Help with graphs

[Lena Gan]

I hope someone can help me with my codes as I am trying to show Gender with different household income (both on x-axis) experiencing various levels of Stress-Scale (y-axis). I tried to do it once and it worked but when I did it again, it failed. Something went wrong with my codes. Please see my excel data with the three variables and graph. In addition, I also want to label 0 as "Male" and 1 as "Female" on the x-axis and on the y-axis, start the Scale on 1 rather than 2.5 and end on 7 rather than 4.5.

Please help me. I am so confused and feel hopeless as I have been going around in circles. thank you so much.

* Example generated by -dataex-. For more info, type help dataex
clear
input str6 Gender str17 HHIncome byte Stress_Scale
"Female" "income<$50K-$100k" 1
"Female" "income<$25K-$50k"  4
"Male"   "income<$25k"       7
"Male"   "income>$100k"      3
"Female" "income<$50K-$100k" 4
"Male"   "income>$100k"      7
"Female" "income<$25K-$50k"  2
"Female" "income>$100k"      4
"Female" "income>$100k"      4
"Female" "income<$50K-$100k" 5
"Female" "income<$50K-$100k" 6
"Female" "income<$25K-$50k"  1
"Male"   "income>$100k"      3
"Female" "income<$25K-$50k"  1
"Female" "income<$50K-$100k" 2
"Female" "income<$25k"       4
"Male"   "income>$100k"      7
"Female" "income<$50K-$100k" 3
"Male"   "income<$25k"       2
"Male"   "income<$25k"       3
"Female" "income<$50K-$100k" 7
"Female" "income<$50K-$100k" 1
"Male"   "income<$25K-$50k"  2
"Female" "income<$50K-$100k" 2
"Female" "income<$50K-$100k" 2
"Female" "income>$100k"      6
"Male"   "income<$25K-$50k"  3
"Female" "income<$50K-$100k" 3
"Male"   "income>$100k"      7
"Male"   "income>$100k"      3
"Male"   "income<$50K-$100k" 1
"Female" "income<$50K-$100k" 2
"Male"   "income<$25K-$50k"  2
"Female" "income<$50K-$100k" 5
"Female" "income<$25K-$50k"  1
"Female" "income<$50K-$100k" 2
"Male"   "income<$50K-$100k" 5
"Female" "income>$100k"      3
"Male"   "income<$25K-$50k"  2
"Female" "income<$50K-$100k" 2
"Male"   "income<$25K-$50k"  1
"Female" "income<$50K-$100k" 1
"Female" "income<$50K-$100k" 3
"Male"   "income<$50K-$100k" 2
"Male"   "income<$25K-$50k"  2
"Female" "income<$50K-$100k" 6
"Female" "income<$25K-$50k"  1
"Female" "income<$25K-$50k"  6
"Female" "income<$25k"       4
"Male"   "income<$25K-$50k"  2
"Male"   "income<$50K-$100k" 2
"Female" "income>$100k"      4
"Female" "income<$50K-$100k" 6
"Female" "income<$50K-$100k" 5
"Female" "income<$50K-$100k" 2
"Female" "income<$50K-$100k" 7
"Female" "income<$25k"       1
"Female" "income<$25K-$50k"  6
"Female" "income>$100k"      1
"Female" "income<$25K-$50k"  4
"Female" "income<$50K-$100k" 1
"Male"   "income<$25K-$50k"  2
"Female" "income>$100k"      4
"Female" "income>$100k"      2
"Male"   "income<$25K-$50k"  1
"Male"   "income<$50K-$100k" 4
"Female" "income<$50K-$100k" 3
"Male"   "income>$100k"      2
"Female" "income<$25K-$50k"  3
"Female" "income<$25K-$50k"  1
"Male"   "income>$100k"      2
"Male"   "income<$25K-$50k"  4
"Male"   "income<$50K-$100k" 2
"Female" "income<$50K-$100k" 1
"Male"   "income<$25K-$50k"  3
"Female" "income<$25k"       1
"Male"   "income<$50K-$100k" 7
"Female" "income>$100k"      4
"Male"   "income>$100k"      2
"Female" "income<$25K-$50k"  2
"Male"   "income<$25K-$50k"  2
"Female" "income<$25K-$50k"  6
"Male"   "income<$50K-$100k" 3
"Male"   "income<$50K-$100k" 1
"Male"   "income<$50K-$100k" 2
"Male"   "income<$25K-$50k"  4
"Male"   "income<$50K-$100k" 4
"Female" "income<$50K-$100k" 6
"Male"   "income>$100k"      5
"Male"   "income>$100k"      7
"Female" "income<$25K-$50k"  1
"Male"   "income<$25K-$50k"  5
"Female" "income<$25K-$50k"  4
"Female" "income>$100k"      4
"Female" "income<$50K-$100k" 6
"Female" "income<$50K-$100k" 5
"Female" "income<$25k"       1
"Male"   "income<$50K-$100k" 6
"Female" "income<$50K-$100k" 3
"Female" "income<$50K-$100k" 3
end

