How do I recode this independent, categorical variable

  • Context: Undergrad 
  • Thread starter Thread starter cachemony
  • Start date Start date
  • Tags Tags
    Independent Variable
Join the discussion
Ask a follow-up here, or get your own question answered by working scientists, mathematicians and engineers — people, not an autocomplete.
Real named experts · corrections over time · the nuance an AI answer skips
4 replies · 6K views
cachemony
Messages
2
Reaction score
0
I’m student at the University of Arizona. In my political science class, I need to use SPSS to calculate bivariate correlation between a categorical independent variable (Presidential or Parliamentary system) and a continuous dependent variable (various measures of democracy)

I’m using the Przeworski data set. My original independent variable in this set is called INST, which designates each state as a dictatorship 0, parliament democracy 1, mixed democracy 2, and a presidential democracy 3.

My question is: How do I recode this independent, categorical variable (INST) into two different variables (Presidential and Parliamentary), so it excludes the mixed democracies and dictatorships and allow me to run a correlation with my continuous dependent variables? So, I want to be able to recode INST into Presidential and Parliament variables and run them separately to see how each one correlates with my continuous dependent variables?



Thanks,
Noah
 
Physics news on Phys.org
If I understand this correctly, you do not want to code INST into 2 different variables. You want a single categorical variable, say, PRES which is = 1 if INST = 3, and PRES = 0 if INST = 1. Before doing this, however, you should exclude all observations where INST = 0 or 2.
 
I want to recode INST so I can run correlations of Presidential and Parliment systems separate from each other against my dependent variables?

Thanks,
Noah
 
If you exclude INST = 0 or 2, you cannot define two independent variables, e.g. PRES and PARL, because they will be perfectly correlated with each other: PRES + PARL = 1 or PARL = 1 - PRES.

OTOH, perhaps you are not aiming to exclude the 0's and the 2's; in which case you can define two independent variables PRES and PARL.
 
Check out www.youtube.com/mycsula for our newly completed series of online SPSS (now IBM Statistics) video tutorials.

Learn how to: define variables, enter data, run a paired samples t test, perform frequency analysis, use scripting and syntax files, create and edit charts and graphs, and much more.

Training modules cover a wide range of topics aimed at beginners to more advanced users. All videos include examples and instructions to help you with whatever area of research or analysis you are working on.

You can find these videos at www.youtube.com/mycsula. Check them out and subscribe to our channel today!

Thanks,

The ITS Online Training team

www.youtube.com/mycsula