Relationship between assessment results and approaches to learning and studying in Year Two medical students

CONTEXT: Students can take different approaches to learning and studying: deep (understanding material); surface (memorizing details), and strategic (motivated by assessments). It is important to know how assessments affect student choices of approach. METHODS: Students' learning approaches in Year 2 of the medical programme were measured using the Approaches to Study Skills Inventory for Students. The course was specifically designed to promote a deep approach and deter a surface approach, with explicit learning objectives and assessment constructively aligned according to Biggs' Structure of the Observed Learning Outcome (SOLO) taxonomy. The marks of individual students in different components of the assessment were compared with their scores for the 3 learning approaches. RESULTS: Marks correlated positively with deep and strategic approaches and negatively with surface approach across a range of assessment methods (relatively well with modified essays and multiple-choice questions, but poorly with in-course assignments). Peer assessment correlated particularly strongly with strategic approach. DISCUSSION: The correlations met our expectations in terms of direction, but were weaker and less consistent than anticipated. Possible reasons include the drive to test basic (core) material, the use of questions that may limit students' scope of expression and markers' ability to detect a deep approach. It is, however, important to refine medical programmes, particularly assessments, so that they concur with and do not adversely affect students' learning approaches.     

Does feedback matter? Practice‐based learning for medical students after a multi‐institutional clinical performance examination

CONTEXT: Achieving competence in 'practice-based learning' implies that doctors can accurately self- assess their clinical skills to identify behaviours that need improvement. This study examines the impact of receiving feedback via performance benchmarks on medical students' self-assessment after a clinical performance examination (CPX). METHODS: The authors developed a practice-based learning exercise at 3 institutions following a required 8-station CPX for medical students at the end of Year 3. Standardised patients (SPs) scored students after each station using checklists developed by experts. Students assessed their own performance immediately after the CPX (Phase 1). One month later, students watched their videotaped performance and reassessed (Phase 2). Some students received performance benchmarks (their scores, plus normative class data) before the video review. Pearson's correlations between self-ratings and SP ratings were calculated for overall performance and specific skill areas (history taking, physical examination, doctor-patient communication) for Phase 1 and Phase 2. The 2 correlations were then compared for each student group (i.e. those who received and those who did not receive feedback). RESULTS: A total of 280 students completed both study phases. Mean CPX scores ranged from 51% to 71% of items correct overall and for each skill area. Phase 1 self-assessment correlated weakly with SP ratings of student performance (r = 0.01-0.16). Without feedback, Phase 2 correlations remained weak (r = 0.13-0.18; n = 109). With feedback, Phase 2 correlations improved significantly (r = 0.26-0.47; n = 171). Low-performing students showed the greatest improvement after receiving feedback. CONCLUSIONS: The accuracy of student self-assessment was poor after a CPX, but improved significantly with performance feedback (scores and benchmarks). Videotape review alone (without feedback) did not improve self-assessment accuracy. Practice-based learning exercises that incorporate feedback to medical students hold promise to improve self-assessment skills.     

The relationship between professional behaviour grades and tutor performance ratings in problem-based learning

PURPOSE: We investigated the influence of harsh grading by tutors on tutor performance rating by students. METHODS: A total of 187 tutors assessed students' professional behaviour in tutorial groups. Students rated tutor performance after receiving their grades for professional behaviour. In addition, students were asked to indicate whether they perceived their professional behaviour grades as too positive, adequate or too negative. This was considered to reflect tutors' harshness of grading. Students also rated the quality of the feedback they received from tutors with respect to their grades. RESULTS: Professional behaviour grades that students perceived as too negative, adequate or too positive were associated with tutor performance ratings of 7.4 (SD = 0.9, scale 1-10, n = 33), 7.7 (SD = 0.9, scale 1-10, n = 95) and 7.5 (SD = 0.8, scale 1-10, n = 59), respectively. Harshness of grading did not influence tutor performance ratings significantly. Tutor ratings were predicted more effectively by the quality of the feedback tutors provided on grades than by the harshness of grading. CONCLUSIONS: Tutor performance ratings were not related significantly to harshness of grading. Two explanations can be given: (1) tutor performance ratings were based on rating by groups of students and (2) the percentage of tutors who rated students' professional behaviour as unsatisfactory was low. The strong relationship between tutor performance ratings and the adequacy of the feedback given by tutors suggests that the tutor performance ratings collected in this study are a valid measure of the quality of their teaching, although, for a full picture of teaching quality, more measures will be needed.     

