Date of Award

Spring 2026

Language

English

Embargo Period

4-16-2026

Document Type

Master's Thesis

Degree Name

Master of Arts (MA)

College/School/Department

Department of Psychology

Program

Psychology (Master's)

First Advisor

Drew Anderson

Committee Members

Drew Anderson, Mitch Earleywine

Keywords

Compulsive exercise, Eating pathology, Exercise motivation

Subject Categories

Clinical Psychology

Abstract

Compulsive exercise (CE) appears frequently in eating disorder (ED) samples but also occurs in nonclinical populations, raising the possibility that different motivational processes drive the behavior, perhaps depending on ED pathology. This study examined the relative contribution of exercise motives to CE while separating motivational variance from ED symptom severity. Undergraduate women (N = 376; 38.3% White, 28.8% Black) completed the Exercise Motivations Inventory–2 (EMI-2), the Compulsive Exercise Test (CET), and the 7-item Eating Disorder Examination Questionnaire (EDE-Q7). An exploratory factor analysis of EMI-2 subscales yielded four factors: Affective Well-being (Enjoyment, Revitalization, Stress Management), Health/Function (Ill-health Avoidance, Nimbleness, Positive Health, Strength/Endurance), Social/Achievement (Affiliation, Challenge, Competition, Social Recognition), and Aesthetic (Appearance, Weight Management). Dominance analysis estimated the relative importance of these factors across three models: motivations-only, covariate-control including ED symptom severity, and residualized-control (CET residualized on EDE-Q7). In the baseline model (R2 = .444), Aesthetic motivation explained the largest share of CE variance (LMG weight = .193). Adding ED symptom severity shifted the pattern of the covariate-control model (R2 = .539): ED pathology explained the most variance (LMG weight = .167), and Aesthetic and Affective Well-being motives contributed similar amounts (LMG weights = .132 and .129, respectively). Removing variance shared with ED pathology in the residualized-control model (R2 = .331) produced a different pattern. Affective Well-being motivation predicted the largest share of remaining CE variance (LMG weight = .160), while the contribution of Aesthetic motivation further declined (LMG weight = .052). These findings support two motivational pathways to CE, the first tied to ED-related appearance and weight concerns and the second linked to affect regulation that persists independent of ED symptom severity. Assessment of motivational context alongside ED pathology might improve screening and intervention for CE.

License

This work is licensed under the University at Albany Standard Author Agreement.

Download

Share

COinS