BALL STATE UNIVERSITY

DEPARTMENT OF POLITICAL SCIENCE

RESEARCH METHODS IN POLITICAL SCIENCE I (POLS 625)

(TERM PAPER)

FAVORABLE OUTCOME OF VOTER DECISION FOR PRESIDENTIAL CANDIDATES

BY:

GODWIN MARCUS OKYERE

TONY LIPITT

NII NARKU NORTEY

(MPA STUDENTS)

DECEMBER, 2020

1
 CHAPTER ONE

INTRODUCTION

1.1 Voter Behavior and United States Presidential Elections

Historically, research on American voting has focused on social demographics, partisanship, issues

and ideology, and economic conditions. Investigating these viewpoints at a basic level reveals that

the above major individual indicators have a commanding effect on voter outcomes (Dunn &

Jessee 2020; Costa-Font & Cowell 2015; Jacobs Meier 2014; Marshal 2015). Race and ethnicity

common ground between the candidate and voters shows the potential to influence voters since a

"shared link" is evident (Abrajano, Nagler & Alvarez 2005). For the same reason, gender can

influence voters to select one candidate over another (Barnes & Cassese 2017; Cooperman &

Crowder-Meyer 2015). Of course, education, income, voter age, and religious commitment levels

distinguish personal choices in individual preference in being a conservative, moderate, or liberal

voter (Dunn & Jessee 2020). Another critical human trait also shows potential as a dominant reason

voter select one candidate over another. That behavior is observed in the Spatial Voting Theory.

The paper investigates the weight of "Spatial Voting "practices and the other sociodemographic

variables to see if there is a dominant component that ultimately influences the individual voter

behavior decision-making process in American Politics. Religion is another key indicator of the

voting behavior of electorates. Religion has an imperative impact on how the voter looks at explicit

issues, such as homosexuality, abortion, the environment, and economics. Fastow, Grant, and

Rudolph (1999) examined the impact of religion in the House of Representatives to see how

religion affected vote choice. Their investigation shows that "members' votes represent both their

(the representatives') own religious affiliation and the religious groups within their districts.

2
Speaking in political voting terms, America is a republic. Meaning the citizens elect officials to

vote on and manage our local, state, and federal governments. What is the dominating aspect that

motivates individual voters to select one candidate over another candidate? This paper will look at

some main concepts that might be the primary reason for the causation of voter turnout and, more

specifically, their individual candidates' preferences. American political behavior begins with an

elected power system having two primary political parties. The first, the Democratic party, viewed

to a greater extent as the liberal group color associated with "blue." The second, the Republican

party, is considered the more conservative group color associated with "Red." Of course, America's

political election system complex and complied with many levels beyond just blue and red choices.

Modern American liberalism aims to preserve and extend human, social, and civil rights, and the

government guaranteed positive rights (Fisher 2014). American conservatism often refers to a

combination of economic liberalism and social conservatism and, to an extent, libertarianism. It

aims at protecting traditional values (especially on social issues) while promoting the concept of

small government. Based on the above evidence, the paper intends to exam primary factors that

influence Americans to vote in elections and influence their political behavior.

1.2 Research Objectives

This paper intends to show whether "Spatial" self-preservation ideology at the individual and

candidate level, emphasizing specific positions on issues, exceeds other physical voter

characteristics such as; marital status, age, level of education, the region of residence, Christian

fundamentalists, illegal immigrant policy, abortion opinion, gay marriage issues or economic

status regarding dominant causation voter behavior in American political elections. Ideology is

one of the critical factors when looking at individuals' voting behavior on a particular issue. A

3
person's ideology reflects what is vital to that individual, whether it is religious influence,

educational level, marital status, or another factor. Chressanthis, Gilbert, and Grimes (1991)

examined ideology and the voting record of United States Senators on the abortion policy to

illustrate how ideology influences voting behavior. What is "Spatial Voter Theory," and how might

it compare with other prevalent voter theories such as Race Voter Theory or Social-Physiological

