[sperry]

The long time in the Hawthorne calculations wasn't due to PAX or anything... it was due to having to audit all the messed up times in the results vs. the paper. Back then we couldn't rely on AXWare at all, not only did the software itself crash all the time and screw up the results, the user interface was bad enough that user errors were constantly occurring and introducing discrepancies. Plus, IIRC, the software didn't score times the way we did PAX back then either, so there was still another manual step using that software I wrote that took the exported results from AXWare. So there were a ton of steps, none of which had anything really to do with calculating the final average PAX results... the spreadsheet Dean made did that great, assuming the data you had to paste into it was right.