Abstract
Being able to understand and evaluate arguments in different modalities and in different disciplines is thought to be a key component of students’ academic success in college. However, many students do not receive explicit instruction in the basic concepts and rules of argumentation. Using a difference-in-differences approach with a multicohort longitudinal data set of almost 15,000 undergraduates beginning in health and science, technology, engineering, and mathematics (STEM)-related fields at a research university, we examined changes in relative performance of students after enrolling in an introductory logic course. We find that students improved their grade point average (GPA) after taking the course, especially if they begin college with low academic achievement (Cohen’s d = 0.18). Our results are consistent with the idea that acquiring foundational skills, in particular general skills in argumentation, prepares STEM students for future learning.
Notes
1 Full regression results are included in the Supplemental Material.
2 The model met the two main assumptions of this procedure (see Arellano & Bond, Citation1991). First, there was no second-order autocorrelation among the residuals, because the null hypothesis of no autocorrelation was not rejected (z = 0.754, p = .451). Second, we could not reject the hypothesis according to which the overidentifying restrictions were valid, χ2(20) = 22.45, p = .317. That is, the instruments used in the procedure appear to be truly exogenous, because they are not related to the second-stage residuals.
3 The subject-matter classifications are included in the transcripts provided by the university.