Fostering Democratic Participation at Community Colleges: Understanding the Relationship Between Voting and Student,Institutional,and Environmental Factors

Abstract

Objective: The purpose of this study is to understand the relationship between organizational characteristics, state political climate, and student civic engagement at community colleges, operationalized in this article as student voting. Method: Utilizing a unique cross-sectional dataset compiled by the Institute for Democracy & Higher Education for the 2012 and 2016 presidential elections that merges student voting data, enrollment records from the National Student Clearinghouse, and institutional data from the Integrated Postsecondary Education Data System, we employed multi-level binary logistic regression to account for individual and organizational influences. Results: The relationship between voting and student characteristics aligns with national voting trends, in that students who are female, older, and White had higher odds of voting. However, there was one notable exception in 2012 where Black students had higher odds of voting than their White peers. We also found that students enrolled full-time and those enrolled at colleges in electoral battleground states were more likely to vote in 2012 and 2016. Weak or mixed relationships emerged between voting and environmental factors such as campaign spending, ballot initiatives, get-out-the-vote programs, restrictive voter laws, and compositional diversity of the student body. Conclusion: This study provides insight for community college leaders on potential ways to engage students in an effort to promote voter turnout in a presidential election year on their campuses. Specifically, we posit that institutions could highlight the salience of public policy issues for students, lobby for less restrictive voting laws, and implement evaluation programs of their civic engagement initiatives.

Keywords

civic engagement community college student voting multi-level regression presidential election

Cultivating a sense of civic responsibility is fundamental to higher education’s mission (Saltmarsh & Hartley, 2011; Scott, 2006). Yet, when compared with other age groups in the United States, traditionally college-aged Americans between 18 and 29 years old—a range that includes the median community college student age of 24 years old (American Association of Community Colleges [AACC], 2019)—have a lower voter turnout than other age groups. For decades, this group has voted at a rate of 10 percentage points less than the next lowest voting age group (30–44), and almost 30 percentage points behind the age group (60+) with the highest voting rate (McDonald, 2018). This troubling trend calls for researchers and practitioners alike to investigate and inform the discussion on how colleges and universities can best support, encourage, and increase civic engagement through voting. Furthermore, our emphasis should be on understanding and supporting civic engagement across all higher education contexts, particularly at community colleges where students identifying with groups underrepresented in politics, such as Blacks and Latinos, are enrolled at disproportionately high rates, relative to 4-year institutions (AACC, 2019).

Higher education administrators and faculty have made civic engagement a student learning outcome, aiming to foster civic engagement through approaches in and out of the classroom (Bennion & Laughlin, 2018; Colby et al., 2007). In recent years, the higher education community has specifically pushed for increasing voter turnout as an important form of civic participation through new campus get-out-the-vote (GOTV) programs and initiatives. These efforts have included voting competitions with other schools, campus-sponsored voting celebrations, institutionally funded civic engagement centers, dialogue-based classroom pedagogies, and inclusion of student voting in institutional rankings and classification criteria (Glastris, 2018; Swearer Center at Brown University, 2018; Thomas & Brower, 2017). While some research analyzes the influence of these types of initiatives or other campus organizational conditions at 4-year institutions (e.g., Benenson & Bergom, 2016), few studies focus on the civic engagement at community colleges (Kisker, Newell, et al., 2016; Kisker, Weintraub, et al., 2016), whose students account for over 40% of total enrollment at degree-granting institutions of higher education and who enroll marginalized students at a higher rate than their 4-year counterparts (Ginder et al., 2017).

In this study, we explore whether factors identified in prior empirical literature as significant predictors of student voter turnout are salient in the community college context. We utilize a unique dataset compiled by the Institute for Democracy & Higher Education (IDHE) that merges student public voting data with enrollment records from the National Student Clearinghouse (NSC). Our sample for this study includes nearly 1.5 million community college students and accounts for information on student voting behavior in the 2012 and 2016 U.S. presidential elections, institutional characteristics, and state political contexts. Our findings reveal several notable relationships between student voting and our variables of interest (e.g., personal, institutional, and state contexts). We conclude by offering both insight for community college leaders and directions for future inquiry.

Background

There is a broad consensus that increased educational attainment is significantly related to more active political involvement (Campante & Chor, 2012; Hillygus, 2005; Persson, 2015). In this vein, scholars have taken a variety of approaches to attempt to understand how postsecondary education might produce increased civic engagement (e.g., Myers et al., 2019). The extant literature on civic engagement encompasses a broad spectrum of activities, including non-partisan community involvement, service-learning projects, student activism, and partisan political participation such as voting (e.g., see Bowman, 2011; Hillygus, 2005; Lopez & Brown, 2006; Reason & Hemer, 2015; Traver & Katz, 2014). These democracy-building activities (Kisker, Newell, et al., 2016; Kisker, 2016) are an important outcome of higher education (Saltmarsh & Hartley, 2011). While the vast majority of the research on democratic citizenship and civic engagement in American higher education has focused on 4-year institutions (e.g., Finley, 2011; Hillygus, 2005; Kushin & Yamamoto, 2010; Reason & Hemer, 2015), a growing body of scholarship explores and calls for a focus on community colleges and their students (Ginder et al., 2017; Kisker, Weintraub, et al., 2016).

