The change from UMAT to UCAT for undergraduate medical school applicants: impact on selection outcomes.

OBJECTIVES: To assess whether the change from the Undergraduate Medical and Health Sciences Admissions Test (UMAT; 1991-2019) to the University Clinical Aptitude Test (UCAT) for the 2020 New South Wales undergraduate medical degree intake was associated with changes in the impact of sex, socio-economic status and remoteness of residence, and professional coaching upon selection for interview. DESIGN, SETTING, PARTICIPANTS: Cross-sectional study of applicants for the three NSW undergraduate medical programs for entry in 2019 (4114 applicants) or 2020 (4270); 703 people applied for both intakes. Applicants selected for interview were surveyed about whether they had received professional coaching for the selection test. MAIN OUTCOME MEASURES: Scores on the three sections of the UMAT (2019 entry cohort) and the five subtests of the UCAT (2020 entry); total UMAT and UCAT scores. RESULTS: Mean scores for UMAT 1 and 3 and for all four UCAT cognitive subtests were higher for men than women; the differences were statistically significant after adjusting for age, socio-economic status, and remoteness. The effect size for sex was 0.24 (95% CI, 0.18-0.30) for UMAT total score, 0.38 (95% CI, 0.32-0.44) for UCAT total score. For the 2020 intake, 2303 of 4270 applicants (53.9%) and 476 of 1074 interviewees (44.3%) were women. The effect size for socio-economic status was 0.47 (95% CI, 0.39-0.54) for UMAT, 0.43 (95% CI, 0.35-0.50) for UCAT total score; the effect size for remoteness was 0.54 (95% CI, 0.45-0.63) for UMAT, 0.48 (95% CI, 0.39-0.58) for UCAT total score. The impact of professional coaching on UCAT performance was not statistically significant among those accepted for interview. CONCLUSIONS: Women and people from areas outside major cities or of lower socio-economic status perform less well on the UCAT than other applicants. Reviewing the test and applicant quotas may be needed to achieve selection equity.

Does 'online confidence' predict application success and later academic performance in medical school? A UK-based national cohort study.

OBJECTIVES: The UK Clinical Aptitude Test (UKCAT) previously piloted an assessment of 'online confidence', where candidates were asked to indicate how confident they were with their answers. This study examines the relationship between these ratings, the odds of receiving an offer to study medicine and subsequent undergraduate academic performance. DESIGN: National cohort study. SETTING: UK undergraduate medical selection. PARTICIPANTS: 56 785 UKCAT candidates who sat the test between 2013 and 2016 and provided valid responses to the online confidence pilot study. PRIMARY OUTCOME MEASURES: Two measures of 'online confidence' were derived: the well-established 'confidence bias', and; a novel 'confidence judgement' measure, developed using Item Response Theory in order to derive a more sophisticated metric of the ability to evaluate one's own performance on a task. Regression models investigated the relationships between these confidence measures, application success and academic performance. RESULTS: Online confidence was inversely related to cognitive performance. Relative underconfidence was associated with increased odds of receiving an offer to study medicine. For 'confidence bias' this effect was independent of potential confounders (OR 1.48, 1.15 to 1.91, p=0.002). While 'confidence judgement' was also a univariable predictor of application success (OR 1.22, 1.01 to 1.47, p=0.04), it was not an independent predictor. 'Confidence bias', but not 'confidence judgement', predicted the odds of passing the first year of university at the first attempt, independently of cognitive performance, with relative underconfidence positively related to academic success (OR 3.24, 1.08 to 9.72, p=0.04). No non-linear effects were observed, suggesting no 'sweet spot' exists in relation to online confidence and the outcomes studied. CONCLUSIONS: Applicants who either appear underconfident, or are better at judging their own performance on a task, are more likely to receive an offer to study medicine. However, online confidence estimates had limited ability to predict subsequent academic achievement. Moreover, there are practical challenges to evaluating online confidence in high-stakes selection.

Correcting the predictive validity of a selection test for the effect of indirect range restriction.