Voter Theory? At the basic level, Spatial Voter Theory seeks to view voting behavior from the

standpoint of self-preservation as a potential significate motivating reason why voters choose one

individual political candidate over another candidate Gonzalez and Granic (2020). Spatial Voter

Theory places significant value on voters' ideology and the policy-based ideology proposals

initiated by specific candidates to measure necessary public attitudes on particular issues that

dominate American political viewpoints. The specific objectives for the study are:

i. To determine the most strongly related characteristics to the voting behavior and pattern in

U.S. Presidential elections.

ii. To examine the relationship between self-preservation ideologies, voter behavior, and

pattern in U.S. Presidential elections.

1.3 Research Question

i. What characteristics are most strongly related to the voting behavior and pattern in U.S.

Presidential elections?

ii. Does self-preservation ideology affect voter behavior and pattern in the U.S. Presidential

elections?

4
 CHAPTER TWO

BACKGROUND AND THEORY

2.1 The History of the United States Presidential Elections

The 2016 presidential election marks the 58 presidential elections in the history of the United

States. At a point in time, the law's framers were divided on the presidential elections' best

approach. Some supported the idea that Congress should elect the president, while others supported

the president's election through national popular votes. As a result, Article two of the Constitution

specifies the presidential election method, giving birth to the Electoral College's creation. Article

two is a compromise between the constitutional framers who wanted Congress to choose the

president and those who preferred a national popular vote. The first presidential election in 1779

only had 6 of the original 13 states who chose electors to represent them. Today, every state decides

its electors to the national popular votes to represent them in Congress in selecting the United

States president.

The president and vice president of the United States are formally elected through an electoral

college. There is a total of 538 electoral colleges in the United States presidential elections. The

presidential candidate receives an absolute majority; thus, 270 out of the 538 votes becomes the

president. However, if no candidate gets a simple major of the total electoral votes, the House of

Representatives would have the privilege to choose among the candidates with the most electoral

votes, the United States' next president. Until 1937, presidents were not sworn in until March 4.

However, the Twentieth Amendments' passage pushed the United States president-elect's

inauguration to January 20 in the year preceding the general elections.

5
2.2 Factors Influencing the United States Presidential Elections

Several studies have identified sociodemographic such as age, marital status, education (Abrajano,

Nagler, & Alvarez 2005; Barnes & Cassese 2017), economic such as income levels, the standard

of living, and the national economy (Hager 1995; Lanou 1994; Kramer 1983). Others cite national

policies (Lenz 2012; Achen and Bartels 2017) as significant factors influencing voter choice in the

United States presidential elections. Emmenegger and Manow (2014) found that religion and

gender have changed voter from 1970. They argued that before the 1970s, women voter's

conservative, that ideology changed in recent times. Economic voting has been blamed for Bush's

loss in 1992, Carter's defeat in 1980, and Clinton's win in 1996 (Jackson 199). Gilbert and Grimes

(1991) also examined ideology the impact of political ideologies on voting records. They conclude

that party ideology influences voting behavior.

A report by the Atlantic found that Hilary Clinton's loss in the 2016 elections is mostly attributed

to her failure to win over white working-class voters. Clinton also failed to get hold of the color

base people who voted massively for Obama in 2008. According to an analysis by the Democratic

pollster Cornell Belcher, 8 percent of African American voters under 30 chose a third-party

candidate, as did 5 percent of Latinos under 30. He referred to these votes as "protest votes," and

he argued that they were enough to seal Clinton's fate.

A 2020 elections survey by PEW revealed that nearly eight-in-ten registered voters (79%) say the

economy will be vital to deciding whom to vote for in the 2020 presidential election. The other

factors that made the top-four were 68% of voters would vote based on health care, 64% cite

Supreme Court appointments, while 62% of voters say their votes will be influenced based on the

management of the COVID-19.

