+ Reply to Thread
Results 1 to 2 of 2

Thread: Insignificant correlation - significant regression

  1. #1
    Points: 19, Level: 1
    Level completed: 37%, Points required for next Level: 31

    Thanked 0 Times in 0 Posts

    Insignificant correlation - significant regression

    I couldn't find this on the search - maybe it's too basic

    So I have two variables which significantly positively correlate with each other (r = .35, p < .01). Both of these have insignificant correlations (p >.1) with a third variable (the first shows as weakly positive, the second as weakly negative).

    However, on a regression analysis, together the first two predictors show up as significant predictors of the third variable. They both have significant standardised coefficients individually as well (I am using SPSS coefficient table here). However, one has a positive standardised coefficient (.26) and the other a roughly equal negative one (-.24).

    I can see that this might have something to do with multicollinearity or suppressor effects, but I am a bit out of my depth at figuring out exactly what is going on? Could anyone explain in simple terms?

  2. #2
    Omega Contributor
    Points: 38,432, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Not Ames, IA
    Thanked 1,186 Times in 1,147 Posts

    Re: Insignificant correlation - significant regression

    Plot scatterplots for each combination of three variables, than a 4th scatterplot with a x, y, z axis and see what the data pattern looks like.
    Stop cowardice, ban guns!

+ Reply to Thread


Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts

Advertise on Talk Stats