The Use of Decision Tree to Predict Nursing Turnover - A Case Study in a Public Hospital

• Main Authors: Hung Wen Kao, 高鴻文
• Other Authors: S. W. Lin
• Format: Others
• Published: 2012
• Online Access: http://ndltd.ncl.edu.tw/handle/25400362916697581771

碩士 === 長庚大學 === 管理學院碩士學位學程在職專班資訊管理組 === 100 === In recent years, the general public has become more and more demanding for the healthcare quality besides their life quality. They generally go to doctor in the hope receiving proper treatment. The hospital is becoming more and more important in people's mind. The overall healthcare quality would indirectly influence the trust that patients assign to the medical staff of their healthcare professional skills and subsequent medical procedures. Nonetheless, the trust of patients is significantly correlated with the service attitude, patience, and attentiveness of the medical staff. Meanwhile, staff turnover in the healthcare sector would, on the other hand, directly influence the overall healthcare quality. Due to the gap between the ideal and real work environments, some medical workers have suffered long-term maladjustments, both mentally and physically, upon entering this career field. Early prediction of possible medical staff turnover will not only reduce the costs of personnel and training, but also elevate both the healthcare quality and patient satisfaction. This thesis apply decision tree algorithm to basic information of the medical staff to analyzes the possible correlations between each of the medical staff turnover cases. Some medical workers from a publicly-sponsored hospital in Northern Taiwan had agreed to participate in this study as research subjects. The sampling period ended in July of 2011, whereas a total of 2351 samples were collected. The experimental results, along with the optimal combination of parameters, demonstrate an accuracy of 78.81% with the decision tree algorithm. This thesis is expected to provide advice of how to prevent high-quality medical workers from leaving their jobs and provide help to recruit newcomers.