It takes 3 chapters of "LoadRunner for the WEB 7.0" to elaborate what/how correlation is/is handled. I found that on-line documents also have good explanation.
"Correlated data is data which is sent to the client from the server, and later sent back to the server by the client" - per Mercury.
Correlation is a method or a way in which a conncection or link to header and detail line.If you have an account no and a name as a record and this is the header then all the detail lines or the transactions for the account need to be correlated.
If I find a script I will send it to you.The Mannual is also fine but you can visit a forum where you meet people.Ask some organizations where they have the QA meetings.
Write back if you have anymore problems.Also see the loadrunner group one is an Advanced one.You can be a member ...there.
You may be correct for your application. Correlation is both application and protocol specific. In general, correlation is required whenever an application hands a piece of information to a client which hass to be handed back as part of a subsequent request. This information can take the form of (but is not limited to) session information, user information, object (record) information or simply a sequence number that must be properly incremented with each pass of a conversation.
Are we correct in assuming this is a classical client-server/two-tier database application and not a web one? If so, then correlation is handled differently than it is in the web universe.