Evaluating the validity of an integrity-based situational judgement test for medical school admissions

BACKGROUND: While the construct of integrity has emerged as a front-runner amongst the desirable attributes to select for in medical school admissions, it is less clear how best to assess this characteristic. A potential solution lies in the use of Situational Judgement Tests (SJTs) which have gained popularity due to robust psychometric evidence and potential for large-scale administration. This study aims to explore the psychometric properties of an SJT designed to measure the construct of integrity. METHODS: Ten SJT scenarios, each with five response stems were developed from critical incident interviews with academic and clinical staff. 200 of 520 (38.5 %) Multiple Mini Interview candidates at Dundee Medical School participated in the study during the 2012–2013 admissions cycle. Participants were asked to rate the appropriateness of each SJT response on a 4-point likert scale as well as complete the HEXACO personality inventory and a face validity questionnaire. Pearson’s correlations and descriptive statistics were used to examine the associations between SJT score, HEXACO personality traits, pre-admissions measures namely academic and United Kingdom Clinical Aptitude Test (UKCAT) scores, as well as acceptability. RESULTS: Cronbach’s alpha reliability for the SJT was .64. Statistically significant correlations ranging from .16 to .36 (.22 to .53 disattenuated) were observed between SJT score and the honesty-humility (integrity), conscientiousness, extraversion and agreeableness dimensions of the HEXACO inventory. A significant correlation of .32 (.47 disattenuated) was observed between SJT and MMI scores and no significant relationship with the UKCAT. Participant reactions to the SJTs were generally positive. CONCLUSIONS: Initial findings are encouraging regarding the psychometric robustness of an integrity-based SJT for medical student selection, with significant associations found between the SJTs, integrity, other desirable personality traits and the MMI. The SJTs showed little or no redundancy with cognitive ability. Results suggest that carefully-designed SJTs may augment more costly MMIs.