TRANSFORMING VARIABLE FOR ANCOVA

Home Forums Methodspace discussion TRANSFORMING VARIABLE FOR ANCOVA

Viewing 1 post (of 1 total)
  • Author
    Posts
  • #2091
    semi0sane
    Member

    I have a question about using transformed data in an Ancova (and similar methods). I have Two groups (Patient vs. controls) and I have six experimental tasks. I would like to do a mixed factorial Ancova (controlling for age, between factor: Group; within-grouops facto: Task). Two of the task distributions are non-normal and I have successfully SQRT transformed them. My question is, do I have to transform the remaining 4 other Task DV distributions for Ancova comparisions?  My pitiful logic says yes, but I have been assured by other much brighter than me, that this is not neccessary?! Please help, anyone 🙂

Viewing 1 post (of 1 total)
  • You must be logged in to reply to this topic.