Comparisons of statistical techniques to assess age-related skeletal markers in bioarchaeology.
In this technical note by Cheverko and Hubbe explores the impact different statistical tests have on determining patterns and trends of osteoarthritis in past populations. This study aims to contribute to the analysis of any type of osteological marker based collected as present or absent. They compares the results of Chi-Square, the Fisher’s Exact, Odds Ratio tests, and ANCOVAs. The study included individuals from seven archaeological sites and examined the prevalence of osteoarthritis in the hip, knee, shoulder, and elbow. The comparison between tests shows that for larger sample sizes all tests provide similar accuracy, with the ANCOVA presenting the important advantage of being able to correct for the effect of age in the osteological markers. Therefore, even though binary data violates some of the assumptions behind ANCOVAs, this Cheverko and Hubbe argue this test should be used more often in bioarchaeological studies.