I agree test proceedure and control over variables are very important.
I think correction factors are most useful when comparing data on the same dyno on different days. I also think correction factors are more accurate when they don't have to correct very much. Unless you are testing under conditions close to 60 degrees, 29.92 pressure and 0% moisture, why wouldn't you use a correction factor closer to the actual conditions you are testing at? I can think of a few reasons and these are not without some merit.
1. Everybody else (my competitors) uses it.
2. That's how I have always done it.
3. I can't or don't know how to change it.
4. I didn't know there was an alternative.
5. It gives me or my customer bigger numbers.
Just to name a few I'm sure there are other reasons.
Just so you don't think I am "correction factor Shaming". I have three correction factors loaded into my dyno program. I do all my testing with the CF that is closest to my conditions J1985. If it is a street or non competition type motor I will also offer (with a shortish explanation) the "higher" numbers with the J607 standard. I ask the customer which they would like me to print out for them. Guess which one they choose? If I were them I would probably pick the bigger numbers. It's human nature I think.