Reassertion of the paradoxical power of tests of significance based on unreliable difference scores

Reassertion of the paradoxical power of tests of significance based on unreliable difference scores

Points out that the negative relationship between power of tests of significance and the reliability of difference scores involves a continuous function in which zero reliability is the most extreme case. It is not necessary to assume zero true interaction in order to recognize a paradoxical increas...

Full description

Saved in:
Bibliographic Details
Published in:Psychological bulletin Vol. 83; no. 5; pp. 776 - 777
Main Authors: Overall, John E, Woodward, J. Arthur
Format: Journal Article
Language:English
Published: Washington, etc American Psychological Association 01-09-1976
American Psychological Association, etc
Subjects:
Human
Statistical Measurement
Statistical Reliability
Statistical Significance
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Points out that the negative relationship between power of tests of significance and the reliability of difference scores involves a continuous function in which zero reliability is the most extreme case. It is not necessary to assume zero true interaction in order to recognize a paradoxical increase in likelihood of significant treatment effects to be associated with low reliability of difference scores on which all calculations are based.
ISSN:0033-2909
1939-1455
DOI:10.1037/0033-2909.83.5.776