Page 291 - Statistics II for Dummies

Chapter 16

Going Nonparametric

In This Chapter

▶ Understanding the need for nonparametric techniques
▶ Distinguishing regular methods from nonparametric methods
▶ Laying the groundwork: The basics of nonparametric statistics

any researchers do analyses involving hypothesis tests, confidence
Mintervals, Chi-square tests, regression, and ANOVA. But nonparametric
statistics doesn’t seem to gain the same popularity as the other methods. It’s
more in the background — an unsung hero, if you will. However, nonparametric
statistics is, in fact, a very important and very useful area of statistics because
it gives you accurate results when other, more common methods fail.

In this chapter, you see the importance of nonparametric techniques and
why they should have a prominent place in your data-analysis toolbox.
You also discover some of the basic terms and techniques involved with
nonparametric statistics.

Arguing for Nonparametric Statistics

Nonparametric statistics plays an important role in the world of data analysis
in that it can save the day when you can’t use other methods. The problem is
that researchers often disregard, or don’t even know about, nonparametric
techniques and don’t use them when they should. In that case, you never
know what kind of results you get; what you do know is they could very well
be wrong.

In the following sections, you see the advantages and the flexibility of using a
nonparametric procedure. You also find out just how minimal the downside
is, which makes it a win-win situation most of the time.

24_466469-ch16.indd 275 7/24/09 9:53:54 AM