Effect of performance appraisal training on rehabilitation supervisor's knowledge and discomfort in appraisal of subordinate's performance

Community rehabilitation organizations have evolved in the past two decades from a position of having little, if any, competition from other rehabilitation organizations for clientele, to the current position where there may be competition for clientele and funding sources. Organizations that hope to gain an advantage over competition can only do so with top quality staff. However, supervisors may not be comfortable appraising performance of subordinates. A connection between the technical knowledge of performance appraisal and discomfort with completing performance appraisals has not been explored. This dissertation explores the idea of deceasing a supervisor's discomfort with appraising subordinate's performance, through increased knowledge of performance appraisal.; The study consisted of an experiment and two replications that addressed the research questions: (a) Does performance appraisal training decrease the supervisor's self-reported discomfort level with conducting performance appraisal on subordinates? (b) Does performance appraisal training increase the supervisor's knowledge of conducting performance appraisal on subordinates? and (c) Is there a negative relationship between supervisor's level of knowledge of effective performance appraisal practice and level of discomfort with performance appraisal?; Results were analyzed using both ANCOVA and t-tests. The findings from the experiment and replications indicated that discomfort with appraising performance appraisal does not occur following performance appraisal training. Further findings indicated that there was a significant increase in knowledge of performance appraisal training in only one of the research groups. And finally, that there is no significant relationship between supervisor's level of knowledge of effective performance appraisal practice and level of discomfort with performance appraisal. The results may have been affected by low power in the independent variable and one of the dependent measures.