The lack of a consensus definition on civic engagement and the wide variation in postsecondary institutional characteristics make identifying the specific programs, policies, and experiences that produce civic engagement among students (and graduates) a challenging task. This is especially true for community college researchers where populations and institutions are less homogeneous than 4-year institutions. This results in unclear trends and complicated implications for practice. For example, in research focusing on the relationship between service-learning coursework and political engagement, researchers have observed, at best, mixed results (e.g., Elder et al., 2007; Prentice, 2011; Ronan & Kisker, 2016). Other approaches to increase civic engagement that mimic the tactics of 4-year institutions have been attempted at community colleges and have been found to lead to positive results, though further research is needed. One approach focuses on the incorporation of references to civic engagement in institutional missions and strategic plans. Another centered on the inclusion of civic engagement in courses, opportunities for civic dialogue, and extracurricular programming (Kisker, Newell, et al., 2016). All of these tactics have been found to increase the odds of students reporting civic behaviors, such as voting in student elections or discussing politics with friends, though further research is needed to strengthen the findings of this work (Kisker, Newell, et al., 2016).

As mentioned, the very definition of what constitutes civic engagement activities leads to unclear findings and mixed results in the research. As a corrective measure, scholars might focus their attention on specific forms of civic engagement (e.g., voting) within specific organizational contexts (e.g., community colleges) using nationally representative data that allows for student-, institution-, and state-level comparisons. There is a critical need to inform institutional leaders of the potential relationship between organizational conditions within schools, through programming and compositional diversity, to guide institutional decision making around enhancing civic engagement (Kisker, 2016; Mathews, 2016). A focused analysis on community college student voter turnout is necessary because past analyses indicated that community college students were less likely to vote in presidential elections than students at 4-year institutions (Thomas et al., 2017), even after controlling for several student and institution characteristics (IDHE Staff, 2017). In the following sections, we review the extant literature on student voter behavior that informs our variable selection.

Student Characteristics

Much of the student civic engagement research focuses on the relationship between student demographics and voter turnout (e.g., Kisker, Newell, et al., 2016; Thomas et al., 2017). Many scholars have observed significant relationships between a student’s gender, race, and socioeconomic status and their odds of voter turnout (Benenson & Bergom, 2016; Blais, 2000; Hillygus, 2005; Ro & Bergom, 2017; Sax, 2004; Thomas et al., 2017). Furthermore, Blais (2000) argues that family relationships and education level all lead to life experiences that inform voting behavior, which aligns with the work of other researchers (Bartels, 2000; Timpone, 1998).

Student voting trends by demographics generally align with overall trends in voting behavior in the U.S. electorate (McDonald, 2018). Female students are more likely to vote than men, and Whites and Blacks tend to have higher voting rates than Asians and Latinx students (File, 2013; Thomas et al., 2017). Researchers have also analyzed college student voting behavior based on major (Bergom & Ro, 2018; Hillygus, 2005; Nie & Hillygus, 2001; Ro & Bergom, 2017), finding that science, technology, engineering, and math (STEM) and business majors are less likely to vote than other majors (Nie & Hillygus, 2001; Thomas et al., 2017).

Institutional Characteristics

Recent research on student voting behavior suggests a significant relationship between voting and campus organizational conditions (Benenson & Bergom, 2016; IDHE Staff, 2017; Niemi & Hanmer, 2010; Vogelgesang & Astin, 2005). Vogelgesang and Astin (2005) found that students at public institutions were slightly more likely to report voting in state or local elections than students at private institutions. Niemi and Hanmer (2010) observed that campus location predicts student voting, and students whose hometown or college town was located in an electoral battleground state were more likely to vote.

The organization of campus life, which can vary significantly by institutional type, also affects students’ voting behavior. Using a sample of community college students, Kisker, Newell, et al. (2016) found increased levels of voter participation when students had the opportunity to engage in campus activities through involvement with campus clubs and participation in campus events focused on political issues. Students at institutions with constrained opportunities had lower voter turnout (Lopez & Brown, 2006). This may be exacerbated at community colleges where students are potentially influenced more by their community context than students at 4-year schools, due to the majority of community college students being geographically place bound, but less likely to live on campus. Given the different structures and contexts across higher education institutions, this analytical component related to organizational goals and structure is helpful to guiding research in higher education and is represented in our model by the number of GOTV programs on campus during the 2016 election.

State Political Environment

For the purpose of this study, we define the state political environment as the policies, resources, political party division, and economy of a given state. Given the complexity of capturing the many facets of the political environment of a state, it is not surprising that this is the least understood aspect of student civic engagement in higher education (see Feddersen & Pesendorfer, 1997). In a review of research focused on political participation, Blais (2000) finds that one of the most common predictors of engagement is whether the election is contested or not competitive. Voters are more likely to turn out for a contested election (Blais, 2000). Contested elections often involve greater spending which correlates with voter turnout (Gerber & Green, 2017). Very few scholars have attempted to understand the relationship between differences in political contexts of an election and community college voter turnout.

Purpose of the Study

In light of existing research, this study examines whether community college students’ individual decisions to vote in a presidential election year are related to their individual identities, experiences, and beliefs; the organizational context of their institution and the opportunities for involvement in campus life; and state-level politics. Said another way, we take a multi-level (i.e., individual, college, and state) approach to better understand how individual experiences that are embedded within organizational and state contexts are associated with student voting behavior. This multi-level approach allows us to account for the unique contexts of each student, and additional details are provided below.

Given the lack of research focusing on community college students’ participation in democracy-building activities and the low rate of voter turnout among community college students, there is a critical need to understand why some community college students vote and others do not (AACC, 2019). Utilizing the current conceptual understanding of student voting behavior, this study seeks to inform practitioners and researchers about the relationship between individual characteristics, organizational conditions, political environment, and civic engagement. Our intent is to further the understanding of how community college practitioners can promote voter engagement and policymakers can develop programming to increase voter turnout and civic engagement. The following research questions guide our study:

Research Questions 1: How do voters and non-voters in community colleges differ on key demographic characteristics (i.e., age, sex, race, major, enrollment status), holding organizational characteristics constant?

