Can I do a regression

Home Forums Methodspace discussion Can I do a regression

Viewing 2 posts - 1 through 2 (of 2 total)
  • Author
    Posts
  • #1803
    chella
    Participant

    Hello Experts,

    Can you please let me know if I can do a regression test for the sample size 40 with 8 predictor variables?

    #1804
    Dave Collingridge
    Participant

    Typically it is best to have at least 20-30 participants for each predictor variable. In your case you have too many predictors relative to the number of cases and so you are likely to overfit the model which means that you are modeling random error. Some people will do this when *exploring* relationships, but in such cases it is probably better to use correlation rather than regression.

Viewing 2 posts - 1 through 2 (of 2 total)
  • You must be logged in to reply to this topic.