Re:I was part of a human subject research study without my c...
by dcow on Sunday December 19, 2021 @11:31AM
(#29615241)
I don’t think it’s accurate to say that any experiment where a human is simply involved qualifies as a human subject experiment. The IRB guidelines are pretty clear about the distinction between simply gathering data from a human and gathering data about a human.
Re:I was part of a human subject research study without my c...
by SeanLuke on Sunday December 19, 2021 @12:10PM
(#29615764)
Sure. But the study is quite specific:
> The study aims to understand how websites would respond to real users, while accommodating websites that may have less capacity to respond.
That is, they're not interested in what the website's policies are via questionnaires. They're prodding them under false pretenses to see what would happen. This is definitely human subjects research.
The bar is very high, and with good reason: you're experimenting on humans. One of the foremost requirements is the notion of informed consent: you can't experiment on people who don't know or understand, or who didn't give you permission.
This experiment very clearly involved human subjects and very clearly violated informed consent. If the student involved did not realize this, certainly his advisor did or he'd have to be considered incompetent. And if the experiment was indeed placed before the IRB and was given an exemption, as was implied, then Princeton's IRB made a glaring and incredible error.
Had this student simply sent out a questionnaire from his Princeton email address, rather than conduct an experiment, he might have reasonably obtained IRB exemption or permission. But as I understand it, he (1) hid behind aliases so they did not know they were being experimented on, and (2) threatened legal deadlines in responding to his questions. There is no way this was ethical by IRB standards.
One thing that really bothers me is that people are attacking the blogger for using the word "consent", suggesting it's hyperbole in the sense of date rape or whatnot. This is terrible: "consent" is the standard and correct term used in human subjects research, and is a factual description of exactly what occurred.