6
2.3 Socioeconomic Factors Affecting the United States Presidential Elections

The paper intends to show whether "Spatial" self-preservation ideology affects voter behavior in

American political elections. Therefore, for these studies, self-preservation ideology will be treated

as the independent variable. In contrast, race, gender, age, religion, income, residency, and

economic status would be treated as control variables.

2.3.1 Ideological Self-Placement

Article Review: Demographic Moderation of Spatial Voting in Presidential Elections (Dun and
Jessee 2020)
The spatial model's key indicator requires direct measurement of both the voter's position and that

of the candidate while using the same scale. The article's three significant findings highlight that

most voters have ideologically organized voting preferences; voters rely heavily on individual

candidate positions and political behaviors to play a role in their vote choices. The third point is

that some but not all voters displayed spatial voting tendency during the Kerry vs. Bush 2004

presidential election, with independence exhibiting the most considerable spatial voting portion.

The spatial voting theory will most likely be observed in more informed citizens than with voters

who have lower knowledge levels about the individual candidates. It requires citizens to base their

choices on how individual candidates select their position on political topics. Thus, less informed

citizens must rely more heavily on political parties' positions than individual candidates' positions.

Therefore, we hypothesized that;

H1: Political ideologies have a significant effect on voter choice in the U.S. presidential elections

7
2.3.2 Age and Race

Article Review: A Natural Experience of Race-Based and Issue Voting: The 2001 City of Los
Angeles Elections (Abrajano, Nagler, and Alvarez 2005)
Relationship between American Voters over the age of 65+ as an old age benefits voting bloc. The

article's outcome indicates an old-age benefit voting bloc when it comes to social security and

Medicare. However, old age is just one of many factors that impact how older people vote, such

as their race, ethnicity, and gender. Although it is assumed that people over 65 are conservative in

generalization, mature people have not predominantly committed to one political party. The article

was written at the start of the Baby Boomers induction into the 65+ age bracket (2010-2011) and

addressed the possibility that new battle lines might cause old and young to clash in a new "class

war" replacing low-income versus the wealthy. Additionally, data clearly shows that the elderly

vote in at a higher rate from 1972-2008 than other groups; their total impact was only accounted

for 16 percent of the options in the 2008 presidential election in the United States. Therefore, we

hypothesized that:

H2: Older people tend to lean towards Republican presidential candidates

The article uses a qualitative view to locate support for either theory. With two races in the same

election, a White and Latino liberal candidate runs for mayor and city attorney. The matchup

presents a clear view of race theory or spatial theory to excel with the same voting population in

the same election. The results were mixed. In the mayor's race, the white candidate Hahn beat the

Latino candidate Villaraigosa with the economy, law, and order, and education being the heavy

topics. However, it was not the white vote that won the day for the white candidate but rather the

Black, Asian moderate, and conservative voting majorities making the difference. As for the city

attorney race, where law and order played the dominant role, the Latino candidate secured the

victory by securing the Black, moderate, and conservative voting majorities. More white voters
8
displayed a willingness to cross over and vote for the Latino candidate at 42 and 39 percent

compared to 18 and 22 percent between the white and Latino voting. However, the spatial theory

seemed to carry the vote, with liberal voters holding a 60 percent vote for both candidates that lost

their elections and more moderates and conservatives voting for both candidates that won their

elections.

H3: Political ideologies have a significant effect on voter choice in the U.S. presidential elections

2.3.3 Income, Religion, and gender

Article Review: Demographic Gaps in America Political Behavior (Fisher, Patrick 2014)

Gapology looks at the political disparities between groups Blue vs. Red, Race, Religion, Gender,

Age, Income/Education, and Geography. By College-age, most Americans are stable in adopting

"lesser of two evils" political views heavily imprinted by family, friends, and social groups' beliefs.

2000-2012 40 of 50 states support the same party in presidential elections with the trend indicating

blue states are getting more blue and red states are getting more red. Could polarization then lead

to a cultural war? Partisan Stance on policies: "Economic Policies," e.g., government services and

oversite of private business. "Domestic and Foreign Policies" Immigration and Gun Control

"Social Issues" Gay married, Abortion policies, and the Death penalty. Impact groups of people

both as a group and at the individual levels.

