Can Uninformed Voters Pick the Right Party? Theory, Assessment, and Solutions

“How does political knowledge matter for democratic choices?” My thesis contemplates this question. It focuses on political knowledge and how it conditions the voting calculus. I argue that political knowledge impacts how voters make their voting decision and that voters with different degrees of political knowledge rely on different criteria when casting a vote during an election. This project links together and contributes to the literatures on vote choice, political knowledge, and democratic theory.

The normative preference that democratic theorists have for democracy rests in good part on the assumption that citizens can identify which party or candidate best represents their interest. If this is the case, then their vote is meaningful and provides an informed choice regarding who should govern the polity. If it is not, then it is unclear what is the meaning of a democratic vote and how legitimate elections can be as a method to select governing actors. Some political scientists have claimed that deficiencies in political knowledge are not truly a threat to democracy. They argue that even politically uninformed voters can make correct voting decisions by relying on cues or shortcuts (Popkin, 1991; Sniderman et al. 1993, Lupia 1994, Lupia and McCubbins 1998). Others disagree with this statement, finding that the evidence for the presence of shortcuts is lacking and identifying clear discrepancies between the behavior of the informed and the uninformed (Bartels 1996, Carpini and Keeter 1996, Brennan 2012).

My thesis investigates this question in three ways. First, I discuss in detail the concept of political knowledge itself. Little fundamental work has been conducted on political knowledge. While many use the variable as part of their analyses, the concept is undertheorized. I develop a framework that focuses on knowledge heterogeneity and the intrinsic benefits of political knowledge. This stands in contrast to previous approaches which stressed the instrumental benefits of political knowledge (Converse 1964, Luskin 1990, Lupia and McCubbins 1998). This model provides a more detailed explanation for the high variance of political knowledge in the population and reconciles the theory with new findings from the discipline of psychology. The predictions of this theoretical model are tested using data from Canada and the United States. The theoretical model is confirmed by the analysis. It also identifies that linguistic minorities suffer from lower levels of political knowledge. This finding raises concerns regarding the democratic representation of minorities.

Second, I investigate how political knowledge impacts the political behavior of voters. I conducted an original experiment to identify the impact of political knowledge on the voting decision. I randomized information given to participants as a way to experimentally manipulate their perceptions of party labels, parties’ economic performance, and candidates’ ideology. I then examine how these randomly determined changes impact the voting decision of voters with different levels of political knowledge. I find that ideology has a greater impact on the vote of informed voters while party identification and economic considerations play the same role for all voters regardless of their level of political knowledge.

Third, I conduct a similar analysis in comparative scope using data from the Comparative Study of Electoral Systems (CSES). I employ the comparative component of this analysis to also investigate how different electoral institutions impact the voting calculus. Results obtained in the comparative analysis confirm the results obtained in national scope, which is that ideological voting is stronger among knowledgeable voters. I also identify electoral institutions which accentuate or mitigate this effect. I find that proportional electoral systems accentuate the gap in ideological voting across voters with different levels of political knowledge, while majoritarian systems diminish it. I conclude that, while the problem of incorrect voting due to low political knowledge can be remedied by increasing political education among the population at the individual level, it can also be remedied by adopting institutions that make it easier for citizens to cast their vote in a more informed manner.