How to design graph for multiple choice question?

[Rabab Al hasni]

Dear Statalist,

I hope this finds you well.

I would like to design a graph (e.g. pie chart or bar graph) for a question that has multiple responses. The table below shows an example of the impact of lack of bank finance on business growth:

Firm	no effect	production decrease	Sales decrease	pressure on cash flow	Informal finance
A	1	.	.	.
B	.	1	1	.	1
C	.	.	1	1	1
D	.	1	1	1	.
E	1	.	.	.	.
F	.	1	1	.	1
G	.	.	.	1	1
H	.	1	.	1	1
I	.	.	1	.	.
J	.	.	1	1	1

Could you please help on how to design a graph for the multiple responses by firms that reflects the effect of lack of bank finance on their growth? I am using Stata MP17

Greatly appreciate your support

Best regards,
Rabab

[Andrew Musau]

After 57 posts, we should not be asking you to present data examples using the dataex command. Read FAQ Advice #12 if you have not familiarized yourself with how to use the command. One possibility is to aggregate the responses and graph percentages. You should fix the axis labels, either using variable labels or by relabeling within the command. Below I use myaxis from SSC, by Nick Cox.

* Example generated by -dataex-. To install: ssc install dataex
clear
input str1 firm byte(noeffect productiondecrease salesdecrease pressureoncashflow informalfinance)
"A" 1 . . . .
"B" . 1 1 . 1
"C" . . 1 1 1
"D" . 1 1 1 .
"E" 1 . . . .
"F" . 1 1 . 1
"G" . . . 1 1
"H" . 1 . 1 1
"I" . . 1 . .
"J" . . 1 1 1
end

mvencode noeffect -informalfinance, mv(0)
rename (noeffect -informalfinance) response=
reshape long response, i(firm) j(which) string
replace response= response*100
*ssc install myaxis
myaxis order=which, sort(sum response) descending
gr hbar response, over(order) scheme(s1mono) ytitle("Percent")

[Rabab Al hasni]

After 57 posts, we should not be asking you to present data examples using the dataex command. Read FAQ Advice #12 if you have not familiarized yourself with how to use the command. One possibility is to aggregate the responses and graph percentages. You should fix the axis labels, either using variable labels or by relabeling within the command. Below I use myaxis from SSC, by Nick Cox.

* Example generated by -dataex-. To install: ssc install dataex
clear
input str1 firm byte(noeffect productiondecrease salesdecrease pressureoncashflow informalfinance)
"A" 1 . . . .
"B" . 1 1 . 1
"C" . . 1 1 1
"D" . 1 1 1 .
"E" 1 . . . .
"F" . 1 1 . 1
"G" . . . 1 1
"H" . 1 . 1 1
"I" . . 1 . .
"J" . . 1 1 1
end

mvencode noeffect -informalfinance, mv(0)
rename (noeffect -informalfinance) response=
reshape long response, i(firm) j(which) string
replace response= response*100
*ssc install myaxis
myaxis order=which, sort(sum response) descending
gr hbar response, over(order) scheme(s1mono) ytitle("Percent")

[ATTACH=CONFIG]n1626420[/ATTACH]

Dear Andrew Musau,

Many thanks for your information.

I have tried to install the dataex but it did not work with me. Thank you again for your kind efforts to explain to me how to improve designing the requested graph. I will try to apply it.


Best regards,
Rabab