The Science Journal of the American Association for Respiratory Care
Introduction:. Peers are often in a better position to evaluate peer job performance than are supervisors. The Central Limit Theorem implies that the more measurements we take, the more likely it is that the sample mean will equal the true mean. We have instituted a system of computer based peer evaluations. A committee of staff and management developed evaluation criteria and assigned weights to each criterion based upon departmental goals. If restructuring is ongoing, a higher weight may be given ?Adapts to rapid changes easily?. If absenteeism is a problem, it might be assigned 7% of the overall evaluation. Staff members prepare a statement factors they wish taken into account by their evaluators. This is distributed to the latter. A spreadsheet template is copied to a floppy disk or network. The spreadsheet consists of evaluation criteria in the A column, percent weight in B and names in column C. The screen is split and frozen keeping names in the first row and evaluation criteria in the first column. Staff members are taught to 1) open the spreadsheet 2) enter the evaluation data and 3) save the spreadsheet with an access code. Very high or low scores requires a justifying comment. Management later produces a hard copy for review and signature. A person in management is given the disk or accesses the spreadsheet. A macro copies the scores for each therapist to an individual's array in a master spreadsheet, criteria in rows and evaluators columns. All scores are averaged and the standard deviation computed. The array is copied again by a conditional statement which blanks out all individual scores which are more than two standard deviations from the mean. Those scores are multiplied by assigned weight to produce a total score. Management edits the comments and individual therapists are given the average score in each category, comments and the overall score. Scores determine merit increases. Anyone who rates ?Unsatisfactory? is not allowed to evaluate others the following year. Peer evaluations resulted in major behavioral changes with a 17% improvement in overall scores between year 1 and year 2. There were fewer complaints about favoritism. Peers were more critical of peers than was management. Confidentiality became an issue as some tried to find out the sources of low scores.