Dealing with assumptions underlying statistical tests
Authors:
- Craig S. WellsCWCraig S. Wells
- John M. HintzeJHJohn M. Hintze
- 3 April 2007
- journal article
- research article
- Published by Wiley in Psychology in the Schools
- Vol. 44 (5) , 495-502
Abstract
The validity of a hypothesis test is partly determined by whether the assumptions underlying the test are satisfied. In the past, a preliminary analysis of the data has been suggested prior to the use of the statistical test. In this article, the authors describe several limitations of preliminary tests (e.g., influence on significance levels). Another strategy that uses theoretical knowledge in conjunction with prior empirical evidence and reason, prior to data collection, is described. © 2007 Wiley Periodicals, Inc. Psychol Schs 44: 495–502, 2007.Keywords
This publication has 15 references indexed in Scilit:
- Some Properties of Preliminary Tests of Equality of Variances in the Two-Sample Location ProblemThe Journal of General Psychology, 1996
- The earth is round (p < .05).American Psychologist, 1994
- Using significance tests to evaluate equivalence between two experimental groups.Psychological Bulletin, 1993
- The two-sample t test versus satterthwaite's approximate f testCommunications in Statistics - Theory and Methods, 1989
- The unicorn, the normal curve, and other improbable creatures.Psychological Bulletin, 1989
- Rationality in psychological research: The good-enough principle.American Psychologist, 1985
- On approximation of power of a test procedure based on preliminary tests of significanceCommunications in Statistics - Theory and Methods, 1981
- Testing the validity conditions of repeated measures F tests.Psychological Bulletin, 1980
- Hypothesis Testing Incorporating a Preliminary Test of SignificanceJournal of the American Statistical Association, 1970
- Analysis and Inference for Incompletely Specified Models Involving the Use of Preliminary Test(s) of SignificanceBiometrics, 1964