ResearchPDF Available

¡Vota! Testing the Impact of Spanish-Language Ballots on Election Results and Preferences



Present research related to racial group interactions has pointed to the existence of a group status threat; when a majority group feels threatened, it takes action to protect its status. This backlash has been observed in white people when primed with an increase in Hispanic immigration or demographic statistics indicating that white people as a racial group will no longer be in the numerical majority. Given these demographic trends and prior literature, this study investigates if backlash can occur against Spanish-speaking populations in elections when voters are exposed to an English-Spanish bilingual ballot. Utilizing Amazon’s Mechanical Turk, 723 participants voted in a mock election with either a monolingual English or bilingual ballot. The ballot contained two races: a mayoral race with a Democrat and Republican and a proposition to support Deferred Action on Childhood Arrivals (DACA). Participants then answered a series of questions about their political ideology and perceptions of group status. The results from the election and post-election questionnaire indicate that white, non-Spanish-speaking conservatives report a higher group status threat in the post-election survey, in line with previous literature. In terms of altering election results, we found that the mayoral race was not significantly affected by the presence of Spanish. However, white, non-Spanish-speaking moderates tended to vote against DACA when exposed to Spanish. As demographics change in the United States, there may be an increased need for bilingual ballots as stipulated under the Voting Rights Act, and given the results of this study, that increase could have electoral consequences.
¡Vota! 1
¡Vota!: Testing the Impact of Spanish-Language Ballots on Election
Results and Preferences
Connor Rothschild, Sarah Berton, Maddy Scannell, Eric Stone
April 24, 2019
Present research related to racial group interactions has pointed to the existence of a group status
threat; when a majority group feels threatened, it takes action to protect its status. This backlash
has been observed in white people when primed with an increase in Hispanic immigration or
demographic statistics indicating that white people as a racial group will no longer be in the
numerical majority. Given these demographic trends and prior literature, this study investigates if
backlash can occur against Spanish-speaking populations in elections when voters are exposed to
an English-Spanish bilingual ballot. Utilizing Amazon’s Mechanical Turk, 723 participants voted
in a mock election with either a monolingual English or bilingual ballot. The ballot contained
two races: a mayoral race with a Democrat and Republican and a proposition to support Deferred
Action on Childhood Arrivals (DACA). Participants then answered a series of questions about
their political ideology and perceptions of group status. The results from the election and
post-election questionnaire indicate that white, non-Spanish-speaking conservatives report a
higher group status threat in the post-election survey, in line with previous literature. In terms of
altering election results, we found that the mayoral race was not significantly affected by the
presence of Spanish. However, white, non-Spanish-speaking moderates tended to vote against
DACA when exposed to Spanish. As demographics change in the United States, there may be an
increased need for bilingual ballots as stipulated under the Voting Rights Act, and given the
results of this study, that increase could have electoral consequences.
¡Vota! 2
The population of Americans who are not proficient in English has steadily increased
over time (Zong and Batalova, 2017). Thus, a large group of eligible voters may face challenges
in obtaining information about voter registration, precincts, and candidates, as well as difficulties
in simply completing a ballot. The barriers to voting for language minorities have been
recognized in some respects, as demonstrated by Section 203 of the Voting Rights Act (VRA),
which requires a covered state or jurisdiction to provide registration and voting information,
materials, and assistance in the language of the covered group (U.S. Department of Justice,
2018). Provisions such as bilingual election materials and oral assistance have proven helpful in
increasing registration and turnout among language minorities in jurisdictions covered by the
language provisions of the VRA. However, little is known about how increased accessibility for
language minorities—and more specifically, how the increased presence of languages other than
English on election materials—impacts the attitudes and policy preferences of other groups. In
Daniel Hopkins’ 2011 study on the impact of bilingual election materials on turnout of language
minorities and overall election outcomes, the author specifically calls for future work which
explores the potential for backlash of other groups as a result of seeing Spanish on a ballot.
These potential backlash effects are important to study given the continued expansion of
provisions and initiatives which reduce barriers to registration and voting for language
minorities. We must understand the potential effects of such an expansion, with particular
attention to the several theoretical arguments suggesting that white people may act to mitigate
any perceived threat to their dominant status in society.
With this in mind, the present article will explore the concept of intergroup emotions
theory as it relates to election behavior and policy preferences for non-Spanish speaking voters.
More specifically, we will investigate if exposure to Spanish on an ordinarily monolingual
English ballot has a measurable effect on election results and preferences for conservative
policies. The paper will begin by providing a background on the language provisions of the VRA
and other language assistance initiatives as well as a review of the relevant literature regarding
the impact of these provisions. We will then present an overview of the relevant theories of
intergroup emotions theory before developing our own hypotheses. After outlining our
methodology for conducting this investigation, we will present and analyze the results of our
study. We end with by outlining potential limitations of our study, key findings, and suggestions
for future research.
