I am trying to use the SGPLOT procedure to create a stacked horizontal bar chart with four categories and apply a color scheme using the attrid statement.
In order to get the four categories in the proper order (rather than alphabetical), each category was assigned a numeric variable (1-4) and a format statement is used to convert the numeric variable back into the category label.
The issue is that I can either get the format to work but the attrid statement won't work to assign the appropriate colors or the attrid statement works to assign the colors and I am left with (1,2,3,4) in my legend.
The first statement produces an output with the appropriate values for label but not the colors specified by the evh type
proc sgplot data=combined5 dattrmap=formatdw.evhstyle;
format value valfmt.;
hbar &sumvar. / response=pcs1 stat=sum
group=value groupdisplay=stack nooutline attrid=evh;
title "Primary Care Services by Type";
run;
The second statement produces a graph with the appropriate colors but leaves (1,2,3,4) in the legend as the values used to order the stacked bars
proc sgplot data=combined5 dattrmap=formatdw.evhstyle;
hbar &sumvar. / response=pcs1 stat=sum
group=value groupdisplay=stack nooutline attrid=evh;
title "Primary Care Services by Type";
run;
Ideally, I'd like the bars stacked in the proper order, with the color scheme applied and the service types labeled in the legend. Thanks in advance for any help.
Use the formatted group values in your dattrmap dataset, sort your data by your numeric variable, and use grouporder=data in the hbar statement.
I don't know the answer, and there is not enough code here for me to easily reconstruct what you are doing. However, I can say that one of the standard legend tricks is to make an invisible plot that has the legend you want and overlay it with a visible plot that makes the graph that you want (and don't show the legend for it).
Also, in either the latest release or the second latest, there is a legenditem statement that is very cool and lets you control precisely what is in the legend. If no one posts better ideas, you might try those.
Use the formatted group values in your dattrmap dataset, sort your data by your numeric variable, and use grouporder=data in the hbar statement.
Registration is now open for SAS Innovate 2025 , our biggest and most exciting global event of the year! Join us in Orlando, FL, May 6-9.
Sign up by Dec. 31 to get the 2024 rate of just $495.
Register now!
Learn how use the CAT functions in SAS to join values from multiple variables into a single value.
Find more tutorials on the SAS Users YouTube channel.
Ready to level-up your skills? Choose your own adventure.