Yes, it was my typo!  I got the same results too.    And my real data is like NEXJ SELK YVRC SFOA (not real codes, but something like these).. ... View more

My real variable names are 4-character codes like AAAA instead of country names.  When I submit the program, I do not get any errors.  Simply the country in the results is empty like this (please see below)  while country for ID1 should be CCCC JJJJ, country for ID2 should be KKKK JJJJ, and country for ID3  should be 'space'.  The codes are all capital letters and separated by one space if there are multiple codes.  In reality there are only 3 codes I want to keep just like my country name example, and there are about 50 codes that I want to replace with space. Lastly, my data is huge.  Not sure if these matter..   Obs ID country keepers1 keepers2 keepers3 original 1     1                KKKK      CCCC      JJJJ        CCCC UUUU JJJJ 2     2                KKKK      CCCC      JJJJ        KKKK CCCC 3     3                KKKK      CCCC      JJJJ        DDDD   Yoko ... View more

My real variable has 4 alphabets (e.g., AAAA) instead of country names.  When I submit the program, there is o error message. Simply, the variable 'country' is emply like this:    Obs ID country keepers1 keepers2 keepers3 original 1     1                JJJJ         CCCC     UUUU JJJJ 2 2 KKKK CCCC KKKK CCCC JJJJ KKKK CCCC 3 3 KKKK CCCC JJJJ DDDD ... View more

Thank you for your suggestion.  It worked!  I also tried slightly differently to be similar to what I actually have to do.   This worked too (please see below).  But, when I tried with my real data, the variable equivalent to 'country' was empty.  Do you have any ideas regarding what I should be careful or should pay attention when applying this to a real data set?    #### DATA myData; input ID country $40.; DATALINES; 1 China USA Japan 2 Korea China 3 Canada ; run;   data keeps(keep=country rename=(country=del)); input country $ 1-20; datalines; Korea China Japan ; run ;   data newData(drop=del); set myData; array keepers(3) $; retain keepers; if _n_ eq 1 then do i = 1 to n; set keeps point=i nobs=n; keepers(i)=del; end; original=country; i=1; do while(scan(original,i,' ') ne ''); if scan(original,i,' ') not in keepers then country = tranwrd(country, scan(original,i,' '), ''); i+1; end; run; proc print; run; ###   Thank you!   Yoko ... View more

Hello,    Thank you for the program. I run it and I got the same result as the one using the 'dropping' method.    But, I realized that the data set called 'keeps' has only Korea and China.  But, Japan was still kept as well.  The program is a bit more complicated for me to tell why Japan was still kept.  (This is what I want, though) Would you mind explaining this?    Thank you,    Yoko   ... View more

Thank you for taking time to explain the program to me.  I now understand your method.  I also tried the program based on your explanation (I copied what I wanted tried based on your explanation below)       data countries(keep=country rename=(country=del)); input country $ 1-20; datalines; USA Canada ; data x(drop=del); input country $1-40; do i = 1 to n; set countries point=i nobs=n; country = tranwrd(country, trim(del), ' '); end; datalines; China USA Japan Korea China Canada ; run; proc print; run;    Obs country 1 China Japan 2 Korea China 3   This was what I wanted to see.      Thank you!   Yoko       ... View more

Hello,    Thank you for your reply.  Let me write what I understand.    You first created a data set which contains only country names.  They were supposed to be country names you want to replace with space in the second data set.    Sorry, but I could not understand what you did after that.  Do you mind explaining it please? It looks that you created a different data set using a data set, 'countries'.    Thank you,    Yoko ... View more

Thank you, it worked!   Yoko ... View more