BACKGROUND: The validity of selection tests is underestimated if it is determined by simply calculating the predictor-outcome correlation found in the admitted group. This correlation is usually attenuated by two factors: (1) the combination of selection variables which can compensate for each other and (2) range restriction in predictor and outcome due to the absence of outcome measures for rejected applicants. METHODS: Here we demonstrate the logic of these artifacts in a situation typical for student selection tests and compare four different methods for their correction: two formulas for the correction of direct and indirect range restriction, expectation maximization algorithm (EM) and multiple imputation by chained equations (MICE). First we show with simulated data how a realistic estimation of predictive validity could be achieved; second we apply the same methods to empirical data from one medical school. RESULTS: The results of the four methods are very similar except for the direct range restriction formula which underestimated validity. CONCLUSION: For practical purposes Thorndike's case C formula is a relatively straightforward solution to the range restriction problem, provided distributional assumptions are met. With EM and MICE more precision is obtained when distributional requirements are not met, but access to a sophisticated statistical package such as R is needed. The use of true score correlation has its own problems and does not seem to provide a better correction than other methods.

Is perfect good? - Dimensions of perfectionism in newly admitted medical students.

BACKGROUND: Society expects physicians to perform perfectly but high levels of perfectionism are associated with symptoms of distress in medical students. This study investigated whether medical students admitted to medical school by different selection criteria differ in the occurrence of perfectionism. METHODS: Newly enrolled undergraduate medical students (n = 358) filled out the following instruments: Multidimensional Perfectionism Scale (MPS-H), Multidimensional Perfectionism Scale (MPS-F), Big Five Inventory (BFI-10), General Self-Efficacy Scale (GSE), Patient Health Questionnaire 9 (PHQ-9), and Generalized Anxiety Disorder 7 (GAD-7). Sociodemographic data such as age, gender, high school degrees, and the way of admission to medical school were also included in the questionnaire. RESULTS: The 298 participating students had significantly lower scores in Socially-Prescribed Perfectionism than the general population independently of their way of admission to medical school. Students who were selected for medical school by their high school degree showed the highest score for Adaptive Perfectionism. Maladaptive Perfectionism was the strongest predictor for the occurrence symptoms of depression and anxiety regardless of the way of admission. CONCLUSIONS: Students from all admission groups should be observed longitudinally for performance and to assess whether perfectionism questionnaires might be an additional useful instrument for medical school admission processes.

General Ability or Distinct Scholastic Aptitudes? A Multidimensional Validity Analysis of a Psychometric Higher-Education Entrance Test.

The present study explored the construct validity of the Psychometric Entrance Test (PET) for higher education in Israel, as represented by the factorial structure of the scholastic aptitudes it measures, and focused on whether the test presents a single measure of overall ability or a measure of the fields of knowledge that are being tested. In Study 1, we used Exploratory Factor Analysis to generate hypotheses regarding the factorial structure of the test. In Study 2, Confirmatory Factor Analysis was carried out to compare competing models that were constructed based on theoretical considerations and the results of Study 1. The findings indicated that a two-layered hierarchical model, encompassing both a General Ability factor and three scholastic domain-specific factors (Verbal Reasoning, Quantitative Reasoning, and English), showed the best fit. Within the framework of the CFA, several statistical procedures were applied to assess reliability (indicator and complexity) and validity (convergent and divergent.).

Association Between Medication Use and Performance on Higher Education Entrance Tests in Individuals With Attention-Deficit/Hyperactivity Disorder.

Importance: Individuals with attention-deficit/hyperactivity disorder (ADHD) are at greater risk for academic problems. Pharmacologic treatment is effective in reducing the core symptoms of ADHD, but it is unclear whether it helps to improve academic outcomes. Objective: To investigate the association between the use of ADHD medication and performance on higher education entrance tests in individuals with ADHD. Design, Setting, and Participants: This cohort study observed 61 640 individuals with a diagnosis of ADHD from January 1, 2006, to December 31, 2013. Records of their pharmacologic treatment were extracted from Swedish national registers along with data from the Swedish Scholastic Aptitude Test. Using a within-patient design, test scores when patients were taking medication for ADHD were compared with scores when they were not taking such medication. Data analysis was performed from November 24, 2015, to November 4, 2016. Exposures: Periods with and without ADHD medication use. Main Outcomes and Measures: Scores from the higher education entrance examination (score range, 1-200 points). Results: Among 930 individuals (493 males and 437 females; mean [SD] age, 22.2 [3.2] years) who had taken multiple entrance tests (n = 2524) and used ADHD medications intermittently, the test scores were a mean of 4.80 points higher (95% CI, 2.26-7.34; P < .001) during periods they were taking medication vs nonmedicated periods, after adjusting for age and practice effects. Similar associations between ADHD medication use and test scores were detected in sensitivity analyses. Conclusions and Relevance: Individuals with ADHD had higher scores on the higher education entrance tests during periods they were taking ADHD medication vs nonmedicated periods. These findings suggest that ADHD medications may help ameliorate educationally relevant outcomes in individuals with ADHD.

