Combining multiple variables into new dummy variable?

[smallembo]

Hello,

I'm a relatively new SAS user so I'm not sure how to do this but it seems like it should be possible.

I have variables for all 50 states and want to split them into a new variable depending on whether or not they have expanded Medicaid coverage or not. Then, I want to be able to make the Medicaid variable into a binary 0/1 in order to run a diff-in-diff model. Is that possible? In my head it would look something like:

ACAyes = (Arizona, Arkansas, California);  

ACAno = (Alabama, Florida, Georgia);

Medicaid = 0;

if ACAyes = 1 then Medicaid = 1;

if ACAno = 1 then Medicaid = 0;

Thanks for your help!

[PeterClemmensen]

So just to be clear, you have variables named Arizona, Arkansas, California and so on, correct? What do those variables contain?

[smallembo]

Yes, I should have clarified that! I created each variable with a state name from an original variable, _STATE, which coded each state according to a number. So, now that you point that out, I should probably go back to the original variable to make the Medicaid variable. Perhaps something like:

Medicaid = 0;

if _STATE = (1, 2, 3) then Medicaid = 1;

if _STATE = (4, 5, 6) then Medicaid = 0;

[Shmuel]

Coding: if _STATE = (1, 2, 3) is wrong sas syntax. what you mean is:

if _STATE in (1  2  3) then Medicaid = 1; else Medicaid = 0;

assuming that (1 2 3) are the only those with Medicaid = 1;

[smallembo]

Ah, thank you for pointing out my syntax. As far as the if/else statement, there are a few states and territories in the original _STATE variable that I don’t want to be in either grouping and want them to be disregarded for the diff-in-diff model. How would you suggest I code the statement so that the states I want to disregard don’t wind up in the model? Would it work to sssign them to another number?

[PaigeMiller]

As far as the if/else statement, there are a few states and territories in the original _STATE variable that I don’t want to be in either grouping and want them to be disregarded for the diff-in-diff model.

In your data step or PROC, if for example you don't want states 5 and 27 in the analysis, then

where state not in (5,27);

[smallembo]

Thank you! Can I ask one more question? I have tried this code:

data newCOMBO where _STATE not in (2 22 23 49 51 66 72 78);
set COMBO;
Medicaid = 0;
if _STATE = (4 5 6 8 9 10 11 15 17 18 19 21 24 25 26 27 30 31 32 33 34 35 36 38 39 41 42 44 50 53 54) then Medicaid = 1;

run;

But keep running into these errors:

WARNING: The data set WORK.NEWCOMBO may be incomplete. When this step was stopped there were 0
observations and 160 variables.
WARNING: The data set WORK.WHERE may be incomplete. When this step was stopped there were 0
observations and 160 variables.
WARNING: Data set WORK.WHERE was not replaced because this step was stopped.
WARNING: The data set WORK._STATE may be incomplete. When this step was stopped there were 0
observations and 160 variables.
WARNING: Data set WORK._STATE was not replaced because this step was stopped.
WARNING: The data set WORK.NOT may be incomplete. When this step was stopped there were 0
observations and 160 variables.
WARNING: Data set WORK.NOT was not replaced because this step was stopped.

Everything works fine up until this point so I'm thinking this step is definitely written wrong. 

[PaigeMiller]

if _STATE in (4 5 6 8 9 10 11 15 17 18 19 21 24 25 26 27 
    30 31 32 33 34 35 36 38 39 41 42 44 50 53 54) then Medicaid = 1;

[smallembo]

Ack, thank you!!