Biology Reference
In-Depth Information
BOX 2.1
(
cont'd)
representative of the entire population, bias
notwithstanding, because that
interested in studying) and then draw the
sample. Ideally the sample should be drawn
in such a way that the entire population is
characterized, e.g., having an equal number
of males and females and all
age cohorts
represented. However, the sample you
analyze is obviously limited by the skeletons
available in the collection. Further, if the
sample originated from a cemetery,
remember that the age-at-death profile does
not mirror the proportions from the living
population (because the cemetery represents
those who died). See DiGangi and Moore
(Chapter 1) and Smith (Chapter 7) for more
information on this part of the osteological
paradox.
Each collection is the result of a mixture of
the sources of bias discussed above, in addi-
tion to the respective social and political
climate present during its initial curation. For
example, some modern collections are
composed of individuals who were impov-
erished during life. This makes extrapolating
conclusions about the population in general
from such a specific group problematic given
the particular environmental conditions they
may have been overly exposed to as a result
of their impoverishment, disproportionately
altering their skeletal features. As another
example, the prehistoric skeletal collection
curated by the Archaeological Museum in La
Serena, Chile has one archaeological site
represented by only skulls, as the archaeolo-
gists who did the excavations in the 1940s did
not think recovering the rest of the bones was
important (not an uncommon practice in the
early twentieth century given the emphasis
on
typologyd
see DiGangi and Moore
[Chapter 1]; DiGangi and Hefner [Chapter 5];
and Smith [Chapter 7], this volume).
is all
that
exists.
Taphonomic bias results from everything
that happens to remains after the moment of
death. A variety of factors influence remains
preservation
d
some will not be preserved at
all, some will be deteriorated, and others will
have excellent preservation. In some cases,
this may affect the preservation of remains
from a single site (i.e., infant remains not
being preserved as well as adult
remains
d
see
Walker et al., 1988
). Refer to
Marden et al. (Chapter 9), this volume for
more information on taphonomy.
There are other sources of bias. Modern
collections may be disproportionately
composed of individuals falling into one
demographic versus another because there
are cultural reasons for individuals being in
the collection (e.g., who decides to donate
themselves and why; whose decision has
been made for them by someone else and
why).
The Sample and Bias
We use samples to make statements about
populations. A “
population
” is a group of
individuals who are contemporaneous,
occupy relatively the same area geographi-
cally, have a shared culture (language, tradi-
tions, belief systems, etc.), and who tend to
find mates from within the same group. An
ideal sample is both (1) smaller than the
population and (2) representative of the
population; and is used to derive conclusions
about the entire population. This is a neces-
sary strategy in skeletal biology and biology
in general, because we could never analyze
every individual in the population.
It is therefore first necessary to identify the
population (which group of people you are
