Why Planned Comparisons Are Good

In general, planned comparisons are good for two distinct reasons. First of all, this kind of comparison is usually more powerful than is the omnibus F-test. In other words, Type II error risk is typically lower when planned comparisons are examined instead of the generic one-way ANOVA F-test. Second, well-trained researchers usually have justifiable hypotheses (gleaned from their own earlier studies and their knowledge of others' research findings), and this expertise shows through better in a set of planned comparisons than it does in the two-step strategy of doing a one-way ANOVA followed by a post hoc investigation.Therefore, give researchers a double set of bonus points when you see them discuss planned comparisons in their technical written reports or in their oral presentations.

(From Chapter 12 in the 6th edition, p. 269)

Copyright © 2012

Schuyler W. Huck
All rights reserved.

| Book Info | Author Info |

Site URL: www.readingstats.com

Top | Site Map
Site Design: John W. Taylor V