Literature Review
Voting Provisions
The most salient piece of legislation affecting voter access to the polls in the 20th century
was surely the Voting Rights Act of 1965 (VRA) and its amendments. After years of judicial
challenges and amendments to the law, the VRA outlawed literacy tests as a prerequisite for
¡Vota! 3
voting, laying the groundwork for multilingual voting. In 1975, Congress amended the VRA to
add Section 203, which laid out language assistance provisions. The amendment, triggered by the
presence of specified levels of low-English proficiency (LEP) citizens of voting age on a
jurisdiction-by-jurisdiction basis, prohibits certain jurisdictions from conducting elections
exclusively in English. The jurisdiction in which the election is conducted is allowed to
determine the manner in which language assistance is provided, but the VRA requires that “all
voting materials,” including “registration or voting notices, forms, instructions, assistance, or
other materials or information relating to the electoral process, including ballots” be provided in
the languages of the minority groups that triggered VRA coverage in addition to English (Tucker
Previous research has indicated that language provisions established in the VRA and
additional steps taken in certain jurisdictions to provide bilingual election materials have
increased accessibility to the polls for voters with limited English skills. The effectiveness of
language provisions is especially important given the United States’ current landscape, in which
a growing number of citizens are the native-born children of immigrants or are naturalized
immigrants themselves. The United States is home to more than 38 million immigrants, eight
million of whom speak little to no English (Hopkins 2011). Additionally, a large portion of these
immigrants are Spanish speakers (Parkin and Zlotnick 2011)—in fact, the presence of a large
number of Spanish-speaking citizens in Texas was part of the rationale for the 1975 addition of
Section 203 to the VRA (Tucker 2006). Recent research has consistently found a relationship
between the ability to speak and read English and one’s likelihood of electoral participation
(Hopkins 2011; Parkin and Zlotnick 2011), suggesting that language differences present a
significant barrier to political involvement for a substantial and growing portion of the United
States citizenry.
There exists a large body of inquiry into the impact and effectiveness of provisions for
language minorities on two aspects of electoral participation: registration and turnout. Previous
research has generally found that bilingual election materials have a significant and positive
effect on the participation of language minorities, and Spanish speakers in particular (Hopkins
2011; Jones-Correa 2005). This means that the impact of language provisions is most clear in
increasing basic access to registration and voting for those not proficient in English, while results
are mixed regarding effects on policy, representation, or efficacy for the entire racial group.
Recently, Hopkins (2011) examined California’s Proposition 227, a 1998 initiative which
restricted bilingual education. The study found that neighborhoods with greater
Spanish-language assistance (including bilingual ballots) were, as predicted, more supportive of
bilingual education. In other words, support for the ballot initiative was markedly lower in
jurisdictions with many Spanish speakers enfranchised by the VRA’s language provisions. This
suggests that language provisions, once proven to be effective in providing access to the polls,
may actually influence policy preferences and election outcomes. Ultimately, it is clear that
language provisions mitigate the barrier for voters with low English proficiency in receiving
¡Vota! 4
election information, instructions, and sample ballots. Several studies even suggest that this
effect increases as English proficiency in an area decreases (Parkin and Zlotnick 2011; Hopkins
2011). More specifically, while language provisions may have a positive impact on Latinos in
general, their influence is more substantial in communities where there are many
low-English-proficiency Spanish speakers. Thus, if the goal of election administrators is to
increase turnout among limited English proficiency (LEP) populations, it is of critical
importance to utilize language-assistance procedures when possible.
It is difficult to measure the direct symbolic and instrumental effects of language
provisions on the electoral participation of language minorities, and results have at times been
mixed. However, in general, the literature finds a positive treatment effect of language provisions
(Fraga and Merseth 2016). Despite this, barriers to registration and turnout remain relatively high
for populations with low English skills, suggesting that Section 203 is insufficient for mitigating
these barriers entirely, especially in jurisdictions without coverage in which non-English
speakers tend to be the most isolated (Parkin and Zlotnick 2011). Thus, there is a definite need
for greater expansion and incorporation of bilingual election materials and language assistance in
covered and
non-covered jurisdictions. When considering the benefit such an expansion might
have, it is just as important to consider any consequences that might occur as a result of
increased language accessibility. Research has yet to explore the effects language-assistance
procedures may have on collateral groups. In fact, in his conclusion, Hopkins called for “future
work at the individual level...” to explore “... the conditions under which seeing Spanish is likely
to produce symbolic effects among Latinos or backlash effects among other groups” (2011, p.
827). Thus, before issuing a wide call for the expansion of bilingual election materials as a
method of minority enfranchisement, we must examine its impacts on the majority in the ballot
box. This study hopes to explore these potential backlash effects.
Group Status Threat
Research has consistently demonstrated that white Americans respond negatively to the
threat of status loss (Wetts and Willer 2018). Extant literature has also suggested that those
perceiving threat are more likely to express support for conservative policies and candidates.
Extensive research into concepts such as mortality salience (Landau et al. 2004; Pyszczynski et
al. 2006) and system instability (Jost et al. 2003) has consistently shown that Americans become
more conservative when they perceive threat. Recent research, for example, has revealed that
group status threat—not economic hardship—was responsible for the rise and eventual
presidency of Donald Trump (Mutz 2018).
The growing enfranchisement of racial minorities in American society may be a trend
which increases perceptions of group status threat. Indeed, studies suggest that the presence of
minorities and the concept of growing diversity trigger anxiety among white Americans. Whites
living in areas with more Latinos are less supportive of liberal immigration reform and more
likely to peddle negative stereotypes about Latinos (Burns and Gimpel 2000). Outside of actual
¡Vota! 5
shifts in racial demography, the simple perception of a growing racial minority population
decreases whites’ support for immigration related to those populations (Alba, Rumbaut, and
Marotz 2005). Increases in immigration are also linked to subsequent increases in hate crimes
against the immigrating population (Stacey, Carbone-Lopez, and Rosenfeld 2011; Citrin et al.
