st: excluding single measurements in time

[John Singhammer <singhammer@gmail.com>]

Dear statalisters

I am investigating growth in physical fitness among 58.000 children in
206 schools.

Individual level data has been collected in 2009, 2010 and 2011.
individuals are nested within classes, within schools.
The schools volunteered to participate in the study and hence, data
are unbalanced as can be seen from the output below.

For a particular analysis, I want to select the schools (children
within schools) with repeated measures. That is, I want to exclude
schools with measurements at only a single point in time.

How do I come about that in stata?

Any comments and advices are appreciated..

Kind regards,
John


. xtdescribe

     pnr:  137400, 347246, ..., 1.231e+10                    n =      58041
     syr:  2009, 2010, ..., 2011                             T =          3
           Delta(syr) = 1 unit
           Span(syr)  = 3 periods
           (pnr*syr uniquely identifies each observation)

Distribution of T_i:   min      5%     25%       50%       75%     95%     max
                         1       1       1         2         2       2       3

     Freq.  Percent    Cum. |  Pattern
 ---------------------------+---------
    27306     47.05   47.05 |  .11
    16079     27.70   74.75 |  ..1
    11318     19.50   94.25 |  .1.
     2109      3.63   97.88 |  111
      597      1.03   98.91 |  1.1
      346      0.60   99.51 |  1..
      286      0.49  100.00 |  11.
 ---------------------------+---------
    58041    100.00         |  XXX
--
*
*   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/