st: graphical output of metan contains fictional data in studyid

[Michael McCulloch <mcculloc@berkeley.edu>]

I am experiencing difficulty with graphical output of metan.Strange numbers 
are appearing where the study name should be, in both the text output 
results and the graphical output. I have defined four variables, resp1 
resp0 noresp1 noresp0, which are the four 2x2 cells.

"Help metan" specifies that I should use the option (namvar=namvar), and in 
my case namvar=studyid.There are some missing data in a covariate, here 
called covar, and so I specify the command for covar!=. I have tried this 
with or without any of the options or "if" command, and this is still 
happening. I have made sure that studyid is a string variable. The numbers 
which appear in the Study column of output do not represent the sort order, 
or any of my data cells. By the way, this problem does not appear when I 
use meta.

Any suggestions?

(my command and output follow below)

metan resp1 resp0 noresp1 noresp0 if covar!=., random xlab(.1,.5,1,2,10) 
counts texts(.7) label(namvar=studyid) by(covar)

           Study     |     RR    [95% Conf. Interval]     % Weight
---------------------+---------------------------------------------------
     covar==1
2                    |    1.159     0.761     1.764         11.83
7                    |    1.706     1.150     2.530         13.46
9                    |    1.189     0.853     1.657         18.99
11                   |    1.347     0.830     2.187          8.90
12                   |    1.316     0.954     1.814         20.27
13                   |    1.385     0.863     2.224          9.33
 Sub-total           |
  D+L pooled RR      |    1.328     1.133     1.557         82.78
---------------------+---------------------------------------------------
     covar==2
14                   |    1.350     0.950     1.919         16.90
15                   |    0.333     0.026     4.268          0.32
 Sub-total           |
  D+L pooled RR      |    1.166     0.489     2.780         17.22
---------------------+---------------------------------------------------
Overall              |
  D+L pooled RR      |    1.326     1.147     1.532        100.00
---------------------+---------------------------------------------------


________________________________________
Michael McCulloch, LAc, MPH
Doctoral Candidate, Division of Epidemiology
University of California at Berkeley

Research Office:
124 Pine Street
San Anselmo, CA 94960-2674
Tel 	(415) 407-1357
Fax 	(415) 485-1065
email	mcculloc@berkeley.edu
URL	www.medepi.org/meta/

*
*   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/