Nesting variables in a bar graph?

[RVA]

I wanted to produce a bar graph as the following (graph 1).  You can see that on the horizontal axis, the "student_class" variable is nested into the "seen_notseen" variable.  However, I could not figure out how to do this.  The only thing I could do was produce two separate graphs, one for "seen" students and one for "notseen" students (see graphs 2a and 2b).  Do you have any ideas on how I can produce a graph that looks like g

Graph 1:

Graphs 2a and 2b:

Below are the codes I used to produce graph 2a (and 2b):

/* Set the graphics environment */
goptions reset=all cback=white border htitle=15pt htext=12pt;
/* Define the axis characteristics */
axis1 value=none label=none;
axis2 label=(angle=90 'Number of Students');
axis3 label=none;
/* Define the legend options */
legend1 frame label=none;

/* Generate the graph */
title1 'Academic Standing by Class (Seen Students)';

proc gchart data=students_b;

where class in ('FR','SO','JR','SR') and acadstanding not in (' ','N/A') and meetingcount > 0;

format acadstanding $standing_b. class $class_b.;

vbar acadstanding /

subgroup=acadstanding group=class type=freq
legend=legend1 space=0 gspace=4 width=5
maxis=axis1 raxis=axis2 gaxis=axis3;
run;
quit;

[DanH_sas]

I'm not sure of your SAS version, but you can use PROC SGPANEL create this chart. Below is a similar example. If you want the headers are the bottom, use COLHEADERPOS=BOTTOM on the PANELBY statement. Hope this helps!

proc sgpanel data=sashelp.heart;

panelby sex / novarname layout=columnlattice onepanel;

vbar weight_status / response=cholesterol stat=mean group=bp_status

     groupdisplay=cluster;

run;

[RVA]

Many thanks, Dan.  Unfortunately, I am still running SAS 9.2, so the groupdisplay= option is not available; otherwise, it would be perfect.  Do you have any work-around for me?  Thank you.

[DanH_sas]

Unfortunately, no. I had another way to approach it using overlays and DISCRETEOFFSET, but that approach requires at least SAS 9.3. Do you have access to a more recent version of SAS?

[RVA]

Thank you Dan.  Let me ask the IT for an upgrade and will get back to you later.  Thank you.