Lazer. 2017. “Improving Election Prediction
Internationally.” Science 355(6324): 515–20.

Kenski, Henry. 1995. “Catholic and Evangelical Voting: 1992 and 1994.” International Journal
of Social Economics 22(9/10/11): 149–64.

Key, Jim. 2020. “USC Dornsife Election Poll That Made Headlines in 2016 Relaunches with
Changes.” USC Dornsife News. http://dornsife.usc.edu/news/stories/3295/2020-daybreak-poll-
qna-with-jill-darling-survey-director/ (May 2, 2022).

King, Patricia A. 1992. “Twenty Years after. The Legacy of the Tuskegee Syphilis Study. The
Dangers of Difference.” https://drum.lib.umd.edu/handle/1903/23029 (April 25, 2022).

Kintisch, Eli. 2013. “Climate Study Highlights Wedge Issue.” Science 339(6116): 128–29.

Kulas, John T., David H. Robinson, Jeffrey A. Smith, and Donald Z. Kellar. 2018. “Post-
Stratification Weighting in Organizational Surveys: A Cross-Disciplinary Tutorial.” Human
Resource Management 57(2): 419–36.

Lauderdale, Benjamin E., Delia Bailey, Jack Blumenau, and Douglas Rivers. 2020. “Model-
Based Pre-Election Polling for National and Sub-National Outcomes in the US and UK.”
International Journal of Forecasting 36(2): 399–413.

Levy, Gilat, and Ronny Razin. 2019. “Echo Chambers and Their Effects on Economic and
Political Outcomes.” Annual Review of Economics 11: 303–28.

Lewis-Beck, Michael S., and Mary Stegmaier. 2014. “To Improve Their Predictions, Election
Forecasters Should Look to Other Disciplines like Meteorology.” LSE American Politics
and Policy.

Lienesch, Michael. 1982. “Right-Wing Religion: Christian Conservatism as a Political
Movement.” Political Science Quarterly 97(3): 403–25.

Litton, Edward et al. 2021. “Increasing ICU Capacity to Accommodate Higher Demand during
the COVID‐19 Pandemic.” Medical Journal of Australia 215(11): 513–17.

Lockerbie, Brad. 2013. “Race and Religion: Voting Behavior and Political Attitudes.” Social
Science Quarterly (Wiley-Blackwell) 94(5): 1145–58.
81

Luskin, Robert C. 1990. “Explaining Political Sophistication.” Political Behavior 12(4): 331–61.

MacWilliams, Matthew. 2015. “Forecasting Congressional Elections Using Facebook Data.” PS:
Political Science and Politics. https://www.cambridge.org/core/journals/ps-political-science-
and-politics/article/abs/forecasting-congressional-elections-using-facebook-
data/A39F358F7A8AE6B39B19C2535F013765 (May 2, 2022).

Madson, Gabriel J., and D. Sunshine Hillygus. 2020. “All the Best Polls Agree with Me: Bias in
Evaluations of Political Polling.” Political Behavior 42(4): 1055–72.

Marchlewska, Marta, Aleksandra Cichocka, and Małgorzata Kossowska. 2018. “Addicted to
Answers: Need for Cognitive Closure and the Endorsement of Conspiracy Beliefs: Need for
Cognitive Closure and Conspiracy Beliefs.” European Journal of Social Psychology 48(2):
109–17.

Margolis, Michele F. 2020. “Who Wants to Make America Great Again? Understanding
Evangelical Support for Donald Trump.” Politics and Religion 13(1): 89–118.

Marti, Gerardo. 2019. American Blindspot: Race, Class, Religion, and the Trump Presidency.
Rowman & Littlefield.

Masuoka, Natalie, and Jane Junn. 2013. The Politics of Belonging: Race, Public Opinion, and
Immigration. University of Chicago Press.

McCarthy, Molly, Kristina Murphy, Elise Sargeant, and Harley Williamson. 2022. “Examining
the Relationship between Conspiracy Theories and COVID-19 Vaccine Hesitancy: A
Mediating Role for Perceived Health Threats, Trust, and Anomie?” Analyses of Social
Issues and Public Policy 22(1): 106–29.

