Abstract
A rationale and method for predicting mean scores on personality scales of the True-False type, based upon social desirability considerations, has been described by Edwards ( 1961). He also suggested that the procedure used in predicting group means might be extended to the prediction of individual scores. To use his prediction equation for individual scores, an estimate of the individual's probability of a socially desirable response to a personality item is needed along with knowledge of the number of items keyed for socially desirable responses in the scale on which the score is to be predicted. In order ro test the adequacy of the prediction equation for the individual case, the MMPI records of 120 male college students were scored for 58 scales derived from the MMPI. Ss were divided into High, Middle, and Low groups on the basis of their scores on the 39-item Social Desirability (SD) scale (Edwards, 1957). Ten Ss were drawn at random from each of these three groups for the present analysis. For each scale the number of items keyed for socially desirable and socially undesirable responses was determined. An estimate of the probability of a socially desirable response for each S was taken as the ratio of his score on the SD scale to the total number of items in the SD scale. Edwards' prediction equation was then used to obtain a predicted score on each of the 58 MMPI scales for each of the 30 Ss. The product-moment correlations between observed and predicted scores for the 30 Ss are presented in Table 1. The correlations given in the table for TABLE 1
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have