RESUMEN

BACKGROUND: Situational judgment tests (SJT) have gained popularity as a standardized assessment of nontechnical competencies for applicants to medical school and residency. SJT formats range from rating the effectiveness of potential response options to solely open response. We investigated differences in test-taking patterns between responders and nonresponders to optional open response SJT questions during the application process. METHODS: This was a prospective multi-institutional study of general surgery applicants to seven residency programs. Applicants completed a 32-item SJT designed to measure ten core competencies: adaptability, attention to detail, communication, dependability, feedback receptivity, integrity, professionalism, resilience, self-directed learning, and team orientation. Each SJT item included an optional, nonscored, open response space for applicants to provide a behavioral response if they desired. Trends in applicant gender, race, ethnicity, medical school ranking, and USMLE scores were examined between the responder versus nonresponder group. RESULTS: In total, 1491 general surgery applicants were invited to complete the surgery-specific SJT. Of these, 1454 (97.5%) candidates completed the assessment and 1177 (78.9%) provided additional responses to at least one of the 32 SJT scenario sets. There were no differences in overall SJT performance, USMLE scores (Step 1: 235, SD 14, Step 2: 250, SD 11), race and/or ethnicity between the responder and nonresponder groups. Responders were more likely to be from a top 25 medical school (p < 0.05) compared to the nonresponder group. Among applicants who completed any open response questions, women completed a significantly higher number of questions compared to men (7.21 vs 6.07, p = 0.003). The number of open responses provided correlated with higher scores on SJT items measuring dependability (r = 0.07, p = 0.007). CONCLUSIONS: SJT design and format has the potential to impact test-taker response patterns. SJT developers and adopters should ensure test format and design have no unintended consequences prior to implementation.

Asunto(s)

RESUMEN

Background and Objectives: Studies have found that older people value care workers' character traits and interpersonal judgment even more highly than their technical skills. Yet identifying these traits at recruitment is challenging. This study aimed to evaluate the first situational judgement tests (STJs) for direct care workers. Research Design and Methods: Online tests were conducted with 251 care workers and members of the public in England. Participants evaluated the appropriateness of 61 potential behavioral responses to 11 "critical incidents," each depicting challenging care work scenarios. Data collection included a measure of personality traits. A subsample of 72 participants completed a second "test-retest" assessment. Results: A majority of test-takers (53%) found the test easy/very easy to complete, and most (55%) participants who worked in care reported the scenarios were highly realistic. Psychometric tests were positive. Test scores were unidimensional under a Confirmatory Factor Analysis (RMSEA = 0.038), and items broadly fit Rasch assumptions. Test-retest reliability (rho = 0.77) was acceptable, and for the general public sample, a modest increase in perceptions of the social standing of care work was observed. Test scores were positively correlated with 2 personality traits: agreeableness (r = 0.250, p < 0.001) and openness to experience (r = 0.179, p = .005). Test scores were not related to age, gender, or education level. Discussion and Implications: The findings indicate support for the use of SJTs in direct care work. Its psychometric properties appear satisfactory, and collectively give confidence in the use of SJTs for assessing the suitability of candidates during recruitment. Further research should corroborate these findings in a new sample, and examine the relationship between test scores and job performance.

