A Question of Values:
Are Conservative Latinos Less Inclined than Their White Counterparts to Support a Conservative Candidate Tied to Racial Issues?
Introduction:
A lot of discussions have centered on whether or not conservative politics and politicians have a race issue. This ongoing question of race in regard to the Republican party and conservative issues has taken on new meaning in the Trump era of politics and policies.6 This study does not seek to answer whether or not conservatives or conservative candidates are racist but rather to study how their perception can influence otherwise conservative individuals away from voting for and supporting conservative candidates if they perceive a racial issue at play.
We question how discussions and perceptions surrounding race may be influencing individuals who would be otherwise inclined to lean conservative. In particular, we would like to know the impact it had on Latinos who consider themselves independent and Republican when presented with information that may tie a Republican candidate to an issue connected to racism.6 This group is a quickly growing member of the electorate, with Pew Research Center noting their consistent, rapid growth and only recent slow.5 American Latinos are noted for strong family ties and majority Catholic beliefs. Both indicators that the rising level of eligible Latino voters in the U.S. should spell great success for the Republican Party, which is also noted for its deep ties to traditional family values and strong ties to religion.6
In the past, conservative candidates have had difficulty attracting people of color to support them and their platforms. Black voters, for example, have long been considered an important and loyal stronghold for the Democratic Party, as such, Republicans and more conservative candidates have been known to suffer with this demographic.3 However, recent history seems to show the potential for disillusionment with the Democratic Party. This brings us to the interesting possibility that the Democratic Party may no longer have the same appeal to people of color that it once did, creating a potential opening for the Republican Party to widen its appeal and its base.7
The reader will note that the terms “conservative” and “Republican” are being used somewhat interchangeably both here and in the information presented to the test subjects (found in the appendix section below). This is intentional, as conventional wisdom holds that the Republican party is perceived as connecting to traditional and “conservative” values. For the purpose of this study, where we are interested in perception, there will be no distinction between the terms or attempt to differentiate the two.
Hypotheses:
Political ideologies hold strong across a number of topics and stressors; however, group identity is also a strong motivator for individuals when it comes to their political preferences.3 There are several things that can be posited from these two understandings. The first is that we can anticipate partisan ideology to outweigh moral issues associated with racism, to an extent.1 This study does not question an individual’s links to racist actions but rather questionable rhetoric; for instance, no one is concerned that the candidate in question is involved in a hate crime. The scenario proposed to the subject involves a racially charged issue with plausible deniability, namely the notion that the candidate’s PR firm may be responsible for the comments and not the candidate himself. It is anticipated that this plausible deniability should be enough to reassure strong ideologues and may even help moderate partisans maintain the same or similar level of support; the question is whether or not this deniability will be enough for those that lean conservative and those who consider themselves independent.3 For those who identify as conservative, the question becomes whether or not their link to an ethnic or racial minority outweighs a partisan preference. It could also be assumed that strong liberal-minded individuals (those identifying as Strong Democrats) will see no significant change as they are already inclined to not support the candidate based on the other information given. Given what is understood about the nature of partisanship as well as group identities, there are a few hypotheses worth examining in this experiment:
Hypothesis 1a: Strong conservative ideologues are less likely to care about perceived racial issues associated with a candidate.
Hypothesis 1b: Strong liberal ideologues are unlikely to be significantly affected by the treatment.
Hypothesis 2: People of color are less likely to support a candidate accused of racism than white individuals.
Hypothesis 3: Independents who lean Republican will be less supportive of a conservative candidate who is tied to accusations of racism than those who identify as Republican or Strong Republican.
Experimental Method & Design:
The method of this study is that of a survey format presented to two groups of student subjects. This is a multi-level design that consists of two independent variables that will be examined.2 The survey will begin by establishing these independent variables before the framework for the stimuli is introduced. The first question consists of a 7-point party identification scale. This 7-point interval scale is self-reported before the candidate information is presented. Likewise, the 7-point nominal scale used for race/ethnicity identification is also presented prior to information on the candidate. Both the question of party identification and race/ethnicity are self-reported. Although a longer survey could be conducted to parse out values and opinions on issues of political significance, for the nature of this study, self-identification should be sufficient. It is intentional and critical that the self-identification questions come before the information about the candidate is introduced. The main purpose of subjects self-identifying prior to receiving information is to allow for honest self-identification. Although research indicates that party identification is stable over time, there is a risk that individuals in the test group who have read about racial bias and insensitivity may be less likely to report their true affiliation.3 In an attempt to avoid this potential issue, the identification questions are asked at the opening of the experiment.
While there is a survey design for this experiment, this is also a control group design in which we look at two randomly assigned groups; one subject pool is presented with treatment, and one is not.2 Because of the nature of this experiment, student subjects will be utilized for the subject pool. Much concern has been made over the potential issues with student subjects. For instance, there is a particular concern with age; students tend to lean younger than the general population. Along with this concern about age is a concern about conservative/liberal identification, as younger generations tend to be more liberal than older generations. This concern, though prevalent, has largely been blown out of proportion.4 Although student samples are convenience samples, there is a growing number of studies and data showing that they are sufficient for our purposes, and the findings can be applied to the general public. Student subjects have been seen to have high external validity, which is the main concern for this particular research question.4 In addition to external validity, there is a desire for control and reliability that can be offered through a student subject pool.2 The same process can be replicated, both using student subjects and crowdsourcing methods, should the results prove significant.