Instructional methods and cognitive and learning styles in web-based learning: report of two randomised trials

CONTEXT: Adapting web-based (WB) instruction to learners' individual differences may enhance learning. Objectives This study aimed to investigate aptitude-treatment interactions between learning and cognitive styles and WB instructional methods. METHODS: We carried out a factorial, randomised, controlled, crossover, post-test-only trial involving 89 internal medicine residents, family practice residents and medical students at 2 US medical schools. Parallel versions of a WB course in complementary medicine used either active or reflective questions and different end-of-module review activities ('create and study a summary table' or 'study an instructor-created table'). Participants were matched or mismatched to question type based on active or reflective learning style. Participants used each review activity for 1 course module (crossover design). Outcome measurements included the Index of Learning Styles, the Cognitive Styles Analysis test, knowledge post-test, course rating and preference. RESULTS: Post-test scores were similar for matched (mean +/- standard error of the mean 77.4 +/- 1.7) and mismatched (76.9 +/- 1.7) learners (95% confidence interval [CI] for difference - 4.3 to 5.2l, P = 0.84), as were course ratings (P = 0.16). Post-test scores did not differ between active-type questions (77.1 +/- 2.1) and reflective-type questions (77.2 +/- 1.4; P = 0.97). Post-test scores correlated with course ratings (r = 0.45). There was no difference in post-test subscores for modules completed using the 'construct table' format (78.1 +/- 1.4) or the 'table provided' format (76.1 +/- 1.4; CI - 1.1 to 5.0, P = 0.21), and wholist and analytic styles had no interaction (P = 0.75) or main effect (P = 0.18). There was no association between activity preference and wholist or analytic scores (P = 0.37). CONCLUSIONS: Cognitive and learning styles had no apparent influence on learning outcomes. There were no differences in outcome between these instructional methods.     

Conditional reliability of admissions interview ratings: extreme ratings are the most informative

CONTEXT: Admissions interviews are unreliable and have poor predictive validity, yet are the sole measures of non-cognitive skills used by most medical school admissions departments. The low reliability may be due in part to variation in conditional reliability across the rating scale. OBJECTIVES: To describe an empirically derived estimate of conditional reliability and use it to improve the predictive validity of interview ratings. METHODS: A set of medical school interview ratings was compared to a Monte Carlo simulated set to estimate conditional reliability controlling for range restriction, response scale bias and other artefacts. This estimate was used as a weighting function to improve the predictive validity of a second set of interview ratings for predicting non-cognitive measures (USMLE Step II residuals from Step I scores). RESULTS: Compared with the simulated set, both observed sets showed more reliability at low and high rating levels than at moderate levels. Raw interview scores did not predict USMLE Step II scores after controlling for Step I performance (additional r2 = 0.001, not significant). Weighting interview ratings by estimated conditional reliability improved predictive validity (additional r2 = 0.121, P < 0.01). CONCLUSIONS: Conditional reliability is important for understanding the psychometric properties of subjective rating scales. Weighting these measures during the admissions process would improve admissions decisions.     

Evaluation of a surgical simulator for learning clinical anatomy

BACKGROUND: New techniques in imaging and surgery have made 3-dimensional anatomical knowledge an increasingly important goal of medical education. This study compared the efficacy of 2 supplemental, self-study methods for learning shoulder joint anatomy to determine which method provides for greater transfer of learning to the clinical setting. METHODS: Two groups of medical students studied shoulder joint anatomy using either a second-generation virtual reality surgical simulator or images from a textbook. They were then asked to identify anatomical structures of the shoulder joint as they appeared in a videotape of a live arthroscopic procedure. RESULTS: The mean identification scores, out of a possible score of 7, were 3.1 +/- 1.3 for the simulator group and 2.9 +/- 1.5 for the textbook group (P = 0.70). Student ratings of the 2 methods on a 5-point Likert scale were significantly different. The simulator group rated the simulator more highly as an effective learning tool than the textbook group rated the textbook (means of 3.2 +/- 0.7 and 2.6 +/- 0.5, respectively, P = 0.02). Furthermore, the simulator group indicated that they were more likely to use the simulator as a learning tool if it were available to them than the textbook group was willing to use the textbook (means of 4.0 +/- 1.2 and 3.0 +/- 0.9, respectively, P = 0.02). CONCLUSION: Our results show that this surgical simulator is at least as effective as textbook images for learning anatomy and could enhance student learning through increased motivation. These findings provide insight into simulator development and strategies for learning anatomy. Possible explanations and future research directions are discussed.     