McDaniel, Eric L. 2008. Politics in the Pews: The Political Mobilization of Black Churches.

McKenzie, Brian D., and Stella M. Rouse. 2013. “Shades of Faith: Religious Foundations of
Political Attitudes among African Americans, Latinos, and Whites: Shades of Faith.”
American Journal of Political Science 57(1): 218–35.

Meijer, Erik. 2016. “Weighting the Daybreak Poll.”

Mendez, Michael. 2005. “Latino New Urbanism: Building on Cultural Preferences.” Opolis 1(1).
https://escholarship.org/uc/item/0mz4k5pb (April 26, 2022).

Michelson, Melissa R. 2003. “Getting Out the Latino Vote: How Door-to-Door Canvassing
Influences Voter Turnout in Rural Central California.” Political Behavior 25(3): 17.

Michelson, Melissa R. 2005. “Meeting the Challenge of Latino Voter Mobilization.” The Annals
of the American Academy of Political and Social Science 601(1): 85–101.
82

Michelson, Melissa R. 2006. “Mobilizing the Latino Youth Vote: Some Experimental Results.”
Social Science Quarterly (Wiley-Blackwell) 87: 1188–1206.

Miller, Joanne M. 2020. “Do COVID-19 Conspiracy Theory Beliefs Form a Monological Belief
System?” Canadian Journal of Political Science/Revue canadienne de science politique
53(2): 319–26.
Miller, Joanne M., Kyle L. Saunders, and Christina E. Farhart. 2016. “Conspiracy Endorsement
as Motivated Reasoning: The Moderating Roles of Political Knowledge and Trust:
Conspiracy Endorsement as Motivated Reasoning.” American Journal of Political Science
60(4): 824–44.

Min, Seong Jae. 2021. “Who Believes in Conspiracy Theories? Network Diversity, Political
Discussion, and Conservative Conspiracy Theories on Social Media.” American Politics
Research 49: 415–27.

Monforti, Jessica Lavariega, and Gabriel R. Sanchez. 2010. “The Politics of Perception: An
Investigation of the Presence and Sources of Internal Discrimination among Latinos.” Social
Science Quarterly: 245–65.

Monforti, Jessica Lavariega. 2017. “The Latina/o Gender Gap in the 2016 Election.” Aztlán: A
Journal of Chicano Studies 42:2.

Morgan, David R, and Kenneth J. Meier. 1980. “Politics and Morality: The Effect of Religion on
Referenda Voting.” Social Science Quarterly 61(1): 144-48

Mude, William et al. “Racial Disparities in COVID-19 Pandemic Cases, Hospitalizations, and
Deaths: A Systematic Review and Meta-Analysis.” Journal of Global Health 11: 05015.

Myers, Cayce, and Karen Russell. 2019. “Image Repair in the Aftermath of Inaccurate Polling:
How the News Media Responded to Getting It Wrong in 1948 and 2016.” Journal of
Political Marketing 18(1–2): 148–77.

Newman, Brian. 2009. “Review of A Matter of Faith: Religion in the 2004 Presidential
Election.” Presidential Studies Quarterly 39(1): 155–57.

Nuño, Stephen A. 2007. “Latino Mobilization and Vote Choice in the 2000 Presidential
Election.” American Politics Research 35(2): 273–93.

Nuño, Stephen A. 2007. “Latino Mobilization and Vote Choice in the 2000 Presidential
Election.” American Politics Research 35(2): 273–93.

Oliver, Eric J and Wood, Thomas. 2014. “Medical Conspiracy Theories and Health Behaviors in
the United States | Cancer Screening, Prevention, Control.” JAMA: Internal Medicine.

83

Oliver, J. Eric, and Thomas J. Wood. 2014. “Conspiracy Theories and the Paranoid Style(s) of
Mass Opinion.” American Journal of Political Science 58(4): 952–66.

