Biology Reference
In-Depth Information
With our example of osteophyte formation, possible co-variates could include sex, activity
patterns, and diet (and possibly others). While we probably cannot control for the effect
activity and diet may have had on the osteophyte formation (since that type of antemortem
information may not be available), we can control for sex by analyzing males and females
separately. Your construction of your hypothesis test needs to be done carefully so that co-
variates and how to handle them are considered. Later the critical analysis of the results
should examine whether any co-variates had an effect.
Null and Alternative Hypotheses
After you have the variables figured out, you will further divide your hypothesis into
null
and
alternative
hypotheses. The null hypothesis is stated in negative, or rather, conservative
terms and will be that any changes in the independent variable will not lead to changes in the
dependent variable (
Marder, 2011
). Essentially with the hypothesis test we are challenging
the conservative viewpoint, namely, that one variable has no effect on the other. The alterna-
tive hypothesis will be the opposite: that changes in the independent variable will lead to
changes in the dependent variable. With our example, the null hypothesis is that osteophyte
formation on the lumbar vertebrae is not age-related. The alternative hypothesis is that osteo-
phyte formation on the lumbar vertebrae is age-related.
You will either reject or fail to reject the null hypothesis. The term “fail to reject” in this sense
is another way of saying “accept”; however, we usually do not use the term “accept” because
that indicates some measure of complete truth, which may not necessarily be the case as other
unknown factors may be at play
d
or the hypothesis may be accepted in one situation given
certain variables but a slight variable change may change the outcome, even though the
hypothesis itself has not changed. In addition, keep in mind that failure to reject the null
hypothesis does not necessarily mean that the alternative hypothesis is the correct explana-
tion. Another alternative hypothesis may exist. Remember that you cannot show that some-
thing is always true, but you can show that it is not always true. The logic here is that there
may be some evidence yet undiscovered or a test yet undeveloped that would negate
previous results. For example, skeletal biologists used to infer that pits on the dorsal side
of the pubic symphysis, commonly seen in female adults, meant that a female had given
birth. Even though not everyone was in agreement about this in the past, today the consensus
is that further research is necessary given the fact that pitting can occur in males and there
may be other causal factors at play (
Ubelaker and De La Paz, 2012
). Refer to
Figure 2.1
for
a flowchart diagram of hypothesis testing.
Step 4: Methods Development/Research Design
Once you have formulated your hypothesis and question (or determined what type of
descriptive analysis you will undertake), it is time to set up your methods and research
design. This is a critical step. Descriptive analysis is primarily used for paleopathology,
trauma, or taphonomy case studies (see Smith [Chapter 7]; Kroman and Symes [Chapter
8]; and Marden et al. [Chapter 9], this volume). However, while case studies are valuable
for describing novel findings, you should focus on a project that involves hypothesis testing
for your thesis or dissertation so that you hone all of the associated research skills. Based on
your library research and conversations with your advisor, you have probably already
