By: Arjun Vishwanath, Vanderbilt University

The following blog post is a summary of the research that won the Midwest Political Science Association’s Best Paper in American Politics Award for research presented at the 2023 MPSA Annual Conference.


Voters’ issue stances are central to representation, as they provide guidance to legislators on what policies ought to be adopted. As a result, it is crucial to understand the origins of these stances. Two commonly posited sources are a voter’s left-right ideological identification or her partisanship. However, existing research suggests that many voters do not hold the understandings of the terms liberal and conservative that politicos do, and as a result, they may not use these ideas to make sense of policy particulars. If voters instead rely on partisanship and partisan cues to determine their issue stances, though, then representation becomes circular. In other words, if Republicans adopt anti-Obamacare attitudes because Republican politicians tell them that that is the Republican position, then voters do not have independent attitudes that politicians can represent.

This article evaluates the role of another potential driver of issue stances: political values, or beliefs about the ways in which society should be ordered and the priorities it should have. Some commonly studied values include moral traditionalism, egalitarianism, and individualism. Past work has established correlational relationships between these types of values and issue stances using surveys at a single point in time—for example, people who are more traditionalist are more likely to be pro-life—but these surveys suffer from omitted variable bias. Perhaps someone holds anti-abortion attitudes not because of her traditional values but because her parents instilled the position in her at a young age, for example. If this is the case, then the observed correlation between values and issue stances may not be causal. To deal with this, I use panel studies, which interview the same respondents over time. These studies also let us compare the role of values on issue stances to those of partisanship and ideology, which past work using panel studies does not do.

I use data from six ANES and GSS panel studies that range from 1992 to 2020. I leverage the two values that regularly appear in these panels: traditionalism and egalitarianism. With this data, I test four hypotheses. First, there is a within-individual relationship between values and issue stances. At points in time when a given individual holds more traditionalist values, she is also more likely to adopt more traditionalist policy stances (e.g., anti-abortion, anti-same sex marriage) relative to the other times in which she espouses less traditionalist values. Second, a given respondent’s values at t1 will influence her policy stances at t2 after accounting for her policy stances at t1; in other words, her policy stances at t2 will move to become congruent with her values over time. Third, when a respondent’s values shift between t1 and t2, her policy stances will shift in t3 to reflect that value change. For example, if a voter becomes more egalitarian in her values over a two year period, her policy stances will subsequently move towards in that same direction. Finally, on two emergent and rapidly changing issue domains—welfare reform in the 1990s and transgender policies in the 2010s—voters’ values will be a key driver of their issue attitudes.

I find statistically significant roles for values in each of these four tests, even after controlling for partisanship and ideology. This suggests that values exert independent force and are not merely consumed by the effects of partisanship or ideology. I also find that the magnitude of values’ effects on issue stances are consistently larger than those of partisanship and ideology. That is, when a voter’s issue stances change, it happens because she is an egalitarian moreso than because she is a Democrat or a liberal.

What does this mean for our understanding of voters’ attitudes? Even if voters do not hold sophisticated understandings of ideology, as other work has found, they can still use principled beliefs to drive their attitudes towards particulars. Voters’ attitudes are not random noise, nor are they merely adopting the issue stances of their own political party. Instead, it appears that voters rely on a different set of principled beliefs—values—to make sense of the world around them.