Customized Peer-Modeled Mindset Interventions Promote Persistence in Undergraduate Science Courses

• In Event: Motivating Effective Strategy Use Through Interventions

Abstract

Objective & Theoretical Framework
Undergraduate STEM majors often struggle in science courses and drop out or change majors before completing college (see Seymour & Hunter, 2019). Many students in these courses rely on surface-level study strategies that fail to develop their conceptual knowledge and result in underachievement (Karpicke et al., 2009). Students often misattribute struggle in these courses to a lack of potential, rather than to their use of ineffective study strategies (see Wilson & Linville, 1985), leading them to leave the sciences (Rosenzweig et al., 2021).

To address this issue, we developed a protocol for creating “peer-modeled” mindset interventions. These interventions draw on stories from previous students within a class to help current students identify ineffective study strategies as a common reason for struggle and motivate them to update their approach to studying.

Methods & Data
We developed peer-modeled mindset interventions for two science courses in which students often struggle: Introductory Biology I and Organic Chemistry I. To develop the interventions, we interviewed past students who had initially struggled but ultimately succeeded in the courses. We then selected video clips from these interviews that conveyed how these students’ initial struggles were attributable to ineffective study strategies rather than a lack of potential.

We randomly assigned incoming students within Introductory Biology I (Study 1; n = 917; 70% women, 64% white or Asian, 29% first-generation college student) and Organic Chemistry I (Study 2; n = 1,420; 69% women, 70% white or Asian, 25% first-generation college student) to complete three intervention activities (including the selected video clips and a writing exercise) or three control activities for course credit. We assessed students’ reliance on old strategies, experimentation with new strategies, and persistence into the second course in the sequence (Introductory Biology II, Organic Chemistry II) the following semester.

We analyzed intervention effects using conservative Bayesian models that intentionally shrink effect sizes toward zero (Hahn et al., 2020). We report standardized mean differences (SMD) in the outcome between treatment and control and Bayesian probabilities that the difference was greater (or less than) zero (pr(SMD>0)). Probabilities less than .75 are not interpreted as meaningful; higher probabilities reflect greater confidence in the effect (see Hecht et al., 2023).

Results
The intervention reduced students’ reliance on ineffective strategies they had used in earlier courses (Study 1: SMD = -0.15, pr(SMD<0) = .98; Study 2: SMD = -0.09, pr(SMD>0) = .95) and increased the extent to which they experimented with new strategies (Study 1: SMD = 0.11, pr(SMD>0) = .94; Study 2: SMD = 0.08, pr(SMD>0) = .92). The intervention also increased students’ enrollment in the second course in the sequence (Study 1: SMD = 0.04, pr(SMD>0) = .78; Study 2: SMD = 0.05, pr(SMD>0) = .86).

Significance
Previous students can help current students see struggles as attributable to their choice of strategies rather than a lack of potential to succeed. Peer-modeled mindset interventions, which distill these stories, can motivate current students to change their approach to studying and become more likely to persist in the sciences.

Authors

• Cameron Hecht, University of Rochester

• Nirel Jones Mitchell, University of California - Los Angeles

• Anita Latham, University of Texas at Austin

• Zachariah A. Page, University of Texas at Austin

• David Yeager, University of Texas at Austin