Policies Gap Matters more to rich than low income seen in voter turnout perhaps linked to class

bias and loss of union jobs. The race Gap heavily influences the income gap. Although Democrats

represent the "gray" wealthiest people in America, they are seen as the party of the people, too,

and have made strides of winning over more middle-class voters in the last few decades.

H4: Income gaps have a significant effect on voter choice in U.S. presidential elections

9
2.3.4 Religious Affiliation

America is becoming more secular, a current trend that bods well for the Democrats. Republican

receive their strength from Christians that regularly attend church. With evangelicals, Mormons,

and white-Catholics being strong support for Republicans, the weakness is that this base is

shrinking under current trends. All regular attending churchgoers vote at least 10 points higher

than those who do not attend church weekly. People of color Protestant and Catholics, as well as

Jews, make up the Democrat's spiritual support along with other non-Christians or anthesis. All of

this is adding to a more massive divide between the two main political parties in America.

H5: Religious affiliation has a significant effect on voter choice in U.S. presidential elections

2.3.5 Gender Gap

Over studied, no significant impact equals 8-10 % difference after other social demographic

factors, Hispanic women are more likely to favor democrats than Hispanic men. White Men more

likely to favor Republicans than white women. The widening gap between young voters over older

voters shows this is another growing issue; however, it is not as pronounced as other demographic

gaps such as race and religion. However, President Obama did win 10 million more female voters

than male in 2012.

H6: Gender has a significant effect on voter choice in U.S. presidential elections

10
 CHAPTER THREE

OPERATIONALIZATION AND METHODS OF DATA ANALYSIS

3.1 Description of Data

The study examines socioeconomic and demographic factors that influence voter behavior in the

United States Presidential Elections. The paper intends to investigate whether "Spatial" self-

preservation ideology and other physical voter characteristics such as marital status, age, level of

education, the region of residence, Christian fundamentalists, illegal immigrant policy, abortion

opinion, gay marriage issues, economic status have any causation for voter behavior in American

political elections. Therefore, the unit of analysis is individual voters and political candidates

during an election. The case choice is registered voters and why their self-preservation and that of

a particular party (candidate) would influence their votes. The case choices are influenced by the

fact the researchers believe specific self-preservation ideology at the individual and candidate

level, emphasizing particular positions on issues, including other physical voter characteristics,

determine how individuals vote. Therefore, we expect the features of case choices to shed light on

aspects of human thinking and behavior that are most favorable and unfavorable to voter choices

during national elections. The outcome can help generate new ideas that might be tested by other

methods. The study's total number of cases was 4,271 obtained from the 2016 American National

Election Studies (NES) Dataset.

3.2 Operationalization of Concepts

For this study, the dependent is the outcome of the election (voter behavior). The leading

independent variable is individuals' ideologies and party candidates on national issues that shape

people's voter patterns. Thus, the extent to which liberals and conservative principles influence

11
individuals/candidates position on national policies. Race, gender, age, education, economic status

is included in the study as control variables because many studies the selected control variables

have shown to be confounding variables (Abrajano, Nagler, & Alvarez 2005; Barnes & Cassese

2017; Lanou 1994; Kramer 1983). Thus, they have influenced both the dependent and independent

variables in similar studies in the past. Since confounding variables are a form of systemic error,

it is likely to give a biased outcome and therefore need to be controlled. Most of the control

variables in this study have been found to have both positive and negative associations on why

voters vote for a particular candidate. The study used the 2016 American National Election Studies

and post-election from (NES) Dataset. The data was obtained from the United States Census

Bureau, 2017.

3.3 Labeling, measuring, and coding

The dependent variable Feeling Thermometer, Trump Post, has its measurement scale value (0-

100) and interval level. Feeling thermometer: Christian Fundamentalists, one of the independent

variables, also has a measurement scale value (0-100) and Interval level. Feeling thermometer:

Illegal Immigration, another independent variable is an interval level variable which measures (0-

100) value and the last independent variable, Gay Marriage Allowed; measurement scale value (1-

3) is an Interval level. Respondent Age, a control variable, is the last Interval level variable among

the list.

