The effects of sex and sex-typing on motivation, delay, and distortion in negative feedback situations.

Personnel evaluation reports are an important tool in the work place. Unfortunately, much research suggests not only that performance evaluations themselves are often not accurate, but also that the results may be distorted and feedback delayed. Research has focused on many possible determinants of accuracy in performance ratings. Landy and Farr (1980) have suggested that focusing on the techniques of evaluation is not sufficient; we need to have a better understanding of the processes involved in appraisal in order to advance in this area. If the factors that influence inaccurate evaluations could be isolated, then proactive measures could be taken to circumvent or reduce the likelihood of their occurring. Benedict and Levine (1988), examined a group of business students and found that females have more difficulty with negative feedback than do males, positively distorting results more and further delaying feedback. Building upon this work, the purpose of this research is to further examine the relation between sex, gender or sex-typing, effectance motivation, and delay and distortion in personnel evaluations. It was hypothesized that it is the sex-typing or gender of the supervisors, and not their biological sex per se, that influences the extent to which they delay and distort. It was expected that in comparison with instrumental supervisors, expressive supervisors would positively distort and delay more on the feedback of poor performance results, and that female supervisors would also have lower levels of self-efficacy expectancy, outcome expectancy, and outcome value for the delivery of prompt and accurate negative feedback. Attribution factors, supervisory concern factors, and the impact of training were also considered. For the most, results did not support the hypotheses, with almost no significant differences between sexes or between sex-types on the variables examined. The findings are significant because they set limits on the results of Benedict and Levine (1988). Using a sample of supervisors from the Federal Service, no significant sex differences were found, suggesting that, within their work context, these individuals act as supervisors first and as males and females second. Interestingly, there was nonetheless a high level of distortion of negative performance results on the part of all supervisors. Training appeared to have little impact on the extent to which supervisors delayed or distorted, and only minimal impact on motivational factors. There was some suggestion that sex-type, as opposed to biological sex, could account for some previously measured sex differences in performance evaluations