Sex estimation accuracies from variables of the index and ring fingers in a Ghanaian population: Absolute lengths versus length ratios

There are variabilities in the sex and height estimation accuracies using variables of the index and ring fingers. This study sought to compare the performance of the absolute length and the length ratio variables of the index and ring fingers in sex and height estimation. The study was cross-sectional from January to June 2021 at the University for Development Studies. The participants were 216 (males=96, females=120) between 18 and 32 years. The index (IF), the ring (RF) finger lengths and their ratio (IF: RF) were obtained using computer-assisted analysis. Logistic regression (LogR) models were formulated for sex estimation while linear regression (LR) models were formulated for height estimation. Males had longer IF and RF than females (P < 0.001) but no sex differences were observed in the IF: RF ratio. The average sex estimation accuracies for all variables ranged from 54.2 % to 66.2 %. The absolute length of IF (61.6 %) and RF (63.9 %) were better than IF: RF (54.2 %) for sex estimation. The index finger distal phalanx (IFDP) was the most sexually dimorphic trait (66.2 %). Models for height estimation were significant except IF: RF in males which also had the largest standard error of estimation in females. The models for estimating height were more precise in females than males. There are substantial overlaps in variables of the index and ring fingers between males and females, such that there are less suitable for sex estimation but could be used for height estimation in a Ghanaian population.