Psychometric properties of the Ekman 60-Faces Test in clinical populations: A systematic review and meta-analysis

The Ekman 60-Faces Test is widely used to assess facial emotion recognition, but information on its validity and reliability is scarce. This systematic review evaluated the reporting of psychometric properties across clinical populations. A total of 159 studies (n = 5978) were included: case-control (n = 109), single cohort (n = 37) and review/norm studies (n = 13). Data were extracted on the reporting of structural validity, internal consistency, measurement invariance, test-retest reliability, criterion validity and construct validity. A meta-analysis was performed, assessing effect sizes (Hedges' d) to summarize differences in performance on the Ekman 60-Faces Test in different categories of clinical groups compared with control participants. Results showed that the majority of studies used the validated version by Young et al. (2002; 42%). Fewer than 10% of the studies addressed criterion validity, internal consistency, measurement invariance, structural validity or test-retest reliability. Associations with demographic variables were reported in 23% of studies. Meta-analysis revealed significantly larger effects in neurodegenerative populations (d = -1.09) than in psychiatric (d = -.70, Q_m = 5.06, p < .05) and acquired brain injury (d = -.78, Q_m = 5.45, p < .05) populations.