1997). While interactions with people of diverse backgrounds may make people more
sympathetic toward and supportive of those populations (Pettigrew 1998; Dixon 2006; Berg
2009), simply living in close proximity to those populations does not result in increased
supportiveness—instead, the aforementioned studies suggest the opposite.
Experimental studies have also supported these findings. Enos (2014) has shown the
simple presence of two additional Spanish speakers on one’s daily commute can evoke marked
increases in anti-immigrant policy preferences. One novel study provided subjects an unedited
photo of President Barack Obama and one with artificially darkened skin; survey respondents in
the dark skin condition were significantly more supportive of the Tea Party than the control
(Willer, Feinberg, and Wetts 2016). Exposing white Americans to population projections in
which they become a numerical minority engenders feelings of anger and fear toward minority
groups (Outten et al. 2018). Interestingly, the same conservative shift is seen among
non-Hispanic racial minorities when they perceive an increase in Hispanic populations (Craig &
Richeson 2017).
Intergroup Emotions Theory (IET) describes “the uniquely group-level nature of the
categorization, identification, and appraisal antecedents of intergroup emotions as well as their
consequences for intergroup relations” (Mackie and Smith 2015, p. 264). In other words,
individuals tend to interpret events in the context of their social identities and the implications
for their ingroup, rather than individual-level implications. Related to IET is the hypothesis that a
dominant ingroup may perceive the growing clout of outgroups as a threat and respond
accordingly (Smith 1993). A host of literature has followed which documents a variety of
responses to outgroups. In sum, the attitudes and behaviors affected by intergroup contact vary
contextually—depending on the circumstances, different emotions will evoke different actions
(Mackie et al. 2008, p. 1875). Indeed, research has found that intergroup contact reduces an
ingroup’s levels of anger and fosters improved attitudes toward the outgroup under certain
circumstances (Tam et al. 2007).
Still, many studies have shown how anger toward an outgroup predicts support for
collective hostile action against that group (Van Zomeren et al. 2007; Miller et al. 2009). Mackie
and colleagues (2008) have argued that anger driven by outgroup hostility and intergroup anxiety
results in attempts to subdue the relevant outgroup, perhaps taking form in physical force or
hostile government policies (the authors use the example of immigration policy). This follows
the research of Smith (1993), in which it is shown that where appraisals of an outgroup lead to
prejudiced emotion toward that outgroup which eventually leads to discrimination.
Moreover, we know that these attitudes are likely to be expressed at the ballot box. As
early as 1949, V.O. Key’s research revealed a relationship between the presence of blacks in a
¡Vota! 6
community and increases in white turnout and votes for segregationist candidates (Key 1949).
Substantiating Key’s research three decades later, Knoke and Kyriazis (1977) explored the
successive presidential campaigns of George Wallace and found voting tendencies among
Southern whites were intended to suppress and dominate black citizens. Combined with
Hopkins’ 2011 call to research, we choose to examine if Spanish language provisions on ballots
create outgroup backlash in an experimental setting.
Based upon our research question and our understanding of the literature, we propose three
hypotheses. Hypothesis one is that the presence of Spanish on a traditionally monolingual
English ballot will mediate a shift toward conservative policy preferences, with this shift most
directly manifesting itself in anti-immigrant sentiment. Hypothesis two is that white,
non-Spanish-speaking voters who vote on a bilingual ballot will perceive a greater threat to their
racial identity than those who vote on a monolingual English ballot. Finally, hypothesis three is
that this effect will also be seen, albeit less pronounced, among non-Hispanic racial minorities.
We conducted our study using Amazon’s Mechanical Turk (MTurk), an online survey
platform (n = 723; 371 men, 352 women; mean age = 40.18 years). Participants from MTurk
were directed to a survey hosted on Qualtrics ( Respondents were paid
between $0.20 and $1.00 for their participation (with the variation depending on activity on the
site). The majority were paid $0.50 for their participation.
Amazon’s Mechanical Turk has been used by a variety of researchers in the social
sciences. Studies have shown that data gathered from MTurk is reliable (Clifford, Jewel, &
Waggoner 2015) and that its participant pool is “generally more diverse” than in-person
convenience samples (Cassese et al. 2013, p. 776). To enhance our results, we utilized
TurkPrime ( which allows for the fine-tuning of MTurk study details (e.g.
the distribution of bonuses, demographic balancing, etc.) and thus enhances the quality of data
(Litman, Robinson, & Abberbock 2017).
Participants agreed to an electronic consent form prior to beginning the survey. We
restricted eligibility to people located in the United States who were aged 18 or over in order to
match eligibility requirements for voting in the United States. The survey was conducted from
November 18th to December 10th.
Survey Design
¡Vota! 7
Participants were randomly redirected to one of two Qualtrics surveys corresponding to
the treatment conditions. One survey was entirely in English (monolingual condition), and the
other included a bilingual ballot with English and Spanish (bilingual condition).
We modeled the bilingual ballot off of a typical two column design, with English on the
left and Spanish on the right.
Ballot Measures
There were two measures on the ballot: a mayoral election and a proposition. The first
ballot measure was the mayoral election with two fictional candidates, a Democrat and a
Republican. Prior to asking respondents to vote, we presented a voter guide for each candidate.
We included their education and professional biographies, as well as a fictional conservative or
liberal rating. Figure 1 shows the voter guides as they appeared on the ballots. Figures 2 and 3
show the mayoral race and DACA proposition, respectively, in the bilingual condition.
