I would use NOT EXISTS, but I'm not sure if the performance would be any different.

            
March 21, 2019 at 5:08 am
            
#2025090

            
            
                                                
        

Another way to do it with a CASE statement is to look for when there are 3 or more columns with the value NULL as there are only 4 combinations...

If you have 4 items, then to check at least 2 are not null you could do:

CASE WHEN (S155.Student_PersonID IS NOT NULL AND S160.Student_PersonID IS NOT...

This will prevent all visits that have a visit within 30 days of the previous, but it will omit blocks of close visits that might have more than 30 days...

If the problem is the space, you could REPLACE the spaces with an unused character (say '~') before you split the string:

myStringSplitFunction(REPLACE(PRODUCT,' ', '~'), '~')

Or you could use Jeff's favourite method: 😛

declare @trend table (keyinstn INT, ratingdate datetime, rating varchar(10))

declare myCursor cursor for
select distinct t.keyinstn,convert(date,ratingdate) ratingdate1
 ...

Or using an inline table:
select c.keyinstn, c.ratingdate, c.rating
from (select t.keyinstn,ratingdate, rating,
               ROW_NUMBER() OVER (PARTITION BY convert(date,ratingdate) ORDER...