"Bi-Modal" Distribution
If
we were to sample a population of high school seniors on hand grip
strength, we would probably get a distribution that looks like the
following:

Frequency
Distribution Diagram showing a bi-modal distribution indicating two separate underlying populations.
This
distribution is typical of a "bi-modal" distribution and
indicates there are really two separate underlying populations and
not one. In this instance, the two populations are: male and female
with different distributions for each. As many statistical tests
are valid only for normal (uni-modal) distributions, it is important
to test our data distributions before proceeding further with statistical
assessment and analysis or risk erroneous conclusions.
Understanding
individual differences and the distribution of human variability
through rigorous statistical testing are "core" concepts
within Human Factors. To illustrate how these concepts are
applied, we will next address anthropometrics.
|