Temporal and group-related trends in peer assessment amongst medical students

Peer assessment has been increasingly recommended as a way to evaluate the professional competencies of medical trainees. Prior studies have only assessed single groups measured at a single timepoint. Thus, neither the longitudinal stability of such ratings nor differences between groups using the same peer-assessment instrument have been reported previously. Participants were all members of 2 consecutive classes of medical students (n = 77 and n = 85) at the University of Rochester School of Medicine and Dentistry who completed Years 2 and 3 of medical school consecutively. All participants were evaluated by 6-12 classmates near the end of both Years 2 and 3. Main outcome measures were mean numerical ratings on peer-assessed scales of professional work habits (WH) and interpersonal attributes (IA). Both scales had high internal consistencies in both years (Cronbach's alpha 0.84-0.94). The IA and WH scales were moderately correlated with one another (r = 0.36 in Year 2, r = 0.28 in Year 3). Year 2 scores were predictive of Year 3 scores for both scales (WH: r = 0.64; IA; r = 0.62). Generalisability and decision analyses revealed that 1 class was consistently more discriminating with the WH scale, while the other was more discriminating with of the IA scale. Depending on the class, year and scale, the number of raters needed to achieve a reasonable reliability ranged between 7 and 28. Although Year 3 peer ratings were consistently higher than Year 2 peer ratings for both WH and IA, individual scores were highly correlated across the 2 years, despite the fact that different individuals were chosen as peer raters. Abilities appear to be stable between Years 2 and 3 of medical school. Groups may differ in their ability to discriminate different kinds of skills. Generalisability analysis can be used to discover these patterns within groups.     

Pre-test and post-test in a medicine clerkship

Offered the opportunity to take the previous quarter's post-test as a pre-test at the beginning of their 3-month medicine clerkship with the understanding it would have no effect on their final grades, 132 of 148 students (89%) took this opportunity. Post-test means for each quarter's group were significantly greater than pre-test means ( P = 0.0001). Pre-test means did not increase over the four quarters of the year of this study despite increasing student experience in other clinical disciplines. For this entire year correlations were as follows: Pre-test vs post-test r =0.591 ( P =0.0001), pre-test vs clinical r =0.306 ( P =0.0004), and post-test vs clinical r =0.329 ( P =0.0001). By quarter these correlations were generally consistent. An interesting and unexpected finding was that students in the autumn quarter had significantly lower pre-test and clinical scores than students in the winter quarter. This may indicate that students highly motivated to medicine take this as their first clinical clerkship and those least motivated as their last first clinical-year clerkship.     

Factors in faculty evaluation of medical students' performance

CONTEXT: Faculty members often use global rating scales as a method of assessing various characteristics of medical students' clinical performance. The purpose of this study was to determine if some performance characteristics are more highly associated with the overall faculty grade than others. METHODS: The clinical performance of 211 surgery clerkship students was evaluated by 2 or 3 faculty preceptors. Faculty rated students on 10 specific performance characteristics, using a 5-point scale. Faculty then assigned a numerical grade summarising the faculty's view of the student's performance. Reliability of the ratings was estimated by the intraclass correlation, and 1-way (analysis of variance) anova was used to test for differences among the students' mean ratings. Logistic regression was employed to determine the accuracy of each performance measure in predicting students' grades (A or B). Stepwise logistic regression was used to determine if there was a combination of performance characteristics that best predicted students' grades. RESULTS: The inter-rater reliabilities were low (相似文献   

Ultrasound anatomy: a practical teaching system in human gross anatomy

Cross-sectional imaging such as CT and ultrasound have rapidly developed a central role in diagnostic medicine. However, cross-sectional anatomy has lagged behind in its incorporation into medical school education, often leaving students without even a basic understanding of these images. At Hannover Medical School, we have successfully combined hands-on ultrasound workshops performed by medical students on other students with our anatomy course. This interactive component of our anatomy courses improves our students' understanding of clinical anatomy and introduces them to ultrasound imaging.     

