Purpose: This paper has two aims: to investigate the relationship of self-efficacy beliefs in terms of research on publication output; and, to identify the relationship of self-efficacy beliefs about research to the publishing outputs of neophyte lecturers. Design/methodology/approach: A questionnaire was utilised to obtain responses from lecturers working full-time at two large Australian universities (n=343). The data from this sample were analysed using factor analysis, correlation, and multiple regression analysis. Data from two sub-samples of neophyte lecturer (n[subscript 1]=47; n[subscript 2]=78) were then subjected to a multivariate analysis of variance. Findings: Four research self-efficacy subscales were derived from a factor analysis. These subscales were positively and significantly related and accounted for 46 percent of the total variance in total publications accrued. Significant differences were found between two groups of neophyte lecturer on nearly all items forming the respective research self-efficacy subscales. And, group membership accounted for 45.4 percent of the total variance. Originality/value: The findings have implications both theoretically and practically. Theoretically, the research self-efficacy construct was shown to have four underlying dimensions and to be highly predictive of a measure of publication output. From a practical perspective, the items forming the research self-efficacy subscales could be a useful tool to promote discussion about the tasks a lecturer may need to perform during an academic career. Further, the items could be ranked in terms of their discriminative capacity and, as a result, be used as the basis for researcher development and interventions to promote improved research self-efficacy and therefore increased publication output. (Contains 7 tables.)