Olson, Laura, and John Green. 2006. “The Religion Gap.” PS: Political Science & Politics 39:
455–59.

Olsson, Henrik, W. Bruine de Bruin, M. Galesic, and D. Prelec. 2021. “Election Polling Is Not
Dead: A Bayesian Bootstrap Method Yields Accurate Forecasts.” Preprint at https://doi.
org/10.31219/osf. io/nqcgs.

Olsson, Henrik, Wandi Bruine de Bruin, Mirta Galesic, and Drazen Prelec. 2019. “Harvesting
the Wisdom of Crowds for Election Predictions Using the Bayesian Truth Serum.”

Ornelas, India J., and Gbenga Ogedegbe. 2021. “Listening to Latinx Patient Perspectives on
COVID-19 to Inform Future Prevention Efforts.” JAMA Network Open 4(3): e210737.

Pachon, Harry, and Louis DeSipio. 1992. “Latino Elected Officials in the 1990s.” PS: Political
Science and Politics 25(2): 212–17.

Palfrey, Thomas R. “A Mathematical Proof of Devuger’s Law.” Cal Tech. 1988.

Park, David K., Andrew Gelman, and Joseph Bafumi. 2017. “Bayesian Multilevel Estimation
with Poststratification: State-Level Estimates from National Polls.” Political Analysis 12(4):
375–85.

Parsons, Sharon et al. 1999. “A Test of the Grapevine: An Empirical Examination of Conspiracy
Theories Among African Americans.” Sociological Spectrum 19: 201–22.

Parsons, Timothy. 2010. The Rule of Empires: Those Who Built Them, Those Who Endured
Them, and Why They Always Fall. Oxford University Press.

Patrikios, S. (2013). “Self-stereotyping as ‘Evangelical Republican’: An Empirical Test.” Politics
and Religion, 6(4), 800-822. doi:10.1017/S1755048313000023

Pennycook, Gordon, and David Gertler Rand. 2021. “Examining False Beliefs about Voter Fraud
in the Wake of the 2020 Presidential Election.” Trends in Cognitive Sciences.

Persily, Nathaniel, and Joshua A. Tucker. 2020. Social Media and Democracy: The State of the
Field, Prospects for Reform. Cambridge University Press.

Peterson, Robert A., and Steven P. Brown. 2005. “On the Use of Beta Coefficients in Meta-
Analysis.” Journal of Applied Psychology 90(1): 175–81.

Petrocik, John Richard. 2009. “Measuring Party Support: Leaners Are Not Independents.”
Electoral Studies 28(4): 562–72.

84

Podewils, Laura Jean et al. 2020. “Disproportionate Incidence of COVID-19 Infection,
Hospitalizations, and Deaths Among Persons Identifying as Hispanic or Latino — Denver,
Colorado March–October 2020.” Morbidity and Mortality Weekly Report 69(48): 1812–16.

Pollock III, Philip H., and Barry C. Edwards. 2019. The Essentials of Political Analysis. CQ
Press.

Post, David. 1990. “College-Going Decisions by Chicanos: The Politics of Misinformation.”
Educational Evaluation and Policy Analysis 12(2): 174–87.

Reverby, Susan M. 2009. Examining Tuskegee: The Infamous Syphilis Study and Its Legacy.
Univ of North Carolina Press.

Richard A. Friedman. 2021. “Why Humans Are Vulnerable to Conspiracy Theories.” Psychiatry
online. https://ps.psychiatryonline.org/doi/epub/10.1176/appi.ps.202000348 (May 2, 2022).
Riker, William H. 1982. “The Two-Party System and Duverger’s Law: An Essay on the History
of Political Science.” American Political Science Review 76(4): 753–66.

“Risk for COVID-19 Infection, Hospitalization, and Death by Race/Ethnicity.” 2022. Centers for
Disease Control and Prevention. https://www.cdc.gov/coronavirus/2019-ncov/covid-
data/investigations-discovery/hospitalization-death-by-race-ethnicity.html (March 26,
2022).

