Association for Public Policy Analysis & Management

One way to measure engagement in democracy in higher education is by whom, among college students, votes. Voting is not the only or even best indicator of student political engagement, but it is fundamental and can be measured objectively. However, researchers lack data based on objective measures rather than possibly socially desirable responses to surveys and also information on individual-level data broken down by higher education institution. In the area of youth voting—and political behavior research more generally—scholars use nationally representative surveys that ask about voting along with other forms of political and civic engagement, demographics, and experiences and opinions. Some of these surveys are longstanding and large; for example, the American National Election Studies (ANES) have been conducted in every national election since 1948, and the U.S. Census, through the Current Population Survey (CPS), asks a robust set of civic and political engagement questions. Because these surveys collect numerous indicators from each individual and have large, nationally representative samples, they have historically been the best tools for documenting broad trends such as the relationship between voting and other factors. Yet, the field of higher education lacks the assessment tools needed to measure its effectiveness as civic educators.

This paper introduces and demonstrates the value of the National Study of Learning, Voting, and Engagement (NSLVE), a partial but critical data source to fill research gaps not only about college student voting, but college student political learning and engagement more broadly. Other than NSLVE, colleges and universities have no objective measure of college student political participation. NSLVE provides an opportunity to re-examine the traditionally explored demographic factors (e.g., age, gender) viewed as relevant to student voting. More importantly, because data are collected and maintained at the institutional level, NSLVE provides the first opportunity to examine the institutional-level characteristics of colleges and universities with student voting rates (e.g., percent of Pell grant recipients, Carnegie classification).

This paper first describes the reliable and confidential process we developed for merging student-level enrollment data and publicly available voting records to generate a database of nearly 8 million student records at over 800 participating colleges and universities. The process included (1) recruiting college and university campuses to obtain permission to use student data, (2) partnering with the National Student Clearinghouse (“the Clearinghouse”)  to obtain student enrollment records, (3) purchasing publicly voting records from an organization called Catalist, and (4) working with the Clearinghouse to merge and de-identify student enrollment and voting records. This process resulted in the creation of the NSLVE database.

The paper then demonstrates the ways the dataset can be merged with other data sources such as data from the U.S. Department of Education’s Integrated Postsecondary Education Data System (IPEDS) and other institution- and state-level datasets (e.g., National Student Loan Data System, American Community Survey). Doing so allows us to understand the contribution this dataset can make toward better understanding political learning and engagement in a variety of policy and educational contexts.