RESUMEN

Although there is extensive evidence confirming the predictive validity of situational judgement tests (SJTs) in medical education, there remains a shortage of evidence for their predictive validity for performance of postgraduate trainees in their first role in clinical practice. Moreover, to date few researchers have empirically examined the complementary roles of academic and non-academic selection methods in predicting in-role performance. This is an important area of enquiry as despite it being common practice to use both types of methods within a selection system, there is currently no evidence that this approach translates into increased predictive validity of the selection system as a whole, over that achieved by the use of a single selection method. In this preliminary study, the majority of the range of scores achieved by successful applicants to the UK Foundation Programme provided a unique opportunity to address both of these areas of enquiry. Sampling targeted high (>80th percentile) and low (<20th percentile) scorers on the SJT. Supervisors rated 391 trainees' in-role performance, and incidence of remedial action was collected. SJT and academic performance scores correlated with supervisor ratings (r = .31 and .28, respectively). The relationship was stronger between the SJT and in-role performance for the low scoring group (r = .33, high scoring group r = .11), and between academic performance and in-role performance for the high scoring group (r = .29, low scoring group r = .11). Trainees with low SJT scores were almost five times more likely to receive remedial action. Results indicate that an SJT for entry into trainee physicians' first role in clinical practice has good predictive validity of supervisor-rated performance and incidence of remedial action. In addition, an SJT and a measure of academic performance appeared to be complementary to each other. These initial findings suggest that SJTs may be more predictive at the lower end of a scoring distribution, and academic attainment more predictive at the higher end.

Asunto(s)

RESUMEN

This paper has two objectives: (1) presenting recent advances in personality theory whereby personality traits are conceptualized within a framework that focuses on the dynamic interactions of behaviour, biology, context, and states, and (2) discussing the implications of these developments for measurement and medical selection. We start by presenting evidence that traits are no longer regarded as stable deterministic predictors of behaviour. Instead, traits are found to change across generations, the life span, and in response to environmental contingencies. Thus, there is an urgent need to explore how traits change as function of medical education. Second, drawing on recent theory and research (behavioural reaction norms and the density distribution model) we highlight evidence to show how the expression of trait relevant behaviour is dependent on context, and is distributed with an average (typical behaviour or personality) and a variance (plasticity or adaptability), with traditional personality measure associated with typical responding. Third, we demystify that some traits are better than others showing that so-called "good" traits have a dark-side. Fourth, we show how these developments impact on how personality might be assessed, thereby presenting recent evidence on the use of contextualized personality measures, situational judgment tests, other reports, and implicit measures. Throughout the paper, we outline the key implications of these developments for medical selection practices.

Asunto(s)

RESUMEN

Traditional methods of assessing personality traits in medical school selection have been heavily criticised. To address this at the point of selection, "non-cognitive" tests were included in the UK Clinical Aptitude Test, the most widely-used aptitude test in UK medical education (UKCAT: http://www.ukcat.ac.uk/ ). We examined the predictive validity of these non-cognitive traits with performance during and on exit from medical school. We sampled all students graduating in 2013 from the 30 UKCAT consortium medical schools. Analysis included: candidate demographics, UKCAT non-cognitive scores, medical school performance data-the Educational Performance Measure (EPM) and national exit situational judgement test (SJT) outcomes. We examined the relationships between these variables and SJT and EPM scores. Multilevel modelling was used to assess the relationships adjusting for confounders. The 3343 students who had taken the UKCAT non-cognitive tests and had both EPM and SJT data were entered into the analysis. There were four types of non-cognitive test: (1) libertariancommunitarian, (2) NACE-narcissism, aloofness, confidence and empathy, (3) MEARS-self-esteem, optimism, control, self-discipline, emotional-nondefensiveness (END) and faking, (4) an abridged version of 1 and 2 combined. Multilevel regression showed that, after correcting for demographic factors, END predicted SJT and EPM decile. Aloofness and empathy in NACE were predictive of SJT score. This is the first large-scale study examining the relationship between performance on non-cognitive selection tests and medical school exit assessments. The predictive validity of these tests was limited, and the relationships revealed do not fit neatly with theoretical expectations. This study does not support their use in selection.

Asunto(s)

RESUMEN

From senior school through to consultancy, a plethora of assessments shape medical careers. Multiple methods of assessment are used to discriminate between applicants. Medical selection in the UK appears to be moving increasingly towards non-knowledge-based testing at all career stages. We review the evidence for non-knowledge-based tests and discuss their perceived benefits. We raise the question: is the current use of non-knowledge-based tests within the UK at risk of undermining more robust measures of medical school and postgraduate performance?

Asunto(s)