Towards the identification of student learning problems: the development of a diagnostic inventory

An understanding of how students approach their learning has important implications for medical education. Of particular interest is the fact that the approach students use in their study has a significant impact on both the quality of the learning and their academic success. It would clearly be of value to identify students whose approach to learning was predictive of unsatisfactory performance. This paper describes the initial development of two versions of an inventory (questionnaire) which was designed to aid in the diagnosis of student learning problems. Preliminary information is given on the reliability and validity of these instruments. Factor analyses support the underlying design. Correlation of subscales with academic performance has provided encouraging evidence of the potential of these inventories in identifying students with specific learning or study problems.     

Analytic global OSCE ratings are sensitive to level of training

PURPOSE: There are several reasons for using global ratings in addition to checklists for scoring objective structured clinical examination (OSCE) stations. However, there has been little evidence collected regarding the validity of these scales. This study assessed the construct validity of an analytic global rating with 4 component subscales: empathy, coherence, verbal and non-verbal expression. METHODS: A total of 19 Year 3 and 38 Year 4 clinical clerks were scored on content checklists and these global ratings during a 10-station OSCE. T-tests were used to assess differences between groups for overall checklist and global scores, and for each of the 4 subscales. RESULTS: The mean global rating was significantly higher for senior clerks (75.5% versus 71.3%, t55 = 2.12, P < 0.05) and there were significant differences by level of training for the coherence (t55 = 3.33, P < 0.01) and verbal communication (t55 = 2.33, P < 0.05) subscales. Interstation reliability was 0.70 for the global rating and ranged from 0.58 to 0.65 for the subscales. Checklist reliability was 0.54. CONCLUSION: In this study, a summated analytic global rating demonstrated construct validity, as did 2 of the 4 scales measuring specific traits. In addition, the analytic global rating showed substantially higher internal consistency than did the checklists, a finding consistent with that seen in previous studies cited in the literature. Global ratings are an important element of OSCE measurement and can have good psychometric properties. However, OSCE researchers should clearly describe the type of global ratings they use. Further research is needed to define the most effective global rating scales.     

An investigation of medical student reactions to feedback: a randomised controlled trial

BACKGROUND: Medical educators have indicated that feedback is one of the main catalysts required for performance improvement. However, medical students appear to be persistently dissatisfied with the feedback that they receive. The purpose of this study was to evaluate learning outcomes and perceptions in students who received feedback compared to those who received general compliments. METHODS: All subjects received identical instruction on two-handed surgical knot-tying. Group 1 received specific, constructive feedback on how to improve their knot-tying skill. Group 2 received only general compliments. Performance was videotaped before and after instruction and after feedback. Subjects completed the study by indicating their global level of satisfaction. Three faculty evaluators observed and scored blinded videotapes of each performance. Intra-observer agreement among expert ratings of performance was calculated using 2-way random effects intraclass correlation (ICC) methods. Satisfaction scores and performance scores were compared using paired samples t-tests and independent samples t-tests. RESULTS: Performance data from 33 subjects were analysed. Inter-rater reliability exceeded 0.8 for ratings of pre-test, pre-intervention and post-intervention performances. The average performance of students who received specific feedback improved (21.98 versus 15.87, P<0.001), whereas there was no significant change in the performance score in the group who received only compliments (17.00 versus 15.39, P=0.181) The average satisfaction rating in the group that received compliments was significantly higher than the group that received feedback (6.00 versus 5.00, P=0.005). DISCUSSION: Student satisfaction is not an accurate measure of the quality of feedback. It appears that satisfaction ratings respond to praise more than feedback, while learning is more a function of feedback.     

Portfolios for assessment of paediatric specialist registrars

INTRODUCTION: In 1997 the Royal College of Paediatrics and Child Health introduced portfolios to guide and monitor the learning of specialist registrars. We studied their value for assessment. METHODS: Using Bigg's SOLO criteria we devised a marking scheme based on 6 domains of competence: clinical, communication, teaching and learning, ethics and attitudes, management and evaluation, and creation of evidence. We rated portfolios according to quality of evidence presented and expectations by year of training. We similarly assessed trainee performance in the annual record of in-training assessment (RITA) interview. Specific advice based on the results of the first portfolio assessments was circulated to all trainees, instructing them to increase the structure and decrease the bulk of portfolios. A second sample of portfolios was reviewed a year later, using similar evaluations, to determine the effects. RESULTS: A total of 76 portfolios were assessed in year 1 by a single rater; 30 portfolios were assessed in year 2 by 2 independent raters. The quality of documentation improved from year 1 to year 2 but there was no significant increase in portfolio scores. The inter-rater correlation coefficient of the portfolio assessment method was 0.52 (Cohen's kappa 0.35). The inter-rater correlation coefficient of the RITA interview was 0.71 (Cohen's kappa 0.38). There was moderate inter-assessment correlation between portfolios and RITA interviews (kappa 0.26 in year 1 and 0.29 in year 2). Generalisability analysis suggested that 5 successive ratings by a single observer or independent ratings by 4 observers on the same occasion would be needed to yield a generalisability coefficient > 0.8 for overall portfolio rating. CONCLUSIONS: This method of portfolio assessment is insufficiently reliable as a sole method for high stakes, single-instance assessment, but has a place as part of a triangulation process. Repeated portfolio assessment by paired observers would increase reliability. Longer term studies are required to establish whether portfolio assessment positively influences learner behaviour.     

