Date of Award

Spring 5-12-2017

Degree Type


Degree Name

Doctor of Philosophy (PhD)


Educational Psychology and Special Education

First Advisor

MIles A. Irving, Ph.D

Second Advisor

Anne C. Kruger, Ph. D

Third Advisor

Rebecca Ellis, Ph.D

Fourth Advisor

Lynda D. Woodruff, P.T., Ph.D


This manuscript style dissertation is composed of three chapters. Chapter 1 is an introduction of the research problem and introduces some of the main areas that can impede African Americans from regularly participating in physical activity such as cultural & regional issues, presence of chronic diseases, and other issues. Chapter Two is a literature review on how autonomy support can facilitate self-determined motivation within individuals. It introduces how an individual can receive autonomy support from one source, multiple sources, or from an environment. Chapter Three is an empirical study assessing the relationship between self-determination theory and physical activity participation variables to exercise self-definition. One hundred and forty nine African American retirees who lived independently completed four outcome measures: Exercise Self-Definition (ESD) scale, Behavioral Regulations for Exercise Questionnaire-2/Relative Autonomy Index (BREQ2RAI), Godin Leisure Time Activity Scale (GLTEQ), and Basic Psychological Needs for Exercise Scale (BPNES). The investigator ran a correlation matrix to demonstrate relationships between the variables and a hierarchical linear regression analysis to determine significant predictors of physical activity participation (GLTEQ). The results revealed all the variables had statistically significant correlations, however, GLTEQ had the highest correlation with ESD (0.318). None of the psychological predictors was significant predictors of GLTEQ. In the final model, the ESD was two one-hundredths of a point over the criterion p-value (.05) for showing statistical significance in predicting GLTEQ among African American retirees. For the overall regression model, all factors explained 16.2% of the total model variance.