Oh how I love citing articles from and about Harvard, and this is a good one: a new lab at Harvard, the Harvard Digital Labs for the Social Sciences, ‘coud be a suitable replacement for MTurk as a way of gathering data, and is great news for researchers currently using DLABSS or considering using it in the future.” The article also refers to MTurk as the “primary source of online subjects among social science researchers”.
What is DLABBS? It is Harvard’s version of MTurk, with workers earning the prestige of filling out surveys for Harvard professors and the chance to win a $50 Amazon gift certificate every month>
The Harvard Digital Lab for the Social Sciences (DLABSS) is an online experiments and survey community for social science research. With the Harvard Digital Lab, you can participate in scientific research from your own computer. You can participate only once, or many times easily, thanks to our unique structure. By participating in our Master DLABSS Survey, you can join our research pool, which lets you participate in many other surveys without providing your basic information each time. Interested in getting more involved in the research? Join our subject pool and we will reach out to you when there is a survey or experiment that matches your interests and background.
To show our appreciation for your participation, we will raffle a $50 Amazon gift card every month. The more surveys you complete, the more times you are entered into the lottery. The winner will receive the gift card by email.
Apparently more than 1300 people have signed up. In the study (which the lead researcher paid people $1 for a nine minute survey on MTurk), the researcher found that the DLABSS panel was more liberal and better educated than MTurk (who tend to be more liberal and better educated than the population as a whole). DLABSS also had more women.
But anyway—the notable thing is that someone (and a someone at Harvard) is trying to show that their sample is as good as MTurk, which says very positive things about the validity of MTurk. But. If all kinds of universities set up these panels, what would happen to MTurk? And is it really as valid?