Virtual Assessment Center - A Media Comparison Study on the Diagnostic Value of Online Role Play for Social Competence Assessment

The assessment center (AC) method is often used in personnel selection and development to assess job applicants' or employees' social competences. The aim of this thesis was to evaluate whether virtual ACs ad-ministered via the Internet would lead to the same results as traditional face-to-face (FTF) administered ACs. Based on a definition of social competence, an operational framework is outlined for its assessment with emphasis on the behavioral manifestation of social skills. After dis-cussing the AC method with focus on role-play as a common type of AC simulation exercise, current approaches to virtual ACs are presented, which are still in their infancy. Based on a discussion of the characteristics of FTF and computer-mediated communication, and quality standards for assessment procedures, 16 presumptions are derived concerning the expected similarities or differences between FTF and online assessment of social skills. Assessment outcome and quality of assessment da-ta are evaluated in a media comparison study using a within-subject de-sign comparing FTF and text-based, anonymous online role-play. Inter-rater reliability was found to be higher in FTF assessment while concur-rent validity, as measured by correlations with self-rated personality traits, was higher in online assessment. Moderate convergent validity was found between the two media indicating the equivalence of assessment results. Used as a pre-selection instrument, online role-play has the potential to reduce the number of candidates invited to costly ACs as it is able to predict AC role-play performance better than chance. Further-more, judgments of social skills based on anonymous online interactions are less biased compared to FTF observation. Nevertheless, chat experience and typing skills have been found to be a critical factor influencing skill ratings in online assessment but their influence can be reduced by modifying the assessment task. Despite relatively high ecological validity and lower levels of negative affective reactions during online role-play, participants showed higher preference for FTF assessment than for online assessment. Women and experienced chat users, however, rated online role-play as an assessment tool more favorably than men and in-experienced chat users did. Based on theoretical considerations and the results of the empirical study, recommendations are derived for human resource professionals and organizational psychology researchers. This thesis is a first step towards the validation of virtual ACs using interactive web-based exercises and lays the corner stone for further media comparison studies concerning their diagnostic value. The practicability of virtual ACs, however, has yet to be evaluated.