RESUMO
BACKGROUND: Dropout and poor academic performance are persistent problems in medical schools in emerging economies. Identifying at-risk students early and knowing the factors that contribute to their success would be useful for designing educational interventions. Educational Data Mining (EDM) methods can identify students at risk of poor academic progress and dropping out. The main goal of this study was to use machine learning models, Artificial Neural Networks (ANN) and Naïve Bayes (NB), to identify first year medical students that succeed academically, using sociodemographic data and academic history. METHODS: Data from seven cohorts (2011 to 2017) of admitted medical students to the National Autonomous University of Mexico (UNAM) Faculty of Medicine in Mexico City were analysed. Data from 7,976 students (2011 to 2017 cohorts) of the program were included. Information from admission diagnostic exam results, academic history, sociodemographic characteristics and family environment was used. The main dataset included 48 variables. The study followed the general knowledge discovery process: pre-processing, data analysis, and validation. Artificial Neural Networks (ANN) and Naïve Bayes (NB) models were used for data mining analysis. RESULTS: ANNs models had slightly better performance in accuracy, sensitivity, and specificity. Both models had better sensitivity when classifying regular students and better specificity when classifying irregular students. Of the 25 variables with highest predictive value in the Naïve Bayes model, percentage of correct answers in the diagnostic exam was the best variable. CONCLUSIONS: Both ANN and Naïve Bayes methods can be useful for predicting medical students' academic achievement in an undergraduate program, based on information of their prior knowledge and socio-demographic factors. Although ANN offered slightly superior results, Naïve Bayes made it possible to obtain an in-depth analysis of how the different variables influenced the model. The use of educational data mining techniques and machine learning classification techniques have potential in medical education.
Assuntos
Estudantes de Medicina , Humanos , Teorema de Bayes , Escolaridade , Logro , Redes Neurais de ComputaçãoRESUMO
El objetivo del presente estudio fue conocer las propiedades psicométricas del Inventario de Estilos de Aprendizaje de Kolb y del Cuestionario de Felder-Silverman en una muestra de estudiantes de educación física de Chile. Para ello se aplicó ambos instrumentos a 141 estudiantes de dicha carrera de la Universidad Católica Silva Henríquez, siendo el 34,8% mujeres y el 65,2% hombres. Los resultados muestran que las cuatro subescalas del inventario de Kolb y las 4 subescalas del cuestionario de Felder-Silverman poseen índices adecuados de validez, sin embargo, los índices de confiabilidad obtenidos con alfa de Cronbach se ubican bajo los niveles aceptables en todas las subescalas. Se recomiendan más estudios con muestras de diversas universidades para confirmar las propiedades psicométricas de estos instrumentos en este tipo de población.
The aim of the present study was to know the psychometric properties of the learning styles inventory of Kolb and Felder-Silverman's questionnaire in a students' sample of physical education of Chile. For it both instruments were applied to 141 students of the above mentioned career of the Universidad Católica Silva Henríquez, being 34,8% ladies and 65,2 % males. The results show that four sub-scales of Kolb's inventory and 4 sub-scales of Felder-Silverman's questionnaire possess suitable indexes of validity, nevertheless, the indexes of reliability obtained with Cronbach's alpha are located under the acceptable levels in all the subscales. More studies are recommended by samples of diverse universi-ties to confirm the psychometric properties of these instruments in this type of population.
