Premium
Age Estimation in Forensic Anthropology: Quantification of Observer Error in Phase Versus Component‐Based Methods
Author(s) -
Shirley Natalie R.,
Ramirez Montes Paula Andrea
Publication year - 2015
Publication title -
journal of forensic sciences
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.715
H-Index - 96
eISSN - 1556-4029
pISSN - 0022-1198
DOI - 10.1111/1556-4029.12617
Subject(s) - forensic anthropology , kappa , cohen's kappa , statistics , component (thermodynamics) , intraclass correlation , observer (physics) , computer science , reliability (semiconductor) , measure (data warehouse) , mathematics , data mining , psychometrics , anthropology , physics , sociology , power (physics) , geometry , quantum mechanics , thermodynamics
The purpose of this study was to assess observer error in phase versus component‐based scoring systems used to develop age estimation methods in forensic anthropology. A method preferred by forensic anthropologists in the AAFS was selected for this evaluation (the S uchey– B rooks method for the pubic symphysis). The S uchey– B rooks descriptions were used to develop a corresponding component‐based scoring system for comparison. Several commonly used reliability statistics (kappa, weighted kappa, and the intraclass correlation coefficient) were calculated to assess observer agreement between two observers and to evaluate the efficacy of each of these statistics for this study. The linear weighted kappa was determined to be the most suitable measure of observer agreement. The results show that a component‐based system offers the possibility for more objective scoring than a phase system as long as the coding possibilities for each trait do not exceed three states of expression, each with as little overlap as possible.