Does Education Increase Political Participation?
Alexander K. Mayer
Abstract
Alexander K. Mayer
Abstract
The consensus among scholars has long held that educational advancement causes greater political participation. Recent research, however, argues that previous empirical analyses of this relationship cannot be used to draw causal inferences. This recent work strongly suggests that selection mechanisms confound previous results, and it employs propensity score matching to argue that education has no effect. In this article I show how propensity score matching, as recently applied to this question, introduces bias by creating poorly matched treatment and control groups. I analyze the same data, realigning the treatment with theories of political participation, and I use genetic matching to create balanced treatment and control groups. I then subject the results to a variety of tests, providing a stronger foundation for causal inference. The estimates provide consistent evidence that, contrary to recent results, educational advancement increases political participation.
OpenAlex reports 164 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
The consensus among scholars has long held that educational advancement causes greater political participation. Recent research, however, argues that previous empirical analyses of this relationship cannot be used to draw causal inferences. This recent work strongly suggests that selection mechanisms confound previous results, and it employs propensity score matching to argue that education has no effect. In this article I show how propensity score matching, as recently applied to this question, introduces bias by creating poorly matched treatment and control groups. I analyze the same data, realigning the treatment with theories of political participation, and I use genetic matching to create balanced treatment and control groups. I then subject the results to a variety of tests, providing a stronger foundation for causal inference. The estimates provide consistent evidence that, contrary to recent results, educational advancement increases political participation.
Key concepts: Causal inference, Propensity score matching, Matching (statistics), Selection bias, Politics, Treatment and control groups, Selection (genetic algorithm), Variety (cybernetics)