Evaluation of an index to detect inaccurate respondents to a task analysis inventory

An infrequency index assesses the tendency of respondents on job analysis questionnaires to rate job-irrelevant tasks as important. The purpose of the study was to evaluate whether this index could detect respondents who made rating errors on job-relevant tasks. The respondents were incumbents of three jobs: mental health workers (N=254), clerical workers (N=102), and police corporals (N=118). Differences in interrater reliability between low and high scorers on the infrequency index were found for the mental health workers on one of two response scales. For all three samples, individuals who scored higher on the infrequency index rated job-relevant tasks significantly higher. It was concluded that the index can detect respondents making constant error due to response sets and distortions. However, the usefulness of the index to detect individuals making random errors may be limited. On the basis of these results and on conceptual grounds, it was argued that job analysts should incorporate the infrequency index as part of their task analysis questionnaires.