Research Questions 2: What organizational conditions of community colleges are associated with student voter turnout, holding demographic factors constant?

Research Questions 3: Is there a relationship between the state political environment and community college student civic engagement, holding organizational and demographic characteristics constant?

Data and Method

Data for this study were drawn from the National Study of Learning, Voting, and Engagement (NSLVE), which was administered by the IDHE at the Tisch College of Civic Life at Tufts University in 2012, 2014, and 2016. The NSLVE database contains publicly available voting records collected by Catalist, a company that maintains a national database of registered voters in the United States, merged with student enrollment records from the NSC. The NSC removed identifying information before sharing the student data with researchers at IDHE. We supplemented these data by merging it with information from multiple data sources, including

the Integrated Postsecondary Education Data System (IPEDS), which captured institutional characteristics such as enrollment and structural diversity;

the Brennan Center for Justice, which identified states with recently enacted restrictive voter laws;

the New York Times, which identified 2012 and 2016 electoral battleground states;

the Federal Election Commission, which compiled data on campaign spending by state;

Ballotpedia, which identified the number of ballot initiatives by state; and

IDHE data, which identified community colleges where external GOTV organizations had a presence in 2016.¹

See Table 1 for a complete list and source information.

Table 1.

Variables and Data Source.

Variable(s)	Data source^a	Link to information
Student Voting Information	Catalist	https://catalist.us/data
Individual Characteristics (e.g., age, sex, race, major, enrollment status)	National Student Clearinghouse	https://studentclearinghouse.org/
Organizational Characteristics (e.g., total enrollment, urbanicity, percent women, percent students of color)	IPEDS	https://nces.ed.gov/ipeds/use-the-data
Organizational Characteristics (e.g., presence of get-out-the-vote programs)	IDHE	https://idhe.tufts.edu/
Political Context (e.g., ballot initiatives)	Ballotpedia	https://ballotpedia.org/Main_Page
Political Context (e.g., total campaign spending by state)	Federal Election Commission	https://www.fec.gov/
Political Context (e.g., electoral battleground state)	New York Times	https://www.nytimes.com
Political Context (e.g., recently enacted restrictive voting laws)	Brennan Center for Justice	http://www.brennancenter.org

Note. IPEDS = Integrated Postsecondary Education Data System; IDHE = Institute for Democracy & Higher Education.

All data collected were for the 2012 and 2016 years.

Sample

Our study design included two independent analytical samples with close to 1.5 million student records divided between the 2012 and 2016 datasets. Almost 60% of our sample was women, with the average age being 26. The majority were White students, and the most prevalent major was humanities. Just under half voted each election year, and there was little variation between the descriptive statistics for the 2012 sample and the 2016 sample (see Table 2 for descriptive information).

Table 2.

Descriptive Statistics of Students in the Sample.

Variable	2012	2016
Proportion of voters	0.44	0.48
Average age^a	26.84 (9.91)	26.18 (10.2)
Enrollment status
Part-time	0.63	0.60
Sex
Female	0.58	0.57
Male	0.42	0.43
Race
White	0.54	0.47
Black	0.18	0.17
Latinx	0.23	0.27
Asian	0.03	0.04
Other race	0.02	0.05
Major
STEM	0.12	0.13
Humanities	0.46	0.45
Social sciences	0.11	0.11
Health professions	0.18	0.18
Business	0.12	0.12
Number of students	408,928	740,957

Note. STEM = science, technology, engineering, and math.

Standard deviation for age is in parentheses.

Table 3 contains the descriptive information for the institutions and state political context of our sample. The 2012 panel from the NSLVE study contains 68 community colleges in 22 states, and the 2016 panel from the NSLVE study contains 129 community colleges in 28 states. There was a total of 62 institutions present in both panels. Similar to the student characteristics, there is not a lot of significant variation between the two samples with a slightly higher proportion of students in the 2016 panel voting than in the 2012 panel. In addition, the average campaign spending on the election was higher in 2012, but the variation of spending by state increased in 2016.

Table 3.

Descriptive Statistics for the Colleges in the Sample.

Variable	2012	2016
Proportion of voters	0.44	0.48
Proportion of voters	0.5	0.5
Proportion of students of color	0.42	0.46
Proportion of students of color	0.01	0.01
Proportion of women	0.58	0.57
Proportion of women	0.01	0.01
Total ballot initiatives	4.7	4.06
Total ballot initiatives	5.09	5.39
Log election spending^a	US$5,265.33	US$4,991.64
Log election spending^a	(US$2,055.09)	(US$2,791.83)
Total enrollment	28,600	22,230
Total enrollment	2,521	2,060
In battleground state	0.31	0.42
In battleground state	0.44	0.60
In state with recent restrictive voting law	0.43	0.26
In state with recent restrictive voting law	0.6	0.7
Get-out-the-vote programs (0–7)	N/A	1.7
Get-out-the-vote programs (0–7)	N/A	1.63

Number of colleges	68	129

Note. The numbers in Italics are Standard Deviations.

Standard deviation for spending is in parentheses.

Variables

Many studies of college student voting behavior are based on student self-reports (Hillygus, 2005; Nie & Hillygus, 2001; Niemi & Hanmer, 2010; Vogelgesang & Astin, 2005). One strength of this study is that our measure of voting is a publicly available record of whether a student actually voted in the election. Because both student and institutional characteristics are associated with student behaviors and outcomes, we included both types of variables in the model. At the student level, we included age at time of election, gender, race or ethnicity, enrollment status (part-time or full-time), and primary field of study. Previous analyses of college student voting indicate that these variables are related to voting (e.g., Bergom & Ro, 2018; Thomas et al., 2017). At the institution level, we included total enrollment and proportion of women and students of color enrolled. We incorporated these variables to account for differences between organizational contexts. We also included characteristics of the surrounding area, including urbanicity of the institution’s location (i.e., city, rural, suburb, town), whether the college is located in an electoral battleground state or a state that recently enacted restrictive voter laws. Finally, we included a variable indicating the total number of GOTV organizations (of the seven we analyzed) that had a presence on the student’s campus in the lead-up to the 2016 election. This variable was collected as part of the construction of the voting database. While we know how many GOTV programs were on a campus in 2016, details about the goals or specific activities of these individual programs were unavailable.