¡Vota! 8
Figure 2: Mayoral Race in Bilingual Spanish Condition
Figure 3: DACA Proposition in Bilingual Spanish Condition
¡Vota! 9
We intentionally chose a proposition regarding Deferred Action on Childhood Arrivals (DACA)
for several reasons. First, DACA has high public support among Americans , particularly for an
immigration policy. Thus, we would expect any shifts in support for the proposition to be
attributable to group status threat in the bilingual condition. Second, we wanted a policy about
immigration on the ballot itself to directly measure any shifts in voting outcomes.
Thus, our research design seeks to understand how the presence of Spanish on a
traditionally English ballot may impact policy preferences (in their support of the DACA
proposition) and partisan affiliation (in their support of candidates).
Following their responses to the mock election, participants were directed to a
post-survey which captured basic demographic data. The post-survey also asked questions which
measured political affiliation and ideology. The baseline ideology measure was a question which
asked the respondents to describe their political affiliation (1 = strongly liberal, 5 = strongly
conservative). Borrowing from Outten et al. (2018, p. 4), we included two questions which
functioned as proxies for respondents’ perceived legitimacy of status
and their appraisals of
intergroup threat
. To access feelings related to status legitimacy, we asked respondents to what
extent they agreed with the following statement: “It is quite justified that white Americans have
higher status in society than other racial groups in the U.S.” (1 = strongly disagree, 7 = strongly
agree). Intergroup threat appraisals were measured with responses to the following: “My racial
group should be worried about its place in the future of the U.S.” (1 = strongly disagree, 7 =
strongly agree).
Following these questions, participants indicated other policy preferences by answering
one question related to immigration (“How do you perceive the impact of current levels of
immigration on the United States?”) and two masking questions related to religious expression
and international trade. The immigration question was reverse-coded on a scale of 1, strongly
positive, to 5, strongly negative.
We removed 7 respondents who had IP addresses that indicated they were not from the
United States and removed an additional 29 respondents who failed to complete every part of the
survey, resulting in a final sample of 723 respondents.
We begin by presenting the demographic data for our respondents. Our respondents
roughly mimic the demographic distribution of the United States electorate, with an even gender
distribution (48.5% female, 51.5% male). Our sample underrepresents racial minorities (21.8%
non-white compared to 31% non-white in the 2016 election). Compared to the 2016 election,
our sample underrepresents Hispanic voters by 7.2% and underrepresents African American
1 As of June 2018, 73% of Americans supported allowing Dreamers to stay in the United States, according to a Pew Research Center report.
2 In 2016, the Pew Research Center reported that 69% of the U.S. electorate was white.
¡Vota! 10
voters by 4.5%. Our participant pool is also more left-leaning than the U.S. as a whole, with just
over half of our respondents describing themselves as “strongly” or “moderately” liberal.
Table 1 summarizes the demographic information for our overall participant pool, and Table 2
shows these demographic details broken down by condition.
3 The same Pew report recorded the Hispanic and Black populations’ vote share to be 12% each.
¡Vota! 11
Hypothesis Testing
Hypothesis one and hypothesis two were partially supported, while there was no evidence
to substantiate hypothesis three. Our results indicate that bilingual ballots have mixed effects on
voters depending on race and pre-existing ideology.
Policy Preferences
The results partially support our first hypothesis relating to a rightward shift in policy
preferences upon seeing Spanish on one’s ballot. We began our analysis by examining the
difference in means between the two conditions without controls or subsets. This test failed to
show a statistically significant difference between the two conditions. The results are shown in
Table 3.
We subsequently ran logit models with controls to more accurately detect the effect of
our treatment. In terms of how ballot language would affect an election, we did not find any
significant effect for preferences in the mayoral race. However, support for the proposition
related to DACA was affected by the presence of Spanish. Certain groups of white respondents
who were not proficient in Spanish tended to be more likely to vote against DACA when
exposed to Spanish on a ballot. Notably, there was a 14.47% difference in support for DACA
between white self-described moderates in the bilingual condition and those in the monolingual
¡Vota! 12
condition (M0 = 21.15%; M1 = 35.59%; p = 0.0915). Table 4 illustrates a logit model run on the
interaction between ballot condition and support for DACA among white moderates with little to
no English proficiency (n = 111), with positive values indicating a downward shift in support for
DACA. Figure 4 visually depicts the discrepancy.
Figure 4: DACA Support Among White Self-Described Moderates
This effect was not seen among conservative nor liberal respondents, as Tables 5 and 6 illustrate.
¡Vota! 13
Thus, the results partially support hypothesis one. When exposed to bilingual ballots,
some subgroups of participants (white, self-described moderates) shifted rightward. The finding
was not borne out among any political group other than moderates.
¡Vota! 14
Threat Perception
Hypothesis two predicted that white people would perceive increased threat to their racial
identity upon seeing a bilingual ballot; this hypothesis was also partially supported. We first
examined the simple difference in distributions in our two conditions; the results returned
nothing significant and can be seen in Table 7.
We then ran tests on specific subsets. These tests yielded some significant results.
A chi-squared test revealed white self-identified conservatives voting on bilingual ballots
expressed threat at a level notably higher level (M = 3.93) than those voting on monolingual
English ballots (M = 3.69); this result was significant at the 0.1 level (χ2 = 11.052; df = 6; p =
0.08678). Again, these results were not consistent among all partisan groups: results were not
significant for white liberals (p = 0.3469) nor white moderates (p = 0.94). Individual chi-squared
test outputs are broken down by subset in Table 8.
