IC Value
About Us
Editorial Board
Contact Us

A comparative approach between multiplication factor & linear regression model in predicting stature from dimensions of hand.

Authors:Sangeeta Dey , A.K. Kapoor
Int J Biol Med Res. 2015; 6(3): 5072-5077  |  PDF File


Forensic investigation requires examination of evidences to determine the identity of the individual to whom the evidence belongs. Thus forensic identification of unknown evidences, mutilated & skeletal remains becomes the main stay in forensic science, criminal proceedings and in the court of law. The prediction of stature is one of the important parameter along with age, sex when rendering biological profile of an individual. Forensic experts use various approaches to predict stature along with anatomical method. However, in many cases they are forced to use only mathematical model or formulae to derive stature due to non – availability of entire skeleton. Two most widely used mathematical model for predicting stature are multiplication factor method and linear regression analysis method. The approach of the present study is to compare between the two methods which can better predict stature and to derive prediction error to demonstrate the reliability and accuracy of using these methods. The research study is based on 180 individuals with equal percentage of males and females within age group 18 to 60 years. Stature, Right Hand Length (RHL), Left Hand Length (LHL), Right Hand Breadth (RHB), Left Hand Breadth (LHB) were taken as anthropometric measurements from each individual. Multiplication factor and linear regression equation were derived for prediction of stature from hand dimensions. Derived factors and equations were applied to the actual hand dimensions collected in the study. Then the comparison was made between the actual stature and the predicted stature from Multiplication factor (MF) and Linear Regression equation (RG) to find the prediction error. The results were analyzed statistically using IBM SPSS Version 20.0 computer software. Statistical analysis revealed that sex differences were found to be significant for all the variables at p < 0.001 by student’s t-test. It is evident from the study that the range of prediction error in case of RG analysis is lesser as compare to MF analysis. This shows that the prediction of stature is more reliable and accurate in case of regression analysis and hence confirming the fact that regression analysis method can better predict stature with minimum range of error in comparison with multiplication factor method.