Analytical Approach

We developed a binary logistic regression model with two levels, including a random intercept for the institution, to determine factors that were significantly related to community college student voting in the 2012 or 2016 U.S. presidential election. The inter-class correlation coefficient (ICC) for each model is negligibly different from zero. However, we estimated a two-level random intercept model because when we examined the community-level residuals (variation at the organizational level), there does appear to be a trend in voter turnout rate that is related to organizational differences. We also estimated a three-level model with a random intercept for state, but the significant results did not change. As there were a handful of states with only one institution in our dataset, we ultimately opted not to include the state-level predictor. We present two analytical samples because slightly different data sources were available for each election. We report the results in odds ratios (ORs) to ease interpretation.

We estimated a random intercept for each institution to identify potential average differences among individual institutions. The structural equation for the multi-level logistic regression using the Bernoulli sampling model is expressed as follows:

Prob (Υ_{i j k} = | β_{j k}) = ϕ_{i j k} .

(1)

We also estimated a model for between- and within-institution factors (see Titus, 2006, for a discussion), which is expressed as follows:

\begin{array}{l} η_{i j k} = \log [\frac{ϕ_{i j k}}{1 - ϕ_{i j k}}] = π_{0 j} + π_{1 j} \times {(Individual Characteristics)}_{i j} + \\ π_{2 j} \times {(Electoral Factors)}_{i j} . \end{array}

(2)

In this model, $η_{i j}$ represents the estimated log odds that an individual student would vote in the election, and $ϕ_{i j k}$ is the predicted probability ranging from 0 to 1. The outcome is estimated for individual i within community college j, based on the average outcome in j.

The between-institution model accounts for the variance in intercepts among institutions:

π_{0 j} = β_{00 j k} + 01 \times {(Organizational Characteristics)}_{j} + r_{o j k},

(3)

where j denotes that the institution and organizational characteristics are a vector of the college-level variables.

Findings

Based on our descriptive analysis, we did not observe any immediate trends in differences between voters and non-voters on the individual level (see Table 4). In general, students were less likely to turn out to vote in 2016, with the exception of White students who turned out at higher levels than their peers. The largest decrease in voter turnout was among Black students, who had significantly higher voter turnout in 2012. Figure 1 provides a graphical depiction of voting behavior by state in 2012 and 2016.

Table 4.

Individual Characteristics by Voting Behavior.

	2012		2016
Variable	Non-voter	Voter	Non-voter	Voter
Age at time of election^a	25.52(8.56)	28.49(11.16)	27.5(11.6)	24.87(8.4)
Race
White	52%	56%	42%	51%
Black	14%	21%	16%	17%
Latinx	26%	18%	29%	23%
Asian/Asian Americans and Pacific Islanders	4%	1%	5%	2%
Other race/ethnicity	2%	2%	5%	4%
Major
STEM	12%	12%	14%	13%
Humanities	45%	45%	44%	45%
Social sciences	12%	11%	10%	10%
Health professions	11%	12%	18%	18%
Business	18%	19%	12%	12%
Enrollment status
Part-time	17%	17%	16%	17%
Sex
Women	56%	61%	54%	59%
Men	44%	39%	45%	40%

Note. STEM = science, technology, engineering, and math.

Standard deviation for age is in parentheses.

Figure 1.

Two-year college proportion of the sample who voted in the respective year’s election by state.

Table 4 presents a descriptive analysis of voters and non-voters for the 2012 and 2016 elections. The first panel includes students who were attending one of 68 community colleges in the sample for the 2012 election year (n = 408,928 students and 68 institutions). The average age of student voters in 2012 was 28, compared with 25 for non-voters and 26 for all students in the panel.

Regression Results

To understand the relationship among individual-, organizational-, and state-election-level factors, we now turn our attention to the results of the multi-level logistic regression model (see Table 5).

Table 5.

Fixed Effects Logistic Regression Models for 2012 and 2016 Election Voting.

