U.S. flag

An official website of the United States government, Department of Justice.

Forensic anthropology casework performance: Assessing accuracy and trends for biological profile estimates on a comprehensive sample of identified decedent cases

NCJ Number
Journal of Forensic Sciences Volume: 66 Issue: 5 Dated: 2021 Pages: 1602-1616
Date Published
15 pages

The purpose of this study was to investigate the accuracy rates and trends in forensic anthropology casework concerning the estimation of the biological profile (sex, age, ancestry, and stature).



Identified cases from the Forensic Anthropology Database for Assessing Methods Accuracy (FADAMA; n = 359) were analyzed to explore the following: accuracy rates per biological profile component, case-level performance in assessing the biological profile, and factors related to inaccuracy rates. Accuracy rates for the four biological profile components ranged from 83% to 98%, with sex estimation performing the best and stature performing the poorest. While the overall sex estimation inaccuracies were the lowest of any biological profile component, we found that females are missexed approximately ten times more often than males. Inaccurate age estimates were more frequently the result of overestimation than underestimation, while the trends are reversed for stature estimation. Regarding ancestry estimation performance, African American/Black and White decedents had the lowest inaccuracy rates, while Hispanic and Asian/Pacific Islander decedents demonstrated greater inaccuracy rates. When examining accuracy rates for each case, 81% of cases had no inaccurate biological profile estimates, while 17% and 2% inaccurately estimated one and two biological profile components, respectively. The demographic trends of identified forensic anthropology cases reflect the national unidentified decedent demographics. Biological profile accuracy rates were generally comparable to previous studies. The findings highlight the current status of forensic anthropologists’ casework performance, with a greater amount of case-level inaccuracy rates than previously thought, and demonstrate the potential methodological and sampling strategies that could improve accuracy rates. (Publisher Abstract)

Date Published: January 1, 2021