In general, when should a nonparametric technique be used instead ...

[AEWBW]

In general, when should a nonparametric technique be used instead of its parameter counterpart?

Question 2

The chapter lists two major disadvantages to nonparametric testing. What are they?

[rleezy04]

Answer to Question 1

A nonparametric test should be used instead of its parametric counterpart whenever: (a) data are of the nominal or ordinal scales of measurement, or (b) data are of the interval or ratio scales of measurement but one or more other assumptions, such as the normality of the underlying population distribution, have not been met.

Answer to Question 2

a) Compared to parametric test, information in the data is used less efficiently, and the power of the test will be lower. For this reason, a parametric test is preferable whenever its assumptions have been met. b) Nonparametric testing places greater reliance on statistical tables, if a computer statistical package or spreadsheet is not being used.

[AEWBW]

Gracias!

[scottmt]

TYSM