	2012				2016
	Odds ratio	95% CI		p	Odds ratio	95% CI		p
Variable	Odds ratio	LL	UL	p	Odds ratio	LL	UL	p
(Intercept)	0.11	0.05	0.26	<.001***	0.06	0.04	0.08	<.001***
Age	1.03	1.03	1.03	<.001***	1.03	1.03	1.03	<.001***
Enrollment status (Full-time enrolled)
Part-time	0.96	0.95	0.98	<.001***	0.97	0.96	0.98	<.001***
Sex (Male)
Women	1.18	1.16	1.19	<.001***	1.15	1.14	1.17	<.001***
Race/ethnicity (White)
Asian/Pacific Islander	0.34	0.33	0.36	<.001***	0.83	0.82	0.84	<.001***
Black	1.37	1.35	1.40	<.001***	0.72	0.71	0.73	<.001***
Hispanic	0.71	0.70	0.73	<.001***	0.35	0.34	0.36	<.001***
Other race	0.84	0.81	0.88	<.001***	0.75	0.73	0.77	<.001***
Academic field of study (Science, technology, engineering, and math)
Business	1.00	0.97	1.02	.747	1.09	0.98	1.02	.854
Health professions	1.13	1.10	1.16	<.001***	1.14	1.12	1.16	<.001***
Humanities	1.08	1.06	1.11	<.001***	0.97	0.96	0.98	<.001***
Social sciences	0.98	0.96	1.01	.211	1.14	1.12	1.16	<.001***
Total enrolled (per 1,000)	1.00	0.99	1.00	.625	1.00	1.00	1.01	.046
Urbanicity (City)
Rural	0.84	0.69	1.03	.099	0.86	0.75	0.98	.026
Suburb	0.91	0.77	1.07	.261	1.10	0.98	1.22	.093
Town	0.78	0.63	0.95	.015*	0.79	0.69	0.91	.001**
Proportion of students of color
Proportion of women	2.97	1.61	5.49	<.001***	2.30	1.64	3.21	<.001***
Ballot initiatives^a	1.02	1.00	1.03	.026*	1.01	1.00	1.02	.196
Election spending^b	1.06	0.96	1.17	.277	1.21	1.15	1.27	<.001***
Restrictive voting laws	1.16	1.00	1.33	.042*	0.88	0.78	0.98	.024*
Battleground states	1.22	1.04	1.43	.015*	1.16	1.05	1.27	.003**
Get-out-the-vote programs (1–7)					0.99	0.95	1.03	.551
Random intercept	2012				2016
τ_00,opeid	0.058				0.056
N_opeid	64				124
ICC_opeid	.017				0.017
Observations	408,928				740,957

Note. CI = confidence interval; LL = lower limit; UL = upper limit; ICC = inter-class correlation coefficient.

Clarifications: ^a Total number of ballot initiatives in the state. ^b Per capita spending on the election (logarithmic transformation).

p < .05. **p < .01. ***p < .001.

We find some support that students’ demographic characteristics and academic major predict voter turnout among community college students. The regression results suggest that voting trends among community college students resemble national trends in voting. As age increased, students’ odds of voting increased slightly in both the 2012 and 2016 models (OR = 1.03/1.03, p < .001 for both); women had higher odds of voting than men in both models (OR = 1.18/1.15, p < .001); and Latinx, Asian, and students of Other races had lower odds of voting than White students. However, Black students turned out at higher rates in the 2012 election than White students (OR = 1.37, p < .001) and lower rates than White students in 2016 (OR = 0.72, p < .001).

Regarding students’ areas of study, we found that students studying health professions had higher odds of voting than students in STEM fields. Students studying humanities had higher odds of voting than STEM students in 2016 and lower odds in 2012. Students studying social sciences had higher odds of voting in 2016 than students in STEM. In addition, we found that part-time students had lower odds of voting when compared with full-time students in both models (OR = 0.96/0.97, p < .001).

We found little evidence that institutional urbanicity is significantly related to students’ odds of voting. Students enrolled at colleges in towns,² as defined by the National Center for Educational Statistics, were less likely to turn out in 2012 and 2016, relative to students enrolled in cities. For both elections, we found no statistically significant difference in turnout between students enrolled in rural communities and suburban communities, compared with students enrolled in cities (OR = 0.78/0.79).

We found some support for the notion that compositional diversity is significantly related to differences in voter turnout. As the percentage of students of color enrolled increased, odds of turnout decreased (OR = 0.61, p < .01), but this relationship was statistically significant only in 2012. For both election cycles, a higher percentage of women enrolled at the institution was associated with higher odds of voting (OR = 2.97/2.30, p < .001 for both).

We found mixed results when we examined factors related to state political context and voter turnout. Students attending community college in states with recently enacted restrictive voting laws were less likely to turn out in 2016 (OR = 0.88, p < .05) and more likely to turn out in 2012 (OR = 1.16, p < .05), relative to students in other states. An increase in the number of ballot initiatives was related to a slight increase in odds of voting in 2012 (OR =1.02, p < .05), but not in 2016. Students attending college in electoral battleground states were more likely to turn out in both elections (OR = 1.22/1.16, p < .05 and p < .01). Finally, per capita spending on election campaigns was significantly related to increased turnout in 2016 (OR = 1.21, p < .001), but not in 2012.

Discussion and Implications

Given the prevalence of community college enrollment (over 12 million total students) in higher education in the United States (AACC, 2019), and the substantial number of students from marginalized communities (two out of every five students) who make up enrollments at community colleges, the lower voter turnout rate among community college students should be a serious area of concern for leaders and policy makers interested in civic engagement. This study builds on past research that has only recently begun to analyze and provide insight into factors associated with increased civic engagement behaviors of community college students (Kisker, Newell, et al., 2016; Thomas et al., 2017). We contribute to that literature by offering a multi-level perspective on significant influences on student voting behavior. We also offer evidence from voting records as opposed to self-reported voting behaviors. While prior research has been descriptive of trends in community college student voting, our research offers insight into areas of concern and opportunities for future research.

Our findings align with past literature that suggests lower rates of voter turnout among men and students of color, compared with women and White students, respectively (Benenson & Bergom, 2016; Hillygus, 2005). We also provide further evidence of differences in voting behavior by college majors, in that students who major in social sciences tend to have higher odds of voting than STEM or business students (Bergom & Ro, 2018; Nie & Hillygus, 2001). However, we did find one anomaly in that Black students had higher odds of voting when compared with their White peers in 2012, but lower odds of voting in 2016. This could be associated with the presidential campaign of Barack Obama, the first African American president of the United States, which may have increased the impact of the varied aspects of political mobilization in the African American community, particularly among religious congregations (Calhoun-Brown, 1996; Djupe & Neiheisel, 2019; Smith & Walker, 2013). Further research could focus on the 2012 election year, given its unique significance for increased participation among Black voters. Specifically, scholars could tease apart the impact of racial group consciousness (Austin et al., 2012; Smith & Walker, 2013), from the role of community cultural engagement (Calhoun-Brown, 1996; Djupe & Neiheisel, 2019), particularly among the community college student population.

