Predicting who will drop out of nursing courses: A machine learning exercise

• Published in: Nurse education today Vol. 28; no. 4; pp. 469 - 475
• Main Authors: Moseley, Laurence G., Mead, Donna M.
• Format: Journal Article
• Language: English
• Published: Scotland Elsevier Ltd 01.05.2008; Elsevier Science Ltd
• Subjects: Algorithms; Analysis of Variance; Artificial Intelligence; Attrition; Causality; Chi-Square Distribution; College students; Data mining; Decision Trees; Dropping out; Education, Nursing, Baccalaureate; Forecasting; Humans; Logical Thinking; Logistic Models; Machine learning; Neural Networks (Computer); Nursing; Nursing education; Nursing Education Research; Prediction; Predictive Value of Tests; School dropouts; Sensitivity and Specificity; Student Attrition; Student drop outs; Student Dropouts - psychology; Student Dropouts - statistics & numerical data; Student Records; Students; Students, Nursing - psychology; Students, Nursing - statistics & numerical data; Studies; Time Factors; Wales; Wales; Data mining; Student drop outs; Machine learning; Prediction
• ISSN: 0260-6917; 1532-2793
• DOI: 10.1016/j.nedt.2007.07.012

The concepts of causation and prediction are different, and have different implications for practice. This distinction is applied here to studies of the problem of student attrition (although it is more widely applicable). Studies of attrition from nursing courses have tended to concentrate on causation, trying, largely unsuccessfully, to elicit what causes drop out. However, the problem may more fruitfully be cast in terms of predicting who is likely to drop out. One powerful method for attempting to make predictions is rule induction. This paper reports the use of the Answer Tree package from SPSS for that purpose. The main data set consisted of 3978 records on 528 nursing students, split into a training set and a test set. The source was standard university student records. The method obtained 84% sensitivity, 70% specificity, and 94% accuracy on previously unseen cases. The method requires large amounts of high quality data. When such data are available, rule induction offers a way to reduce attrition. It would be desirable to compare its results with those of predictions made by tutors using more informal conventional methods.