Testing the guidelines of the New Coimbra Method for recording entheseal changes via interobserver agreement in a Bronze Age skeletal sample from Kultepe-Kanesh, Turkiye

Kale D., ÜSTÜNDAĞ H., Ozen S., Ozgu D. C.

INTERNATIONAL JOURNAL OF OSTEOARCHAEOLOGY, vol.33, no.3, pp.479-488, 2023 (AHCI) identifier identifier

  • Publication Type: Article / Article
  • Volume: 33 Issue: 3
  • Publication Date: 2023
  • Doi Number: 10.1002/oa.3215
  • Journal Indexes: Arts and Humanities Citation Index (AHCI), Social Sciences Citation Index (SSCI), Scopus, Academic Search Premier, Periodicals Index Online, Anthropological Literature
  • Page Numbers: pp.479-488
  • Keywords: activity, Coimbra Method, entheseal changes, interobserver agreement, kappa
  • Anadolu University Affiliated: Yes


The New Coimbra Method is the most recently proposed standardized method to assess entheseal changes (EC). However, the method's developers acknowledge the inadequacy of their guidelines and state that those who will use the method need in-person training from them. This study aims to determine the applicability of the method's guideline by testing the interobserver agreement (IOA) with observers who were not trained by the developers. The second objective of this study is to apply the New Coimbra Method to a relatively small archeological sample, in this case a Bronze Age population from Kultepe-Kanesh, Turkiye. Four observers (the authors) scored EC in three entheses of the right upper limb of 52 adults. IOA was analyzed via Fleiss' kappa (FK) and percentage of agreement (PA) tests. FK test results indicate moderate overall agreement (0.458), while PA implies significantly high agreement (95%). The results of the FK and PA tests contradict each other due to the known limitations of the latter. However, FK values among features vary broadly; some of the features indicate substantial agreement and surpass the required benchmark, while the rest remains moderate or lower. The features that revealed substantial agreement indicate the sufficiency of the method's guidelines. However, the guidelines for the other features appear to be inadequate. Further revision of the guidelines could make the method more applicable for all researchers with no requirement of in-person training.