Well it looks like you will first have to clean the data.  Then the query I first provided will do the job just fine.

Start a new thread if you need...

Do you have a secondary date on the same record, an auditing table or an identity column?

So what did you do to solve the remaining of the problem?

can we see the code that leads to this result?

HTH .

Looks to me like the client thinks that the procedure is name

"dbo.usp_AT @ Prog, @Chapt" instead of named dbo.usp_AT with 2 parameters.

Also is this a typo (@ Prog) >>...

This should get you started.

EXEC sp_msForEACHDB 'IF EXISTS (Select * from ?..SysObjects WHERE Name = ''TriggerName'')

BEGIN

USE ?

;

ALTER TRIGGER script

END'

SCAN  : reads all rows.

SEEK :  binary search algorithm to find only the required row.

Could you reindex and / or update the statistics on the table.

If that doesn't help you...

This is taken from the source code of the ROUTINE view :

 ROUTINE_DEFINITION   = convert(nvarchar(4000),

   (SELECT TOP 1 CASE WHEN encrypted = 1 THEN NULL ELSE com.text END

    FROM syscomments com WHERE com.id=o.id...

Agreed here.  But I would think that the best possible documentation can hardly be as accurate and always up to date as the actual used code.  That's why I have built a...

Your guess will be correct 99.9% of the time .

Is it possible that you have more than one record per day per SSN and that you would be in need of some sort of tie breaker?

Can you run this...

4th or 5th time already in the last 2-3 months... This one comes up all the time.

I was just demonstrating with a sample dataset that AVG and sum/count() handle nulls differently.  Now to think of it I would assume that the avg would be too low and...

Can you show us the some sample data and the required output from the update query... I'm sure there's a way to avoid the spool altogether.