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])

Reply via email to