I wouldn't assume the in office workers get more done because they go to the office, correlation != causation.
It seems just as likely to me that the most dedicated, hard working employees are the ones willing to commute in to the office in the first place, and they would get more work done than others regardless of where they work.