The identification of people for whom job descriptive index scores are inappropriate

Abstract

Occasionally, personal and situational factors can adversely affect measurement accuracy of job satisfaction questionnaires. Fatigue, misunderstood instructions, and poor reading comprehension are examples where at least some observed responses to questionnaires may not be meaningfully related to an individual's existing level of job satisfaction. Appropriateness indices (M. V. Levine & D. Rubin, Journal of Educational Statistics, 1979, 4, 269–290) are a method of quantifying the probability of response patterns in order to identify those questionnaires containing unexpected responses. The indices in this study, based on a two-parameter item response model, showed good hit rates for identifying experimentally distorted response patterns. Further predictions concerning lower scale variance in samples with lower appropriateness indices were also substantiated in three large samples of survey data. These results are discussed in relation to potential improvements in the appropriateness index, the systematic study of undesirable response distortion, and applications of the research.