Cost Perceptions in STEM: Examining the Effects of Effort and Emotional Cost Reduction Interventions

• In Event: Cognitive, Metacognitive, and Motivational Interventions Supporting Student Success in STEM: What Works, When, for Whom?

Abstract

Background/Aims

According to Situated Expectancy-Value Theory (SEVT), students’ expectancies, values, and costs are related to academic choices, effort, and achievement (Eccles & Wigfield, 2020). Cost negatively influences students’ achievement-related outcomes (Perez et al., 2019), and recent work has thus explored the effectiveness of cost-reduction interventions to support students’ academic success (Rosenzweig et al., 2020). The present study aimed to replicate Rosenzweig et al.’s cost-reduction work and to further examine dimension-specific cost interventions in college physics. We targeted both effort and emotional costs to investigate whether separate intervention conditions reduced cost perceptions, and whether effects differed by gender. Analyses were also conducted to determine whether cost perceptions late in the semester were related to achievement outcomes. This work is important in educational STEM settings where cost perceptions are often high (Benden & Lauermann, 2022; Gaspard et al., 2017), and women are typically underrepresented (National Science Foundation, 2021).

Method

Students in introductory college physics (N = 563) were randomly assigned to an effort cost intervention, emotional cost intervention, or control condition (Table 1). Intervention materials were modeled after Rosenzweig et al. (2020): quotations from students regarding overcoming course challenges were collected and examined to ensure that they reflected relevant cost dimensions. Students in intervention conditions read quotations regarding assigned cost dimensions and composed advice to future students about overcoming course difficulties (Table 2). Intervention and control materials were delivered in two sessions. Course-specific expectancies and costs were assessed early in the semester, after the intervention sessions, and at the end of the semester (Figure 1; Table 3). Exam scores and course grades were also collected.

Results

Path analyses with orthogonal coding for intervention and control conditions revealed a marginally positive indirect effect of the intervention-versus-control contrast on course grades, after accounting for covariates: initial exam, expectancies (T1), and each cost facet (T1). Utilizing a serial-mediation path model, students in the intervention conditions were found to have marginally higher effort cost perceptions (T4) than those in the control group, and those cost perceptions subsequently related to course grades, with higher effort cost perceptions predicting higher grades. Emotional cost perceptions (T4) were significantly and negatively related to course grades but were unrelated to contrast codes. That is, greater perceptions of emotional cost predicted lower course grades for all students, regardless of condition. Multiple-groups analysis was employed to examine gender moderations, but no significant differences were found.

Discussion

Our methods closely replicated and extended the initial cost-reduction study; however, results were more similar to Rosenzweig et al.’s (2020; 2022) recent findings, which only partially replicated earlier results. Instead of reducing costs, intervention students’ effort, but not emotional, costs were increased by the end of the semester. Importantly, however, these effort costs were subsequently associated with higher course grades. Additionally, while male students showed more adaptive expectancies at baseline and female students had higher effort and emotional costs at T1, our outcomes did not significantly differ by gender. Future work should continue to examine cost-reduction mechanisms to target costs more precisely, especially emotional costs.

Authors

• Elise C. Allen, The Ohio State University
  Presenting Author

• Shirley L. Yu, The Ohio State University
  Non-Presenting Author

• Michelle R. Richard, The Ohio State University
  Non-Presenting Author

• Andrew Holmes Perry, The Ohio State University
  Non-Presenting Author

• Kimiko Ching, The Ohio State University
  Non-Presenting Author

• Adriana I. Martinez Calvit, The Ohio State University
  Non-Presenting Author

• Huy Nguyen, The Ohio State University
  Non-Presenting Author

• Arianna Henning, The Ohio State University
  Non-Presenting Author

• Hyewon Lee, University of California - Irvine
  Non-Presenting Author