Quick and dirty if you are hady with a wordprocessing program. If you just want it for display crosstabs tables= sic_code by occ_code /cells=count/statistics=none. crosstabs tables= sic_code by occ_code /cells=column /statistics=none. then cut and paste into your favorite wordprocessing program. you probably want to use a fixed font. (you might want to do draft output depending on how your wordprocessing program works). If you want to make it a data file, set the table look under <options> <pivot table> to "report" so there isn't a lot to remove. "Patrick E. Burns" wrote:

> Dear SPSS'ers, > > I have two very long columns of data that I am seeking to > restructure/aggregate into a matrix of sorts. The data looks like this: > > "sic_code" "occ_code" > 0111 111140 > 0111 111140 > 0111 111140 > 0112 111140 > 0112 111140 > 0112 111140 > 0112 111160 > 0112 111160 > 0112 111160 > 0112 111160 > 0112 111160 > 0112 111160 > 0115 111110 > 0115 111110 > 0115 111150 > 0115 111150 > 0115 111150 > 0115 111150 > 0115 111150 > 0115 111150 > 0116 111110 > 0119 111120 > 0119 111120 > 0119 111120 > > I would like to create two databases from this original data set, one > capturing the raw cross-tab counts between the two variables, and another > capturing the column frequency percentages. They would look like this: > > sic_code 111110 111120 111140 111150 111160 > 0111 0 0 3 0 > 0 > 0112 0 0 3 0 > 6 > 0115 2 0 0 6 > 0 > 0116 1 0 0 0 > 0 > 0119 0 3 0 0 > 0 > > sic_code 111110 111120 111140 111150 111160 > 0111 0 0 .5 0 > 0 > 0112 0 0 .5 0 > 1.0 > 0115 .66 0 0 1.0 > 0 > 0116 .33 0 0 0 > 0 > 0119 0 1.0 0 0 > 0 > > What is the best way to go about this aggregation/restructuring of a database? > > PATRICK > > Patrick Burns, Senior Researcher > Economic Roundtable > Los Angeles > Email: patrickburns@economicrt.org > http://www.economicrt.org