Originally Posted by BDL
Hi Martin :-)
Glad you enjoy the calculations side, let me know if you need a hand with any excel or pivot tables as it's part of daily life in my job...
Hopefully you'll crack a solution, I think it's a limited number that need addressing but definitely worth recording as a reference.
Well I've moved away from Mean and Median onto P Factor and Weighted Fades working on a curve to remove outliers, mostly because this method doesn't affect higher grades and balances out C a bit more over time. It's only in C that you really see so that many outliers, pretty much everyone seems to balance out after they escape C.
The problem with averages is there is no real "one' answer if there's a great degree of difference in the dataset, if there was I'd be typing this from a beach with seven female experts as I would have applied it to stock market shares
The above won't help us with Sans Cardage tho, I still feel we should choose to make it visible to see if it's a problem first.
Not sure about regional and national grades either, I'd kinda like it be as even as possible as I tend to measure against other people in the same grade.