Of concern to practitioners and program developers interested in promoting civic engagement activities, such as voting, we did not find a relationship between GOTV programs at community colleges and voting (Kisker, Weintraub, et al., 2016). Much of the work focused on GOTV initiatives has been conducted at 4-year institutions. Our findings may stem from the structural and cultural differences between 4-year institutions and community colleges. For example, community college students tend to be commuters who live within the community but not necessarily on their campus, which may reduce the effectiveness of a campus-based program. Furthermore, community college faculty are more likely to be part-time or contingent (McNaughtan et al., 2017), which could also limit the exposure that even the faculty would have with these programs on community college campuses. We should also note that our ability to draw inferences about GOTV programs is limited as we know only how many programs an institution had but not the scope or nature of their activities on campus. Future researchers should turn their attention to the qualitative aspects that differentiate programs, and how community college students respond. The Community College Civic Outcomes Survey has started to explore this important work (Kisker et al., 2017).

Moving forward, we suggest researchers and practitioners interested in encouraging voter turnout pay specific attention to the nature of the election in which students will be participating. Restrictive voting laws, battleground election states, and to some extent the presence of multiple ballot initiatives were significantly related to turnout. It may be that researchers in higher education underestimate the importance of how students perceive the opportunity to vote (i.e., Is it difficult? Will my vote actually make a difference?). In addition, community college students, while far from homogeneous, may represent a special interest group with overlapping policy concerns. Programming around ballot initiatives aimed at the issues they care about may reach students on campus before an election to encourage participation and engagement.

One of the most interesting contributions of this study to research on community college student voter turnout is focused on the relationship between state political climate and student voting. While most of our results in this area are mixed between the two analyses, we found that students in battleground states (in 2012 and 2016) and students in states with more ballot initiatives (in 2012) had higher odds of voting. Further analysis should be conducted to better understand whether students’ desire to vote increases as additional ballot initiatives related to their industry of interest are proposed.

Implications for Community College Leaders

This study has a number of potential implications for community college leaders. First, institutions need to help students identify both with candidates running for office and the issues invoked by elections, as our findings indicate this may increase their likelihood of voting. Civic engagement programs should focus efforts on helping students to understand how the issues on the ballot and the views of the candidates running are connected to their lives. For example, additional emphasis on candidates’ stance on Pell grants, or industry-specific legislation that could affect the region where the college is located, may increase student voting. This finding aligns with the work of Kisker, Weintraub, et al. (2016) who argue that increased political dialogue on campus can increase civic engagement.

Second, institutions could lobby for less restrictive voting laws. In our 2016 analysis, we found that students in states with recently enacted restrictive voting laws have lower odds of voting (although the opposite was true in 2012). Given the already low levels of voter turnout among college students, especially at community colleges, institutions need to be engaged at the state level in supporting measures that can help their students stay civically engaged. This is especially critical given than many restrictive voting laws negatively affect students of color disproportionately (Bentele & O’Brien, 2013).

Third, institutional leaders should implement evaluation programs of their civic engagement initiatives, particularly GOTV programs. This study did not reveal a statistically significant relationship between these programs and student voting. As institutions increase their investment in civic engagement, they will need to ensure that their programs are achieving their goals.

Directions for Future Inquiry

This study provides insight into a number of potential directions for future inquiry. First, the analytical framework outlined for this analysis could be applied to additional measures of student civic engagement. In this study, we utilized students’ voting records; however, institutional leaders could identify other measures, including community service, participation in community events, or engagement in community government. These measures of civic engagement may provide insight into how different individual, organization, or state contextual factors are connected with different aspects of civic engagement.

Second, future research could focus on the role of local elections in civic engagement. Given the focus on community colleges in this study, one of the limitations of our analysis is that we focus on the state-level political context when community colleges are often closely tied to their local municipalities. Future studies in this area could investigate whether local ballot initiatives or campaign spending at the local level follow the same relationships as found at the state level. Examining specific ties between candidates and the college may also yield potentially useful findings.

Third, qualitative inquiry to investigate what motivates students to be civically engaged would be informative. The quantitative nature of this analysis allows us to discuss odds of voting for specific students by characteristics but understanding why requires robust qualitative research. The qualitative approach could include interviews with students before or after an election to better understand what institutions can do to promote civic behavior.

Fourth, this study involved a cross-sectional approach to 2012 and 2016 data. Future analysis could take a more longitudinal approach to see how voting rates at institutions change over time. This could be especially important for institutions that add new civic engagement programs or devote institutional resources to support new programs aimed at increasing specific civic engagement interests. The data in this study retrieved from IDHE included 62 institutions that participated in both the 2012 and 2016 iterations of the study. Utilizing a more in-depth approach that could account for year to year or municipal elections could help to better understand how structural and programmatic changes at the campus level are associated with voter turnout.

Finally, this study did not suggest a statistically significant relationship between a campus having at GOTV program and student voting, but future research could differentiate programs and examine whether the amount of money, staff, and so forth, predicts increases in student voting. Additional analysis might discover that institutions with specific fully functioning GOTV programs have success, and that some programs are ineffective regardless of the resources used to support them.

Conclusion

Through this study, we sought to enhance our understanding of the relationship between organizational characteristics, state political climate, and student civic engagement at community colleges. We employed a binary logistic regression technique to analyze a unique cross-sectional dataset for the 2012 and 2016 presidential elections and found a number of statistically significant relationships. In summary, we found that student civic engagement is associated with some organizational characteristics and features of state political contexts, such as restrictive voter laws and being located in an electoral battleground state. Specifically, our results indicate that students at institutions with a higher proportion of women had higher odds of student voting. The relationship between voting and student characteristics generally aligns with past research on national voting trends, in that students who are female, older, and White tend to have a higher odds of voting. These findings provide insight for community college leaders to utilize as they work to promote civic engagement on their campuses through targeted programming and staff engagement with students.

