# Thread: How to test for difference between 2 R-Squared in Binary Hierarchical Regression?

1. ## How to test for difference between 2 R-Squared in Binary Hierarchical Regression?

I ran a BINARY hierarchical regression in SPSS. However, I could not find an option to test change in R-Squared. So is there a way to test it in SPSS?

I found a formula to test change in R-squared.

F = [(R^2_Full - R^2_Reduced) / (the number of variables in mode 2 - the number of variables in mode 1) ] / [ (1-R^2_Full) / ( N - the number of variables in mode 2 - 1) ].

But how do I know whether the F value is significant or not?

Thanks a lot

2. ## Re: How to test for difference between 2 R-Squared in Binary Hierarchical Regression?

R-Square's interpretation and calculation in logistic regression is something of an enigma or at least up for debate. This is probably why you can't find an answer to you question. People typically don't do this and if they did the explanation of the value may be dubious.

Why do you want to do this?

You are probably better off looking at the Area Under the Curve values!

3. ## Re: How to test for difference between 2 R-Squared in Binary Hierarchical Regression?

Thanks for your reply. I am investigating the effect of task complexity on decision (yes or no), controlling for the effect of self-efficacy. So I want to know whether the task complexity can moderate the effect of self-efficacy on decision. Is there a better way to test it?

4. ## Re: How to test for difference between 2 R-Squared in Binary Hierarchical Regression?

Are all variables binary? If so, you can include the multiplicative interaction term of the two predictor variables.

 Tweet

#### Posting Permissions

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