Statistical Models of Self-Efficacy in STEM Students

Student's Name

Sarah Painter, Minnesota State University, MankatoFollow

1st Student's Major

Mathematics and Statistics

1st Student's College

Science, Engineering and Technology

Students' Professional Biography

Sarah Painter is a junior, double-majoring in Mathematics and French. As an MSU student, she has worked as a tutor for the mathematics department and a research assistant for the computer science department, working with STEM education research. She has been involved with the MSU Honor’s Program, MAX Scholar seminar, the math club, and the French club. She plans to enroll in a mathematics PhD program after graduating from MSU, in the hopes of becoming a mathematics professor.

Mentor's Name

Rebecca Bates

Mentor's Email Address

rebecca.bates@mnsu.edu

Mentor's Department

Integrated Engineering

Mentor's College

Science, Engineering and Technology

Abstract

Persistence through undergraduate education may be explained by self-efficacy. It is the belief in one’s self to persevere through challenges. Bandura stated four areas that are thought to influence self-efficacy: mastery experience, social persuasion, vicarious experience, and physiological state. In this study, we focused on general and academic self-efficacy in STEM students, in the hopes of learning more about the relationships between Bandura’s categories, demographics, and self-efficacy. Data was taken from two institutions: one, a large research focused university, and the other, a smaller teaching focused university. In the first phase, surveys on general self-efficacy were taken at both institutions by 118 students. In the second, academic self-efficacy data was taken from 599 students. These surveys included questions concerning demographics, Bandura’s categories, and self-efficacy. Scores were summed for constructs relating to one of Bandura’s four categories. We used Cronbach’s alpha as a measure of internal reliability within each of the constructs. Correlation and linear regression analyses were used to study the data. Dummy variables for demographic data were created and used in the regression models. The best model found for general self-efficacy, including all phase 1 constructs and dummy variables, has an R2 value of 0.558. For academic self-efficacy, our best model includes all constructs and dummy variables and has an R2 value of 0.526. The goal of this work is to find factors that may potentially influence self-efficacy, in the hopes that they may be used in further research aimed at ensuring persistence of STEM students.

Recommended Citation

Painter, Sarah (2012) "Statistical Models of Self-Efficacy in STEM Students," Journal of Undergraduate Research at Minnesota State University, Mankato: Vol. 12, Article 7.
DOI: https://doi.org/10.56816/2378-6949.1015
Available at: https://cornerstone.lib.mnsu.edu/jur/vol12/iss1/7

Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License