Brilliant! Done and Done! ... View more

Thanks Ballardw,   This very close to what I need.  I did not know about the SPLITCHARNODROP.  Very useful.  Only one problem remains: I need the program to treat a *space* as a break as well as the *hyphen* character.  I need the labels to split into three lines.  Is there a way to specify more than one SPLITCHAR?   I can't rotate the value labels as the y-axis space is at a premium and x-axis categories can be wordy.   data g; input Cat$ 16. Value; cards; OneH-TwoH Three1 4 One Two Three2 3 OneH-TwoH Three3 2 One Two Three4 1 OneH-TwoH Three5 4 One Two Three6 5 OneH-TwoH Three7 2 One Two Three8 3 ; proc sgplot data=g; vbar Cat / response=Value; xaxis fitpolicy=splitalways splitchar='-' splitcharnodrop; run; ... View more

Yes, thanks.  EscCharacter is what I am using now on an ad hoc basis.  But, as you noted, that's not what I need.  In my case, what is printed is a format for a numeric value.  I need it to work across multiple different deployments.  Forcing a line-break is not what I want when it is used as a row header. ... View more

Multiple procs.  Specifically, I used in PROC SGPLOT as a value label, PROC REPORT as a column header as well as a row header, PROC ODSTEXT, etc.  If there are PROCs that support this kind of functionality and others that don't, I'd like to know that too.   EDIT: so, apparently, both PROC PRINT and PROC report do this correctly with text in the cells--they both break a line at the dash.  PROC SGPLOT does not break at the dash when labeling values on x axis.  I will need to double-check how dash is treated in PROC REPORT column headers. ... View more

You have to be careful with the interpretation of the multi-level random effects. Just like the fixed effects, they are additive: i.e., your facility-level effect will not contain the county-level. In other words, if you're looking for a facility-level effect, you will need to add the county in which it is. ... View more

If I understand you correctly, random intercepts already give you what you seek—covariate standardized deviations of each random effect from sample average. If you need an estimate at a different standard than GLIMMIX default, you can use the ESTIMATE statement to obtain whatever estimate you need. ... View more

Use nested SQL queries. select * form A where date_A GT (select max(date_B) from B) having date_A=min(date_A) AND grade = '.' You probably meant to group by ID because the last condition does not make sense other wise. ... View more