Hello, I have 3 mutually exclusively groups within my data set, each marked by a variable called "group" where group = 1,2, or 3. I want to run a proc freq looking at age: proc freq data = dataset;      where group = 1;      tables age; run; proc freq data = dataset;      where group = 2;      tables age; run; proc freq data = dataset;      where group = 3;      tables age; run; Is there a  more efficient way to do this using a macro? At first I had separated the groups into 3 mutually exclusive data sets and i was able to write a macro for this just changing the name of the data set. but now that it is in one data set, how can I accomplish this? Or would it be done using an array?thank you ... View more

Hello, My data set has the following variables: event1 t1a0 t1a1 t1a2 .... t1a18 event2 t2a0 t2a1 t2a2 t2a3 .... t2a18  ......... event20 t20a0 t20a1 t20a2 t20a3 ..... t20a18 So each variable is binary. the child either did or did not receive event 1 (1=yes, 0=no), ..., the child either did or did not receive event 20 (1=yes, 0=no). the child either did or did not receive event1 at age0 (t1a0=1 if yes, t1a0=0 if no), the child either did or did not receive event1 at age1 (t1a1=1 if yes, t1a1=0 if no), . . . the child either did or did not receive event1 at age18 (t1a18=1 if yes, t1a18=0 if no), . . . the child either did or did not receive event20 at age0 (t20a0=1 if yes, t20a0=0 if no), the child either did or did not receive event20 at age1 (t20a1=1 if yes, t20a1=0 if no), . . . the child either did or did not receive event20 at age18 (t20a18=1 if yes, t20a18=0 if no), My question is this - I would like to present the mean duration of years the child experienced each of the 20 events and the standard deviation, as well as the number of kids who experienced each type of event and the percentage of those who did so and put together in a nice table as such: Header 1 Header 2 Header 3 event type       endorsed: % (n)        duration: mean (SD) event 1 55.44 (2550)        1.89 (1.94) event 2 etc.           etc ...... .....         ... event 20 etc          etc I was able to calculate the % and n endorsed for each event. However, I'm not sure how to calculate the mean duration of years that each child experienced the event. For example, the child may have had experienced event 1 at ages 0, 1, 5, 6, 16, 17, 18 and event 5 at ages 0, 5, 6, 7. Each child can experience multiple events. I was trying to do an array but I don't think my way is correct... Given that the names of the variables are similar (t[event #]a[age]), how could I calculate this? For event 2 i could calculate the # of years each child experienced the event as duration2= sum(of t2:); But now I want to do this for each event. I dont' want to make 20 arrays b/c there are 20 events, and in my array statement I'd have 20*18=360 variables which is too long.. any suggestions as to how I could calculate the mean duration of each event and the standard deviation? I can clarify if this is too confusing.. And then what, in your opinion is the best way to present this in a table? I'd like to just ods output it, or use ods trace to obtain values to put into a proc report. Thanks, Gina ... View more