Hi You have called the proposed tests post-hoc, which implies that you did not expect/predict an interaction between the two factors? If you DID predict an interaction, then you would be justified to do the tests proposed by Stephen B., arguably without any correction and without the interaction itself being significant. That is, the tests would simply be simple effects tests.
The situation is somewhat more complicated if you did not predict/expect an interaction. You could do the tests as proposed by Stephen B., but only if the interaction is significant (or approaches significance?? ... the latter qualification because tests of interaction are very insensitive except for pure cross-over interactions). In general you are better off if you have predicted the pattern of results (and that pattern occurs, of course). That gives more freedom to the kind of tests that are warranted. Take care Jim James M. Clark Professor of Psychology 204-786-9757 204-774-4134 Fax [email protected] Department of Psychology University of Winnipeg Winnipeg, Manitoba R3B 2E9 CANADA >>> <[email protected]> 07-Jan-09 4:02 PM >>> (note change of subject header: cross-cultural scientific screw-up is not what this is about, for sure) On 7 Jan 2009 at 14:54, Steven Specht wrote: > What are TIPSters views of various post hoc tests after doing > a 2 X 2 ANOVA with repeated measures on one of the variables. > Tukey's HSD isn't really appropriate as it would adjust for all four > comparisons when I am only interested in comparing across the repeated > > measures variable (that is, a total of two comparisons rather than \> four). I'd go with two separate paired t-tests, with a Bonferroni correction (instead of testing at p = .05, do it at p = .025). Easy, quick, conservative. Stephen ----------------------------------------------------------------- Stephen L. Black, Ph.D. Professor of Psychology, Emeritus Bishop's University e-mail: [email protected] 2600 College St. Sherbrooke QC J1M 1Z7 Canada Subscribe to discussion list (TIPS) for the teaching of psychology at http://flightline.highline.edu/sfrantz/tips/ ----------------------------------------------------------------------- --- To make changes to your subscription contact: Bill Southerly ([email protected]) --- To make changes to your subscription contact: Bill Southerly ([email protected])
