Integrated k-means clustering with data envelopment analysis of public hospital efficiency


HEALTH CARE MANAGEMENT SCIENCE, vol.23, no.3, pp.325-338, 2020 (SSCI) identifier identifier identifier

  • Publication Type: Article / Article
  • Volume: 23 Issue: 3
  • Publication Date: 2020
  • Doi Number: 10.1007/s10729-019-09491-3
  • Journal Indexes: Social Sciences Citation Index (SSCI), Scopus, ABI/INFORM, Business Source Elite, Business Source Premier, CINAHL, EconLit, EMBASE, MEDLINE, Public Affairs Index
  • Page Numbers: pp.325-338
  • Keywords: K-means clustering, Data envelopment analysis, Public hospital, Efficiency, Turkey, HEALTH-CARE PERFORMANCE, DEA, TURKEY, MODEL
  • Hacettepe University Affiliated: Yes


The goal of this study is to integrate k-means clustering with data envelopment analysis to examine technical efficiencies in public hospitals in Turkey. A two-step analysis procedure involving provinces and public hospitals is applied in this study. The first step examines similar provinces in terms of welfare state indicators by using k-means clustering and silhouette (Sil) cluster validity index measures. Then, the efficiencies of public hospitals in different groups of provinces are determined. The data are taken from the Turkish Statistical Institute and the 2017 Public Hospitals Statistical Year Book for eighty-one provinces and 688 public hospitals. Study results show that, relative to similarities of welfare state indicators, there are five province groups (Sil = .58). The number of technically inefficient public hospitals is greater than the number of technically efficient public hospitals in all groups. Study results emphasize that incorporated methodology of k-means clustering with data envelopment analysis is useful to identify efficiencies of public hospitals located in provinces that have similar welfare status.