Effects on Non-Hispanic Racial Minorities
The results presented in Table 8 also indicate that bilingual ballots did not have a
significant effect on the threat perception of racial minorities. This could be in part due to the
¡Vota! 15
small sample of survey respondents who were non-Hispanic minorities (n = 113). Regardless,
our third hypothesis is not supported.
Given the growing non-white and immigrant populations in America, the need for
bilingual ballots is likely to increase in coming years. Our results indicate that bilingual ballots
may have significant electoral consequences, which may influence election administration
decisions. For example, some voting machines offer respondents the opportunity to select their
preferred language on the voting device at the beginning of the process, which would minimize
an English-speaking voter’s exposure to Spanish.
While exposure to Spanish had an effect on white moderates’ propensity to vote against
DACA, no effect was seen on the mayoral race. This disparity may indicate that exposure to
Spanish has a larger effect on voters’ preferences for policies linked to a racial status threat than
on voters’ preference for political party.
Further, our finding that the electoral preferences of conservatives and liberals were not
significantly affected by the presence of Spanish does not support the finding of Enos (2014) that
a fleeting exposure to Spanish produces a shift towards anti-immigrant preferences. This could
be a product of our small sample size—the effect on these populations could be too small for our
study to detect—but it may imply that exposure to a bilingual ballot only produces an effect in
those without pre-existing strong political preferences. Partisans’ policy preferences are
well-established and do not appear to be influenced by the presence of Spanish on a ballot.
Moderates, perhaps because they are not beholden to party loyalty or identification, are more
susceptible to a shift in policy preferences upon seeing a bilingual ballot.
Our third hypothesis about the responses of non-Hispanic racial minorities was not
supported; the backlash effect seen in subgroups of white respondents was not observed among
other races in this study. Although this was due in part to a small sample size, our literature
review indicates that majority groups are most susceptible to group status threat. Simply,
non-Hispanic minority groups may not perceive Spanish-speakers as a threat.
There are limitations to the methods and statistical significance of our study. First and
foremost, while we were able to detect effects at the p
level, none of our findings were
significant at the conventional p
= .05
Though we have set certain checks on respondents to ensure their responses are complete,
legitimate, and that the respondents themselves are within our target population, there are
inherent limits to running experiments on an online platform. Horton et al. (2010) illuminate two
such challenges that are relevant to our study. In an online platform, it is more difficult to
establish a basis of knowledge for all participants, and it is impossible to address questions or
¡Vota! 16
misunderstandings on the part of respondents. Thus, protection against errors as a result of
misunderstanding the instructions and questions cannot be guaranteed. Additionally, on online
platforms, the precise identity of respondents cannot be pinpointed. Though it is unlikely, we
cannot guarantee that respondents worked independently on the survey.
Another limitation concerns the statistical power of our experiment. In particular, this
study found relatively few participants with relatively weak political preferences (lean
conservative, moderate, and lean liberal). Additionally, of our 723 total respondents, only 51
reported speaking Spanish. This makes detecting an effect on Spanish-speaking voters a difficult
task. Finally, the relatively small number of non-Hispanic minority respondents limits our ability
to evaluate our third hypothesis.
Future Research
One of the most important aspects of future research on this topic is the sample size. Our
survey had 723 respondents, but given the small percentage differences we were attempting to
detect, we would need more respondents to see significant results for many of our variables.
Given the real world implications, in which 0.5% differences can win elections, detecting small
effect sizes is crucial.
Though we did not find significant results for non-Hispanic racial minorities, the
contrasting perspectives on the responses of racial minorities to intergroup contact in prior
research offers some insight. Craig and Richeson’s (2017) study, as referenced in our literature
review, suggested that non-Hispanic groups respond negatively to the increasing Hispanic
population. But in 2011, the same authors published a paper discussing intra-minority intergroup
relations, suggesting that minority groups may coalesce against the threat of the majority racial
group (Richeson and Craig 2011). Future research, with a larger sample size, may offer
statistically significant results regarding the question of minority competition or coalition.
Our ballot measure about Deferred Action on Childhood Arrivals was also specifically
chosen in order to target feelings about Hispanic populations and for its general popularity.
Future research should vary ballot measures to include policies less explicitly connected to
Spanish-speaking populations to test the effect of bilingual ballots more generally. This research
design could also be applied to the California proposition about bilingual education studied by
Hopkins (2011) for a more explicit connection.
Other research designs could be effective. Amazon’s Mechanical Turk, although an
effective way to recruit a large, diverse sample of participants, has limitations. Further research
could utilize an in-person experiment to mitigate some of these concerns. Another possibility is
applying a statistical analysis of results in real elections across precincts that offer bilingual or
monolingual ballots.
While this study focused on the presence of Spanish against a monolingual control, future
research may focus on whether varying the amount of Spanish to which a voter is exposed
affects their preferences. Also interesting may be whether the presence of more than two
¡Vota! 17
languages affects voter preferences. A third or fourth language may have a mediating or
compounding effect. More research is needed to model this common real-world situation.
In all, the results of this study suggest that bilingual ballots can have unintended electoral
consequences. Investigating these effects more robustly is critical for increasing accessibility
while limiting undue electoral influence.
¡Vota! 18
Alba, R., Rumbaut, R. G., & Marotz, K. (2005). A Distorted Nation: Perceptions of
Racial/Ethnic Group Sizes and Attitudes Toward Immigrants and Other Minorities.
