Gender differences in locus of control and student performance in the South African context of accounting studies

Abstract

Purpose – This paper aims to test the theory that predicts differences in locus of control (LOC) by gender and the relationships between LOC and the performance of accounting students in a large South African university. Design/methodology/approach – Confirmatory factor analysis was applied to test the applicability of Spector’s (1988) LOC scales as a first-order construct in this context. An exploratory factor analysis was then performed to provide a more fine-grained analysis of subordinate constructs. Three component categories were found to emerge from a test of this widely used LOC questionnaire. These component categories were classified as beliefs about the effectiveness of agency, beliefs about chance and beliefs about networks in the contribution to the attainment of outcomes in working contexts. Findings – Further tests revealed that female accounting students demonstrate higher LOC in all the three categories. Females were therefore found to have significantly higher levels of both LOC and student performance; yet, the majority of tested items were not significant in their associations, and total LOC was not found to be associated with higher performance for female or male students. Certain individual items were, however, found to be associated with performance for male students. It is concluded that despite the predictions of seminal theory that predicts convergence around gender, or more egalitarian outcomes in high-skilled contexts over time, accounting student performance in this context might currently be dominated by females. This reflects a current general dominance of females in higher educational attainment and in employment numbers in educational contexts. Research limitations/implications – Limitations of the study include: first, the use of a single university; second, a sample of only first-year accounting students. It is not known whether these findings generalise beyond accounting students with similar university environments. This research is also not causal in nature. The statistical testing used in this study cannot indicate causality. Originality/value – It is recommended that further research investigate the more fine-grained dimensions of LOC that can contribute to accounting student performance and that further qualitative or causal research is performed to “surface” the causal mechanisms that underlie these findings. The value of this research is in the fact that it tests theory that predicts differences in LOC and the relationships between LOC and performance in an important formative context of accounting.