catplot alternatives?

[Guillaume Pousse]

Hi all!

I have worker level data which I have split into two groups: those that are motivated, and those that are not.

I want to make a two tables for each category, one which looks at the workers' demographic composition, and the other at the workers' job characteristics.

I found a great example in the Tazhitdinova (2022) paper on moonlighting, attached below. I tried using catplot, but that gives me more of a horizontal bar chart rather than what I'm looking for.

Any tips on alternative routes?

Thanks

[Maarten Buis]

https://doi.org/10.1177/1536867X0800800208 has various ideas

[Nick Cox]

catplot is from SSC. It can do bars or dots; horizontal or vertical displays;; stacked or unstacked bars; and so on.

I agree with @Maarten Buis's suggestion, unsurprisingly if you look at his link. In addition, designplot is of some relevance.

SJ-14-4 gr0061  Design plots for graphical summary of a response given factors
        (help designplot if installed)  . . . . . . . . . . . . . .  N. J. Cox
        Q4/14   SJ 14(4):975--990
        produces a graphical summary of a numeric response variable
        given one or more factors

The most recent update is documented at https://journals.sagepub.com/doi/pdf...6867X241297954

. net sj 24-4 gr0061_4

-------------------------------------------------------------------------------------------------------------
package gr0061_4 from http://www.stata-journal.com/software/sj24-4
-------------------------------------------------------------------------------------------------------------

TITLE
      SJ24-4 gr0061_4. Update: Design plots for graphical ...

DESCRIPTION/AUTHOR(S)
      Update: Design plots for graphical summary of a
        response given factors
      by Nicholas J. Cox, Department of Geography,
           Durham University, Durham, UK
      Support:  [email protected]
      After installation, type help designplot
      DOI:  10.1177/1536867X241297954

INSTALLATION FILES                             (type net install gr0061_4)
      gr0061_4/designplot.ado
      gr0061_4/designplot.sthlp

ANCILLARY FILES                                (type net get gr0061_4)
      gr0061_4/designplot_examples.do
---------------------------------------------------

The design in #1 does not look quite so good once you notice that there isn't enough space to give details for the more unusual categories.

[Chen Samulsion]

The paper you cited probably draw plots by using Stata (Tazhitdinova, 2022). https://www.nber.org/system/files/wo...726/w27726.pdf

* Example generated by -dataex-. To install: ssc install dataex
clear
input float(demographic categoryv percent)
1 1 45
1 2 55
2 1 18
2 2 23
2 3 34
2 4 14
2 5 11
3 1 83
3 2 17
4 1 93
4 2  7
5 1 12
5 2 74
5 3 14
6 1  8
6 2 29
6 3 29
6 4 17
6 5 17
7 1 16
7 2 22
7 3 42
7 4 20
8 1  8
8 2 20
8 3 11
8 4 25
8 5 36
9 1 40
9 2 26
9 3 22
9 4  5
9 5  6
end
label values demographic Demographic
label def Demographic 1 "1. Gender", modify
label def Demographic 2 "2. Age", modify
label def Demographic 3 "3. Geo", modify
label def Demographic 4 "4. Nationality", modify
label def Demographic 5 "5. Education", modify
label def Demographic 6 "6. Monthly Income", modify
label def Demographic 7 "7. Wage at firm", modify
label def Demographic 8 "8. Firm: size", modify
label def Demographic 9 "9. N minijobs at firm", modify

graph hbar percent, over(categoryv) over(demographic) nofill asyvars stack bar(1, color(eltblue)) bar(2, color(gold)) bar(3, color(ltblue)) bar(4, color(orange)) bar(5, color(olive_teal)) blabel(bar, position(inside)) ytitle("percent") legend(off)

[Guillaume Pousse]

Thanks all!