Footnotes

Acknowledgements

The authors would like to acknowledge the consistent support of Inger Bergom who was instrumental in the development and completion of this study.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Jon McNaughtan

Notes

Author Biographies

Jon McNaughtan is an assistant professor of Higher Education In the College of Education at Texas Tech University in Lubbock, Texas, where he also serves as the co-director of Education Policy Fellowship Program. His research focuses on the role of higher education leaders to empower faculty, staff, and students.

Michael Brown is an assistant professor in Student Affairs and Higher Education in the School of Education/College of Human Sciences at Iowa State University. His research focuses on the impact of social and academic peer relationships on undergraduate student outcomes and on the mediating role of digital technologies in undergraduate curriculum, teaching, and learning.

References

American Association of Community Colleges. (2019). Fast facts 2018. https://www.aacc.nche.edu/wp-content/uploads/2019/05/AACC2019FactSheet_rev.pdf

Austin

S. D. W.

Middleton

R. T.

Yon

(2012). The effect of racial group consciousness on the political participation of African Americans and Black ethnics in Miami-Dade County, Florida. Political Research Quarterly, 65(3), 629–641. https://doi.org/10.1177/1065912911404563

Bartels

L. M.

(2000). Partisanship and voting behavior, 1952-1996. American Journal of Political Science, 44(1), 35–50. https://doi.org/10.2307/2669291

Benenson

Bergom

(2016, November 19–22). Political participation for all? Institutional predictors of college student voting and the role of student socioeconomic status [Paper presentation]. Association for the Study of Higher Education Annual Meeting, Columbus, OH, United States.

Bennion

E. A.

Laughlin

X. E.

(2018). Best practices in civic education: Lessons from the Journal of Political Science Education. Journal of Political Science Education, 14(3), 287–330. https://doi.org/10.1080/15512169.2017.1399798

Bentele

K. G.

O’Brien

E. E.

(2013). Jim Crow 2.0? Why states consider and adopt restrictive voter access policies. Perspectives on Politics, 11(4), 1088–1116. https://doi.org/10.1017/s1537592713002843

Bergom

(2018, January 29). Why don’t STEM majors vote as much as others? The Conversation. https://theconversation.com/why-dont-stem-majors-vote-as-much-as-others-89015

Blais

(2000). To vote or not to vote? The merits and limits of rational choice theory. University of Pittsburgh Press.

Bowman

N. A.

(2011). Promoting participation in a diverse democracy: A meta-analysis of college diversity experiences and civic engagement. Review of Educational Research, 81(1), 29–68. https://doi.org/10.3102/0034654310383047

10.

Calhoun-Brown

(1996). African American churches and political mobilization: The psychological impact of organizational resources. The Journal of Politics, 58(4), 935–953. https://doi.org/10.2307/2960144

11.

Campante

F. R.

Chor

(2012). Schooling, political participation, and the economy. The Review of Economics and Statistics, 94(4), 841–859. https://doi.org/10.1162/rest_a_00206

12.

Colby

Beaumont

Ehrlich

Corngold

(2007). Educating for democracy: Preparing undergraduates for responsible political engagement. Jossey-Bass.

13.

Djupe

P. A.

Neiheisel

J. R.

(2019). Political mobilization in American congregations: A religious economies perspective. Politics and Religion, 12(1), 123–152. https://doi.org/10.1017/s1755048318000512

14.

Elder

Seligsohn

Hofrenning

(2007). Experiencing New Hampshire: The effects of an experiential learning course on civic engagement. Journal of Political Science Education, 3(2), 191–216. https://doi.org/10.1080/15512160701338379

15.

Feddersen

T. J.

Pesendorfer

(1997). Voting behavior and information aggregation in elections with private information. Econometrica, 65(5), 1029–1058. https://doi.org/10.2307/2171878

16.

File

(2013). The diversifying electorate—Voting rates by race and Hispanic origin in 2012 (and other recent elections) (Current Population Survey Reports, P20-569). U.S. Census Bureau.

17.

Finley

(2011). Civic learning and democratic engagements: A review of the literature on civic engagement in post-secondary education. Association of American Colleges & Universities. https://www.aacu.org/sites/default/files/files/crucible/LiteratureReview.pdf

18.

Gerber

A. S.

Green

D. P.

(2017). Field experiments on voter mobilization: An overview of a burgeoning literature. In Duflo

Banerjee

(Eds.), Handbook of economic field experiments (Vol. 1, pp. 395–438). North-Holland.

19.

Ginder

S. A.

Kelly-Reid

J. E.

Mann

F. B.

(2017). Enrollment and employees in postsecondary institutions, fall 2016, and financial statistics and academic libraries, fiscal year 2016: First look (provisional data) (NCES 2018-002). National Center for Education Statistics, U.S. Department of Education.

20.

Glastris

(2018, March). Colleges that encourage students to vote help the country—And now their rankings. Washington Monthly. https://washingtonmonthly.com/2018/03/05/colleges-that-encourage-students-to-vote-help-the-country-and-now-their-rankings/

21.

Hillygus

D. S.

(2005). The missing link: Exploring the relationship between higher education and political engagement. Political Behavior, 27(1), 25–47. https://doi.org/10.1007/s11109-005-3075-8

22.

Institute for Democracy & Higher Education Staff. (2017). How institutional context matters for college student voting: An analysis of 2012 voting rates. https://idhe.tufts.edu/sites/default/files/idhe_regression_analysis_2012.pdf

23.