The remaining other independent variables, including the controlled variables, are nominal. The

leading independent variable, Ideological self-placement, has its measurement scale value (1-2),

Abortion Opinion; an independent variable with measurement scale value (1-4). The variables

below are the control variables and are nominal level. Level of education; measurement scale value

12
(1-4), Is R married?; measurement scale value 0-1, Region of Residence; measurement scale value

(1-4) and Income Gap; measurement scale value -9 – 5.

3.4 Coding & Recoding

3.4.1 Independent Variable

Ideological Self-Place (Nominal Level) 1. Liberals & Moderates, 2. Conservatives

Gay marriage Allowed: value 1-2 (Interval level) 1. Allowed, 2. Not allowed

Abortion Opinion: value 1-2 (Nominal level) 1. Not agreed, 2. Agreed

3.4.2 Control Variable

Respondent Age: value 1-2 (Interval level) 1 = "18-49" , 2 = ”50-above"

Level of education: value 1-4 (Nominal level) 1 = “HS or less" , 2 = ”sm. collage to Grad Sch"

Is R married?: value 0-1 (Nominal level) 0 = "No", 1 = "Yes"

3.5 Data Analysis

The Statistical Package for the Social Sciences (SPSS) was used to compute the statistical figures

to test the hypotheses. For inferential statistics, correlation and regression will be computed to

examine the relationship between socioeconomic, demographic, and voter behavior in the United

States national elections. The dependent variable, coded as Thermometer Trump Post 2016, is

measured at the interval level. Linear regression is used in computing the figures since the

dependent variable is an interval measurement.

13
 CHAPTER FOUR
ANALYSIS AND DISCUSSION

Table 4.1 Descriptive Statistics

Variables (N) Min Max Mean Std. Deviation

Region of Residence 4271 0.00 1.00 .3816 .48585
Income Level 4069 1.00 2.00 1.5933 .49128
Educational Level 4227 1.00 2.00 1.7417 .43777
Abortion Opinion 4208 1.00 2.00 1.6055 .48880
Feeling Therm: Donald Trump 2016 3633 0 100 42.09 34.921
Self-Ideological Placement 4185 1.00 2.00 1.4435 .49686
Gay Marriage Allowed 3435 1.00 2.00 1.2859 .45190
Respondent Age 4150 1.00 2.00 1.5101 .49996
Marital Status 4103 1.00 2.00 1.489 .43840
Feeling Therm: Illegal Immigrants 3581 0 100 46.34 38.471
Feeling Therm. Christian Fund. 4138 -7 100 41.79 32.509
Valid N (Listwise) 2644

Table 4.1 above is the descriptive statistics of research variables. The mean for the region of a

resident was .3816 with a standard deviation of .485845. The mean spread indicates that there were

more non-southern states than southern states. Since the standard deviation is larger than the mean,

it implies widespread data from the mean.

Income level obtained a mean of 1.59. It is closer to the maximum value, indicating that most of

the respondents are middle to higher degree earners. Similarly, the majority of the respondents are

college or higher degree graduates. The majority of the respondents supported the choice to have

an abortion or not. Furthermore, the majority of the respondents are anti-Trump. Self-ideological

placements leaned towards conservative, an indication most of the respondents support

14
conservative ideologies. Respondents were indecisive as to whether gay marriage should be

allowed or not. The mean age indicates that the majority of the respondents are old. Also, the

majority of the respondents are married, considering a mean of 1.48. Respondents were somehow

also indecisive on illegal immigrant policies. Finally, the majority of the respondents do not follow

Christian fundamentals.

