+ Reply to Thread
Results 1 to 4 of 4

Thread: Multiple regression

  1. #1
    Points: 18, Level: 1
    Level completed: 35%, Points required for next Level: 32

    Posts
    2
    Thanks
    1
    Thanked 0 Times in 0 Posts

    Multiple regression




    Is there any way you can control for a variable using multiple regression? In order to see if the relationship reduces or disappears between an Iv and DV?
    Last edited by LLH1994; 01-04-2017 at 03:53 PM.

  2. #2
    Omega Contributor
    Points: 38,432, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    7,006
    Thanks
    398
    Thanked 1,186 Times in 1,147 Posts

    Re: Multiple regression

    Yes, this sounds like a case for multiple regression.


    Is it thought that they both explain the same variability in the dependent variable? Or is it that they are both on the same causal path. And if so, does the further up stream variable have its own direct effects on the dependent variable or is it mediated through the intermediate variable. If it is this later scenario, you may need to run mediation analysis to parcel out the direct and indirect effects of the variables. Which this includes running a couple of regression models.


    Lastly, you sound like you don't know how to do multiple logistic regression. What is your background and confidence level? Also, how are all of these variables formatted, are they categorical variable, continuous, or ordinal?


    Thanks!
    Stop cowardice, ban guns!

  3. The Following User Says Thank You to hlsmith For This Useful Post:

    LLH1994 (01-03-2017)

  4. #3
    Points: 18, Level: 1
    Level completed: 35%, Points required for next Level: 32

    Posts
    2
    Thanks
    1
    Thanked 0 Times in 0 Posts

    Re: Multiple regression

    Yes because the question suggests that when we control for it
    Last edited by LLH1994; 01-04-2017 at 03:53 PM.

  5. #4
    Omega Contributor
    Points: 38,432, Level: 100
    Level completed: 0%, Points required for next Level: 0
    hlsmith's Avatar
    Location
    Not Ames, IA
    Posts
    7,006
    Thanks
    398
    Thanked 1,186 Times in 1,147 Posts

    Re: Multiple regression


    If they are both in the model, you are controlling for them. One thing that some people neglect to do, is that if you have an interaction term in the model you still need to keep the two base terms in the model as well. So if you have: y-hat = Bo + B1(X1) + B2(X2) + B3(X1*X2) model you should be good. I don't work with many continuous variables, but I believe it is good practice to center your continuous predictors when you have an interaction term in the model (and some may say to do it all the time).


    Feel free to upload your outputted results to better help us understand what you have done and what the results looked like.


    Thanks.
    Stop cowardice, ban guns!

+ Reply to Thread

           




Tags for this 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