Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)



First Advisor

Chad D. Ellett


The purpose of this study was to explore relationships between several variables which help to explain the process by which adolescents decide to pursue a college education. Previous models were enhanced by including important theoretical constructs well documented in social cognitive and attributional theories as elements of human agency. The psychological constructs, self-efficacy and locus of control were the primary focus of attention as independent variables and for their significance as mediating variables affecting the relationship between previously identified factors attributed to students' postsecondary attendance decisions, and students' college aspirations and expectations. Particular attention was given to the college choice process for members of minority groups, as previous research has not adequately identified the variables which motivate these individuals to pursue a college education. The study also explored the conceptualization of self-efficacy to provide a better understanding of the construct's generalizing nature and to discern the relationship between the capabilities and the persistence notions of the construct. The study sample consisted of 1076 ninth-grade students attending public high schools in the Miami-Dade County Public Schools. Parts of three measures were used for data collection: the Internal External Locus of Control Scale (Rotter, 1966), the academic sections of the Children's Self-Efficacy Scale, (Bandura, 1989), and the National Education Longitudinal Study of 1988 (U.S. Department of Education, 1992). Major findings of the study showed that: (a) locus of control is not a significant factor in the college choice process although low reliability in the data made this finding inconclusive; (b) there is evidence that academic self-efficacy is both directly related to college aspirations and expectations and mediates the linkages between academic achievement and aspirations and expectations; (c) the models of college choice are different for members of minority groups than for White students; (d) to some extent an individual's self-efficacy can be generalized both across academic domains and within academic domains; and (e) self-efficacy beliefs about capabilities to execute academic behaviors and beliefs about academic task persistence can be independently measured and are moderately related within self-efficacy theory.