Assessment centers versus cognitive ability tests: Challenging the conventional wisdom on criterion-related validity.

Separate meta-analyses of the cognitive ability and assessment center (AC) literatures report higher criterion-related validity for cognitive ability tests in predicting job performance. We instead focus on 17 samples in which both AC and ability scores are obtained for the same examinees and used to predict the same criterion. Thus, we control for differences in job type and in criteria that may have affected prior conclusions. In contrast to Schmidt and Hunter's (1998) meta-analysis, reporting mean validity of .51 for ability and .37 for ACs, we found using random-effects models mean validity of .22 for ability and .44 for ACs using comparable corrections for range restriction and measurement error in the criterion. We posit that 2 factors contribute to the differences in findings: (a) ACs being used on populations already restricted on cognitive ability and (b) the use of less cognitively loaded criteria in AC validation research. (PsycINFO Database Record

The impacts of computer adaptive testing from a variety of perspectives.

Computer adaptive testing (CAT) is a kind of tailored testing, in that it is a form of computer-based testing that is adaptive to each test-taker's ability level. In this review, the impacts of CAT are discussed from different perspectives in order to illustrate crucial points to keep in mind during the development and implementation of CAT. Test developers and psychometricians often emphasize the efficiency and accuracy of CAT in comparison to traditional linear tests. However, many test-takers report feeling discouraged after taking CATs, and this feeling can reduce learning self-efficacy and motivation. A trade-off must be made between the psychological experiences of test-takers and measurement efficiency. From the perspective of educators and subject matter experts, nonstatistical specifications, such as content coverage, content balance, and form length are major concerns. Thus, accreditation bodies may be faced with a discrepancy between the perspectives of psychometricians and those of subject matter experts. In order to improve test-takers' impressions of CAT, the author proposes increasing the target probability of answering correctly in the item selection algorithm even if doing so consequently decreases measurement efficiency. Two different methods, CAT with a shadow test approach and computerized multistage testing, have been developed in order to ensure the satisfaction of subject matter experts. In the shadow test approach, a full-length test is assembled that meets the constraints and provides maximum information at the current ability estimate, while computerized multistage testing gives subject matter experts an opportunity to review all test forms prior to administration.

Widening access in selection using situational judgement tests: evidence from the UKCAT.

CONTEXT: Widening access promotes student diversity and the appropriate representation of all demographic groups. This study aims to examine diversity-related benefits of the use of situational judgement tests (SJTs) in the UK Clinical Aptitude Test (UKCAT) in terms of three demographic variables: (i) socio-economic status (SES); (ii) ethnicity, and (iii) gender. METHODS: Outcomes in medical and dental school applicant cohorts for the years 2012 (n = 15 581) and 2013 (n = 15 454) were studied. Applicants' scores on cognitive tests and an SJT were linked to SES (parents' occupational status), ethnicity (White versus Black and other minority ethnic candidates), and gender. RESULTS: Firstly, the effect size for SES was lower for the SJT (d = 0.13-0.20 in favour of the higher SES group) than it was for the cognitive tests (d = 0.38-0.35). Secondly, effect sizes for ethnicity of the SJT and cognitive tests were similar (d = ~ 0.50 in favour of White candidates). Thirdly, males outperformed females on cognitive tests, whereas the reverse was true for SJTs. When equal weight was given to the SJT and the cognitive tests in the admission decision and when the selection ratio was stringent, simulated scenarios showed that using an SJT in addition to cognitive tests might enable admissions boards to select more students from lower SES backgrounds and more female students. CONCLUSIONS: The SJT has the potential to appropriately complement cognitive tests in the selection of doctors and dentists. It may also put candidates of lower SES backgrounds at less of a disadvantage and may potentially diversify the student intake. However, use of the SJT applied in this study did not diminish the role of ethnicity. Future research should examine these findings with other SJTs and other tests internationally and scrutinise the causes underlying the role of ethnicity.

Do later wake times and increased sleep duration of 12th graders result in more studying, higher grades, and improved SAT/ACT test scores?

