Extracting word

[Ivy]

Dear all, 

I would like to extract the part between end of  first word and before "vs" .   How I can do it ?

For example, I would like to get "Black or African American" this part out.

Thank you.

[nehalsanghvi]

Is this what you need?

WantString = substr(HaveString,1,indexw(string,'vs')-1);

Edited to include data step code to show how my statement works:

data Have;
length HaveString $100.;
HaveString='Black or African American vs White'; output;
HaveString='Hispanic or Latino vs White';output;
HaveString='Other Race        vs White';output;
;
run;

data Want;
set Have;
WantString = substr(HaveString,1,indexw(HaveString,'vs')-1);
run;

Editing again as I noticed that words such as 'Race' appear to be in the same column as the value for Race. In this case, this code will work:

data Have;
length HaveString $100.;
HaveString='Race   Black or African American vs White'; output;
HaveString='Race   Hispanic or Latino vs White';output;
HaveString='Race        Other Race        vs White';output;
;
run;

data Want;
set Have;
WantString = substr(HaveString, indexw(HaveString,scan(HaveString,2)), indexw(HaveString,'vs')-length(substr(HaveString,1,indexw(HaveString,scan(HaveString,2)))));
run;

[collinelliot]

I've only eyeballed the data since I could not copy it and test it, but I think something like this will work with the cases you've shown. 

data want;
have = 'Race Black or African American vs White';
want = catx(' ', scan(have, 1, ' '), scan(have, -1, 'vs'));
run;

[art297 · Accepted solution]

data want;
  set have;;
  CALL SCAN(string, 2, start, length);
  want = substr(string, start, index(string, 'vs')-start);
run;

Art, CEO, AnalystFinder.com

[nehalsanghvi]

I did the same thing in a very complicated manner, haha! Yours is elegant 🙂

[Ivy]

Thank you all very much for the disuccsion !

Ivy