It turns out that particular way of conducting the experiments may have led to__ 14__ interpretation of what happed.__ 15___ , lighting was always changed on a Sunday .When work started again on Monday, output __16___ rose compared with the previous Saturday and__ 17 _to rise for the next couple of days.__ 18__ a comparison with data for weeks when there was no experimentation showed that output always went up on Monday. Workers__ 19__ to be diligent for the first few days of the weeking week in any case , before __20 __a plateau and then slackening off. This suggests that the alleged “Hawthorne effect “ is hard to pin down.