Hi there, I am working with publication data. Starting with a set of cited publications, I am trying to characterize the set of citing publications that cite the cited publications. I am trying to define the following variables. New Key Word: defined as new if a key word has never before appeared in the key word list of the cited article or previous citations to the cited article; old otherwise. Total New Key Words: total number of new keywords in citing article. DATA cited; INFILE CARDS; INPUT CITED_ID KW_1 $ KW_2 $ KW_3 $; CARDS; 1000 xx1 xx2 xx3 1001 aa1 aa2 aa3 ; run; DATA citing; INFILE CARDS; INPUT CITED_ID CITING_ID KW_1 $ KW_2 $ KW_3 $; CARDS; 1000 2000 xx1 xx2 1000 2001 yy1 xx2 yy2 1000 2002 xx3 yy1 yy2 1000 2003 yy2 xx2 zz1 1001 2004 aa1 aa2 xx1 1001 2005 aa1 bb1 bb2 1001 2006 aa1 bb2 cc1 ; run; Desired Data CITED_ID CITING_ID NEW_KW TOT_NEW_KW 1000 2000 0 0 1000 2001 1 2 1000 2002 0 0 1000 2003 1 1 1001 2004 1 1 1001 2005 1 2 1001 2006 1 1 Thanks, Gary
... View more