Stature estimation and calibration: Bayesian and maximum likelihood perspectives in physical anthropology

Stature Estimation

“…When Pearson first introduced regression for stature estimation, he also set the trend for using the inverse calibration method (Pearson, 1899). Depending on the context, other statistical methods may actually be better (Konigsberg et al, 1998). When stature is used as the y variable, it implies that the bone length measured causes stature, which is an improbable causal relationship.…”

“…According to this same statistical review (Konigsberg et al, 1998), classical calibration is the regression of bone length on stature, and then solving for stature or x ¼ (y-b)/m. Stature is essentially estimated by ordinary least squares.…”

“…Konigsberg and colleagues (Konigsberg et al, 1998;Hens et al, 2000) tested all of the above statistical models on a case with extreme extrapolationdestimating the stature of the Australopithecus afarensis fossil AL 288-1, also known as "Lucy." Konigsberg and colleagues (Konigsberg et al, 1998;Hens et al, 2000) tested all of the above statistical models on a case with extreme extrapolationdestimating the stature of the Australopithecus afarensis fossil AL 288-1, also known as "Lucy."…”

“…Problems resulting from error in bone measurement (especially the tibia) and in the accuracy of the antemortem data have been discussed in several studies (Giles and Hutchinson, 1991;Willey and Falsetti, 1991;Ousley, 1995;Wilson et al, 2010). More recently, it has been noted that there is a high error when the reference sample and the estimated sample are genetically distinct, requiring modification of the statistical methods (Holliday and Ruff, 1997;Konigsberg et al, 1998;Hens et al, 2000;Ross and Konigsberg, 2002). Thus, despite numerous stature equations that have been developed for populations around the world, there is always the need for additional population-specific reference data.…”

“…In 1899, the Englishman Carl Pearson developed the modern method of using statistical regression (Pearson, 1899). This is the predominant mathematical model used today (also known as inverse calibration), though it may be important to consider whether the individual's stature in question is likely from the population from which the formula was derived (Konigsberg et al, 1998).…”