Caveats in Using Trotter and Gleser's (1958) Asian Equations for Stature Estimation

Abstract

Trotter and Gleser’s (1958) Asian equations have most frequently been used for stature estimation of Korean skeletal remains. However, limitations or caveats in using those equations have rarely been argued. This study reviews five issues frequently overlooked in applying the equations and interpreting the outcomes. First, Trotter and Gleser (1958) multiply the standard errors by 2 to obtain the 95% prediction interval (PI). However, there is discrepancy between their calculation and actual PIs, and thus correct method for PI calculation is recommended. Secondly, given the uncertainty about the tibia length measurement, there is a possibility that the tibia-related equations yield biased estimates. Thirdly, since a mathematical error was incorporated in the development process of the combined equations, caution should be taken. Fourthly, extrapolation may cause additional unexpected error in applying regression equations. Therefore, it is necessary to check if a target sample falls within the reference sample range prior to using the equations. Lastly, applying Trotter and Gleser’s (1958) equations to female samples should be avoided because they produce highly biased estimates. The issues discussed in this study will contribute to reducing potential errors associated with application of these equations, and eventually enhancing accuracy of the final stature estimates. If errors are anticipated but unavoidable, discussions about the potential errors should be made to minimize misunderstanding about the outcomes.