Dear Experts,
I would like to estimate influence of a score to a questionnaire on brain functioning using ANCOVA analysis.
First, can you let me know if I am correct.
I used 2 different options that seem to be identical, all within group:
- I ran a multiple regression analysis with my 1st level contrast of interest and the score to the questionnaire as the covariate (no mean centering);
- I ran a one sample t-test with my 1st level contrast of interest the same score as covariate (no mean centering)
And used the contrasts [0 1] and [0 -1] to evaluate (positive and negative, respectively) correlations between score at the questionnaire and brain functioning.
Is that right that these 2 analyses are identical? Is it the way to do this kind of analysis?
Now, I already have group comparisons analyses, with a certain threshold.
Should I use a more stringent threshold for this analysis? Or should I report my results with the same threshold than my group comparisons analyses?
Thanks for your help.
Cheers,
Yann
|