collapse (mean) Stress_Scale , by(Gender)
format Stress_Scale %9.1f
separate Stress_Scale, by(HHIncome) veryshortlabel
twoway connected Stress_Scale0 Stress_Scale1 Stress_Scale2 Stress_Scale3 Gender, mlab(Stress_Scale Stress_Scale Stress_Scale Stress_Scale) mlabpos(12 12) xlab(0/1) plotregion(margin(large))

[Matthew Alexander]

Hi Lena,
Can you explain what the problem is exactly?
If it is just that you want to recode the x-axis and constrain the range of the y-axis, then the following should work
For the former

xlabel(0 "Male" 1"Female")

For the latter

ylabel(1(1)7)

Simply add the above lines to your graph code.
All the best,
Matt

[Øyvind Snilsberg]

you need to make some modifications to your code,

gen female = Gender == "Female"
collapse Stress_Scale, by(female HHIncome)
separate Stress_Scale, by(HHIncome) veryshortlabel

tw connected Stress_Scale1-Stress_Scale4 female ///
,xlab(0 "Male" 1 "Female") xti("") yla(1(1)7) ysc(r(1 7))

[Lena Gan]

I modified the first line of the code ie gen female = Gender == "Female" but received an error msg "mismatch r (109). Did I type it correctly? Thank you.

[Jared Greathouse]

No. No idea what context this is in but what you might mean is

g female = Gender if Gender == "Female"

Whether this makes sense or not is your call.

[Øyvind Snilsberg]

does the error occur if you run the code in #3 on the data you provided in #1?
is Gender string or numeric in your dataset?

[Lena Gan]

Gender is Male or Female, string. And yes, the error occurs on the third line
separate Stress_Scale, by(HHIncome) veryshortlabel

[Nick Cox]

The suggestions in #3 look good to me. I guess you're changing something else.

This works for me. I added a kind of histogram using tabplot from the Stata Journal. Some researchers might feel queasy taking means of an ordinal scale variable.

Notice also the clean-up of what I guess are your income categories. You need to be consistent in use of k or K and inequality signs.