BACKGROUND: The aim of this study was to investigate the relationship between sleep duration, wake time, and hours studying on high school grades and performance on the Scholastic Aptitude Test (SAT)/ American College Testing (ACT) college entrance exams. METHOD: Data were collected from 13,071 recently graduated high school seniors who were entering college in the fall of 2014. A column proportions z test with a Bonferroni adjustment was used to analyze proportional differences. Analysis of covariance (ANCOVA) was used to examine mean group differences. RESULTS: Students who woke up prior to 6 a.m. and got less than 8 h of sleep (27 %) were significantly more likely to report studying 11 or more hours per week (30 %), almost double the rate compared to students who got more than 8 h of sleep and woke up the latest (16 %). Post hoc results revealed students who woke up at 7 a.m. or later reported significantly higher high school grades than all other groups (p < 0.001), with the exception of those students who woke up between 6:01 a.m. and 7:00 a.m. and got eight or more hours of sleep. The highest reported SAT/ACT scores were from the group that woke up after 7 a.m. but got less than 8 h sleep (M = 1099.5). Their scores were significantly higher than all other groups. CONCLUSION: This study provides additional evidence that increased sleep and later wake time are associated with increased high school grades. However, this study also found that students who sleep the longest also reported less studying and lower SAT/ACT scores.

Item exposure control for multidimensional computer adaptive testing under maximum likelihood and expected a posteriori estimation.

Item bank stratification has been shown to be an effective method for combating item overexposure in both uni- and multidimensional computer adaptive testing. However, item bank stratification cannot guarantee that items will not be overexposed-that is, exposed at a rate exceeding some prespecified threshold. In this article, we propose enhancing stratification for multidimensional computer adaptive tests by combining it with the item eligibility method, a technique for controlling the maximum exposure rate in computerized tests. The performance of the method was examined via a simulation study and compared to existing methods of item selection and exposure control. Also, for the first time, maximum likelihood (MLE) and expected a posteriori (EAP) estimation of examinee ability were compared side by side in a multidimensional computer adaptive test. The simulation suggested that the proposed method is effective in suppressing the maximum item exposure rate with very little loss of measurement accuracy and precision. As compared to MLE, EAP generates smaller mean squared errors of the ability estimates in all simulation conditions.

UK medical selection: lottery or meritocracy?

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?

U.S. Dental School Applicants and Enrollees, 2014 Entering Class.

This report analyzes data collected annually by the American Dental Education Association (ADEA) on the characteristics of applicants to and first-year enrollees in dental schools. Among the key findings this year are that, since 2010, there has been a gradual decline in the number of individuals taking the Dental Admission Test (DAT), while the number applying to dental schools has remained relatively flat. During the same five-year time period, the number of first-year dental students has continuously increased. The result is an increasing rate of enrollment among applicants. While the overwhelming majority of dental schools utilize some aspect of a holistic admissions process, the DAT scores and grade point averages of applicants and enrollees continue to rise. Unlike a decade ago, women now account for nearly half of all applicants and enrollees. At least part of the gain has come from a decline in the number of men applicants, rather than a change in the enrollment rate. Among underrepresented minorities, there remains a relatively low number of applicants and a lower than average enrollment rate. Overall, by analyzing data about individuals applying to and matriculating in dental schools, this report provides a window into the future dental workforce.

The stochastically curtailed generalized likelihood ratio: A new termination criterion for variable-length computerized classification tests.

Computerized classification tests (CCTs) are used to classify examinees into categories in the context of professional certification testing. The term "variable-length" refers to CCTs that terminate (i.e., cease administering items to the examinee) when a classification can be made with a prespecified level of certainty. The sequential probability ratio test (SPRT) is a common criterion for terminating variable-length CCTs, but recent research has proposed more efficient methods. Specifically, the stochastically curtailed SPRT (SCSPRT) and the generalized likelihood ratio criterion (GLR) have been shown to classify examinees with accuracy similar to the SPRT while using fewer items. This article shows that the GLR criterion itself may be stochastically curtailed, resulting in a new termination criterion, the stochastically curtailed GLR (SCGLR). All four criteria-the SPRT, SCSPRT, GLR, and the new SCGLR-were compared using a simulation study. In this study, we examined the criteria in testing conditions that varied several CCT design features, including item bank characteristics, pass/fail threshold, and examinee ability distribution. In each condition, the termination criteria were evaluated according to their accuracy (proportion of examinees classified correctly), efficiency (test length), and loss (a single statistic combing both accuracy and efficiency). The simulation results showed that the SCGLR can yield increased efficiency without sacrificing accuracy, relative to the SPRT, SCSPRT, and GLR in a wide variety of CCT designs.

