This study was designed to examine the effects of five variables, educational training in adapted physical education, years experience teaching adapted physical education, comprehensiveness of preservice training, comprehensiveness of professional development activity, and the primary setting of adapted physical education instruction, on the efficacy expectation and outcome expectation beliefs of adapted physical education teachers. Subjects included certified adapted physical education teachers from the state of Wisconsin (N = 227). A descriptive correlational research design was implemented and utilized the Adapted Physical Education Teacher Efficacy Scale, a revised version of the Teacher Efficacy Scale. Factor analysis confirmed The Adapted Physical Education Teacher Efficacy Scale to be capable with measuring self-efficacy from a multidimensional framework and provided support for Bandura's theory that self-efficacy is not a unidimensional construct. Dimensions of self-efficacy which resulted from the factor analysis included efficacy expectation beliefs (r = .76) and outcome expectation beliefs (r = .70). Multiple regression analysis indicated none of the five independent variables in the study explained a statistically significant amount of variance in the adapted physical education teachers' outcome expectation beliefs. Three variables, namely, comprehensiveness of professional development activity, comprehensiveness of preservice training and level of training in adapted physical education each explained a statistically significant proportion of variance in efficacy expectation beliefs when the effects of the other variables were held constant. The linear combination of scores on all five of the independent variables did not adequately explain an adapted physical education teacher's level of efficacy expectation or outcome expectation beliefs.