1

Is it possible to have a IV that is not correlated significantly with the DV but then in the multiple regression analysis this IV explain a significantly % of the unique variance of this DV? If so, how can we explain it?

gung - Reinstate Monica
  • 132,789
  • 81
  • 357
  • 650
Audrey
  • 11
  • 1
  • Yes it's possible. The most basic explanation is that the extra variable absorbs residual variance & increases power (see: [How can adding a 2nd IV make the 1st IV significant?](http://stats.stackexchange.com/q/28474/7290)). It is also possible the variable is a suppressor (see: [Suppression effect in regression: definition and visual explanation/depiction](http://stats.stackexchange.com/q/73869/7290), & [X and Y are not correlated, but X is significant predictor of Y in multiple regression. What does it mean?](http://stats.stackexchange.com/a/34016/7290)). – gung - Reinstate Monica Feb 26 '16 at 21:41
  • I think you will find the information you need in the linked thread. Please read it. If it isn't what you want / you still have a question afterwards, come back here & edit your question to state what you learned & what you still need to know. Then we can provide the information you need without just duplicating material elsewhere that already didn't help you. – gung - Reinstate Monica Feb 26 '16 at 21:42
  • Is it right to say that for testing the suppressor effect of a IV, this one must be significantly correlated to the other IV? – Audrey Feb 27 '16 at 18:09
  • It must be correlated, but it may or may not be 'significant'. (I suppose for *testing* you want it to be significant, of course.) – gung - Reinstate Monica Feb 27 '16 at 18:43

0 Answers0