Collecting identifiable information on Amazon MTurk - Question about Informed Consent

[Marissa4usa]

Hi all,
I'm in the process of preparing my IRB submission, and thus my Informed Consent forms for my dissertation study. I plan on recruiting individuals via Amazon MTurk, for which people will complete a brief screener survey on MTurk (for which they will be paid accordingly via the MTurk payment mechanism), and if they meet eligibility criteria, I would like to invite them to participate in a follow-up study (i.e., the actual research project).

Typically, data collection via MTurk is anonymous, however, in this case, I will obviously need people's email address to contact them for the main study. My understanding is that MTurk is not as secure as, say, Qualtrics (which I have used for research fairly extensively), however, I admittedly can't really find anything on Amazon MTurk's website that describes their data security procedures. I have heard of people who do recruit via Amazon MTurk the same way (though I don't know anyone personally), so it must be possible and IRBs are clearly approving it.

I was wondering if anyone had suggestions for how to best phrase something like this in an Informed Consent form, or better even, if someone was willing to share their Informed Consent form with me.

Thanks, guys!

---

Members don't see this ad.

 

[MCParent]

you can link mturk and qualtrics. They pick the survey on mturk, complete it on qualtrics, and it turns itself around to give them the $$.
http://brentcurdy.net/qualtrics-tutorials/link/

So, if you mean that you have qualtrics access, you can get that data thru qualtrics rather than mturk.

 

[EmotRegulation]

You actually don't HAVE to get their email addresses to contact them for the main study. I've done several followup studies with mTurk that involve anonymously contacting people VIA mturk to tell them that they are eligible for a follow up study. And yes, it does involve typically recruiting through mturk and then sending them to Qualtrics to actually complete the study.

See this article (and their associated information online) for inviting people back:
Shapiro DN, Chandler J, Mueller PA. Using Mechanical Turk to
study clinical populations. Clinical Psychol Sci. 2013;1(2):213–220.
doi:10.1177/2167702612469015.

There are also waaaaaaaaaaaaaaaaaaaay easier ways of inviting people back to Mturk nowadays via R (because let me tell you, the Shapiro et al method worked for us, but there was a lot of swearing and frustrating trying to get it to work, as we are not programmers).

 

[Justanothergrad]

Emot and I had the same experience with trying to get people in for a second study. I used the Shapiro et al method as well. My suggestion is to make sure its not time-lapsed for any period of during between initial and follow-up. The dropout/non-contact rate is very high in that case.

 

[Marissa4usa]

Wow, that is awesome information - I had no idea that is possible! It will make my life exponentially easier. I'm going to look into linking the Qualtrics screener and MTurk.

Thank you!