I should investigate the relationship between the time-course of a variable (assessed by means of growth curve analysis) and the treatment outcome in a sample of about 69 therapies. I was wondering if I should rather (1) create a good vs. poor outcome groups based on the outcome questionnaire scores I have, or (2) to stay with the values of the questionnaire and to insert directly there into the analysis.

Thanks a lot!

More Omar Carlo Gioacchino Gelo's questions See All
Similar questions and discussions