Page 36 - Statistics II for Dummies

20 Part I: Tackling Data Analysis and Model-Building Basics

Nonparametrics

Nonparametrics is an entire area of statistics that provides analysis tech-
niques to use when the conditions for the more traditional and commonly
used methods aren’t met. However, people sometimes forget or don’t bother
to check those conditions, and if the conditions are actually not met, the
entire analysis goes out the window, and the conclusions go along with it!

Suppose you’re trying to test a hypothesis about a population mean. The
most common approach to use in this situation is a t-test. However, to use
a t-test, the data needs to be collected from a population that has a normal
distribution (that is, it has to have a bell-shaped curve). You collect data
and graph it, and you find that it doesn’t have a normal distribution; it has a
skewed distribution. You’re stuck — you can’t use the common hypothesis
test procedures you know and love (at least, you shouldn’t use them).

This is where nonparametric procedures come in. Nonparametric procedures
don’t require nearly as many conditions be met as the regular parametric
procedures do. In this situation of skewed data, it makes sense to run a
hypothesis test for the median rather than the mean anyway, and plenty of
nonparametric procedures exist for doing so.
If the conditions aren’t met for a data-analysis procedure that you want to
do, chances are that an equivalent nonparametric procedure is waiting in the
wings. Most statistical software packages can do them just as easily as the
regular (parametric) procedures.

Before doing a data analysis, statistical software packages don’t automatically
check conditions. It’s up to you to check any and all appropriate conditions
and, if they’re seriously violated, to take another course of action. Many times
a nonparametric procedure is just the ticket. For much more information on
different nonparametric procedures, see Chapters 16 through 19.

05_466469-ch01.indd 20 7/24/09 9:30:48 AM