Robert Courtney Smith. 2017. “‘Don’t Let the Illegals Vote!’: The Myths of Illegal Latino
Voters and Voter Fraud in Contested Local Immigrant Integration.” RSF: The Russell Sage
Foundation Journal of the Social Sciences 3(4): 148.

Rodriguez-Diaz, Carlos E. et al. 2020. “Risk for COVID-19 Infection and Death among Latinos
in the United States: Examining Heterogeneity in Transmission Dynamics.” Annals of
Epidemiology 52: 46-53.

Romer, Daniel, and Kathleen Hall Jamieson. 2020. “Conspiracy Theories as Barriers to
Controlling the Spread of COVID-19 in the U.S.” Social Science & Medicine.

Rosenblum, Nancy, and Russell Muirhead. 2021. “Conspiracy and Democracy: Same as
Always? We Don’t Think So.” The American Journal of Psychology 134(4): 507–10.

Rozell, M.J. (2018). “Donald J. Trump and the Enduring Religion Factor in US Elections.” In:
Religion and the American Presidency: The Evolving American Presidency. Palgrave
Macmillan, Cham. https://doi.org/10.1007/978-3-319-62175-3_14

Rozell, Mark J., and Clyde Wilcox. 1996. “Second Coming: The Strategies of the New Christian
Right.” Political Science Quarterly 111(2): 271–94.

Sanchez, Gabriel R., and Barbara Gomez-Aguinaga. 2020. “The Latino Rejection of the Trump
Campaign: Latinos and the 2016 Election.” Aztlán: A Journal of Chicano Studies : 25–48.
85

Schmidt, Ronald, Edwina Barvosa-Carter, and Rodolfo D. Torres. 2000. “Latina/o Identities:
Social Diversity and U.S. Politics.” PS: Political Science & Politics 33(3): 563–67.

Schmith, Virginia D., Jie (Jessie) Zhou, and Lauren R.L. Lohmer. 2020. “The Approved Dose of
Ivermectin Alone Is Not the Ideal Dose for the Treatment of COVID-19.” Clinical
Pharmacology & Therapeutics 108(4): 762–65.

Schwadel, Philip. 2017. “The Republicanization of Evangelical Protestants in the United States:
An Examination of the Sources of Political Realignment.” Social Science Research 62: 238–
54.

Scoble, Harry M., and Leon D. Epstein. 1964. “Religion and Wisconsin Voting in 1960.” The
Journal of Politics 26(2): 381–96.

Seitz, Amanda and Will Weissert. (2021). “Inside the ‘big Wave’ of Misinformation Targeted at
Latinos” AP News. https://apnews.com/article/latinos-misinformation-election-
334d779a4ec41aa0eef9ea80636f9595 (April 25, 2022).

Seña, Arlene C., and David J. Weber. 2021. “From Health Disparities to Hotspots to Public
Health Strategies: The Impact of the COVID-19 Pandemic in North Carolina.” North
Carolina Medical Journal 82(1): 37–42.

Shaw, Todd, Louis DeSipio, Dianne Pinderhughes, and Toni-Michelle C. Travis. 2018. Uneven
Roads: An Introduction to U.S. Racial and Ethnic Politics. CQ Press.

Sherkat, Darren E. 2017. “Intersecting Identities and Support for Same-Sex Marriage in the
United States.” Social Currents 4(4): 380–400.

Shrestha, Maniz. 2018. “Detecting Fake News with Sentiment Analysis and Network Metadata.”

Silver, Nate. 2016. “2016 Election Forecast.” FiveThirtyEight.
https://projects.fivethirtyeight.com/2016-election-forecast/ (May 2, 2022).

Simmons, William, and Sharon Parsons. 2005. “Beliefs in Conspiracy Theories Among African
Americans: A Comparison of Elites and Masses.” Social Science Quarterly 86: 582–98.

Sinclair, Betsy, Margaret McConnell, and MelissaR. Michelson. 2013. “Local Canvassing: The
Efficacy of Grassroots Voter Mobilization.” Political Communication 30(1): 42–57.

