Graphing by groups of variables

[Yosef Cohen]

* Example generated by -dataex-. To install: ssc install dataex
clear
input str16 provider float(year pet_stress cmr_stress ccta)
"Cardiologist"     2010 0 0 0
"Radiologist"      2010 0 0 0
"Nuclear Medicine" 2010 0 0 0
"Other"            2010 0 0 0
"Other"            2010 0 0 0
"Cardiologist"     2010 0 0 0
"Other"            2010 0 0 0
"Radiologist"      2010 0 0 0
"Cardiologist"     2010 0 0 0
"Other"            2010 0 0 0
"Nuclear Medicine" 2010 0 0 0
"Other"            2010 0 0 0
"Radiologist"      2010 0 0 0
"Cardiologist"     2010 0 0 0
"Nuclear Medicine" 2010 0 0 0
"Cardiologist"     2010 0 0 0
"Other"            2010 0 0 0
"Radiologist"      2010 0 0 0
"Cardiologist"     2010 0 0 0
"Nuclear Medicine" 2010 0 0 0
"Radiologist"      2010 0 0 0
"Cardiologist"     2010 0 0 0
"Other"            2010 0 0 0
"Other"            2010 0 0 0
"Cardiologist"     2010 0 0 0
"Cardiologist"     2010 0 0 0
"Cardiologist"     2010 0 0 0
"Cardiologist"     2010 0 0 0
"Other"            2010 0 0 0
"Radiologist"      2010 0 0 0
"Cardiologist"     2010 0 0 0
"Other"            2010 0 0 0
"Other"            2010 0 0 0
"Cardiologist"     2010 0 0 0
"Other"            2010 0 0 0
"Cardiologist"     2010 0 0 0
"Radiologist"      2010 0 0 0
"Cardiologist"     2010 0 0 0
"Other"            2010 0 0 0
"Radiologist"      2010 0 0 0
"Cardiologist"     2010 0 0 0
"Other"            2010 0 0 0
"Radiologist"      2010 0 0 0
"Other"            2010 0 0 0
"Cardiologist"     2010 0 0 0
"Other"            2010 0 0 0
"Cardiologist"     2010 0 0 0
"Radiologist"      2010 0 0 1
"Cardiologist"     2010 0 0 0
"Radiologist"      2010 0 0 0
"Radiologist"      2010 0 0 0
"Cardiologist"     2010 0 0 0
"Radiologist"      2010 0 0 0
"Cardiologist"     2010 0 0 0
"Radiologist"      2010 0 0 0
"Other"            2010 0 0 0
"Cardiologist"     2010 0 0 0
"Cardiologist"     2010 0 0 0
"Cardiologist"     2010 0 0 0
"Cardiologist"     2010 0 0 0
"Cardiologist"     2010 0 0 0
"Other"            2010 0 0 0
"Nuclear Medicine" 2010 0 0 0
"Cardiologist"     2010 0 0 0
"Other"            2010 0 0 0
"Cardiologist"     2010 0 0 0
"Nuclear Medicine" 2010 0 0 0
"Cardiologist"     2010 0 0 0
"Radiologist"      2010 0 0 0
"Other"            2010 0 0 0
"Other"            2010 0 0 0
"Cardiologist"     2010 0 0 0
"Nuclear Medicine" 2010 0 0 0
"Cardiologist"     2010 0 0 0
"Other"            2010 0 0 0
"Nuclear Medicine" 2010 0 0 0
"Cardiologist"     2010 0 0 0
"Other"            2010 0 0 0
"Cardiologist"     2010 0 0 0
"Cardiologist"     2010 0 0 0
"Nuclear Medicine" 2010 0 0 0
"Cardiologist"     2010 0 0 0
"Cardiologist"     2010 0 0 0
"Radiologist"      2010 0 0 1
"Cardiologist"     2010 0 0 0
"Cardiologist"     2010 0 0 0
"Other"            2010 0 0 0
"Cardiologist"     2010 0 0 0
"Cardiologist"     2010 0 0 0
"Cardiologist"     2010 0 0 0
"Radiologist"      2010 0 0 0
"Other"            2010 0 0 0
"Cardiologist"     2010 0 0 0
"Cardiologist"     2010 0 0 0
"Cardiologist"     2010 0 0 0
"Other"            2010 0 0 0
"Cardiologist"     2010 0 0 0
"Cardiologist"     2010 1 0 0
"Other"            2010 0 0 0
"Nuclear Medicine" 2010 0 0 0
end

Hi,
I am trying to graph variable "provider" by groups of variable "patients" (10-50, 50-100, 100-500, 500-1000, >1000) by variable "year" for each of the variables "cmr_stress" "ccta" and "pet_stress".
How would I do this?

Thanks

[Nick Cox]

3 responses (summarized or raw?) X provider categories X 6 patient categories X (in your concurrent thread) 12 years.

My mind boggles at what should show all this intelligibly.

[Yosef Cohen]

I would do each response in its own graph. I can decrease the number of patient categories to 10-50, 50-500, >500. Would that make more sense?

[Nick Cox]

Sorry, but still too broad a request to tempt me into coding. Note that your data example has constant year and all zeros for the 3 responses.