# RE: st: question about heckprob

 From Rafal Raciborski To statalist@hsphsun2.harvard.edu Subject RE: st: question about heckprob Date Wed, 18 May 2005 19:42:56 -0400

```Mike,

I am not sure whether this will solve the problem but here's a piece of advice:

The coding of your dv2 is problematic.  Note that right now when dv1 = 0, dv2 is
necessarily ALWAYS = 0 as well.  It seems to me that you should

replace dv2 = . if dv1 == 0

I have a dataset on GATT/WTO disputes where dv1 = 0 if a case was withdrawn
before a ruling was issued and 1 if it reached a judgment phase.  Then my
"judgment" variable, dv2, is set to missing if dv1 = 0.  Only when dv1 = 1,
does my dv2 = 0 or 1.  Here is a short list:

. list dv1 dv2 in 21/40

+-----------+
| dv1   dv2 |
|-----------|
21. |   1     0 |
22. |   1     1 |
23. |   1     0 |
24. |   1     0 |
25. |   0     . |
|-----------|
26. |   1     0 |
27. |   1     0 |
28. |   1     1 |
29. |   1     0 |
30. |   1     0 |
|-----------|
31. |   1     1 |
32. |   1     0 |
33. |   1     1 |
34. |   1     0 |
35. |   1     0 |
|-----------|
36. |   0     . |
37. |   1     0 |
38. |   0     . |
39. |   1     0 |
40. |   1     0 |
+-----------+

Think about it this way:  The number of cases in the second stage should not be
equal (or exceed!) the number of cases in the first stage, because the cases
are "selected out" in the first stage.  You do not really observe 0 in the
second stage; the second stage itself is not there.

Hope this helps.

rafal

================
Rafal Raciborski
Department of Political Science
Emory University
301 Tarbutton Hall
1555 Dickey Drive
Atlanta, GA 30322
404-378-9826 (home)
rafal.raciborski@emory.edu (e-mail)
www.roofoos.net

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/
```