Smallpage, Steven M., Hugo Drochon, Joseph E. Uscinski, and Casey Klofstad. 2020. “Who Are
the Conspiracy Theorists?: Demographics and Conspiracy Theories.” In Routledge
Handbook of Conspiracy Theories. Routledge Press.

Smidt, Corwin E., et al. The Disappearing God Gap?: Religion in the 2008 Presidential Election.
Oxford University Press, 2010.
86

Sondheimer, Rachel Milstein, and Donald P. Green. 2010. “Using Experiments to Estimate the
Effects of Education on Voter Turnout.” American Journal of Political Science 54(1): 174–
89. doi: 10.1111/j.1540-5907.2009.00425.x.

Stern, Chadly, Tessa V. West, John T. Jost, and Nicholas O. Rule. 2014. “‘Ditto Heads’: Do
Conservatives Perceive Greater Consensus Within Their Ranks Than Liberals?” Personality
and Social Psychology Bulletin 40(9): 1162–77.

Terhanian, George. 2017. “What Survey Researchers Can Learn From the 2016 U.S. Pre-
Election Polls: How to Fine-Tune Survey Methods and Restore Credibility.” Journal of
Advertising Research 57(2): 182–89.

Thompson, Kenneth W. 1986. “Religion and Politics in the United States: An Overview.” The
Annals of the American Academy of Political and Social Science 483: 12–24.

“USC Dornsife - Election 2016| Understanding America Study.” 2016. https://election.usc.edu/
(May 2, 2022).

“USC Dornsife - Election 2020 | Understanding America Study.” https://election.usc.edu/ (May
2, 2022).

“USC Dornsife 2020 Presidential Poll.” California, Nov. 2020.

Uscinski, Joseph E. et al. 2020. “Why Do People Believe COVID-19 Conspiracy Theories?”
Harvard Kennedy School Misinformation Review.
https://misinforeview.hks.harvard.edu/?p=1314 (April 25, 2022).

Uscinski, Joseph E., and Joseph M. Parent. 2014. American Conspiracy Theories. Oxford
University Press.

Valentino, Nicholas A., John Leslie King, and Walter W. Hill. 2017. “Polling and Prediction in
the 2016 Presidential Election.” Computer 50(5): 110–15.

Van der Linden, Sander, Graham Dixon, Chris Clarke, and John Cook. 2021. “Inoculating
against COVID-19 Vaccine Misinformation.” Clinical Medicine 33: 100772.

Van Prooijen, Jan-Willem, Tom W Etienne, Yordan Kutiyski, and André PM Krouwel. 2022.
“Just a Flu? Self-Perceived Infection Mediates the Link between Conspiracy Beliefs and
Covid-19 Health Beliefs and Behaviors.” Journal of Health Psychology 27(6): 1421–31.

Vargas, Edward D., and Gabriel R. Sanchez. 2020. “COVID-19 Is Having a Devastating Impact
on the Economic Well-Being of Latino Families.” Journal of Economics, Race, and Policy
3(4): 262–69.

87

Voss, Stephen, Andrew Gelman, and Gary King. 1995. “The Polls- A Review: Preelection
Survey Methodology: Details from Eight Polling Organizations, 1988 and 1992.” Public
Opinion Quarterly 59(1): 98–132.

Wald, Kenneth D, and Allison Calhoun-Brown. 2014. Religion and Politics in the United States.
Rowman & Littlefield.

Wang, Wei, David Rothschild, Sharad Goel, and Andrew Gelman. 2015. “Forecasting Elections
with Non-Representative Polls.” International Journal of Forecasting 31(3): 980–91.

Wertheimer, Albert. 2022. “Conspiracy Theories, Misinformation and the Decline for Respect of
Science.” Journal of Pharmaceutical Health Services Research 13(1): 1.

