Hi, I am trying to achieve a specific sort order using custom sort for the variables Names and Codes. I have a sample variable with this order below: Names    Codes Boys        2795                  2796                  2900                  PHA                  Total Girls        2795                  2796                  2900                  PHA                 Total Dads      2795                 2796                 2900                 PHA                Total Moms   2795                2796                2900                PHA                Total   This is the order I wanted Names  Codes Dads      PHA                 2795                 2796                 2900                 Total Moms    PHA                 2795                 2796                 2900                Total Boys      PHA                2795                2796                2900               Total Girls     PHA               2795               2796               2900               Total   ... View more

Hello all,  I have initially posted this inappropriately, am sorry. I am creating a better representation of the issue so I can get help on this.  I have a dataset with category (teachers, students) and a list of forms (form2, event46, case32) which they are expected to fill within 60days after the school resume. Event46 has a group variable (grp=A, B, C). That is someone can be in any of the group. Among the teachers and students, some did not fill the form (0), some filled the form within 30 days (1), and some filled the form within 60 days (2).  The 0,1,2 represents the days that the form was filled below: No form filled=0 Form filled within 0-30days=1 Form filled within 31-60days=2   This is what I want   Table 1:  1. If a teacher or student only filled form2 within 30days and did not fill event46 or case32 they should be in the form2 only column. Same apply for other forms. If they filled the form after 30 days, they will be in the 31-60 days count by category.  2. If anyone filled more than one(1) forms either within 30 days or within 60 days or both, they should be in the multiple column.   Table 2: 3. Finally, I want a total of those who did not fill any form in both category(teacher or student).   I have created a sample output and a demo dataset. Thanks so much for your help. data have; input category $ form2 case32 event46 grp $; datalines; Teacher 1 0 0 . Student 1 0 0 . Teacher 0 1 0 . Teacher 1 0 1 A Teacher 0 0 1 C Teacher 0 0 1 B Teacher 1 0 0 . Student 0 1 1 C Student 0 0 1 C Student 0 0 1 C Student 1 0 0 . Student 0 1 0 . Student 0 2 0 . Teacher 1 1 1 B Teacher 0 0 2 A Teacher 0 0 2 B Teacher 2 0 0 . Student 0 1 1 C Teacher 0 0 0 . Student 0 0 0 . Student 0 0 1 B Teacher 0 0 1 B Teacher 0 2 0 . Teacher 2 2 0 . Student 0 1 0 . Teacher 2 0 1 A Student 0 0 1 A Teacher 0 0 0 . Student 2 0 0 . Teacher 0 0 2 C Student 0 0 2 C Student 0 0 2 B Student 0 0 2 A ; run; The expected output should look like this:   Form2 only Case32 only Event46 A only Event46 B only Event46 C Only Multiple Total Teacher fill form within 0-30 days               Teacher fill form within 31-60 days               Student fill form within 0-30days               Student fill form within 31-60days               Total                    Total Teacher who did not complete any form within 0-60 days   Student who did not complete any form within 0-60 days   Total     ... View more

Hello Everyone, I need help trying to find the count for the multi_event column. Here is a sample code and my output. I was able to get the count for event 1-4 only. I also want to count when event is in more than 1 column. data have; input event1 event2 event3 event4; datalines; . 1 . . 1 1 . . . . 1 . 1 . . 1 1 . . . . 1 . . . . 1 . . 1 1 1 . . . 1 1 . . . . . 1 1 ; run; SAS output   Row 2 and 4 met my criteria for multi_event so they were output in multi_event.  Row 8  and 11 met the criteria but did not output. I want row 8, 11 or any other row with multiple count to be in the multi_event column.  Here is my attempted code (My coding is not classic I will appreciate learning a better way to code it) data want; set have; if event1 ne . and event2=. and event3=. and event4=. then event1only=1; if event2 ne . and event1=. and event3=. and event4=. then event2only=1; if event3 ne . and event1=. and event2=. and event4=. then event3only=1; if event4 ne . and event1=. and event2=. and event3=. then event4only=1; if event1 ne . and (event2 ne . or event3 ne . or event4 ne . ) then multi_event=1; run; This is my final desired output is to be able to count each column like this with proc tabulate or report. Event1only Event2Only Event3Only Event4Only Multi_event 2 2  2 1 3           Thanks ... View more