* Example generated by -dataex-. For more info, type help dataex
clear
input str6 Gender str17 HHIncome byte Stress_Scale
"Female" "income<$50K-$100k" 1
"Female" "income<$25K-$50k"  4
"Male"   "income<$25k"       7
"Male"   "income>$100k"      3
"Female" "income<$50K-$100k" 4
"Male"   "income>$100k"      7
"Female" "income<$25K-$50k"  2
"Female" "income>$100k"      4
"Female" "income>$100k"      4
"Female" "income<$50K-$100k" 5
"Female" "income<$50K-$100k" 6
"Female" "income<$25K-$50k"  1
"Male"   "income>$100k"      3
"Female" "income<$25K-$50k"  1
"Female" "income<$50K-$100k" 2
"Female" "income<$25k"       4
"Male"   "income>$100k"      7
"Female" "income<$50K-$100k" 3
"Male"   "income<$25k"       2
"Male"   "income<$25k"       3
"Female" "income<$50K-$100k" 7
"Female" "income<$50K-$100k" 1
"Male"   "income<$25K-$50k"  2
"Female" "income<$50K-$100k" 2
"Female" "income<$50K-$100k" 2
"Female" "income>$100k"      6
"Male"   "income<$25K-$50k"  3
"Female" "income<$50K-$100k" 3
"Male"   "income>$100k"      7
"Male"   "income>$100k"      3
"Male"   "income<$50K-$100k" 1
"Female" "income<$50K-$100k" 2
"Male"   "income<$25K-$50k"  2
"Female" "income<$50K-$100k" 5
"Female" "income<$25K-$50k"  1
"Female" "income<$50K-$100k" 2
"Male"   "income<$50K-$100k" 5
"Female" "income>$100k"      3
"Male"   "income<$25K-$50k"  2
"Female" "income<$50K-$100k" 2
"Male"   "income<$25K-$50k"  1
"Female" "income<$50K-$100k" 1
"Female" "income<$50K-$100k" 3
"Male"   "income<$50K-$100k" 2
"Male"   "income<$25K-$50k"  2
"Female" "income<$50K-$100k" 6
"Female" "income<$25K-$50k"  1
"Female" "income<$25K-$50k"  6
"Female" "income<$25k"       4
"Male"   "income<$25K-$50k"  2
"Male"   "income<$50K-$100k" 2
"Female" "income>$100k"      4
"Female" "income<$50K-$100k" 6
"Female" "income<$50K-$100k" 5
"Female" "income<$50K-$100k" 2
"Female" "income<$50K-$100k" 7
"Female" "income<$25k"       1
"Female" "income<$25K-$50k"  6
"Female" "income>$100k"      1
"Female" "income<$25K-$50k"  4
"Female" "income<$50K-$100k" 1
"Male"   "income<$25K-$50k"  2
"Female" "income>$100k"      4
"Female" "income>$100k"      2
"Male"   "income<$25K-$50k"  1
"Male"   "income<$50K-$100k" 4
"Female" "income<$50K-$100k" 3
"Male"   "income>$100k"      2
"Female" "income<$25K-$50k"  3
"Female" "income<$25K-$50k"  1
"Male"   "income>$100k"      2
"Male"   "income<$25K-$50k"  4
"Male"   "income<$50K-$100k" 2
"Female" "income<$50K-$100k" 1
"Male"   "income<$25K-$50k"  3
"Female" "income<$25k"       1
"Male"   "income<$50K-$100k" 7
"Female" "income>$100k"      4
"Male"   "income>$100k"      2
"Female" "income<$25K-$50k"  2
"Male"   "income<$25K-$50k"  2
"Female" "income<$25K-$50k"  6
"Male"   "income<$50K-$100k" 3
"Male"   "income<$50K-$100k" 1
"Male"   "income<$50K-$100k" 2
"Male"   "income<$25K-$50k"  4
"Male"   "income<$50K-$100k" 4
"Female" "income<$50K-$100k" 6
"Male"   "income>$100k"      5
"Male"   "income>$100k"      7
"Female" "income<$25K-$50k"  1
"Male"   "income<$25K-$50k"  5
"Female" "income<$25K-$50k"  4
"Female" "income>$100k"      4
"Female" "income<$50K-$100k" 6
"Female" "income<$50K-$100k" 5
"Female" "income<$25k"       1
"Male"   "income<$50K-$100k" 6
"Female" "income<$50K-$100k" 3
"Female" "income<$50K-$100k" 3
end

gen income_str = subinstr(HHIncome, "income", "", .)
replace income_str = subinstr(income_str, "K", "k", .)
replace income_str = subinstr(income_str, "<", "", .) if strpos(income_str, "50")
label def income_str 1 "<$25k" 2 "$25k-$50k" 3 "$50k-$100k" 4 ">$100k" 
encode income_str, gen(income_num) label(income_str)

gen female = Gender == "Female"
label def female 0 Male 1 Female 
label val female female 

tabplot Stress female, by(income_num, note("") row(1) compact) showval yreverse ytitle(Stress scale) name(histogram, replace)

collapse Stress_Scale, by(female income_num)
separate Stress_Scale, by(income_num) veryshortlabel

tw connected Stress_Scale? female , ytitle(Mean stress) xlab(0 "Male" 1 "Female") xti("") yla(1(1)7, ang(h)) ysc(r(1 7)) ///
ms(T O D S) legend(order(4 3 2 1) col(1) pos(11) ring(0)) xsc(r(-0.1 1.1)) name(means, replace)

[Lena Gan]

Dr Cox
Thank you so much for your help. Yes, the graphs you illustrated was what I have been struggling towards since Christmas so thank you for this lovely New Year's gift. I am deeply grateful. Thank you for sharing the codes as well. Yours truly, Lena