I was going to call this series “Tantrums and Tips” because it will address things that piss me off as an MTurk worker and provide insights at how Academic researchers can avoid pissing people off. But I think Trick or Treat just has a more, I don’t know, adult ring to it?
So the problem: I accept a HIT, I click on the link, I answer five OR MORE demo questions, and get screened out.
OK I get it. When we’re trying to get a generalizable sample we don’t want too many people in a specific category in the group (from what I can glean, I don’t get screened out on age or ethnicity, but maybe on gender or education level). But to have me answer more than five questions (yes, five seems to be my tolerance level) and THEN screen me out means a. you just collected some information about me as a Turker that you can use in some way and b. this took me about a minute of my time (if you count finding the HIT, accepting the HIT, loading the survey, and answering the questions and c. (and this is the problem) YOU ARE NOT PAYING ME FOR IT.
I get, again, that you don’t want to pay me 45 cents (or whatever you are offering) if my information won’t be used in this study. But you could at least pay me 5 cents which would equal an hourly wage of $3.00 (and we won’t go into the fact that many Requesters think a minimum of $6 per hour is the right equivalency). But to pay me nothing? That means you are on my ‘do not HIT’ list. As this post explains, MTurk is NOT volunteer work.
Here’s what you can do: pay me for my screening questions. Pay EVERYONE for their screening questions. If I screen in, ask for my MTurk ID and invite me to an ‘invitation only’ survey. You can match my ID to the ID I give at the invitation only site. If I screen out, pay me the small amount so I won’t dislike you.
I know, I know, this takes up more time for Dr. Requester. But do we want to get into an argument on whose time is ‘worth’ more? I don’t think we do.
I’m eager to hear your thoughts on this idea.