 |
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
st: RE: question regarding merging
This may be a very bad idea. The attrition may be related to some variables that are present in the data. I would do the merge without dropping subjects with fewer than 5 data points. Later you can drop if needed, but the longitudinal analyses can proceed even with unequal numbers of observations per panel(subject).
Tony
Peter A. Lachenbruch
Department of Public Health
Oregon State University
Corvallis, OR 97330
Phone: 541-737-3832
FAX: 541-737-4001
-----Original Message-----
From: [email protected] [mailto:[email protected]] On Behalf Of Debs Majumdar
Sent: Thursday, July 16, 2009 2:25 PM
To: [email protected]
Subject: st: question regarding merging
Hi,
I am new to stata and was trying to figure out the best way to solve the following question.
I have a dataset with id's of school children and their demographics at 5 time points.
The 1st timepoint has 800 id's, the second has 650,...the last one has 325 id's. Some id's are common to all timepoints.
How do I create a dataset which will have only those id's which are present at all time points? I have created a variable 'time' which corresponds to the # of id's at each point and I tried with breaking up the dataset into 5 datasets(one for each timepoint) and merging them using nokeep but that didn't work.
Thanks,
Debs
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/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/statalist/faq
* http://www.ats.ucla.edu/stat/stata/