Create Combined data set with variable based on original table

[jmmedina252]

Hello, I need to combine about ten tables all with different names (example: work.weight, work.diagnosis, work.dose) into one table with a variable that specifies their original table name.

[PaigeMiller]

---

@jmmedina252 wrote:

Hello, I need to combine about ten tables all with different names (example: work.weight, work.diagnosis, work.dose) into one table with a variable that specifies their original table name.

---

Could you define "combine"?? Do you want these tables to be combined horizontally, side by side? Or do you want these tables to be combined vertically, one on top of another? Or do you want something else?

[jmmedina252]

stacked

[Reeza]

Then try the first set of code I included.

[jmmedina252]

is there anyway to exclude the libname without saving the data sets to work? instead of source=data.dose I would want source=dose

[Reeza]

Use SCAN() to parse just the second portion.

source =scan(src, 2, ".");

[Reeza · Accepted solution]

Are you stacking the data or merging the data (adding variables)?

data want;
set weight diagnosis dose indsname=src;
source = src;
run;

 Merge (one option)

data want;
merge weight (in=t1) diagnosis (in=t2) dose (in=t3);
by subjid;
src=catx('|', t1, t2, t3);
run;

You likely need to provide more details to get an answer that's more specific to what you need.

[jmmedina252]

I would want to stack the data sets