I don't know. If someone is still employed at that location, presumably a higher percentage can implicitly be expected to vote 5* compared to the average former employee. It's similar to the controversy over mailing ballots to specific neighborhoods based on demographics to try to swing the votes.