Diagnostic accuracy of the Groningen Effort test for detecting underperformance in early retirement Claimants—New analyses in an expanded Austrian sample

Performance validity tests (PVTs), such as the Groningen Effort Test (GET), were designed to assess neuropsychological underperformance. This retrospective study examined the utility of the GET in the detection of underperformance among early retirement claimants and explored the impact of demographic variables. Data from 249 applicants who completed a neuropsychological test battery was analyzed. Based on predefined cutoff scores, participants were categorized as underperformers and non-underperformers. The underperformance group demonstrated significantly poorer outcomes across all GET variables than non-underperformance group. ROC analyses and logistic regression analyses were carried out. The total error count emerged as the most reliable indicator of underperformance. ROC analyses confirmed the diagnostic validity of all GET variables, with total error count yielding the highest discriminative accuracy (AUC = .82). Using the Cutoff score ≥ 24 total error count resulted in a specificity of 0.90 and a sensitivity of 0.56. Stricter definitions of underperformance improved specificity and overall accuracy. Likelihood ratios LR+ = 3.56 and LR- = 0.33 indicated usefulness of the GET total error count for screening purposes. These findings indicate that the GET, specifically the total error count, may be a valid tool for performance validity testing in early retirement assessments.