• Rebecca Turner and Sandra Nicholson.
• Broomfield Hospital, Mid Essex Hospital Trust, UK.
• Med Educ. 2011 Oct 1;45(10):1041-7.

ContextUK medical school traditional selection processes are faltering in their ability to distinguish among highly qualified candidates. New methods of selection, including the UK Clinical Aptitude Test (UKCAT), herald a new era in which candidates are selected on aptitude and also aim to widen participation. However, the predictive validity of UKCAT and its role in the selection process are yet to be defined. This paper examines current selection practices and questions the role that such an aptitude test may take.ObjectiveThis study is intended to determine whether UKCAT can select suitable candidates for interview.MethodsThe study was designed to determine whether selectors rejected fewer candidates with high rather than low UKCAT scores using routine selection techniques. Selector rejection rates for 812 candidates with high UKCAT scores and 200 candidates with low UKCAT scores were compared. Additionally, any relationships among UKCAT and subsequent interview performance scores were explored by examining the correlation coefficients between overall and component UKCAT and interview scores for 637 candidates with high UKCAT scores.ResultsThe rejection rate before interview of candidates with low UKCAT scores was 2.7 times that of candidates with high UKCAT scores. However, no relationship between overall UKCAT score and overall interview score existed within a pre-selected cohort of applicants with high UKCAT scores (Kendall's correlation coefficient, tau b =- 0.004; p = 0.88).ConclusionsUKCAT can facilitate the independent selection of appropriate candidates for interview when used as described. However, UKCAT is not predictive of success at interview. The long-term predictive validity of UKCAT is currently under investigation. Therefore, we conclude that UKCAT is best viewed as a useful adjunct to current selection processes.© Blackwell Publishing Ltd 2011.

21 keywords...

hide…