Whitehead, Andrew L, Samuel L Perry, and Joseph O Baker. 2018. “Make America Christian
Again: Christian Nationalism and Voting for Donald Trump in the 2016 Presidential
Election.” Sociology of Religion 79(2): 147–71.

Wood, Michael J. 2016. “Some Dare Call It Conspiracy: Labeling Something a Conspiracy
Theory Does Not Reduce Belief in It.” Political Psychology 37(5): 695–705.

Ylanan, Aida, Thomas Suh Lauder, and Andrea Roberson. 2020. “2020 Polls: USC Daybreak on
the Election, Trump and Biden.” Los Angeles Times. https://www.latimes.com/projects/usc-
presidential-election-poll-2020/.

Young, Dannagal G., Huma Rasheed, Amy Bleakley, and Jessica B. Langbaum. 2022. “The
Politics of Mask-Wearing: Political Preferences, Reactance, and Conflict Aversion during
Covid.” Social Science & Medicine 298: 114836.

Asset Metadata

Creator Cuellar, Jarred R. (author)

Core Title La eleccion de la pandemia: analyzing Latino political behavior during the 2020 election

Contributor Electronically uploaded by the author (provenance)

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Political Science and International Relations

Degree Conferral Date 2022-08

Publication Date 07/22/2022

Defense Date 05/10/2022

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

Tag 2020,anomie,Biden,conspiracy theory,COVID,Democrats,evangelical,Latino politics,misinformation,oai:digitallibrary.usc.edu:usctheses,OAI-PMH Harvest,pandemic,political behavior,political science,politics,polling,polling techniques,presidential election,Religion,Republican,Trump,USC Poll,voting,voting behavior

Format application/pdf (imt)

Language English

Advisor Hancock Alfaro, Ange-Marie (committee chair), DeSipio, Louis (committee member), Grose, Christian (committee member), Lo, James (committee member), Ruddell, Darren (committee member)

Creator Email jrcuella@usc.edu,jrcuellar@cpp.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC111373887

Unique identifier UC111373887

Legacy Identifier etd-CuellarJar-10902

Document Type Dissertation

Format application/pdf (imt)

Rights Cuellar, Jarred R.

Type texts

Source 20220722-usctheses-batch-960 (batch), University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright. The original signature page accompanying the original submission of the work to the USC Libraries is retained by the USC Libraries and a copy of it may be obtained by authorized requesters contacting the repository e-mail address given.

Repository Name University of Southern California Digital Library

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

Repository Email uscdl@usc.edu

Abstract (if available)

Abstract While the study of American politics has long studied the effects of religion on political behavior, there is a dearth of knowledge on how religion affects Latino political behavior, given the long-held belief that the Latino community is a religious monolith, with Latinos being overwhelming Catholic in faith. This has caused a gap in knowledge in which we know relatively little about how religious denomination may impact voter behavior. In this dissertation, I will go onto show that Evangelical Latinos were significantly more likely to vote for Trump. While those under 45 do not follow this trend, my findings highlight the necessity to stop omitting religious denomination from further Latino politics studies.
Similar to the study of Evangelical Latino voting behavior, there is not much known about Latinos and what drives their belief in conspiracy theories. While Latinos have been found to be more susceptible to conspiratorial thinking, we are still unsure what causes Latinos this phenomenon. 2020 saw the emergence of numerous conspiracy theories related to Covid-19 and the 2020 election. With this said, I use my second article (Chapter 3) to show that college educated Latinos, and those who are associated with the Democratic Party are less likely to subscribe to contemporary conspiracy theories. In contrast, I show that Latinos with an external locus of control are more likely to believe in these Covid related conspiracy theories.
Recently we have seen much of the public lose faith in pre-election polls due to incorrect forecasts. In this article I show that inaccurate survey weighting may not actually be to blame, like many pollsters believe. Rather, in this new era of hyper-partisanship it appears that many Latino respondents do not feel comfortable disclosing their true intentions. I use the experimental “social circle” question from the USC Poll to show that it may be time to ask new questions that more accurately forecast, not only election results, but also the Latino vote.