Predicting performance in anatomy using antecedent academic criteria

Summary. The performance of two multivariate prediction models (equal and differential weights) at forecasting the outcome of the comprehensive examination in preclinical anatomy is reported. The models were devised by regression analysis, using scores in two antecedent examinations which correlated highly with the criterion variable, namely the science aggregate at the Joint Matriculation Examination and the premedical science aggregate. Five years of data were considered and the scores for each year were computed using the formula derived from scores of the preceding year. The average failure rate for the period covered by the study was 31·28%.
The correlation between actual and predicted scores was positive and moderately high (Pearson r = 0·47 and 0·49 for the equal- and differential-weights models respectively). On average, the equal-weights model correctly predicted 33·9% of failures, with a false alarm rate of 28·4%, compared with 44·3% and 38·9% respectively for the differential-weights model. In predicting candidates whose scores would fall in the bottom half and bottom third of the class, no statistically significant difference was noted between the hit rates achieved by both models.
The equal-weights model is simple to formulate and efficient in operation. Although its hit rate at the pass/fail boundary was lower than that of the differential-weights model, it yielded a significantly lower false alarm rate (28·4% vs 38·9%).
The relevance and application of performance prediction to the planning of remedial instruction are discussed.     

The rating of pre-clerkship examination questions by postgraduate medical students: an assessment of quality and relevancy to medical practice

OBJECTIVE: To determine whether postgraduate students are able to assess the quality of undergraduate medical examinations and to establish whether faculty can use their results to troubleshoot the curriculum in terms of its content and evaluation. SUBJECTS: First and second year family medicine postgraduate students. MATERIALS: A randomly generated sample of undergraduate medical examination questions. METHODS: Postgraduate students were given two undergraduate examinations which included questions with an item difficulty (ID) > 0.60. The students answered and then rated each question on a scale of 1-7. RESULTS: The percentage of postgraduate students answering each question correctly correlated significantly with the average perceived relevance (Examination 1: r=0.372; P < 0.05; Examination 2: r=0.458; P < 0.05). Questions plotted for average postgraduate/undergraduate performance ratio versus the average perceived relevance were significantly correlated (Examination 1: r=0.462; P < 0.01; Examination 2: r=0.458; P < 0.05). CONCLUSIONS: This study offers a method of validating question appropriateness prior to examination administration. The design has the potential to be used as a model for determining the relevancy of a medical curriculum.     

The effectiveness of a computer-assisted instruction programme on communication skills of medical specialists in oncology

BACKGROUND: Although doctor--patient communication is important in health care, medical specialists are generally not well trained in communication skills. Conventional training programmes are generally time consuming and hard to fit into busy working schedules of medical specialists. A computer-assisted instruction (CAI) programme was developed -- 'Interact-Cancer' -- which is a time-efficient learning method and easily accessible at the workplace. OBJECTIVE: To investigate the effect of the CAI training, 'Interact-Cancer', on the communication behaviour of medical specialists, and on satisfaction of patients about their physician interaction. DESIGN: Consultations of medical specialists with cancer outpatients were videotaped at 4 specific stages, 2 before and 2 after Interact-Cancer, with intervals of 4 weeks. PATIENTS/PARTICIPANTS: Participants were 21 medical specialists, mainly internists, working in 7 hospitals, and 385 cancer outpatients. METHODS: Communication behaviour was assessed on 23 observation categories derived from the course content. Frequencies were rated as well as judgements about the quality of the performance of each target skill. Satisfaction was measured by the Medical Interview Satisfaction Scale. Data were analyzed by means of multilevel statistical methods. RESULTS: The behavioural assessment showed course effects on ratings of the physicians' quality of performance. No course effects were found on the frequencies of physicians' behaviours and on the patient satisfaction ratings. CONCLUSIONS: CAI is a promising method to supply medical specialists with postgraduate training of communication skills. The application of judgement ratings of communication behaviour proved to be valuable to evaluate course effects in real-life patient encounters.     