Social Forces,
(2), 901-919. doi:10.1353/sof.2006.0002
Berg, J. A. (2009). White Public Opinion toward Undocumented Immigrants: Threat and
Interpersonal Environment. Sociological Perspectives, 52
(1), 39-58.
Burns, P., & Gimpel, J. G. (2000). Economic Insecurity, Prejudicial Stereotypes, and Public
Opinion on Immigration Policy. Political Science Quarterly, 115
(2), 201-225.
Cassese, E. C., Huddy, L., Hartman, T. K., Mason, L., & Weber, C. R. (2013). Socially Mediated
Internet Surveys: Recruiting Participants for Online Experiments. PS: Political Science &
(4), 775-784. doi:10.1017/s1049096513001029
Citrin, J., Green, D. P., Muste, C., & Wong, C. (1997). Public Opinion Toward Immigration
Reform: The Role of Economic Motivations. The Journal of Politics, 59
(3), 858-881.
Clifford, S., Jewell, R. M., & Waggoner, P. D. (2015). Are Samples Drawn from Mechanical
Turk Valid for Research on Political Ideology? Research & Politics, 2
(4), 1-9.
Craig, M. A., & Richeson, J. A. (2017). Hispanic Population Growth Engenders Conservative
Shift Among Non-Hispanic Racial Minorities. Social Psychological and Personality
Science, 9
(4), 383-392. doi:10.1177/1948550617712029
¡Vota! 19
Dixon, J. C. (2006). The Ties That Bind and Those That Don’t: Toward Reconciling Group
Threat and Contact Theories of Prejudice. Social Forces, 84
(4), 2179-2204.
Doherty, C. (2018, June 18). 73% Favor Legal Status for Those Brought to US Illegally as
Children. Retrieved November 4, 2018, from
Enos, R. D. (2014). Causal Effect of Intergroup Contact on Exclusionary Attitudes. Proceedings
of the National Academy of Sciences, 111
(10), 3699-3704. doi:10.1073/pnas.1317670111
Fraga, B. L., & Merseth, J. L. (2016). Examining the Causal Impact of the Voting Rights Act
Language Minority Provisions. The Journal of Race, Ethnicity, and Politics, 1
(1), 31-59.
Hopkins, D. J. (2011). Translating into Votes: The Electoral Impacts of Spanish-Language
Ballots. American Journal of Political Science, 55
(4), 814-830.
Horton, J.J., Rand, D.G. & Zeckhauser, R.J. Exp Econ (2011) 14: 399.
Jones-Correa, M. (2005). Language Provisions Under the Voting Rights Act: How Effective Are
They? Social Science Quarterly, 86
(3), 549-564. doi:10.1111/j.0038-4941.2005.00317.x
Key, V. O. (2006). Southern Politics in State and Nation.
Knoxville: University of Tennessee
Knoke, D., & Kyriazis, N. (1977). The Persistence of the Black-Belt Vote: A Test of Key's
¡Vota! 20
Hypothesis. Social Science Quarterly, 57
(4), 899-906. Retrieved from
Landau, M. J., Solomon, S., Greenberg, J., Cohen, F., Pyszczynski, T., Arndt, J., . . . Cook, A.
(2004). Deliver us from Evil: The Effects of Mortality Salience and Reminders of 9/11 on
Support for President George W. Bush. Personality and Social Psychology Bulletin,
(9), 1136-1150. doi:10.1177/0146167204267988
Litman, L., Robinson, J., & Abberbock, T. (2016). A Versatile Crowdsourcing
Data Acquisition Platform for the Behavioral Sciences. Behavior Research Methods,
(2), 433-442. doi:10.3758/s13428-016-0727-z
Mackie, D. M., & Smith, E. R. (2015). Intergroup Emotions. APA Handbook of Personality and
Social Psychology, 2
, 263-293. doi:10.1037/14342-010
Miller, D. A., Cronin, T., Garcia, A. L., & Branscombe, N. R. (2009). The Relative Impact of
Anger and Efficacy on Collective Action is Affected by Feelings of Fear. Group
Processes & Intergroup Relations, 12
(4), 445-462. doi:10.1177/1368430209105046
Mutz, D. C. (2018). Status Threat, Not Economic Hardship, Explains the 2016 Presidential Vote.
Proceedings of the National Academy of Sciences, 115
Outten, H. R., Lee, T., Costa-Lopes, R., Schmitt, M. T., & Vala, J. (2018). Majority Group
Members Negative Reactions to Future Demographic Shifts Depend on the Perceived
Legitimacy of Their Status: Findings from the United States and Portugal. Frontiers in
Psychology, 9
(79). doi:10.3389/fpsyg.2018.00079
Parkin, M., & Zlotnick, F. (2011). English Proficiency and Latino Participation in U.S. Elections.
¡Vota! 21
Politics & Policy, 39
(4), 515-537. doi:10.1111/j.1747-1346.2011.00302.x
Pettigrew, T. F. (1998). Intergroup Contact Theory. Annual Review of Psychology, 49
(1), 65-85.
Pyszczynski, T., Abdollahi, A., Solomon, S., Greenberg, J., Cohen, F., & Weise, D. (2006).
Mortality Salience, Martyrdom, and Military Might: The Great Satan Versus the Axis of
Evil. Personality and Social Psychology Bulletin, 32
(4), 525-537.
