SAS Day 31: Scatter Plot and Fancy Scatter Plot
Yesterday I found an unwanted Graphics in SAS book; it has a very clean layout so i took it. I found there are many useful tips for creating SAS graphs, i would like to share with you.
We will start with the Scatter Plot, is a two-dimensional data visualization tool good for indicating the relationship between two variables.
We will use the SAS build-in dataset Class to show the height and weight of a fictitious set of children.
PROC SGPLOT: Basic Version Scatter Plot:
proc sgplot data=sashelp.class; title "Weight by Height"; scatter y=weight x=height; run;
2. PROC SGPLOT: Fancy Version of Scatter Plot
proc sgplot data=sashelp.class; title "Weight by Height"; scatter y=weight x=height/ datalabel=name markerattrs=(symbol=circlefilled color=orange size=5px); xaxis offsetmin=0.05 offsetmax=0.05 label="Class Height"; yaxis offsetmin=0.05 offsetmax=0.05 label="Class Weight" values=(50 to 150 by 25); run;
3. Proc Template: Fancy Version of Scatter Plot
Note: Proc Template and Proc Sgrender are twins, they show up together most of the time!
proc template; define statgraph ClassScatter; begingraph; entrytitle "weight by height"; layout overlay/ xaxisopts=(offsetmin=0.05 offsetmax=0.05 label="Class Height") yaxisopts=(offsetmin=0.05 offsetmax=0.05 label="Class Weight" linearopts=(tickvaluesequence=(start=50 end=150 increment=25) viewmin=50)); scatterplot y=weight x=height/ datalabel=name markerattrs=(symbol=circlefilled color=bibg size=5px); endlayout; endgraph; end; run; proc sgrender data=sashelp.class template=ClassScatter; run;
Except for all the lovely colors and labels we can manipulate, what can we conclude from the graph? It doesn’t take much attention to observe that in general, the taller the kids, the heavier the kids’ weight, or vice versa. The data visualization indicates there is a positive correlation between weight and height, we need to run more statistical analysis to prove the relationship!
Creating Statistical Graphics in SAS,
Happy Practicing! 😈