Re: st: Merging databases

[Nick Cox <njcoxstata@gmail.com>]

It's not clear why this is worrying you. -merge- is about joining
observation to observation, but the same syntax may well have
different effects depending on what is being merged. You are just
saying "-merge- in this way" and Stata will do what is needed.

(I know you just have to follow institutional rules, but either your
email footer is ridiculous or you just violated policy by sending a
confidential email to several thousand strangers.)

Nick
njcoxstata@gmail.com

On 14 November 2013 09:26, Polly Casey <pcasey@tccr.org.uk> wrote:

> I've been scratching my head about this for days...
>
> I have two databases which contain therapeutic outcome data, collected at a number of time points during treatment (e.g. 3 months, 6 months, 9 months, and so on). One is updated by staff at site A (master database) and the other is updated by staff at site B. Both databases contain exactly the same variables, and patients have unique ID numbers. I need to be able to merge the database from site B into the master database of site A every week.
>
> However, it is not a simple case of adding new variables, nor is it a simple case of adding new observations. Sometimes it will be a matter of adding new observations, for example when new clients are added to a database, but more often it will be a matter of adding new data to existing cases. For example, a client may come in and complete their 9 month questionnaires. This will then be added to that client's row of data, after their 6 month questionnaire data. It will be 'filling in gaps' in the database, if you like. How do I merge the database from site B into the master database of site A?

*
*   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/