Derivation of regression equations for stature reconstruction from digit lengths of Nigerian Medical Students in University of Lagos

Nandi, Michael ; Okubike, Emeka ; Obun, Obun ; Onen, Onenson (2017-12)


Abstract: Identification of an individual using stature is of great significance in forensic practice. With the increasing frequency of natural and man-influenced disasters, it has become imperative for forensic anthropologists to establish population-specific forensic reference standards. This study aimed to derive predictive equations for stature estimation using five digit lengths. A sample of 230 medical students (100 males and 130 females) of Nigerian parentage, aged 18 to 36 years was recruited for this research. Stature and digit lengths were measured following standard procedures, followed by statistical analysis using SPSS (Version 20 Chicago Inc). Results of descriptive showed average stature of 176.36 ± 8.13cm, 164.38 ± 6.62cm and 169.59 ± 8.79cm for males, females and pooled sample respectively. Sexual dimorphism was observed to be statistically significant (P<0.01) across all the measured parameters, with greater values consistently recorded in males. Pearson correlation (r) had a range of 0.40 to 0.63. The weakest r-value is seen in males and strongest in females. Simple and multilinear regression equations derived showed different values of Coefficient of determination (R2) and standard error of estimate (SEE) at accurate estimation rate of >99%. Digits correlation with stature from this study may be of great relevance in human identification.

المستخلص: يعد التعرف على الأفراد باستخدام طول القامة ذو أهمية كبيرة بالنسبة لممارسة الطب الشرعي، ومع تزايد تواتر الكوارث الطبيعية والكوارث الناتجة عن الفعل البشري أصبح من الضروري أن يضع علماء الأنثروبولوجيا الجنائيون معايير مرجعية للسكان تساعد على الاستعراف خاصة بالطب الشرعي. هدفت هذه الدراسة إلى اشتقاق معادلات تتنبأ بتقدير القامة باستخدام أطوال الأصابع الخمس، ... .