A complete skeleton was found in McAllen. You established that it belongs to a White female. Which of the following regression formulae would yield the best estimate of stature? Justify your answer.


AStature $=86.62+2.523 \times($ Humerus $) \pm($ Standard Error $) 5.32$
BStature $=83.29+3.530 \times($ Radius $) \pm($ Standard Error $) 4.81$
CStature $=82.82+3.346 \times($ Ulna $) \pm($ Standard Error $) 4.51$
DStature $=76.51+2.487 \times($ Tibia $) \pm($ Standard Error $) 3.80$

Verified Answer :

  • The best estimate in forensic regression is determined by the lowest standard error, which minimizes prediction variance.
  • Comparing the provided standard errors (\(5.32, 4.81, 4.51, 3.80\)), we identify the minimum value of \(3.80\).
  • Formula (d) utilizes the tibia bone and provides the most precise estimation due to this lowest standard error.
  • The best regression formula is \(\boxed{d}\).

Explanation

  1. Understanding the Criterion for the "Best" Estimate

In forensic anthropology, stature estimation from skeletal remains is a fundamental task. When multiple regression formulae are available, each derived from a specific bone, we need a way to determine which one provides the most reliable (or "best") estimate. The accuracy and reliability of these estimates are determined by the standard error of the estimate (often denoted as SE). The standard error quantifies how much the observed stature values deviate from the predicted values provided by the regression equation. A lower standard error indicates that the predicted stature is closer to the true stature, thus offering greater precision. Therefore, the "best" estimate is the one with the smallest standard error.

  1. Comparing Standard Errors

We examine the provided standard errors (SE) for each of the four formulae:

ASE = \(5.32\)
BSE = \(4.81\)
CSE = \(4.51\)
DSE = \(3.80\)

Comparing these values, we look for the minimum:
\(3.80 < 4.51 < 4.81 < 5.32\).

The lowest standard error among the provided choices is \(3.80\), which corresponds to the tibia-based formula.

  1. Justification and Conclusion

Since formula (d) has the lowest standard error of \(3.80\), it provides the most precise estimate of stature among the options. In statistical regression, a lower standard error indicates a narrower confidence interval around the prediction, which directly correlates to a more precise estimation of the individual's height. Consequently, formula (d) is the superior choice for estimating stature.

Examples

Forensic anthropologists often use these regression equations to estimate the heights of individuals from skeletal remains, which is a key step in identifying missing persons. By selecting the formula with the lowest standard error, such as the one using the tibia bone, professionals can narrow down the potential height range of the individual with greater confidence, helping to refine the profile used in legal and medical investigations. If, for instance, a tibia measures 35 cm, the estimate is \(76.51 + 2.487 \times 35 \approx 163.56 \text{ cm}\), with a standard error of only 3.80 cm, providing a more reliable range than using an alternative bone with higher error.