Hi:   This question would be easier to answer if we knew what your data looked like. Your PROC REPORT statements are incorrect...you can't have the same variable defined with ACROSS, GROUP and an ANALYSIS (N) usage. So even though PROC REPORT doesn't give you a message in the log, your syntax is confusing. Also, your syntax includes references to LISTING only options (such as HEADLINE and WIDTH= that would not work for ODS HTML, ODS PDF or ODS RTF).   The code for me produced exactly the same table in TABULATE vs REPORT. PROC TABULATE provides the N statistic automatically for CLASS variables. For PROC REPORT, the comma operator is used to place a statistic or a statistic based on another variable "underneath" the across variable values. So, the second PROC REPORT step shows how to get the MEAN of HEIGHT and WEIGHT "underneath" each unique value for the SEX variable. cynthia data history (keep= name smoking alcohol);   set sashelp.class;   ** make some fake data;   if age lt 13 then smoking = 'No ';   else smoking = 'Yes';   if substr(name,1,2) le 'Je' then alcohol = 'No ';   else alcohol='Yes'; run;     ods listing; proc print data=history;   var name smoking alcohol; run; ods listing close;    ods html file='c:\temp\across.html';   proc tabulate data=history; title 'TABULATE method';   class smoking alcohol;   table smoking alcohol;   keylabel n=' '; run;    proc report data=history nowd ; title 'REPORT method';   column smoking,n alcohol,n;   define smoking / across ;   define alcohol / across;   define n / ' '; run;     proc report data=sashelp.class nowd;   title 'Showing different use of the comma operator';   column age sex,(height weight);   define age / group;   define sex / across 'Gender Averages by Age';   define height / mean 'Height' f=comma6.2;   define weight / mean 'Weight' f=comma6.2; run; ods html close; title; ... View more

It is what you are looking for ? data want(keep=found); input; ExpressionID = prxparse('/((\d|\.)+\s*(mm|mM|Mm|MM)\s*\w+)|((\d|\.)+\s*%\s*(\w|\s)+)/o'); start = 1; stop = length(_infile_); call prxnext(ExpressionID, start, stop, _infile_, position, length); do while (position > 0); found = substr(_infile_, position, length); output; call prxnext(ExpressionID, start, stop, _infile_, position, length); end; cards; 20% PEG4000, 200mM Sodiumchloride, 50mM MES, 0.3% dioxane, pH 6.0, VAPOR DIFFUSION, HANGING DROP, temperature 310K PEG3350, LiCl, Tris, pH 9.0, VAPOR DIFFUSION, HANGING DROP, temperature 295.0K PEG3350, MgOAc, Na Cacodylate, TCEP, pH 6.9, VAPOR DIFFUSION, SITTING DROP, temperature 280K CRYSTALLISED FROM HANGING DROP WHICH ALSO CONTAINED 50MM Potassiumphosphate, 5MM DTT AND 30% SATURATED AMMONIUMSULFATE, PH 7.0. THE WELL SOLUTION WAS 45% SATURATED AMMONIUMSULFATE., vapor diffusion - hanging drop ; run; Ksharp ... View more

As Astounding said. SAS order the value based on alphabetical order. You can customize the order by adding a blank before the value to cheat SAS. Because a blank is before than all of other alpha character. Like: .............. proc transpose data=c out= tab4d; run; data tab4d; set tab4d; length col $ 50 ; *make the variate long enough to hold the adding blanks; select(col); when('day3')   col='                       day3'; *day3 is the first value so need the most blanks; when('day7')   col='                   day7'; ......................... Ksharp ... View more

Hi:   Seeing your TABULATE code would be most helpful here and a bit of your data. For example, when you say you want to "remove a field from the list" -- what field and what list? Do you mean the variable name in the VAR statement or do you mean the variable label in the TABULATE final table? You could mean either. You say that the field is not needed in the output, which would be an argument for just taking the variable out of the TABLE statement, but without seeing your actual code, it is hard to envision what you mean.   Next, you asked how to "align the listings under a specific variable in the middle of the table" -- I'm stumped over this one -- what "listings" -- again, seeing your actual TABULATE code might help here. What do you mean by the "middle of the table"? Do you want to alter the justification of the data cells or do you have multiple TABLE statements in one TABULATE job and want all the tables to be justified the same? There may not be just one keyword or code option to do what you want to do -- assuming that we understand what it is you want to do.   Also, what is your destination of interest? LISTING window or output? HTML, RTF, PDF -- many TABULATE options that work in the LISTING destination do not work for other ODS destinations.   If you cannot or are not in a position to post some sample data and your TABULATE code, then you might want to work with Tech Support, as they can look at ALL your code and ALL your data and help you come to the best resolution of your question. cynthia ... View more

Yes, just as you would write your variable names instead of VAR1, VAR2 and VAR3, you would put your statistics in place of STAT1, STAT2 and STAT3. The names I used were just placeholders. The actual code I pasted into the answer (under my name) should be what you follow/use. cynthia ... View more