Ancova

[ez_e]

My DV is performance scores
MY IV is task allocation

Previous research has shown an association of EDUCATION on performace scores but i did not control for it- i randomly allocated particpants between two groups

I found education to be significantly correlated with performance scores-

Should I just be doing an independent t-test between my IV and DV and report the effects and then subsequenty report that EDUCATION is significantly associated with performance

Or should I be doing a ANCOVA where education is a covariate?? which will help explain alot of the variance in my DV-

It makes a big difference including education as a covariate- it results in my a sig main effect between my IV and DV- where without including it as a covariate there is no sig main effect between my IV and DV-

Assumptions aren't violated for ANCOVA- I don't think- but I just don't know if I am using it in the right context- scares me a bit because books say it is a complex analysis and is easily mis used to the detriment of producing reliable and interpretable statistics-

What to do???

Oh also i am conduction some Wilcoxons T- tests and SPSS spits out z scores yet in journals people report a a Wilcoxons T=??? Where do I get or work out this figure? (my sample is 60 people)

Any help asap will be greatley appreciated

Thanks

[JohnM]

Since EDUCATION is a characteristic inherent in the participants in the experiment, you can feel free to include it as a covariate.

The rule of thumb is that you need to have data on the covariate prior to assignment of the participants to treatment groups, so you're OK as far as I can tell.