Although the concerns associated with student subjects are noteworthy, particularly given our interest in conservative views and identification, a sufficiently large sample of students should help mitigate these concerns.2 A larger group of students is certainly more convenient to recruit than other methods of recruiting larger numbers of participants. The cost involved in implementing a large-scale national survey through a company would be far too great for the project at hand and the resources available. Additionally, while crowdsourcing tools like Amazon’s MTurk would be perfectly reasonable for this type of study, although representative, the pool is quite narrow, and we run the risk of individuals who have taken multiple surveys producing inaccurate or biased results in an attempt to please the researcher.2
In this instance, there are two independent variables of interest. The first value of interest is party identification. The second independent variable of interest is that of race/ethnicity. Our dependent variable is the reaction to a candidate based on the introduced stimuli. We anticipate that manipulation of the presented information in regard to the candidate will shift the reaction to the said candidate, as noted in the hypotheses above. Note that in the information presented to the subjects (a copy can be viewed in the appendix attached), the candidate’s race is not mentioned, nor is there a party identification; however, it is strongly implied that that candidate is Republican through the information provided. This is intentional, as the purpose of this study is to examine identification with the individual through perceptions and emotions. The candidate is presented as a member of a particular group, rural conservative, most closely associated with whites and Republicans.8 The candidate is then presented in exactly the same way to the second group but with the added addition of a race-related controversy.
The candidate in question is fictitious. This is done for a few reasons, the primary reason being that we would like to control the information presented. This is also why the fictitious candidate is presented as running for a statehouse position, a local election not likely to have garnered national attention. It is a beneficial control to present a fictitious candidate, as opposed to a real candidate, as subjects will have no prior knowledge or opinions to color their perspectives and opinions.2 The race-related controversy is uniquely related to a particular growing member of the electorate. This study’s chief concern is as a primary source of information on conservative Latino voters’ response to racial issues tied to a conservative candidate and how and if this differs from other individuals of color or conservatives in general. Thus, the stimulus utilized is geared toward subjects of immigration and labor as they relate to American Latinos more than other people of color.
The advantage of the control group design is that we will not be clouding individual opinions as we would in a pretest and post-test model. This is particularly important when looking at Republicans and independents that lean Republican, as in a pretest/post-test design, we run the risk of causing a “dig-in” effect that could impact our results. In the control group experiment, subjects are only presented with either the treatment or not, and their opinion is either shaped by what is presented or not.2 From the perspective of this study, it is more beneficial to have two groups to compare than to risk having individuals, once presented with the new information, see the presented controversy as a partisan attack. The “controversy” has been included in the survey example in the appendix attached to this proposal. Note that the highlighted section is the part of the survey that the control group will not be subjected to; all other information will remain the same. Although there is still a risk that conservative-minded individuals may hold steadfast to their support for a perceived conservative candidate despite the information they are presented, the control group format will allow all information to be presented at once rather than some information, the most inflammatory, being presented separately a questioned in a post-test design.
Conclusion:
As stated in the introduction, the purpose of this study is not to determine whether or not Conservatives and Conservative politicians are racist. Such questions are complicated, nuanced, and beyond the limited scope of this study. The purpose of this experiment is to gain important insight into the very complicated trades that are made during the decision-making process of citizens and how questions of racial bias and insensitivity play into those decisions.
The issue of race is of particular interest in regard to Latinos, as they are a growing base that both liberals and conservatives are vying for. Despite speculation that the Republican Party may struggle with the Latino vote, recent election trends suggest that this large group is still very much up for grabs for either party’s candidates.6 This study is a starting place for consideration when it comes to developing an understanding of how candidates who struggle with racial issues and perceptions may be viewed by this population, particularly in comparison to other people of color and white voters.
References:
Brian Resnick, N. J. (2014, September 24). How Politics Breaks Our Brains. Retrieved from https://www.theatlantic.com/politics/archive/2014/09/how-politics-breaks-our-brains/380600/
Druckman, J. N. (2013). Cambridge handbook of experimental political science. New York: Cambridge University Press.
Introduction to political psychology. (2016). New York: Routledge, Taylor & Francis Group.
Kam, C. D., Wilking, J. R., & Zechmeister, E. J. (2007). Beyond the "Narrow Data Base": Another Convenience Sample for Experimental Research. Political Behavior, 415-440.
Noe-Bustamante, L., Lopez, M. H., & Krogstad, J. M. (2020, July 10). U.S. Hispanic population surpassed 60 million in 2019, but growth has slowed. Retrieved from https://www.pewresearch.org/fact-tank/2020/07/07/u-s-hispanic-population-surpassed-60-million-in-2019-but-growth-has-slowed/
Paz, C. (2020, November 04). What Liberals Don't Understand About Pro-Trump Latinos. Retrieved from https://www.theatlantic.com/politics/archive/2020/10/trump-latinos-biden-2020/616901/
Suro, R., Fry, R., & Passel, J. S. (2019, December 30). II. The Growth of the Hispanic Population and Its Voting Strength. Retrieved from https://www.pewresearch.org/hispanic/2005/06/27/ii-the-growth-of-the-hispanic-population-and-its-voting-strength/
Tuch, S. A., & Hughes, M. (2011). Whites' Racial Policy Attitudes in the Twenty-First Century: The Continuing Significance of Racial Resentment. The Annals of the American Acadamy, 134-152.