Abstract
During the 1970's, causal attribution took its place as a viable explanation for the underrepresentation of women in college level mathematics courses. Based upon the prior reporting of gender differences in attribution patterns, researchers have hypothesized that attributions have a significant mediating effect on the academic choices made by students. This study centered on the problem of attribution as a causal variable in a model of academic choice as proposed by Meece et al. (1982). In order to be considered a causal variable in such a model however, the assertion was made that attributions for women in areas where they do participate, such as English, must be different from attributions in areas where they do not participate, such as Mathematics, and different from those of men. The current study was undertaken to test hypotheses concerning apparent gender differences in attribution patterns for success in two subject areas.An instrument was constructed to measure ten student attributions for success in natural classroom settings. 1,110 undergraduate community college students completed the instrument in either an English, Social Science, or Mathematics class. Analysis of data from the 421 successful students indicated that there are no gender differences in attributions for success, controverting several previous studies of this nature. However, attributions did depend upon the particular course being taken. Thus, attributions for success in Mathematics were different from those in English. The study concludes that causal attribution is probably not a causal variable in the Meece model of academic choice. The role of attribution as a causal variable in a model of academic choice is discussed, and recommendations made for future attribution research.