if nmissT0=110 then the images was only read at T2 if nmissT2=110 then the images was only read at T0 if nmiss is <110, then the image was read at least for some features at one timepoint- not concerned with that part for this question   if nmissT0=110 then category=1; else if nmissT2=110 then category=2; else category=3;   and then   proc format; value catf 1='V0 readings only' ...  ; run;     ... View more

Either attach the right format to the right variable. Or recode the variable.  For your variable it looks a simple subtraction should do the recoding. peace2 = 6 - peace; format peace2 s. peace sr. ; ... View more

Thanks ballardw,, your code works! ... View more

Hi Tom,   That's fantastic, thank you! I really appreciate your excellent help. Maggie   ... View more

Wouldn't it be easier to put the data into a dataset and just merge it? data units; length variable $32 units $20 notes $200; infile cards dsd truncover ; input variable units notes; cards4; abcd,,collected on females only diabp,mm Hg,to the nearest 2 mmHg time,minutes, ;;;; data want; merge have(in=in1) units; by variable; if in1; run; ... View more

First thing I did but I could not find (or maybe understand) how the format information was abstracted ... View more

Thanks for all your help with the code and the caution explanation - much appreciated! ... View more

It looks like your last step is trying to recode VIST from so that should work without any formats.   And since you have recoded VISIT then you do NOT want to use the new values with the old format.  To remove any format that might be attached to VISIT you use a FORMAT statement that lists variable(s) but no format. data want; length visit $10; set have; if visit='offsite' then visit="1:offsite"; else if visit='hospital' then visit="2:hospital"; else if visit='remote' then visit="3:remote"; else visit=.; format visit ; run; ... View more

Thank you Paige for your reply which of course would have been the correct way to do it starting out. Margaret ... View more

Providing the data as a SAS data step (and not in any other format), according to these instructions, would be very helpful.   Also, what is wrong (in your opinion) with this data structure, and also please provide concrete examples of what you mean by "track project tasks and deliverables by analyst". ... View more

I don't understand what TESTM1 means and what the table is counting.   Is it counting individual trial results? Is it counting each TEST only once?  So it has some type of summary of the 3 trial replicates at the TEST level?  Is it counting each individual only once no matter how many tests or trial replicates were done?  If so how is the individual summarized?     ... View more

@urban58 wrote: (...) techid did not sort properly except for the latest timeperiod. There is something with my data which I'm trying to understand. One common reason for strange, unexpected sort orders are blanks or other invisible characters in character variables. Is techid a character variable (according to PROC CONTENTS output)? And, if so, what is the defined length of the variable (again: see PROC CONTENTS output)? Does techid have hundreds of different values or would it be feasible to review a PROC FREQ output showing the distinct values?   What does the improper sort order look like, compared to the proper sort order in the latest timeperiod? Can you show us the relevant portion of your PROC MEANS output or even a sample of your data so that we can reproduce the issue? Obviously, the analysis variables (var1, var2) are irrelevant and can be omitted. ... View more