Hi Gail,
I don't want to sound like I am beating a dead horse and I did read your reply...if that is common knowledge I didn't know.
Your comment
The higher levels of the index are generally not large enough for fragmentation to matter, and they're far more likely to be in memory where fragmentation doesn't matter.
Make sense but how would you determine the threshold of 'large enough' a ratio?
The table I focused on initially was 12 gb's in size, has 132955 pages...
I am not trying to be critical of your comments and hope it doesn't sound that way...I guess I am trying to figure out the 'rule of thumb' when digging through the results.