Status:Closed    Asked:Jan 03, 2018 - 08:27 AM

SDA Regression Analysis

Jeff -- thanks for the response earlier. I'm able to see the recoded questions within the SDA by clicking "list created variables."

My issue is with the analysis. When I recode a variable and it is a categorical/factor, why does SDA treat it as numeric in the analysis? I need an Odds Ratio and CIs created for each level of the factor in the statistical output. Any help here?

Thanks in advance.

Do you have the same question? Follow this Question

Staff Answer


Jeff Bloem


I believe the issue here is that SDA does not have the capability to run ordered probit/logit regression models. So, when you insert a categorical dependent variable in the program it automatically re-codes the variable to be a binary 0 or 1 variable. The SDA system is limited in a couple ways, and this seems to be one of them. If you want to perform analysis using a categorical variable I'd suggest extracting the microdata from the IPUMS online data extract system and using your preferred statistical software.


Jan 03, 2018 - 09:11 AM

Report it


NHIS SDA Regression Question
Health Insurance Unit in ACS versus MEPS
Login   |   Register

Recently Active Members

View More »

Share |