The 81st Annual Meeting of the American Association of Physical Anthropologists (2012)


An analysis of methods and importance of interobserver error in odontometric studies

KATIE ZEJDLIK and ANDREW R. THOMPSON.

Anthropology, Indiana University

Friday All day, Plaza Level Add to calendar

Interobserver error should be a concern for any study utilizing data collected by multiple individuals. The issue is of particular significance for investigations involving odontometrics because there are several different measurement methods commonly used by physical anthropologists. While the underlying information being sought after is often the same, different measurement protocols result in data that are incompatible. Even when attempting to adhere to specific measurement guidelines, published method descriptions tend to be ambiguous and difficult to replicate in practice. This introduces the complication of individual interpretation in determining the location of measurement landmarks, limiting the degree to which future research can build on existing datasets.

The purpose of this paper is to review commonly used methods for measuring dental remains as well as present the results of an interobserver error study completed by the authors. Research was conducted over a period of one year with measurements being recorded on the same specimens by both authors at three separate intervals. Substantial decrease of interobserver measurement error over time highlights the importance that experience plays in obtaining consistent results. Because teeth are small, variable structures, open dialog between observers during the process permitted fine-tuning of the measurement procedure for questionable or difficult specimens. Based on these findings it is suggested that interobserver error is important to consider both at the onset and completion of a project.

This study was funded by the David C. Skomp Fellowship Fund of the Department of Anthropology, Indiana University.

Tweet
comments powered by Disqus