Export table with only one category from one variable

[Maria Ortega]

Dear Statalist,

I am doing a twoway tab between an education variable = {1,2,3,4} and a bnary variable. I want the row percentages for binary_variable==1.

This is my code:

estimates clear
estpost tab educ missjob if tem==2, nototal
eststo tem0

estpost tab educ missjob if tem==1, nototal
eststo tem1


esttab tem0 tem1 using "$OUT/Table_1'.tex", replace unstack booktabs ///
cell(rowpct(fmt(2))) varlabels(`e(labels)') ///
eqlabels(`e(eqlabels)') nonum collabels(none) ///
mtitle("Without permit" "With permit")

but I only want a table with 2 columns, like this:

[Andrew Musau]

estout is from SSC (FAQ Advice #12). Look at how the equation is named and then use the -drop()- option.

sysuse auto, clear
qui sum price, d
gen hiprice= price>r(p50)
eststo m1: estpost tab rep78 foreign if !hiprice, nototal 
eststo m2: estpost tab rep78 foreign if hiprice, nototal
esttab m1 m2, replace unstack cell(rowpct(fmt(2))) nonum collabels(none)  
esttab m1 m2, replace unstack cell(rowpct(fmt(2))) nonum collabels(none)  drop(Domestic:)

Res.:

. esttab m1 m2, replace unstack cell(rowpct(fmt(2))) nonum collabels(none)  

----------------------------------------------------------------
                                                                
                 Domestic      Foreign     Domestic      Foreign
----------------------------------------------------------------
1                  100.00         0.00                          
2                  100.00         0.00       100.00         0.00
3                   88.24        11.76        92.31         7.69
4                   50.00        50.00        50.00        50.00
5                   40.00        60.00         0.00       100.00
----------------------------------------------------------------
N                      34                        35             
----------------------------------------------------------------

. 
. esttab m1 m2, replace unstack cell(rowpct(fmt(2))) nonum collabels(none)  drop(Domestic:)

--------------------------------------
                                      
                  Foreign      Foreign
--------------------------------------
1                    0.00             
2                    0.00         0.00
3                   11.76         7.69
4                   50.00        50.00
5                   60.00       100.00
--------------------------------------
N                      34           35
--------------------------------------

To see how the equation is named, just after the esttab command, run

mat l e(b)

Res.:

. mat l e(b)

e(b)[1,8]
     Domestic:  Domestic:  Domestic:  Domestic:   Foreign:   Foreign:   Foreign:   Foreign:
            2          3          4          5          2          3          4          5
y1          4         12          6          0          0          1          6          6

[Maarten Buis]

Alternatively, you can use the table command if you have Stata 17 or higher

. // open example data
. sysuse nlsw88, clear
(NLSW, 1988 extract)

. // prepare one of the variables
. gen byte urban = c_city + smsa

. label define urban 2 "central city" ///
>                    1 "suburban"     ///
>                    0 "rural"

. label value urban urban

. label variable urban "urbanicity"

.
. // create a basic table
. table (urban) (south union), nototal stat(percent, across(union)) nformat(%9.1f) sformat(%s%%) zerocounts

-----------------------------------------------------
               |           Lives in the south        
               |      Not south            South     
               |    Union worker       Union worker  
               |  Nonunion   Union   Nonunion   Union
---------------+-------------------------------------
urbanicity     |                                     
  rural        |     73.1%   26.9%      87.0%   13.0%
  suburban     |     72.0%   28.0%      86.8%   13.2%
  central city |     63.2%   36.8%      75.1%   24.9%
-----------------------------------------------------

. // modify the table to look the way we want
. collect layout (urban) (south#union[1])

Collection: Table
      Rows: urban
   Columns: south#union[1]
   Table 1: 4 x 2

---------------------------------------------
               |       Lives in the south    
               |     Not south          South
               |  Union worker   Union worker
               |         Union          Union
---------------+-----------------------------
urbanicity     |                             
  rural        |         26.9%          13.0%
  suburban     |         28.0%          13.2%
  central city |         36.8%          24.9%
---------------------------------------------

. collect style header south, title(hide)

. collect style header urban, title(hide)

. collect style header union, title(hide) level(hide)

. collect preview

---------------------------------
             |  Not south   South
-------------+-------------------
rural        |      26.9%   13.0%
suburban     |      28.0%   13.2%
central city |      36.8%   24.9%
---------------------------------
. // we can now use -collect export- to export the table to LaTeX, Word, Excel, ...

[Maria Ortega]

Thank you both! It worked perfectly