Predictability of medical students' performance at the Aga Khan University from admission test scores, interview ratings and systems of education

INTRODUCTION: Performance in an admission test and carefully conducted, structured interview provides a uniform basis for the assessment of applicants prepared in different systems of education for admission to an institute of higher professional learning. We studied the predictability of the system of education followed by the students prior to entrance into the Medical College, the admission test scores, and interview ratings on performance after five trimesters at the Aga Khan University Medical College. METHOD: A cohort of 374 medical students who were admitted during 1989--1994, were considered. The associations between the admission test score, interview ratings, system of education, and the scores obtained for anatomy, physiology, biochemistry and community health sciences examinations held after five trimesters were analysed using appropriate statistical procedures. RESULTS: Interview ratings were associated with the scores in Physiology but not other subjects. The chief finding of this study was the association between the system of education and performance in both the admission test and the examination after five trimesters. Students who followed the British school curricula for 13 years scored significantly higher than those who followed the 12 years of the Pakistani system. When controlled for the admission test score, the difference in mean scores of the two groups was still evident for two subjects; community health sciences and physiology. CONCLUSION: We believe that the evidence indicates differences in learning methods inculcated by the system of education prior to entry into the Medical College, notwithstanding the 1-year difference in duration of education.     

A preliminary study of empathy, emotional intelligence and examination performance in MBChB students

CONTEXT: There is considerable interest in the attributes other than cognitive ability that medical students need in order to be professionally successful, with a particular focus on empathy and emotional intelligence (EI). Selection considerations have also motivated interest in such attributes as predictors of academic success. There are reports of declines in empathy in US medical students, but no comparative information is available for UK students. OBJECTIVES: This study aimed to compare empathy levels in medical students in Years 2, 3 (pre-clinical) and 5 (clinical), to examine gender differences in empathy and EI, and to investigate whether EI and empathy are related to academic success. METHODS: Questionnaires assessing EI and empathy were completed by students. Previous empathy scores for the Year 2 cohort were also available. Empathy trends were examined using anova; trends for the Year 2 group for whom Year 1 scores were available were examined using repeated-measures anova. Associations of EI and empathy with academic success were examined using Pearson correlation. RESULTS: A significant gender x cohort effect was found, with male empathy scores increasing between Years 1 and 2, whilst female scores declined. Peer ratings in Year 2 problem-based learning (PBL) groups were positively correlated with EI. CONCLUSIONS: Trends in levels of empathy differed by gender. The reasons for this require further investigation, particularly in relation to course content. Associations between academic performance and EI were sparse, and there were none between academic performance and empathy, but the effects of EI (and other characteristics) on PBL group functioning represent a promising area for future study.     

Does problem‐based learning lead to deficiencies in basic science knowledge? An empirical case on anatomy

INTRODUCTION: Problem-based learning (PBL) is supposed to enhance the integration of basic and clinical sciences. In a non-integrative curriculum, these disciplines are generally taught in separate courses. Problem-based learning students perceive deficiencies in their knowledge of basic sciences, particularly in important areas such as anatomy. Outcome studies on PBL show controversial results, sometimes indicating that medical students at PBL schools have less knowledge of basic sciences than do their colleagues at more traditional medical schools. We aimed to identify differences between PBL and non-PBL students in perceived and actual levels of knowledge of anatomy. METHODS: Samples of Year 4 students in all eight medical schools in the Netherlands completed a questionnaire on perceived knowledge and took part in a computerised anatomy test consisting of both clinically contextualised items and items without context. RESULTS: Problem-based learning students were found to have the same perceived level of anatomy knowledge as students at other medical schools. Differences in actual levels of knowledge were found between schools. No significant effects on knowledge levels were found for PBL schools versus non-PBL schools. CONCLUSION: The results of this study show that PBL does not result in a lower level of anatomy knowledge than more traditional educational approaches. It remains to be ascertained whether the levels students attain are adequate. Subjects for further study are the desired level of anatomy knowledge at the end of undergraduate medical education and the effectiveness of basic science learning within a clinical context and with repetition over the course of the curriculum.