Making Assignments Count: The Quest for Critical Thinking in Undergraduate Political Theory Essays

ABSTRACT

Political theory lags behind other subfields in political science in rigorously testing what helps foster critical thinking (CT). Yet some of the greatest temptations to engage in motivated reasoning can be found in normative political contexts. This study uses multiple regression analysis to explore nine semesters of data from an introductory course in political theory. Three results stand out: (1) as the stakes of an assignment decrease, so do CT scores; (2) as the number of course assignments increase, CT scores fall; (3) when preparatory exercises matter, theory-specific CT exercises outperform generic ones. A theory of student rationality is put forth. Implications for course design, program assessment, and future research are discussed.

KEYWORDS:

• Assessment
• critical thinking
• epistemic rationality
• instrumental rationality
• low stakes testing

Acknowledgments

The author would like to thank Marsha Lyle-Gonga, Mike Gruszczynski, Jamie Leek, Chase Larson, and Jamaica Macdonald for their help, comments, and support on this project, along with the anonymous reviewers at JPSE for comments and suggestions on earlier versions of this article.

Notes

According to a report from the Association of American Colleges and Universities, the ability for critical analysis and logical thinking is mentioned in fully 25% (75 of 301) of mission statements of the Princeton Review’s “331 Best Colleges and Universities in the United States” that express an educational vision. This places critical thinking in fifth place on the list of most common educational themes in university mission statements (Gaff and Meacham 2006).

A more detailed discussion of the WSU grading rubric and adaptations made for political theory is available in Appendix A.

The thesis score stands out as the most likely candidate for improvement.

The forward stepwise specifications are available upon request. The differences are minor.

Other decision criteria were considered, including minimum AIC and BIC, but were rejected on the grounds that parsimony and fit were less important than determining whether there was some statistically significant relationship that could be detected for the purposes of future investigation.

These figures are an average for all topics. Some individual topics may account for more, others less—hence the need for more fine-grained analysis than I can give the topic here.

A related measure, the percentage of readings accessed since the last assignment was also computed, but dropped from the study since it produced very similar overall results.

A regression using a semester dummy variable and a standard set of individual- and assignment-level controls finds that semester-level effects are significant in the aggregate and account for around 3.5% of the variation in aggregate critical thinking scores.

Additional information

Notes on contributors

John LaForest Phillips

John L. Phillips is an Assistant Professor of Political Science at Austin Peay State University in Clarksville, TN. His research normally focuses on the philosophy, politics, and economics of natural resources.