The contribution of general cognitive abilities and approximate number system to early mathematics.

BACKGROUND: Math learning is a complex process that entails a wide range of cognitive abilities to be fulfilled. There is sufficient evidence that both general and specific cognitive skills assume a fundamental role, despite the absence of shared consensus about the relative extent of their involvement. Moreover, regarding general abilities, there is no agreement about the recruitment of the different memory components or of intelligence. In relation to specific factors, great debate subsists regarding the role of the approximate number system (ANS). AIMS: Starting from these considerations, we wanted to conduct a wide assessment of memory components and ANS, by controlling for the effects associated with intelligence and also exploring possible relationships between all precursors. SAMPLE AND METHOD: To achieve this purpose, a sample of 157 children was tested at both beginning and end of their Grade 1. Both general (memory and intelligence) and specific (ANS) precursors were evaluated by a wide battery of tests and put in relation to concurrent and subsequent math skills. Memory was explored in passive and active aspects involving both verbal and visuo-spatial components. RESULTS: Path analysis results demonstrated that memory, and especially the more active processes, and intelligence were the strongest precursors in both assessment times. ANS had a milder role which lost significance by the end of the school year. Memory and ANS seemed to influence early mathematics almost independently. CONCLUSION: Both general and specific precursors seemed to have a crucial role in early math competences, despite the lower involvement of ANS.

Embarazo deseado y planificado, lactancia previa, aptitud para lactar y conocimientos de las madres sobre lactancia / Planned and wanted pregnancy, early breastfeeding, aptitudes for and mother's knowledge about breastfeeding

INTRODUCCIÓN: la OMS y la Unicef recomiendan lactancia materna exclusiva durante los primeros 6 meses de vida, y complementada, al menos, hasta los 2 años; sin embargo, los resultados no son los esperados. OBJETIVO: evaluar la relación entre la planificación del embarazo, la lactancia en hijos anteriores, la aptitud referida para lactar y los conocimientos sobre lactancia materna de las gestantes en su tercer trimestre, de los Policlínicos "Luis Augusto Turcios Lima" y "Mártires del 9 de Abril", del municipio San José de las Lajas. MÉTODOS: se realizó un estudio analítico y transversal, entre marzo de 2010 y marzo de 2011. De 192 gestantes en su tercer trimestre, y que cumplieron los criterios de inclusión, se estudiaron: hijos previos y si recibieron lactancia materna, planificación, deseos del embarazo, aptitudes referidas, así como conocimientos sobre la lactancia. Se aplicaron 22 preguntas y se estableció un sistema de puntaje. RESULTADOS: algo más de la mitad de las madres se consideraron aptas para la lactancia, y asumieron que poseían los conocimientos necesarios para lograrlo, pero menos de la quinta parte mostró poseerlos. Las gestantes que obtuvieron valores superiores en el cuestionario fueron: las que tenían un hijo previo, las que ya habían ofrecido lactancia materna exclusiva de 4 a 6 meses, y las que deseaban y habían planificado su embarazo. CONCLUSIONES: aunque la mayoría de las gestantes consideraron que estaban aptas para la lactancia materna, no demostraron tener los conocimientos necesarios para asumirla exitosamente.

INTRODUCTION: WHO and UNICEF recommend exclusive breastfeeding for the first six months of life and as supplement for two years at least; however, the final results are not the expected ones. OBJECTIVE: to evaluate the relationship between planned pregnancy, breastfeeding of previous children, aptitudes for breastfeeding and mother's knowledge about this in the third-trimester pregnant women from "Luis Augusto Turcios Lima" and "Martires del 9 de Abril" polyclinics located in San Jose de las Lajas municipality, Mayabeque province. METHODS: a cross-sectional analytical study was carried out from March 2010 through March 2011. One hundred and ninety two women in their third trimester of pregnancy, who met the inclusion criteria, provided information about their early children and their breastfeeding, planning, desires to be pregnant, breastfeeding aptitudes and knowledge. Twenty two questions were posed and then evaluated by a scoring system set for these purposes. RESULTS: over half of mothers regarded themselves as being apt for breastfeeding and they assured that they had the required knowledge for that, but less than one fifth did show it. The pregnant women who scored highest in the questionnaire were those having a child, those having exclusively breastfed for 4 to 6 months and those who had planned and desired their pregnancy. CONCLUSIONS: although the majority of pregnant women considered that they were ready for breastfeeding, they did not show adequate knowledge to succeed in this task.

