Exporting simple diff-in-diff table

[Tomer Stern]

I am trying to export a simple diff-in-diff table. I have included a sample of data down below.

I have tried the "diff" command and the "ieddtab" command. They require outside download, ssc install diff & ssc install ietoolkit


Below, I have an example of the simple DiD table I like and an attempted export. The export "runs" but it looks terrible and is not readable.
Does anyone have a good go-to command they use when exporting these kinds of simple tables?


ieddtab in_laborforce, time(survey_1998) treatment(treat)

asdoc ieddtab in_laborforce, time(survey_1998) treatment(treat), save(simple_DiD.doc)


clear
input float(in_laborforce treat control survey_1998)
0 0 1 0
0 0 1 0
0 1 0 0
0 1 0 0
0 0 1 0
1 0 1 0
1 1 0 0
0 0 1 0
0 0 1 0
0 0 1 0
1 1 0 1
0 0 1 1
0 0 1 1
1 1 0 1
1 0 1 1
1 0 1 1
1 1 0 0
1 0 1 0
1 0 1 0
1 0 1 0
1 1 0 0
1 0 1 0
1 0 1 0
0 1 0 0
0 1 0 0
0 0 1 0
1 0 1 1
0 0 1 1
0 0 1 1
0 0 1 1
0 0 1 1
0 0 1 1
1 1 0 1
0 1 0 1
end

[Attaullah Shah]

Since the two programs you have mentioned are not Stata official programs, asdoc (available on SSC) is not aware of them. Therefore, asdoc tries to export the results of these programs from a log file, which is off course not pretty in this case. The program ieddtab leaves behind a matrix ieddtabResults of all the results, but that is also not well formatted matrix for creating a nice table. For example :

ssc install asdoc
ieddtab in_laborforce, time(survey_1998) treatment(treat)
mat table =  r(ieddtabResults)
matlist table
 matlist table

             | in_labo~e
-------------+-----------
          2D |   .482967
      2D_err |  .3847713
    2D_stars |         0
        2D_N |        34
     2D_clus |         .
         1DC | -.1615385
     1DC_err |  .2119227
   1DC_stars |         0
       1DC_N |        23
    1DC_clus |         .
         1DT |  .3214286
     1DT_err |  .3279757
   1DT_stars |         0
       1DT_N |        11
    1DT_clus |         .
     C0_mean |  .4615385
      C0_err |  .1439099
        C0_N |        13
     C0_clus |         .
     T0_mean |  .4285714
      T0_err |  .2020305
        T0_N |         7
     T0_clus |         .
     C1_mean |        .3
      C1_err |  .1527525
        C1_N |        10
     C1_clus |         .
     T0_mean |       .75
      T0_err |       .25
        T0_N |         4
     T0_clus |         .

If you want to export this matrix, you can try this code:

asdoc wmat, mat(table) replace

If this template is what you frequently use, then I think you can go for asdocx where making a customized template file is a good investment.

[Tomer Stern]

If this template is what you frequently use, then I think you can go for asdocx where making a customized template file is a good investment.

I'm not married to this format. If you know of a way to export a table created with the "diff" command that would also be great. Or really anyway of exporting a simple nice diff-in-diff table

diff in_laborforce, t(treat) p(survey_1998)

[Justin Niakamal]

Have you tried the examle used in the diff (SSC) helpfile using outreg2 (SSC)

diff in_laborforce,  period(survey_1998) treated(treat)



outreg2 using table_diff, ctitle(`r(depvar)') addstat(Mean control t(0), r(mean_c0), Mean treated t(0), r(mean_t0), Diff t(0), r(diff0), Mean control t(1), r(mean_c1), ///
     Mean treated t(1), r(mean_t1), Diff t(1), r(diff1)) label word excel keep(_diff) nocons

[Tomer Stern]

Have you tried the examle used in the diff (SSC) helpfile using outreg2 (SSC)

Yea, I did try that and it "works" but it doesn't look like the classic table we've all seen. I included what I am thinking of below

[Justin Niakamal]

collapse (mean) in_laborforce, by(survey_1998 treat)
reshape wide in_laborforce, i(treat) j(survey_1998)

gen diff = in_laborforce1 - in_laborforce0
set obs `=_N+1'

ds treat, not 
foreach v in `r(varlist)'  {
    replace `v' = `v'[2] - `v'[1] in `=_N'
}
drop treat
ren (in_laborforce0 in_laborforce1) (before after)
gen per = ""
replace per = cond(_n == 1, "Control", cond(_n == 2, "Treatment", "Difference"),"Difference")
order per 


. list , div noobs

  +-------------------------------------------+
  |        per |   before | after |      diff |
  |------------+----------+-------+-----------|
  |    Control | .4615385 |    .3 | -.1615385 |
  |  Treatment | .4285714 |   .75 |  .3214286 |
  | Difference | -.032967 |   .45 |   .482967 |
  +-------------------------------------------+

And for reference

diff in_laborforce,  period(survey_1998) treated(treat)

DIFFERENCE-IN-DIFFERENCES ESTIMATION RESULTS
Number of observations in the DIFF-IN-DIFF: 34
            Before         After    
   Control: 13             10          23
   Treated: 7              4           11
            20             14
--------------------------------------------------------
 Outcome var.   | in_la~e | S. Err. |   |t|   |  P>|t|
----------------+---------+---------+---------+---------
Before          |         |         |         | 
   Control      | 0.462   |         |         | 
   Treated      | 0.429   |         |         | 
   Diff (T-C)   | -0.033  | 0.239   | -0.14   | 0.891
After           |         |         |         | 
   Control      | 0.300   |         |         | 
   Treated      | 0.750   |         |         | 
   Diff (T-C)   | 0.450   | 0.302   | 1.49    | 0.146
                |         |         |         | 
Diff-in-Diff    | 0.483   | 0.385   | 1.26    | 0.219
--------------------------------------------------------
R-square:    0.07
* Means and Standard Errors are estimated by linear regression
**Inference: *** p<0.01; ** p<0.05; * p<0.1

[Tomer Stern]

Wow, that looks great! I had feared that the best way to make this would be using the collapse command. So odd that there are commands for making these tables but not for exporting them!