How to keep same cases in hierarchical regression analysis

[Student073]

I'm working with SPSS to run a hierarchical regression analysis.

First, I introduced the control variables. Then, the independent variables and the moderator, and finally the interaction term.

The problem is that initially I start with N=3500 cases (only introducing the control variables). However, after including the independent ones I get N=2450...

How could I make sure in SPSS that I am always using the same cases in all the models?

Thanks a lot!

[Bruce Weaver]

This is untested (no SPSS on this computer), but I believe it will work.  For every model in the series...

MIXED Y BY {all factors in the final model} WITH {all covariates in the final model}
 /FIXED {only fixed effects in the current model}
 /RANDOM {only random effects in the current model, if any}
 etc...

If that does not work, for some reason, another way is to set a filter before running your series of models.  Something like:

COMPUTE @f = NMISS(comma-separated list of variables in the full model) EQ 0.
USE ALL.
FILTER by @f.

Then run your series of models.  When done, turn the filter off.

USE ALL.
FILTER OFF.


HTH.

Student073 wrote

I'm working with SPSS to run a hierarchical regression analysis.

First, I introduced the control variables. Then, the independent variables and the moderator, and finally the interaction term.

The problem is that initially I start with N=3500 cases (only introducing the control variables). However, after including the independent ones I get N=2450...

How could I make sure in SPSS that I am always using the same cases in all the models?

Thanks a lot!

[Art Kendall]

Did you put all of the terms into REGRESSION and then use a set of /enter specifications?
On 7/27/2013 4:43 PM, Student073 [via SPSSX Discussion] wrote:

I'm working with SPSS to run a hierarchical regression analysis.

First, I introduced the control variables. Then, the independent variables and the moderator, and finally the interaction term.

The problem is that initially I start with N=3500 cases (only introducing the control variables). However, after including the independent ones I get N=2450...

How could I make sure in SPSS that I am always using the same cases in all the models?

Thanks a lot!

---

If you reply to this email, your message will be added to the discussion below:

http://spssx-discussion.1045642.n5.nabble.com/How-to-keep-same-cases-in-hierarchical-regression-analysis-tp5721401.html

To start a new topic under SPSSX Discussion, email [hidden email]
To unsubscribe from SPSSX Discussion, click here.
NAML

[Bruce Weaver]

Oops...for some reason, I concluded upon first (and apparently not very careful reading) that the OP was using MIXED.  If they are using REGRESSION, then Art is spot on.  I apologize for any confusion this may have caused.

Art Kendall wrote

Did you put all of the
        terms into REGRESSION and then use a set of /enter specifications?
      Art Kendall
Social Research Consultants
      On 7/27/2013 4:43 PM, Student073 [via SPSSX Discussion] wrote:
   
     I'm working with SPSS to run a hierarchical
      regression analysis.
     
      First, I introduced the control variables. Then, the independent
      variables and the moderator, and finally the interaction term.
     
     
      The problem is that initially I start with N=3500 cases (only
      introducing the control variables). However, after including the
      independent ones I get N=2450...
     
     
      How could I make sure in SPSS that I am always using the same
      cases in all the models?
     
     
      Thanks a lot!
     
     
     
     
     
        If you reply to this email, your
          message will be added to the discussion below:
        http://spssx-discussion.1045642.n5.nabble.com/How-to-keep-same-cases-in-hierarchical-regression-analysis-tp5721401.html 
     
     
        To start a new topic under SPSSX Discussion, email
        [hidden email] 
        To unsubscribe from SPSSX Discussion, click
          here .
        NAML

[Student073]

Yes, I could solve the problem. Thanks!!