Kisker

C. B.

(2016). An inventory of civic programs and practices. New Directions for Community Colleges, 2016(173), 13–21. https://doi.org/10.1002/cc.20185

24.

Kisker

C. B.

Newell

M. A.

Weintraub

D. S.

(2016). The civic outcomes of community college. New Directions for Community Colleges, 2016(173), 103–110. https://doi.org/10.1002/cc.20195

25.

Kisker

C. B.

Weintraub

D. S.

Newell

M. A.

(2016). The community colleges’ role in developing students’ civic outcomes: Results of a national pilot. Community College Review, 44(4), 315–336. https://doi.org/10.1177/0091552116662117

26.

Kisker

C. B.

Weintraub

Newell

M. A.

(2017). Community college civic outcomes survey, spring 2015. Inter-university Consortium for Political and Social Research [distributor]. https://doi.org/10.3886/ICPSR36833.v1

27.

Kushin

Yamamoto

(2010). Did social media really matter? College students’ use of online media and political decision making in the 2008 election. Mass Communication and Society, 13(5), 608–630. https://doi.org/10.1080/15205436.2010.516863

28.

Lopez

M. H.

Brown

(2006). Civic engagement among 2-year and 4-year college students. Center for Information & Research on Civic Learning & Engagement. https://files.eric.ed.gov/fulltext/ED494028.pdf

29.

Mathews

(2016). The community college’s role in helping to make democracy work as it should. New Directions for Community Colleges, 2016(173), 33–40. https://doi.org/10.1002/cc.20187

30.

McDonald

(2018). Voter turnout demographics. U.S. Election Project. http://www.electproject.org/home/voter-turnout/demographics

31.

McNaughtan

García

H. A.

Nehls

(2017). Understanding the growth of contingent faculty. New Directions for Institutional Research, 2017(176), 9–26. https://doi.org/10.1002/ir.20241

32.

Myers

C. B.

Myers

S. M.

Peters

(2019). The longitudinal connections between undergraduate high impact curriculum practices and civic engagement in adulthood. Research in Higher Education, 60(1), 83–110. https://doi.org/10.1007/s11162-018-9504-4

33.

Nie

Hillygus

D. S.

(2001). Education and democratic citizenship. In Ravitch

Viteritti

J. P.

(Eds.), Making good citizens: Education and civil society (pp. 30–57). Yale University Press.

34.

Niemi

R. G.

Hanmer

M. J.

(2010). Voter turnout among college students: New data and a rethinking of traditional theories. Social Science Quarterly, 91(2), 301–323. https://doi.org/10.1111/j.1540-6237.2010.00694.x

35.

Persson

(2015). Education and political participation. British Journal of Political Science, 45(3), 689–703. https://doi.org/10.1017/s0007123413000409

36.

Prentice

(2011). Civic engagement among community college students through service learning. Community College Journal of Research and Practice, 35(11), 842–854. https://doi.org/10.1080/10668920802205014

37.

Reason

R. D.

Hemer

(2015). Civic learning and engagement: A review of the literature on civic learning, assessment, and instruments. Association of American Colleges & Universities. https://www.aacu.org/sites/default/files/files/qc/CivicLearningLiteratureReviewRev1-26-15.pdf

38.

Bergom

(2017, November). Who votes in college? Differences by gender, race/ethnicity, and field of study [Paper presentation]. Association for the Study of Higher Education, Houston, TX, United States.

39.

Ronan

Kisker

C. B.

(Eds.). (2016). New directions for community colleges: Civic learning and democratic engagement (Vol. 173). Jossey-Bass.

40.

Saltmarsh

Hartley

(Eds.). (2011). “To serve a larger purpose”: Engagement for democracy and the transformation of higher education. Temple University Press.

41.

Sax

L. J.

(2004). Citizenship development and the American college student. New Directions for Institutional Research, 122, 65–80. https://doi.org/10.1002/ir.110

42.

Scott

J. C.

(2006). The mission of the university: Medieval to postmodern transformations. The Journal of Higher Education, 77(1), 1–39. https://doi.org/10.1080/00221546.2006.11778917

43.

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. https://doi.org/10.1177/1065912912443873

44.

Swearer Center at Brown University. (2018). Carnegie Foundation for the Advancement of Teaching elective community engagement classification. https://drive.google.com/file/d/1pWgsE4gq4CLdQhWb7wXvEClz9I4eJgBK/view

45.

Thomas

Bergom

Casellas Connors

Gautam

Gismondi

Roshko

(2017). Democracy counts: A report on U.S. college and university student voting. Institute for Democracy & Higher Education, Jonathan M. Tisch College of Civic Live, Tufts University.

46.

Thomas

Brower

(2017). Politics 365: Fostering campus climates for student political learning and engagement. In Matto

E. C.

McCartney

A. R. M.

Bennion

E. A.

Simpson

(Eds.), Teaching civic engagement across the disciplines (pp. 361–374). American Political Science Association.

47.

Timpone

R. J.

(1998). Structure, behavior, and voter turnout in the United States. American Political Science Review, 92(1), 145–158. https://doi.org/10.2307/2585934

48.

Titus

M. A.

(2006). Understanding college degree completion of students with low socioeconomic status: The influence of the institutional financial context. Research in Higher Education, 47(4), 371–398.

49.

Traver

A. E.

Katz

Z. P.

(Eds.). (2014). Service-learning at the American community college: Theoretical and empirical perspectives. Palgrave Macmillan.

50.

Vogelgesang

Astin

(2005). Post-college civic engagement among graduates (Research Report No. 2). Higher Education Research Institute, University of California, Los Angeles.