An examination of the structure and predictability of Myers-Briggs Type Indicator preferences using a job component validity strategy based on the Common-Metric Questionnaire

Abstract

When criterion related validity studies are not feasible due to statistical power problems, a number of alternate methods for validation are available. Indirect validity offers a few of these strategies. One of these is synthetic validity which, in its most generalized form, is the linking of predictor items or tests directly to a job's elements.

Three main operational definitions exist for synthetic validity, but only one, job component validity (JCV), actually specifies minimum scores for a given job. In an era of minimum cutoff scores, having an empirical procedure to determine requisite scores can be quite valuable. JCV can be used to predict personality test scores for jobs, scores which can be used for selection as well as other personnel decisions. While much is known about the process of predicting cognitive ability test scores as well as the efficacy of such predictions, little is known about the use of JCV with personality test profiles. This study sought to examine the efficacy of the ICV model when applied to a personality test, namely the Myers-Briggs Type Indicator, using the work dimension scores of the Common-Metric Questionnaire (CMQ) as predictors. Results indicated that the JCV model is applicable to non-ability tests such as the MBTI. Moreover, prediction of personality test scores is not moderated by the type of work performed. Finally, results supported the MBTI's claim that its scales synergistically interact with each other.