The potential of game- and video-based assessments for social attributes: examples from practice

Abstract

Purpose The purpose of this study is to describe the development and psychometric properties of a novel game- and video-based assessment of social attributes. Despite their increasing adaption, little research is available on the suitability of games and video analytics for measuring noncognitive attributes in the selection context. Design/methodology/approach The authors describe three novel assessments and their psychometric properties in a sample of 1,300 participants: a game-based adaptation of an Emotion Recognition Task, a chatbot-based situational judgment test for emotion management and a video-based conscientiousness assessment. Findings The novel assessments show good to moderate convergent validity for Emotional Recognition (r = 0.42), Emotion Management (r = 0.39) and Conscientiousness (r = 0.21). The video-based assessment demonstrates preliminary predictive validity for self-reported work performance. Novel game-based assessments (GBAs) are perceived as better designed and more immersive than traditional questionnaires. Adverse impact analysis indicates small group differences by age, gender and ethnicity. Research limitations/implications Predictive validity findings need to be replicated using objective measures of performance, such as performance ratings by supervisors and extended to the GBAs. Adverse impact should be evaluated using a real-life applicant pool and extended to additional groups. Practical implications Evidence for the psychometric validity of novel assessment formats supports their adoption in selection and recruitment. Improved user experience and shortened assessment times open up new areas of application. Originality/value This study gives first insights into psychometric properties of video- and game-based assessments of social attributes.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 14 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.