This is the authors' abstract. We don't add key points for this paper.
AIM: To explore historical student data to identify patterns predictive of attrition risk among nursing students, and hence train a predictive model of an individuals' risk of leaving the course. BACKGROUND: The World Health Organization point to an international shortage of trained nurses, which poses a risk for patient safety and care worldwide. The risk is compounded where the workforce is also aging creating additional pressures on the delivery of quality care. To stabilize the workforce, a healthy supply of newly trained registered nurses is necessary; however undergraduate nursing has one of the highest rates of student attrition (approx. 24 %). METHODS: This study follows a knowledge discovery in databases (KDD) methodology performing an observational analysis of routinely collected student data. The data (1840 students, taken from the pre-existing university business intelligence systems) was modelled for three end points; 'attrition in 1st year', 'attrition in 2nd year', and 'failure to complete'. Analysis was performed via step-wise binomial regression. RESULTS: Several attrition factors have been identified by the model (e.g. students who return from periods of intermittence, are Male and/or non-mature have an increased likelihood to leave). CONCLUSION: To our knowledge this is the first study to examine the role of study intermittence on student attrition, or to be built on the pre-existing university business intelligence (BI) systems. The use of pre-existing university BI systems as reported here can serve as the grounding for an individual, tailored approach to retention strategy rather than an approach built on demographic assessment alone.
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Crisp et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: