Dear Sam:
You have your unique identifier for the parent file that is "inherited" by the child file. In the child file your create a unique identifier for that (simple is a construct of the parent identifier plus the date, if each child event happens on a different date). For validation, you compare the two parent files and the two child files, and that's it.
Best regards,
Hans
Hi,
The only problem with this solution is that I then cannot use the double entry method that compares responses as they are keyed in. As far as I know, it is not possible to do that with relational files.
Thanks, Sam
epidata-list@lists.umanitoba.ca