[Sakura-Chan]

Okay so, I know how to run the ANCOVA but I'm kinda struggling with how to actually write about it for my stats exam.

EDIT: Just to clarify, it's not for a results section of a lab report or anything like that. For the exam we just have to ascertain which test needs to be used from an introductory paragraph, run the test and then write up 1) why we've chosen the test, 2) how we ran the analysis including SPSS functions, which I'm fine with, 3) what we found, and conclude.

So far, we've been told we have to have:

• Intro - details about the actual study, need to state why the test you've chosen is suitable etc.

I'm pretty sure we have to include:

• Testing the homogeneity of regression assumption, report interaction effect between IV and covariate (run on custom model)
• If there's no significant effect, run the analysis (run on full factorial)

...And then I get a bit stuck. It's been so long since we covered this in lectures, I've totally forgotten.

Does this sound right for the next things to do?

• Look at covariate and DV and see if there is a significant correlation? And report it to see if it was worthwhile including the covariate?
• Write about the effect of the IV on the DV once the covariate has been eliminated.

In my notes I've written that I need to include a table of adjusted means and standard error to compare to the non-adjusted descriptive stats, and an ANCOVA summary table, followed by conclusions about the effect of the IV on the DV once the covariate has been partialled out, and overall conclusions about the IV.

We've been told that we need to use Bonferroni correction in the analysis, but what do you actually say about that when you're writing it up?

I hope that actually makes sense! I get the feeling I'm missing something/doing something wrong.

[Sakura-Chan]

Anyone?