Year of Publication


Degree Name

Master of Science in Education (MSEd)

Document Type

Master's Thesis




Educational, School, and Counseling Psychology

First Advisor

Dr. Kenneth Tyler

Second Advisor

Dr. Ellen Usher


Ability grouping has been prevalent in American schools for over a century (Burris & Welner, 2005; Museus, Palmer, Davis & Maramba, 2011; Slavin, 1990). Although ability grouping has been studied in terms of student performance, little research has examined the relationship between this practice and student motivation. The purpose of this study was to examine middle school students’ (N = 2,279) mathematics self-efficacy and its sources in ability grouped mathematics courses in the Southeastern U.S. The study also examined whether students in each ability group were represented proportionately by gender, ethnicity, and SES when compared to the full sample. Students responded to Likert scaled items assessing self-efficacy and its four sources (Bandura, 1997). Tests of mean differences in self-efficacy and its sources revealed that students in above-level courses reported significantly higher levels of self-efficacy than students in on- and below-level courses. Regression analysis revealed that mastery experiences, social persuasions, and negative physiological state predicted self-efficacy for above- and on-level students. Only mastery experiences and vicarious experiences predicted mathematics self-efficacy for below-level students. Results imply that teachers who work with students who are struggling in mathematics may find it beneficial to provide ample opportunity to expose students to models in mathematics.