Biology Reference
In-Depth Information
(
Spradley and Jantz, 2011
).
Frutos (2002)
achieved sex estimation accuracy rates (86
95%)
from discriminant function analysis for a Guatemalan sample, including the clavicle
maximum length, clavicle midshaft circumference, and the height and width of the gle-
noid fossa.
e
RADIUS AND HAND
Berrizbeitia (1989)
estimated sex from the head of the radius using multivariate discrim-
inant function analysis on a sample of European and African Americans from the Robert
J. Terry Anatomical Skeletal Collection.
9
The cross-validated results of the maximum and
minimum head diameters combined yielded accuracy of 92% with the left side, 94% with
the right, and 96% accuracy if both bones were used. Barrio and colleagues (2006) conducted
univariate discriminant function analysis using the metacarpals for sex estimation. These
researchers achieved accuracy of between 81% and 91%, with the best accuracy from the
left second metacarpal. They found that the right hand is generally larger than the left and
that the transverse dimensions are more dimorphic than the longitudinal measures (
Barrio
et al., 2006; Albanese et al., 2008
). They concluded that this dimorphism is the result of
extrinsic functional stress and physical activity, which will be highly population specific.
Lower Limb and Pelvic Girdle
PELVIS
The pelvis has had the reputation of having the best success for both sex estimation and sex
assessment due the functional necessity of the female pelvic inlet to be wide for childbirth. A
variety of indices and angles were developed decades ago for the purpose of sex estimation,
but the earlier analyses lacked statistical rigor for successful application. The sexually
dimorphic subpubic angle has been measured metrically and is acute in males and obtuse
in females (
Stewart, 1979
). A study of anteroposterior radiographs (X-rays) of individuals
from Uganda showed significant sexual dimorphism. In this study, the male subpubic angle
mean was 93.86 degrees (SD
17.79 degrees) for
females, though sex discrimination accuracy rates did not exceed 71%, likely because they
did not use a sophisticated statistical analysis (
Igbigbi and Nanono-Igbigbi, 2003
).
The ischiopubic index (length of pubis
¼
21.12 degrees) and 116.11 degrees (SD
¼
100/length of ischium) is lower in males and
was hoped to be a replacement for the use of the subpubic angle in sex estimation accuracy
(
Brothwell, 1963
). This index, however, includes the pubic length, the measurement that has
the highest interobserver measurement error (
Adams and Byrd, 2002
). Another index is in
the sacrum and has been used for sex estimation, called the sacral corporobasal index
(S1 corpus width
100/basal width), but there is extensive overlap between the sexes
(
Stewart, 1979
). Some of these pelvic indices may provide better accuracy for sex estimation
if discriminant function analysis is applied, as a suggestion for further study.
In the studies applying a univariate discriminant function approach, the success of the pelvis
for sex estimation is surprisingly mediocre compared to many other postcranial elements. Pat-
riquin and colleagues (2005) investigated metric sex estimation from the pelvis using discrim-
inant function analysis on a sample of South African Blacks and Whites. The ischial length was
9
This collection is curated by the National Museum of Natural History at the Smithsonian Institution in
Washington, D.C.
