One-to-many database with merge requirement

I am designing a database to hold results of a computerized test. Each person who takes the test will have some identifying information: last and first name, date of birth, and sex. Each person can take the test several times, so a one-to-many relational database seems like the approach to take, with identifying information in one table and test results in another.

However, the person might take the test on different computers, and the databases from various computers might be merged (appended)for analysis. There's no guarantee that the identifying information would be identical on different computers -- William J. Smith on one computer might be Billy Joe Smith on another computer. The test will be administered to children, so using a Social Security number as a unique ID is not practical.

I am leaning toward a model with one record per test, with identifying information included in each record. It would then be up to a human operator to decide which records actually refer to the same person when doing test-retest comparisons.

I'm open to any thoughts/suggestions!

Sent via Deja.com http://www.deja.com/
Before you buy. Received on Mon Apr 17 2000 - 00:00:00 CEST