Testing Means
Prerequisites
Specified in individual
sections
- Single Mean
- t Distribution Demo
- Difference between
Two Means (Independent Groups)
- Robustness Simulation
- All Pairwise Comparisons
Among Means
- Specific
Comparisons
- Difference between
Two Means (Correlated Pairs)
- Correlated t
Simulation
- Specific Comparisons
(Correlated Observations)
- Pairwise
Comparisons (Correlated Observations)
- Exercises
- PDF Files (in .zip archive)
Many if not most experiments are
designed to compare means. The experiment may involve only one
sample mean that is to be compared to a specific value. Or the
experiment could be testing differences among many different
experimental conditions, and the experimenter could be interested
in comparing each mean with each other mean. This chapter covers
methods of comparing means in many different experimental situations.
The topics covered here in sections D, E, G, and
H are typically covered in other texts in a chapter on Analysis
of Variance. We prefer to cover them here since they bear no
necessary relationship to analysis of variance. As has been pointed
out elsewhere, it is not logical to consider the procedures
in this chapter tests to be performed subsequent to an analysis
of variance. Nor is it logical to call them post-hoc tests as
some computer programs do.
|
|