4.2 Regression Analysis

TABLE 4.2 Effects of Ideological Self-Placement on Trump Vote (2016)

Variables Model 1 Model 2 Model 3

Ideological self-placement 36.101 (1.014) *** 20.306 (1.159) *** 20.292 (1.200) ***
Feeling therm: Christian .249 (.020) *** .247 (.021) ***
Fundamentalists
Feeling therm: Illegal -.392 (.019) *** -.392 (.020) ***
Immigrants
Gay Marriage Allowed 7.855(1.201) *** 8.172(1.263) ***
Abortion -6.206 (1.171) *** -5.512(1.217) ***
Age of R -.568(1.026)
Highest Level of Education -1.538 (1.265)
Is R Married -1.982 (1.083) *
Region of Residence 1.847(1.057) *
R Income -1.871 (1.142)
Constant -9.859 (1.546) 15.730 (3.555) *** 19.527 (4.397) ***
(Adj.) R2 .262 .443 .444
N 3,563 2,759 2,617
*: P< .10, **: P< .05, ***: P< .01 Note: Standard Errors are in paratheses
Dependent Variable: Feeling Thermometer: Donald Trump (post, v162079)

15
The study investigated the relationship between ideological self-placement and voter choice in the

2016 United States presidential elections.

The dependent variable is the feeling thermometer: Donald Trump (Post 2016). The leading

independent variable was the self-placement measured by the ideological position (Left-Right) of

the voter. Feeling thermometer: Christian Fundamentalists, feeling thermometer: Illegal

immigrants, Gay marriage allowed, and Abortion opinion were the supporting explanatory

variables. These supporting explanatory variables were selected based on the policy-based

ideology of the Democratic and Republican parties. To control for confounding variables,

respondent age, marital status, the region of residence, income gap, and educational level were

included in the model control variables. The study hypothesized no relationship between

ideological self-placement and voter choice in the U.S. presidential elections.

The result show a statistically significant relationship between ideological self-placement and

voter choice in the 2016 presidential elections even if they controlled other confounding variables.

The relationship was strong and positive, and it was significant at the 99% confidence level in all

the three Models. An interpretation of the result is that an increase in ideological self-placement

increased 36.101 votes (Model 1), 20.306 votes (Model 2), and 20.292 votes (Model 3) for Donald

Trump in the 2016 national elections.

Additionally, all the supporting explanatory variables also passed at the test statistics at the 99%

confidence level. It indicates a strong and statistically significant relationship between all the

supporting explanatory variables and Trump votes in 2016. The directional hypothesis is also

positive.

Surprisingly, none of the sociodemographic factors included in the study affected voter choice in

2016. All the control variables were statistically insignificant to Trump votes in 2016, even though

16
the age of the respondents and region of residence passed the test statistics at the 0.10 significance

level, thus, at the 90% confidence level. However, in this study, we rejected the null hypothesize

for all relationships at the 95% confidence level.

The findings presented here do not question the integrative property of Ideological self-placement.

The analyses confirmed that ideological preferences are related to various voter issues and find

strong support for the hypothesis of a left-right scale on voter choice in the 2016 U.S. presidential

elections. The study further concludes that in the 2016 presidential elections, people voted purely

on policy-based ideologies and not on sociodemographic issues. This is evident from the

regression results, which show that all the policy-based ideologies of the Republicans and Donald

Trump in the 2016 election were statistically with Donald Trump votes. Support for each of these

conservative policies meant higher votes for Trump.

4.3 Policy Implication

The implication is that the policy-based ideologies of a candidate tend to influence voters' behavior

in the national elections. Therefore, the two biggest political parties' presidential candidates must

be conscious of their campaign policies since such policies can influence voters' choice on an

election day.

4.4 Limitation of the Study

The main limitation of the study was the unavailability of the percentage of the rural and urban

population. The study would have wished to control these two variables to determine if they had

any association with Trump's votes in the 2016 election.