Richeson, J., & Craig, M. (2011). Intra-minority Intergroup Relations in the Twenty-First
Century. Daedalus,
(2), 166-175. Retrieved from
Smith, E. R. (1993). Social Identity and Social Emotions: Toward New Conceptualizations of
Prejudice. In D. M. Mackie & D. L. Hamilton (Eds.), Affect, Cognition, and
Stereotyping: Interactive Processes in Group Perception
(pp. 297-315). San Diego:
Academic Press.
Stacey, M., Carbone-López, K., & Rosenfeld, R. (2011). Demographic Change and Ethnically
Motivated Crime: The Impact of Immigration on Anti-Hispanic Hate Crime in the United
States. Journal of Contemporary Criminal Justice, 27
(3), 278-298.
Tam, T., Hewstone, M., Cairns, E., Tausch, N., Maio, G., & Kenworthy, J. (2007). The Impact of
Intergroup Emotions on Forgiveness in Northern Ireland. Group Processes & Intergroup
Relations, 10
(1), 119-136. doi:10.1177/1368430207071345
Tucker, J. T. (2006). Enfranchising Language Minority Citizens: The Bilingual Election
¡Vota! 22
Provisions of the Voting Rights Act. Legislation and Public Policy, 10
(195). Retrieved
U.S. Department of Justice. (2018, February 26). About Language Minority Voting Rights.
Retrieved from
Wetts, R., & Willer, R. (2018). Privilege on the Precipice: Perceived Racial Status Threats Lead
White Americans to Oppose Welfare Programs. Social Forces, 97
(2), 793-822.
Willer, R., Feinberg, M., & Wetts, R. (2016). Threats to Racial Status Promote Tea Party
Among White Americans. SSRN Electronic Journal.
Zomeren, M. V., Fischer, A. H., & Spears, R. (2007). Testing the Limits of Tolerance: How
Intergroup Anxiety Amplifies Negative and Offensive Responses to Out-Group-Initiated
Contact. Personality and Social Psychology Bulletin, 33
(12), 1686-1699.
Zong, J., & Batalova, J. (2017, March 02). The Limited English Proficient Population in the
United States. Retrieved December 8, 2018, from
ResearchGate has not been able to resolve any citations for this publication.
Full-text available
Significance Support for Donald J. Trump in the 2016 election was widely attributed to citizens who were “left behind” economically. These claims were based on the strong cross-sectional relationship between Trump support and lacking a college education. Using a representative panel from 2012 to 2016, I find that change in financial wellbeing had little impact on candidate preference. Instead, changing preferences were related to changes in the party’s positions on issues related to American global dominance and the rise of a majority–minority America: issues that threaten white Americans’ sense of dominant group status. Results highlight the importance of looking beyond theories emphasizing changes in issue salience to better understand the meaning of election outcomes when public preferences and candidates’ positions are changing.
Full-text available
Using concepts from social identity theory (Tajfel and Turner, 1979), we examined whether racial/ethnic majority group members' reactions to future demographic shifts is a function of the degree to which they perceive their ingroup's higher-status in society to be legitimate. In two studies, participants who varied in the degree to which they perceived their group's status to be legitimate were either exposed to real projections for 2060 (i.e., large decline in proportion of population that is the “majority” group), or fake projections for 2060—that resembled current figures (i.e., small decline). In Study 1, White Americans who perceived their status to be highly legitimate expressed greater intergroup threat, and negative feelings (anger and fear) toward minorities after exposure to projections with a large decline in the relative size of the White American population. In contrast, demographic shift condition had no effect on intergroup threat and negative feelings toward minorities among White Americans who perceived their status to be relatively illegitimate; negative feelings and threat remained low across both conditions. Similarly, in Study 2, ethnic Portuguese people in Portugal exposed to projections in which there was a large decline in the relative size of the ethnic Portuguese population experienced more intergroup threat and expressed a greater desire to engage in anti-immigration behaviors. The effect of demographic shift condition on intergroup threat and anti-immigration behaviors was stronger among ethnic Portuguese who perceived their status to be legitimate compared to ethnic Portuguese people who perceived their status to be relatively illegitimate. These results highlight that across different cultural contexts, majority group members' beliefs about the legitimacy of intergroup relations can affect their reactions to the prospect of increased diversity.
Full-text available
In recent years, Mechanical Turk (MTurk) has revolutionized social science by providing a way to collect behavioral data with unprecedented speed and efficiency. However, MTurk was not intended to be a research tool, and many common research tasks are difficult and time-consuming to implement as a result. TurkPrime was designed as a research platform that integrates with MTurk and supports tasks that are common to the social and behavioral sciences. Like MTurk, TurkPrime is an Internet-based platform that runs on any browser and does not require any downloads or installation. Tasks that can be implemented with TurkPrime include: excluding participants on the basis of previous participation, longitudinal studies, making changes to a study while it is running, automating the approval process, increasing the speed of data collection, sending bulk e-mails and bonuses, enhancing communication with participants, monitoring dropout and engagement rates, providing enhanced sampling options, and many others. This article describes how TurkPrime saves time and resources, improves data quality, and allows researchers to design and implement studies that were previously very difficult or impossible to carry out on MTurk. TurkPrime is designed as a research tool whose aim is to improve the quality of the crowdsourcing data collection process. Various features have been and continue to be implemented on the basis of feedback from the research community. TurkPrime is a free research platform.
