I’ve been reading lots of studies about and using MTurk as research for a new book, and it has amused me to see how perspectives have changed over time. Two things to share:
1. Some Institutional Review Boards used to not require IRB approval for MTurk studies since Amazon ‘took care’ of keeping workers anonymous. We know now that isn’t true for all workers .
2. Some researchers (and they will remain nameless) argued that since there wasn’t really ‘interaction’ between Workers and Requesters on MTurk, IRB approval wasn’t required. The argument was that the interface ‘prevented’ and interaction. As anyone who has been a Requester knows, there can be plenty of interaction between Workers and Requesters.
Plus, the worker profile has changed from primarily women to primarily men. Not sure why exactly but interesting to ponder.