17
 REFERENCES
Abrajano, M., Nagler, J., & Alvarez, R. (2005). A Natural Experiment of Race-Based and Issue
Voting: The 2001 City of Los Angeles Elections. Political Research Quarterly, 58(2),
203-218. doi:10.2307/3595623
Amos, B., McDonald, M. P., & Watkins, R. (2017). When boundaries collide: Constructing a
national database of demographic and voting statistics. Public Opinion Quarterly, 81(1),
385.
Barnes, T. D., & Cassese, E. C. (2017). American party women: A look at the gender gap within
parties. Political Research Quarterly, 70(1), 127-141. doi:10.1177/1065912916675738
Cassese, E. C., & Barnes, T. D. (2018;2019;). Reconciling sexism and Women's support for
republican candidates: A look at gender, class, and whiteness in the 2012 and 2016
presidential races. Political Behavior, 41(3), 677-700. doi:10.1007/s11109-018-9468-2
Collingwood, L., Gonzalez O'Brien, B., & Tafoya, J. R. (2019). Partisan learning or racial learning:
Opinion change on sanctuary city policy preferences in C.A. and TX. Journal of Race,
Ethnicity, and Politics, 5(1), 92-129. doi:10.1017/rep.2019.25
Costa-Font, J., &amp; Cowell, F. (2015). SOCIAL IDENTITY AND REDISTRIBUTIVE
Dun, L., & Jessee, S. (2020). Demographic Moderation of Spatial Voting in Presidential Elections.
American Politics Research., 48(6), 750–762.
https://doi.org/10.1177/1532673X20925416
Emmenegger, P., & Manow, P. (2014). Religion and the Gender Vote Gap: Women's Changed
Political Preferences from the 1970s to 2010. Politics & Society, 42(2), 166–193.
https://doi.org/10.1177/0032329213519419
Fisher, P. (2015). Demographic gaps in American political behavior. Westview Press, a member
of the Perseus Books Group.
Jacobsmeier, M. (2015). From Black and White to Left and Right: Race, Perceptions of
Candidates' Ideologies, and Voting Behavior in U.S. House Elections. Political
Behavior, 37(3), 595-621. Retrieved August 29, 2020, from
http://www.jstor.org/stable/43653506
Krupnikov, Y., & Piston, S. (2016). The political consequences of Latino prejudice against blacks.
Public Opinion Quarterly, 80(2), 480-509. doi:10.1093/poq/nfw013

18
Marshall, J. (2016). Education and voting conservative: Evidence from a major schooling reform
in Great Britain. The Journal of Politics, 78(2), 382-395. doi:10.1086/683848
Ordabayeva, Nailya et al. "Better or Different? How Political Ideology Shapes
Page, S. (2004). Churchgoing closely tied to voting patterns; GOP capitalizes on religion gap;
dems debate what to do: FINAL edition. USA Today (Arlington, Va.)
Preferences for Differentiation in the Social Hierarchy." The journal of consumer
PREFERENCES: A SURVEY. Journal of Economic Surveys., 29(2), 357–374.
research. 45.2 (2018): 227–250. Web.
Schaffner, B. F., Macwilliams, M., & Nteta, T. (2018). Understanding white polarization in the
2016 vote for president: The sobering role of racism and sexism. Political Science
Quarterly, 133(1), 9-34. doi:10.1002/polq.12737
Smith, L. E., & Walker, L. D. (2013). Belonging, believing, and group behavior: Religiosity and
voting in American presidential elections. Political Research Quarterly, 66(2), 399-413.
doi:10.1177/1065912912443873
Sondheimer, R. M., & Green, D. P. (2010). Using experiments to estimate the effects of education
on voter turnout. American Journal of Political Science, 54(1), 174-189.
doi:10.1111/j.1540-5907.2009.00425.x
TORGLER, B., STADELMANN, D., & PORTMANN, M. (2020). Church voting
recommendations, voter preferences, and political decisions. Journal for the Scientific
Study of Religion, doi:10.1111/jssr.12669
Weller, N., & Junn, J. (2018). Racial Identity and Voting: Conceptualizing White Identity in
Spatial Terms. Perspectives on Politics, 16(2), 436-448.
doi:10.1017/S1537592717004285

19
20