On robustness and power of the likelihood-ratio test as a model test of the linear logistic test model.

Recently the linear logistic test model (LLTM) by Fischer (1973) is increasingly used. In applications of LLTM, a likelihood-ratio test comparing the likelihood of the LLTM to the likelihood of the Rasch model is the most often applied model test. The present simulation study evaluates the empirical Type I risk, test power, and approximation to the expected distribution in the context of the LLTM. Furthermore, as possible influence factors on the distribution of the likelihood-ratio test statistic, the misspecification of the superior model, the closeness to singularity of the design matrix, and different sorts of misspecification of the design matrix are implemented. In summary, results of the simulations indicate that the likelihood-ratio test statistic holds the fixed Type I risk under typical conditions. Nevertheless, it is especially important to ensure the fit of the superior model, the Rasch model, and to consider the closeness to singularity of the design matrix.

Predictive validity of the UK clinical aptitude test in the final years of medical school: a prospective cohort study.

BACKGROUND: The UK Clinical Aptitude Test (UKCAT) was designed to address issues identified with traditional methods of selection. This study aims to examine the predictive validity of the UKCAT and compare this to traditional selection methods in the senior years of medical school. This was a follow-up study of two cohorts of students from two medical schools who had previously taken part in a study examining the predictive validity of the UKCAT in first year. METHODS: The sample consisted of 4th and 5th Year students who commenced their studies at the University of Aberdeen or University of Dundee medical schools in 2007. Data collected were: demographics (gender and age group), UKCAT scores; Universities and Colleges Admissions Service (UCAS) form scores; admission interview scores; Year 4 and 5 degree examination scores. Pearson's correlations were used to examine the relationships between admissions variables, examination scores, gender and age group, and to select variables for multiple linear regression analysis to predict examination scores. RESULTS: Ninety-nine and 89 students at Aberdeen medical school from Years 4 and 5 respectively, and 51 Year 4 students in Dundee, were included in the analysis. Neither UCAS form nor interview scores were statistically significant predictors of examination performance. Conversely, the UKCAT yielded statistically significant validity coefficients between .24 and .36 in four of five assessments investigated. Multiple regression analysis showed the UKCAT made a statistically significant unique contribution to variance in examination performance in the senior years. CONCLUSIONS: Results suggest the UKCAT appears to predict performance better in the later years of medical school compared to earlier years and provides modest supportive evidence for the UKCAT's role in student selection within these institutions. Further research is needed to assess the predictive validity of the UKCAT against professional and behavioural outcomes as the cohort commences working life.

Automatic item generation implemented for measuring artistic judgment aptitude.

Automatic item generation (AIG) is a broad class of methods that are being developed to address psychometric issues arising from internet and computer-based testing. In general, issues emphasize efficiency, validity, and diagnostic usefulness of large scale mental testing. Rapid prominence of AIG methods and their implicit perspective on mental testing is bringing painful scrutiny to many sacred psychometric assumptions. This report reviews basic AIG ideas, then presents conceptual foundations, image model development, and operational application to artistic judgment aptitude testing.

The Hamburg selection procedure for dental students - introduction of the HAM-Nat as subject-specific test for study aptitude.

INTRODUCTION: The present study examines the question whether the selection of dental students should be based solely on average school-leaving grades (GPA) or whether it could be improved by using a subject-specific aptitude test. METHODS: The HAM-Nat Natural Sciences Test was piloted with freshmen during their first study week in 2006 and 2007. In 2009 and 2010 it was used in the dental student selection process. The sample size in the regression models varies between 32 and 55 students. RESULTS: Used as a supplement to the German GPA, the HAM-Nat test explained up to 12% of the variance in preclinical examination performance. We confirmed the prognostic validity of GPA reported in earlier studies in some, but not all of the individual preclinical examination results. CONCLUSION: The HAM-Nat test is a reliable selection tool for dental students. Use of the HAM-Nat yielded a significant improvement in prediction of preclinical academic success in dentistry.