+ Reply to Thread
Results 1 to 2 of 2

Thread: Comparing linear regression models: a chi-squared test to compare adjusted R-squares?

  1. #1
    Points: 14, Level: 1
    Level completed: 27%, Points required for next Level: 36

    Posts
    1
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Comparing linear regression models: a chi-squared test to compare adjusted R-squares?




    Hi all,

    I used linear regression to analyze the data of one of my experiments, and used 3 predictors in a first model, added 4 more predictors in a second model, and again 3 more in a third model. The second and third model are significant due to one of the added predictors in each model.

    Now I was asked by a reviewer to do a chi-squared test to check whether the change in adjusted R squares between the second and third model (from .08 to .19) is significantly different. I never heard of this test and am unable to contact him/her to ask for more details. I analyzed my data in SPSS and the only option I can find to assess the difference is the F change, which does show a significant F change between model 2 and 3. However, I am by far not an expert on regression analyses and donít know if using a chi-squared test to compare the goodness of fit would be a better way, or how to do such a test.

    I tried to google it, but could not find anything concrete about using the adjusted R squares to compare the models.
    I found you can calculate the chi-square of each model and then check if the difference is significant, but I got stuck at calculating the chi-squares of the models.

    How can I do this? What would you recommend?

    Any help would be much appreciated!
    Thank you for your time.

  2. #2
    Points: 657, Level: 13
    Level completed: 14%, Points required for next Level: 43

    Posts
    57
    Thanks
    1
    Thanked 9 Times in 8 Posts

    Re: Comparing linear regression models: a chi-squared test to compare adjusted R-squa


    As your models are nested, I would use the R-squared change F-test, this should be sufficient to compare models of fit.

+ 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