Do Factors from Admissions and Dental School Predict Performance on National Board Exams? A Multilevel Modeling Study. Issue 10 (1st October 2019)

• Record Type: Journal Article
• Authors: Chahine, Saad; Plouffe, Rachel A.; Goldberg, Harvey A.; Sadler, Kathy; Drosdowech, Nadine; Bohay, Richard; Garcia, Bertha; Hammond, Robert
• Is Part Of: Journal of dental education. Volume 83:Issue 10(2019)
• Page Start: 1213
• Page End: 1223
• Publication Date: 2019-10-01
• Subjects: dental education -- dental students -- dental school admissions -- admissions criteria -- admissions/selection -- assessment -- licensure -- hierarchical linear modeling -- technical skills; Dentistry -- Study and teaching -- Periodicals; Dentistry -- Study and teaching; Education, Dental; Dentisterie -- Étude et enseignement -- Périodiques; Tandheelkunde; Onderwijs; Periodicals; Periodical; Electronic journals; 617.6007
• DOI: 10.21815/JDE.019.111
• Languages: English
• ISSNs: 0022-0337

Abstract

Abstract : The aim of this study was to assess the association among admissions variables, dental school performance, and licensing exam performance for six cohorts of graduates of one dental school. Data from all dental school graduates of Schulich School of Medicine & Dentistry, Western University, Canada, from 2009 to 2014 who had matching National Dental Examining Board of Canada (NDEB) data (N=298) were analyzed. In the results, significant differences between cohorts were found on both the NDEB objective structured clinical examination (OSCE) and written scores. Approximately 18% of the variation in OSCE scores was attributable to cohort differences and 82% to student differences. Approximately 10% of the variation in written scores was attributable to cohort differences and 90% to student differences. Several multilevel models were conducted. The final predictive model for NDEB OSCE scores consisted of age, Canadian Dental Aptitude Test (DAT) reading comprehension scores, year 2 average, and year 4 average. For predicting NDEB written exam scores, the final model consisted of DAT chemistry and year 1, 2, and 4 averages. The findings of this study showed that academic performance on admissions variables and in training predicted performance on dental licensing exams, whereas variables that captured noncognitive or interpersonal skills, such as interview scores, were not predictive. This difference may be due to construct mismatch, such that the outcome variables had no theoretical association with the predictors. Additional outcome measures (including noncognitive) are needed that have greater ecological validity in predicting potential for competence in practice.