The Validity of Single-Response Situational Judgment Tests: A Nomological Network Meta-Analysis

Nearly 15 years after the first empirical validation of the then-novel single-response situational judgment test (SJT) methodology, research using single-response SJTs has proliferated. Single-response SJTs simply feature one edited critical incident that is evaluated by respondents–hence, the term “single-response” SJT. Single-response SJT items bypass the need for experts to generate and evaluate response options, simplifying and reducing the cost of test construction. We report the first meta-analysis of the criterion-related validity of single-response SJTs and explore the nomological network surrounding the procedural knowledge measured by this format. Results from a random-effects meta-analysis (k = 20, N = 3685) demonstrate that associations between antecedents of single-response SJT scores and criteria mirrored those in the multiple-response SJT literature, with positive associations in all cases. The reliability estimates for single-response SJTs ranged from ⍺ = 0.37 to ⍺ = 0.93, with an average of ⍺ = 0.82. The 95% confidence interval for the uncorrected correlation for single-response SJTs (95% CI [0.12, 28]) encompasses the validity correlations for multiple-response SJTs reported by McDaniel et al. (2007) (0.20, 0.26). We found that single-response SJTs correlated 0.18 (uncorrected) and 0.20 (corrected) with job performance. Additionally, we meta-analyze the correlations between single-response SJTs scores, personality and emotional intelligence, and also explore their criterion-related validity. Despite the nascency of this study area and that most studies were conducted in low-stakes lab settings, findings suggest that overall, single-response SJTs may be promising personnel selection tools.