Self-Reports for Employee Selection
construct validity, faking, measurement bias, scale development, self-report, common method variance
Digital Object Identifier (DOI)
Self-reports are used in the assessment of potential predictors for employee selection and placement, most notably personal and work background, and personality. Sophisticated methods exist for the development of quantitative multiitem scales that can demonstrate acceptable reliability and evidence for construct validity. One concern with such scales is the possibility of biases that might distort relationships with other variables, as well as the possibility of common method variance when predictor and criterion are both assessed with self-reports. A concern specifically with personality tests in a selection context is the possibility that applicants will fake their responses. Although there is evidence that faking exists, there is no consensus about the impact it might have on test validity or the accuracy of selection decisions. Factor analysis can provide evidence that a set of scale items reflects multiple constructs, but artifactual factors can be caused by the nature of items rather than the constructs they reflect. Finally, a distinction has been made between scales in which each item is a parallel measure intended to assess the same underlying construct (reflective indicator scale) and those in which items assess different things that are combined into an index (formative indictor scale). Despite limitations, self-reports can be useful tools to assess important predictors, as well as criteria, in selection applications.
Was this content written or created while at USF?
Citation / Publisher Attribution
Self-Reports for Employee Selection, in N. Schmitt (Ed.), The Oxford Handbook of Personnel Assessment and Selection, Oxford University Press, p. 443-461
Scholar Commons Citation
Spector, Paul E., "Self-Reports for Employee Selection" (2012). Psychology Faculty Publications. 601.