The interplay of elicitation and evaluation of trait-expressive behavior: Evidence in assessment center exercises

In assessment centers (ACs), research on eliciting candidate behavior and evaluating candidate behavior have largely followed independent paths. This study integrates trait activation and trait rating models to posit hypotheses about the effects of behavior elicitation via situational cues on key assessor observation and rating variables. To test the hypotheses, a series of experimental and field studies are conducted. Only when trait-expressive behavior activation and evaluation models work in conjunction, increases in observability are coupled with increases in the interrater reliability, convergent validity, discriminant validity, and accuracy of AC ratings. Implications of these findings for AC theory and practice are formulated.