Biology Reference
In-Depth Information
to develop an ordinal scoring system, if the trait is not easily measured metrically. Analysis of
variation (ANOVA) is an acceptable statistical method to calculate whether the sexual dimor-
phism you observe is statistically significant. If the trait demonstrates significant sexual
dimorphism, a discriminant function analysis is not as daunting as you might believe. In
most statistical packages it is as simple as one click to conduct a discriminant function clas-
sification. Although these analyses may be much more feasible today, interpretation of the
results requires a deeper understanding of statistics than is presented in this chapter. I recom-
mend taking statistics courses through the math department of your university (including
multivariate statistics) in addition to any courses that may be offered within the biological
or social sciences. Most graduate programs in anthropology will have this same requirement.
It would behoove you to take a course in osteometrics in addition to osteology and skeletal
biology, if offered.
In some ways, research methods in sex estimation and assessment have changed a great
deal in the last 50 years, yet many of the observations are still the same. Although a great
deal of research has been conducted on the estimation and assessment of sex from the skel-
eton, there is still a need for a number of studies to be developed, as populations continue to
evolve. Albanese and colleagues (2008:1283) state, “A widely accepted but erroneous view
has been and continues to be that morphological methods can be applied across populations
while metric methods are population specific.” As demonstrated in nearly every study
described in this chapter, population variation has a significant impact on the ability to esti-
mate sex accurately from the skeleton. Metric standards must be population specific in both
time and space (
Buikstra and Ubelaker, 1994
). Due to the scarcity of subadult skeletal collec-
tions, molecular methods could be used as the standard for sexing as explained above,
against which metric sex estimation methods can then be validated. Molecular analyses
could be used to develop a larger known sample of subadults in order to test and validate
the use of new metric sex estimation methods in subadults. Other studies in subadults could
compare skeletal and dental development to examine timing differences in growth and
development that could be used to estimate sex.
Quantitative estimation of sex using the sex ratio in demography is yet another area that
would be wise to pursue, in conjunction with other methods of sex estimation and assess-
ment. Pathology is an area that has received very little attention, but will likely never provide
accuracy for sex estimation that would be any better than chance. Sample sizes in this area are
simply too small and confounding variables too large.
I encourage all new students of skeletal biology considering research in sex estimation to
thoughtfully consider the etiology of sexual dimorphism. By investigating the intrinsic and
extrinsic variables that ultimately cause sexual dimorphism, we may better select traits of
importance and gain a deeper understanding of the biocultural variables involved in the
development of the human skeleton.
REFERENCES
Adams, B.J., Byrd, J.E., 2002. Interobserver variation of selected postcranial skeletal measurements. Journal of
Forensic Sciences 47 (6), 1193
1202.
Albanese, J., Eklics, G., Tuck, A., 2008. A metric method for sex determination using the proximal femur and
fragmentary hipbone. Journal of Forensic Sciences 53 (6), 1283
e
e
1288.
