BookmarkSubscribeRSS Feed
Mdormond
Calcite | Level 5

Is it possible to take three dummy variables and recode them into one categorical variable with 4 different categories?

5 REPLIES 5
ballardw
Super User

Yes.

Simple example is 3 variables coded 1 for yes (the dummy value of interest).

ina a data step

data want;

set have;

If var1=1 then Cat=1;

else if var2=1 then Cat=2;

else if var3=1 the cat=3;

else Cat=4;
run;

Jagadishkatam
Amethyst | Level 16

From the above code, since we are using the same cat variable for categorizing the var1,var2,var3 data, the final output will show the cat=3. per the code everytime the if then else condition executes and if the var1,var2 and var3 values are same , the cat value will be replaced or overwritten with 1, 2 and 3. but the final output will show 3.

the above code will work only when the three variables will have different values per row.

Thanks,

jag

Thanks,
Jag
Mdormond
Calcite | Level 5

Is there any way to code it so the final output does not show all 3's? I would like each category to display a unique value.

PGStats
Opal | Level 21

You could use

cat = whichn(1, var1, var2, var3);

PG

PG
ballardw
Super User

Since you didn't provide any example of your dummy variables or data layout I had to guess. If multiples of your dummy variables are 'Yes' then you need to describe the rules for which result you want. If the dummy variables are coded 0/1 and ONLY one is set to yest the code will work as described. Incomplete descriptions=incomplete answers.

hackathon24-white-horiz.png

The 2025 SAS Hackathon has begun!

It's finally time to hack! Remember to visit the SAS Hacker's Hub regularly for news and updates.

Latest Updates

How to Concatenate Values

Learn how use the CAT functions in SAS to join values from multiple variables into a single value.

Find more tutorials on the SAS Users YouTube channel.

SAS Training: Just a Click Away

 Ready to level-up your skills? Choose your own adventure.

Browse our catalog!

Discussion stats
  • 5 replies
  • 1542 views
  • 0 likes
  • 4 in conversation