Themes
First, I am interested in the construct-driven approach to personnel selection. Although personnel selection has traditionally been regarded as an applied area with a heavy emphasis on predictive efficiency, the construct-driven approach has in recent years gained in importance among researchers and practitioners. A tenet of this construct approach is that, whereas high predictive validity of selection practices is certainly desirable, it is also crucial to understand why selection devices work and what constructs they exactly measure. My doctoral dissertation applied this construct-driven approach to assessment centers as high-fidelity simulations. I also investigated under what conditions assessment centers might be valid measures of the constructs. Along these lines, I examined the effects of different assessor training strategies, different types of assessors, and assessee performances. This research is helpful for practitioners because the results show how practitioners can increase the quality of construct measurement of their operational assessment centers. Recent research projects involve applying the logic of trait activation theory to develop specific interventions for improving the reliability and validity of assessment center ratings. Below you can download some relevant studies.
A second area of interest involves the use of low fidelity simulations (i.e., situational judgment tests) in high stakes selection contexts. In recent years, there exists increasing interest to use situational judgment tests as supplements to more traditional cognitive ability tests in high-stakes testing situations (e.g., in the context of college admission). These situational judgment tests present candidates with a hypothetical situation and ask them what they would do in that situation. Research in a personnel selection context has shown that such tests have good criterion-related validity and less adverse impact than their cognitive counterparts. I have addressed key issues in transporting the use of situational judgment tests from a personnel selection context to a high stakes educational context. Specifically, my research dealt with the criterion-related validity, fakability, and coachability of situational judgment tests. I also found that video-based situational judgment tests that were constructed as part of the college admission to medical studies in Belgium became more valid over the years and had incremental validity over traditional cognitive ability tests. Below you can download some relevant studies.
Third, in recent years, I have become interested in hybrids of the two aforementioned research streams (assessment centers and situational judgment tests). In these recent studies, selection procedures are considered to be combinations of "methods" and "constructs". Next, I investigate how different methods of measuring the same construct might impact on key selection outcomes such as validity, adverse impact, and applicant perceptions. Special attention is paid here to response fidelity as a method factor.
