Voter Turnout in the U.S.: A Data-Driven Learning Guide

Data for this exercise come from the 2004 American National Election Study (ANES): Pre- and Post-Election Survey. The ANES grew out of the Survey Research Center and the Center for Political Studies of the Institute for Social Research at the University of Michigan. These organizations, together, have been covering elections since 1952. ANES produces high quality data on voting, public opinion, and political participation to serve the research needs of social scientists, teachers, students, policy makers and journalists who want to better understand the theoretical and empirical foundations of national election outcomes.

Respondents in the 2004 ANES were interviewed before and after the November election. Questions cover a broad range of topics including demographics, attitudes toward candidates and parties, attitudes on different segments of the American public, attitudes on foreign policy matters, and political behavior. The ANES uses random sampling in order to produce the most representative data possible on the American electorate.

Variables used in this exercise include:

• Respondent voted (v045017b)
• Age (v043250)
• Race (v043299)
• Household income (v043293x)
• Education (v043254)

In this exercise, you will examine voter turnout by age, race, income, and education using crosstabulation.

In this dataset, voter turnout is measured in the post-election survey by the variable V045018X. We collapsed the responses into two categories coded as "0" if the respondent did not vote and coded as "1" if the respondent voted. The new variable is called "VOTE."

Age

In the survey respondents' age ranged from 18 to 90 years old. To make this variable appropriate for crosstabulation, we recoded the variable into three age categories: 18-39; 40-59; and 60-90. We called the new variable "AGE."

Run a crosstab of VOTE by AGE. Which age group has the highest voting rate? Is there enough of a pattern to claim that voting differs by age?

Race

The ANES allows respondents to choose from 12 race categories. Respondents may also use a "don't know" response or a "refuse to answer" response. For the purposes of this exercise, we created a new variable called "WHITE" which is coded as "1" if the reported race is white with no other race mentioned and as "0" if the reported race is any of the categories other than white, not including "don't know" or "refused to answer."

Look at the crosstab of VOTE by WHITE. Is voter turnout higher among whites or nonwhites?

Income

Next consider how voter turnout might vary by household income. In the survey, income was measured using 24 categories. We recoded income into four categories: $0-$19,999 (1); $20,000-$39,999 (2); $40,000-$89,999 (3); and $90,000 and over (4), and called the new variable "HHINCOME."

Run a crosstab of VOTE by HHINCOME. Are respondents with higher incomes more or less likely to vote than those with lower incomes?

Education

We recoded the ANES measure of education (V043254) into four categories: less than H.S. diploma (1); H.S. diploma or equivalency (2); some college, no bachelor's degree (3); and bachelor's degree and higher (4). The new variable is called "EDUC4."

Looking at the crosstab of VOTE by EDUC4, do the highly educated appear to have higher voter turnout rates than those with less education? What percentage of people with a high school diploma voted?

Race and Education

Because nonwhites are less likely than whites to be highly educated, and education is related to voting behavior, an analysis between voter turnout and race should control for the confounding effects of education. Re-examine the relationship between race and voter turnout, controlling for education. This approach will create a separate crosstab for each education level. Does the relationship between voting and race differ by education level? At what education level is the difference between whites and non-whites the smallest?

Think about your answers to the application questions before you click through the interpretation guide for help in answering the questions.

Weights (mathematical formulas) are often used to adjust the sample proportions, usually by race, sex, or age, to more closely match those of the general population. The analyses in this guide used weights to increase the generalizability of the findings, so the resulting tables are meant to reflect the relationships we would expect to see in the general population.

Age

Which age group has the highest voting rate? Is there enough of a pattern to claim that voting differs by age?

Race

Is voter turnout higher among whites or nonwhites?

Income and Turnout

Is turnout related to income? Are respondents with higher incomes more or less likely to vote than those with lower incomes?

Education

Do the highly educated have higher voter turnout rates than those with less education? What percentage of people with a high school diploma voted?

Race and Education

Does the relationship between voting and race differ by education level? At what education level is the difference between whites and non-whites the smallest?