Full-text available
Amazon’s Mechanical Turk (MTurk) is an increasingly popular tool for the recruitment of research subjects. While there has been much focus on the demographic differences between MTurk samples and the national public, we know little about whether liberals and conservatives recruited from MTurk share the same psychological dispositions as their counterparts in the mass public. In the absence of such evidence, some have argued that the selection process involved in joining MTurk invalidates the subject pool for studying questions central to political science. In this paper, we evaluate this claim by comparing a large MTurk sample to two benchmark national samples – one conducted online and one conducted face-to-face. We examine the personality and value-based motivations of political ideology across the three samples. All three samples produce substantively identical results with only minor variation in effect sizes. In short, liberals and conservatives in our MTurk sample closely mirror the psychological divisions of liberals and conservatives in the mass public, though MTurk liberals hold more characteristically liberal values and attitudes than liberals from representative samples. Overall, our results suggest that MTurk is a valid recruitment tool for psychological research on political ideology.
Here we integrate prior work to develop and test a theory of how perceived macro-level trends in racial standing shape whites’ views of welfare policy. We argue that when whites perceive threats to their relative advantage in the racial status hierarchy, their resentment of minorities increases. This increased resentment in turn leads whites to withdraw support for welfare programs when they perceive these programs to primarily benefit minorities. Analysis of American National Election Studies data and two survey-embedded experiments support this reasoning. In Study 1, we find whites’ racial resentment increased beginning in 2008, the year of Barack Obama’s successful presidential candidacy and a major economic downturn, the latter a factor previously shown to amplify racial threat effects. At the same time, whites’ opposition to welfare increased relative to minorities’. In Study 2, we sought to better establish the causal effect of racial status threats. We found that experimentally presenting information suggesting that the white majority is rapidly declining increased whites’ opposition to welfare, and this effect was mediated by heightened racial resentment. Finally, in Study 3 we found that threatening whites’ sense of their economic advantage over minorities led whites to report greater opposition to welfare programs, but only if these programs were portrayed as primarily benefiting minorities, not if they were portrayed as benefiting whites. These findings suggest that whites’ perceptions that minorities’ standing is rising can produce periods of “welfare backlash” in which adoption of policies restricting or curtailing welfare programs is more likely.
The racial/ethnic diversity of the United States is increasing, yet recent social psychological research has focused primarily on White Americans’ reactions to this demographic trend. The present research experimentally examines how members of different racial minority groups perceive increasing diversity, driven by Hispanic population growth, focusing on downstream consequences for political ideology and policy preferences. Four studies reveal that making Hispanic population growth salient leads non-Hispanic racial minorities to identify as more conservative and support more conservative policy positions, compared with control information. The policy preferences of Hispanics, however, were not affected by exposure to information about their in-group’s growth. Considered in tandem with previous research, the present studies suggest that Hispanic population growth may motivate greater support for conservative ideology among members of both racial majority and minority groups.
Since its rapid rise in early 2009, scholars have advanced a variety of explanations for popular support for the Tea Party movement. Here we argue that various political, economic, and demographic trends and events--e.g., the election of the first nonwhite president, the rising minority population--have been perceived as threatening the relative standing of whites in the U.S., with the resulting racial resentment fueling popular support for the movement. This "decline of whiteness" explanation for white Americans' Tea Party support differs from prior accounts in highlighting the role of symbolic group status rather than personal experience, or economic competition, with minority group members in generating perceptions of threat. We tested this explanation in five survey-based experiments. In Study 1 we sought to make salient the president's African-American heritage by presenting participants with an artificially darkened picture of Barack Obama. White participants shown the darkened photo were more likely to report they supported the Tea Party relative to a control condition. Presenting participants with information that the white population share (Study 2) or income advantage (Study 3) is declining also led whites to report greater Tea Party support, effects that were partly explained by heightened levels of racial resentment. A fourth study replicated the effects of Study 2 in a sample of Tea Party supporters. Finally, Study 5 showed that threatened white respondents reported stronger support for the Tea Party when racialized aspects of its platform (e.g., opposition to immigration) were highlighted, than if libertarian ones (e.g., reduced government spending) were. These findings are consistent with a view of popular support for the Tea Party as resulting, in part, from threats to the status of whites in America.
The following study investigates the causal impact of the Voting Rights Act (VRA) language minority provisions, which mandate multilingual election assistance if certain population thresholds are met. While lower rates of Latino and Asian American political participation are often attributed to language barriers, scholars have yet to establish a direct impact of the provisions on electoral behavior. Building off of previous state- and county-level analyses, we leverage an individual-level voter file database to focus on participation by Latino and Asian American citizens in 1,465 counties and municipalities nationwide. Utilizing a regression discontinuity design, we examine rates of voter registration and turnout in the 2012 election, comparing individual participation rates in jurisdictions just above and just below the threshold for coverage. Our analysis attributes a significant increase in Latino voter registration and Asian American turnout to coverage under the VRA.
Significance There is generally conflict when members of different social groups, such as racial, ethnic, or religious groups, come in contact in the same geographic area. This phenomenon is commonly observed across a variety of settings. However, the cause of such conflict is poorly understood: Some theorists have argued that contact between groups is insufficient to cause conflict and that, under certain conditions contact may lead to improved intergroup relations. Although most theories of contact propose that repeated contact between individuals is important to the disposition of intergroup attitudes, experimenting on the effects of repeated contact has proven difficult. Here, I report a randomized controlled trial that assigns repeated intergroup contact between members of different ethnic groups. The contact results in exclusionary attitudes toward the outgroup.