Some Requesters believe it is appropriate to pay MTurk workers less for an academic survey than for non-Academic work. I’ve heard several reasons why, including:
-there is greater intrinsic motivations/rewards for academic work (citing Kaufmann, Nicolas; Schulze, Thimo; Veit, Daniel (August 4, 2011). “More than fun and money: Worker Motivation in Crowdsourcing – A Study on Mechanical Turk”. Proceedings of the Seventeenth Americas Conference on Information Systems, Detroit, Michigan as well as Burhmester, Kwang, & Gosling, 2011, Perspectives on Psychological Science. However, both studies are from 2011, and the landscape has changed.
-paying less for academic work reflects the type of compensation we would give students or volunteers in a lab
-there’s often X amount of money to spend on y amount of responses, which results in paying x/y per response.
I’ve discussed this with two experiences workers, and will share their responses tomorrow. But the bottom line is: the Amazon Mechanical Turk agreement doesn’t discriminate between work on, say, a receipt transcription and work on an academic survey. Work is work, period. One question is: do Workers discriminate between two different tasks, and think it is OK to earn ‘less’ on one than the other? My focus group of two says no (as you will see tomorrow). I think this will be an interesting study to conduct to learn more about Turker motivations. I don’t think the answer will be any different than my focus group.