differences in logistic regression output from SPSS 22 and Modeler 16

[Baglivio, Michael]

Hello,

I am new to using Modeler 16 and in trying to get accustomed to it am re-running some old logistic regression analyses. I cannot seem to figure out how to get the Modeler 16 output to mimic the SPSS 22 default output I am used to. In Modeler on the “variables in the equation” section of the logistic regression output, I am getting B, S.E., Wald, etc. for each option of a given covariate, while in SPSS 22 I get one line of output for each covariate (not each option of each covariate).

 

For example, for a variable called “risk” with 4 response options, Modeler will have:

                                B             SE           Wald      df            Sig.         Exp(B)

Risk                                                        36.206   3              .000                       

Risk(1)                  -.182      .046        15.87     1              .000        .834

Risk(2)                  -.151      .052        8.598     1              .003        .86

Risk(3)                  .030        .05          .367        1              .545        1.031

 

While SPSS 22 will have:

                                B             SE           Wald      df            Sig.         Exp(B)

Risk                        .095        .013        49.74     1              .000        1.099                    

 

I understand that the output in Modeler provides some more information (such as the 3^rd option of risk is not sig different from the 4^th), but I would still like to figure out how to get the outputs to match.

Thank you for any insight.

mike

 

 

[David Marso]

"SPSS 22 I get one line of output for each covariate (not each option of each covariate). "
Perhaps it is NOT seeing it as categorical?
I don't have Advanced Stats on this box so can't test that idea.
You should perhaps post your syntax next time around rather than have us try to guess what you are running?
--

Baglivio, Michael wrote

Hello,
I am new to using Modeler 16 and in trying to get accustomed to it am re-running some old logistic regression analyses. I cannot seem to figure out how to get the Modeler 16 output to mimic the SPSS 22 default output I am used to. In Modeler on the "variables in the equation" section of the logistic regression output, I am getting B, S.E., Wald, etc. for each option of a given covariate, while in SPSS 22 I get one line of output for each covariate (not each option of each covariate).

For example, for a variable called "risk" with 4 response options, Modeler will have:
                                B             SE           Wald      df            Sig.         Exp(B)
Risk                                                        36.206   3              .000
Risk(1)                  -.182      .046        15.87     1              .000        .834
Risk(2)                  -.151      .052        8.598     1              .003        .86
Risk(3)                  .030        .05          .367        1              .545        1.031

While SPSS 22 will have:
                                B             SE           Wald      df            Sig.         Exp(B)
Risk                        .095        .013        49.74     1              .000        1.099

I understand that the output in Modeler provides some more information (such as the 3rd option of risk is not sig different from the 4th), but I would still like to figure out how to get the outputs to match.
Thank you for any insight.
mike

[Jon K Peck]

In Statistics you are treating Risk as a continuous covariate while in Modeler you are treating it as categorical.   If you want Modeler to treat it as continuous, set the measurement level to continuous.


Jon Peck (no "h") aka Kim
Senior Software Engineer, IBM
[hidden email]
phone: [phone]




From:        "Baglivio, Michael" <[hidden email]>
To:        [hidden email],
Date:        02/10/2014 08:50 AM
Subject:        [SPSSX-L] differences in logistic regression output from SPSS 22              and              Modeler 16
Sent by:        "SPSSX(r) Discussion" <[hidden email]>

---

Hello,
I am new to using Modeler 16 and in trying to get accustomed to it am re-running some old logistic regression analyses. I cannot seem to figure out how to get the Modeler 16 output to mimic the SPSS 22 default output I am used to. In Modeler on the “variables in the equation” section of the logistic regression output, I am getting B, S.E., Wald, etc. for each option of a given covariate, while in SPSS 22 I get one line of output for each covariate (not each option of each covariate).
 
For example, for a variable called “risk” with 4 response options, Modeler will have:
                                B             SE           Wald      df            Sig.         Exp(B)
Risk                                                        36.206   3              .000                        
Risk(1)                  -.182      .046        15.87     1              .000        .834
Risk(2)                  -.151      .052        8.598     1              .003        .86
Risk(3)                  .030        .05          .367        1              .545        1.031
 
While SPSS 22 will have:
                                B             SE           Wald      df            Sig.         Exp(B)
Risk                        .095        .013        49.74     1              .000        1.099                    
 
I understand that the output in Modeler provides some more information (such as the 3^rd option of risk is not sig different from the 4^th), but I would still like to figure out how to get the outputs to match.
Thank you for any insight.
mike