In elections, the assumption that more choice leads to better outcomes is not always true. A recent study by Spencer Goidel at Auburn University examines the phenomenon of “choice overload” in elections and how it can affect voter behavior. The study suggests that the abundance of candidates in elections leads to higher “rolloff” rates, a situation in which voters participate in an election but do not vote in some of the races on the ballot.
The results were published in Research on American politics.
Political scientists have long been interested in how the structure of elections—such as the number of candidates or the type of primary—affects voter turnout. Goidel’s research focuses specifically on Louisiana’s unique “jungle primary” system, a nonpartisan, first-past-the-post primary held on Election Day. This system can lead to a crowded field of candidates, especially from the same political party, and Goidel wanted to explore whether this abundance of choices could lead to voter turnaround.
The study draws on theories from consumer psychology, particularly the concept of “choice overload,” which states that when people are faced with too many similar options, they may become overwhelmed and either decide against a decision or make a choice they later regret. The aim of the study was to find out whether this concept applies to voting behavior, especially in elections with many candidates.
“After the 2016 Republican presidential primaries, I kept hearing that there were no good options, and that completely baffled me. On the Republican side, there were more than a dozen experienced candidates vying for the nomination in 2016,” said Goidel, an assistant professor of political science at Auburn University.
“Then the same thing happened to the Democrats in 2020. That got me thinking about how large fields of candidates would affect voters. With so many options, it seemed like people would start ‘cancelling’ candidates for the smallest of reasons, and they were hesitant to strongly support any one candidate. At the most basic level, I thought large fields of candidates would cause voters to abstain or drop their ballots at the ballot box.”
To study this, Goidel used a natural experiment created by changes to voting rules in Louisiana due to Supreme Court decisions that led to the abandonment and later reinstatement of the jungle primary system. This unique context allowed the researcher to compare voter behavior under different electoral systems within the same state.
“The case of the Louisiana jungle primary seemed obvious to me because Louisiana voters would be choosing between many candidates at the same time in these elections, while being able to choose between two presidential candidates. This provides a good basis for ‘normal’ voter turnout against which to measure vote share,” Goidel explained.
For his aggregate-level analysis, Goidel used a difference-in-differences design and compared enrollment rates in Louisiana to those in Mississippi, a neighboring state with a different electoral system. He also used a generalized synthetic control model and constructed control districts using data from Arkansas, Mississippi, and Texas to further validate the results. Using these models, Goidel was able to isolate the effects of the jungle primaries on enrollment rates at the district level.
In addition to the overall analysis, Goidel conducted an individual-level analysis using survey data from participants in the Louisiana Cooperative Election Study. This part of the study examined whether voters in congressional districts with more candidates were more likely to not vote in House elections. By combining these two approaches, Goidel was able to assess the impact of candidate diversity on voter behavior from both a broad population-level perspective and a more detailed individual-level perspective.
The results of both analyses were consistent, supporting the hypothesis that a large number of candidates in an election increases the likelihood of no-votes. At an aggregate level, Goidel found that the reinstatement of the jungle primary in Louisiana led to a 5.7 to 7.1 percent increase in voting compared to when the jungle primary was not introduced. This means that a significant portion of voters who were presented with more candidates on the ballot chose not to vote in the House elections, even though they had voted.
Analysis at the individual level further reinforced these findings. The study showed that for each additional candidate in a general election, the probability that a voter would cast their vote increased by one percent. The probability that a voter would abstain in a general election increased significantly with the number of candidates, particularly when that number exceeded two. For example, in elections with up to twelve candidates, the probability that a voter would cast their vote increased from 2.9 percent to 19.1 percent. This suggests that even a moderate increase in the number of candidates can lead to a significant increase in voter defection.
“Our standard response to dissatisfaction with politics is to reform our institutions to give voters more choice,” Goidel told PsyPost. “I would point out that democratizing the process does not always lead to better outcomes. Democratizing the primary process could inadvertently burden and demobilize parts of the electorate.”
Although the study provides insights into the impact of candidate diversity on voter behavior, it is important to consider its limitations. One important limitation is that the study focuses specifically on Louisiana and its unique jungle primary system. Although the results are also relevant for other states with similar electoral systems, such as California and Alaska, the generalizability of the results to other contexts may be limited.
“In Louisiana, ballot entry declined in 2020 when mail-in voting was at an all-time high,” Goidel noted. “Of course, we can’t know what caused this decline, but I think it makes sense that voting at home with internet access and time would be less burdensome. In a state like Washington, where most voters vote by mail, it’s reasonable to assume that the negative impact of congestion would be offset by this more targeted voting method.”
“A polling experiment I conducted shows that large candidate fields overwhelm voters. That paper is in preparation, and I plan to study how primaries for the top two or top four candidates introduced in other states have affected voters.”
The study “More Candidates and Fewer Voters: How an Abundance of Choices Demobilizes the Electorate” was published on August 13, 2024.
