You could impute the missing values, but Stata is doing what it should be doing by dropping cases with missing values on the independent variables. It also seems strange that attendance is being treated as a binary variable.
HTH,
Billy
Sent from my iPhone
On Dec 27, 2012, at 17:19, Mag K <kmagvini@gmail.com> wrote:
> Hi
>
> I have a dataset for children and parents and would need ro run a
> probit model on child school attendance(binary dependent variable)
> with the parents education, age, sex etc as independent variable. If I
> have observations where there is only one parent in the household I
> can't seem to run the probit properly. Here is an example
>
>
> ID Child school Father's education (in
> years) Mother's education (in years)
>
>
> 1 1 5
> 2
> 2 1 4
> 3
> 3 1 2
> 7
> 4 0 5
> -
> 5 1 -
> 3
>
>
>
> In the above example, if I run a probit
>
> probit child_school father_edu mother_edu
>
> on the 5 observations where the first 3 households have both parents
> and the 4th household only has the father and the 5th household only
> has the mother, stata drops the 4th and 5th observation because of the
> missing value. How do I get stata to consider running all the 5
> observations for analysis of 2 parent household and single parent
> household
>
> Grateful for any suggestions. Thanks
>
>
> Mag
> *
> * For searches and help try:
> * http://www.stata.com/help.cgi?search
> * http://www.stata.com/support/faqs/resources/statalist-faq/
> * http://www.ats.ucla.edu/stat/stata/
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/
