One technique being used to reduce employee turnover is the Weighted Application Blank. Data obtained from application blanks are analyzed and weights are assigned to each item. Utilizing these weights, predicted scores are derived and compared to each person's actual tenure to determine the effectiveness of the model. The present study analyzed application blank data from the files of 93 currently employed and 69 terminated female clerical workers. Twelve items were analyzed by means of a stepwise multiple linear regression procedure, with months of tenure being the dependent variable. The five most significant items yielded a multiple correlation of .54. The total sample also was divided randomly into two groups, and cross-group analyses resulted in simple correlations of .56 and .29.

The first purpose was to develop and validate a quantitative selection tool, the weighted application blank, tailored to the nursing home industry. The second purpose of this study was to determine whether data scaling and increased statistical rigor can reduce the frequency of type I and type II errors in the weighted application.