4 ms·

Ethical guidelines on research exist to prevent an adverse impact on participants. This study had adverse impacts: fear, stress, time & money in consulting lawy

Ethical guidelines on research exist to prevent an adverse impact on participants. This study had adverse impacts: fear, stress, time & money in consulting lawyers. It was therefore defacto an unethical research study. Speculation as to why the protocol slipped through the IRB cracks are that the language used in the study proposal (at least the part made public) dehumanized the protocols by referring to "websites" rather than humans that would be responding to the inquiries. The IRB ruled this was not a human subject piece of research, but that is contradicted by the deception protocol. Deception was justified as necessary because people's behavior might change if they knew it was a research request. That acknowledgement made it implicit that human behavior and potential changes to it due to the experiment was a core factor in the study-- ergo, it had human subjects. Behavioral research on human subjects is required to go through a much more rigorous IRB oversight process precisely to anticipate and mitigate potential adverse reactions. Some people are focussing on the deception, but that is, under some circumstances, allowed by research ethics. The more serious problem was adverse impact which, again, is the primary motivator for why we now have laws and regulation-mandated IRB processes to make sure it doesn't become an issue.

belorn 5y ago
I wonder if IRB ruled in this way because of the assumption of algorithmic response for requests like DMCA take down notices. I can imagine that even for GDPR/CCPA requests, there is still no human involved for website like Google, facebook, youtube and other major sites that is primarily operated through automation. If there is no humans involved then there is no humans to have an adverse impact on. But as you said, researchers however must have suspected that responses would be made by humans or else the email would have included the fact that it was a study.
throw10920 5y ago
> Ethical guidelines on research exist to prevent an adverse impact on participants. Not relevant in this case, because (1) it's clearly not human research[1] and (2) all kinds of other clearly-non-human-experiments still have adverse effects on humans tangentially involved, so that's not unique to human research, either. When scientists were testing the first particle accelerators, they caused some people a lot of stress who were worried that they would destroy the world - does that mean that those tests were human experiments? (clearly not) [1] https://news.ycombinator.com/item?id=29656792 https://news.ycombinator.com/item?id=29656792
- ineedasername 5y ago
  I may be wrong, but I'm going to guess that you don't work with IRB's all that often. I do. I asked a colleague their thoughts on this and they were unequivocal in believing this should have been flagged as involving human subjects if the approving IRB had all of the details of the protocol. Their guess is that the proposal's presentation-- perhaps very innocently-- did not fully convey the details that would have resulted in oversight of the research as a human subject project. These are experts on the nuances of these laws. Of course, experts may disagree, so that reference is not definitive. But it is suggestive that dismissals from those here on HN that are arm-chairing this with no-- or minimal-- experience with this sort of research are not basing their opinions on a full understanding of the IRB & research ethics ecosystem. One or two IRB applications for a research project will not convey the understanding needed to evaluate research protocols under the laws & regulations involved. As for definitions of human subject, it seems like you are overlooking part of the regs you want to use to support your argument. Per the link in the comment you cite, it's a human subject research if the research obtains data "through intervention or interaction with the individual, and uses, studies, or analyzes the information" That was clearly part of the research in this case: It involved interaction with individuals. I'm not sure how you can overlook this part of the definition when making your determination. You should also be aware that regulatory laws do not stand alone: The government provides explanatory statements of interpretation and policy guideline. The grand-daddy of these in this case and a fundamental guiding document to anyone in an IRB is the Belmont Report in 1979 that guided the development of modern regulated IRB's. It is pretty clear: participants should "undertake activities freely and with awareness of possible adverse consequence" It's understandable that plenty of folks here on HN are not familiar with the body of clarifications and case-law that that guide interpretation of the law's statutes, but researchers & especially IRB members are supposed to know this stuff inside and out. Next: Deception to avoid advanced consent is allowed in only very limited circumstances, and is a significant red flag that an IRB needs to bring more scrutiny to the research protocols. I don't know how you (or the IRB if it was doing its job properly) can say that human subjects were not involved if the research protocol relied on the deception of human subjects to see their behaviors. All of which is somewhat besides the point: This entire research ethics process, independent of specific statutes, exists to prevent adverse impacts on human subjects. This research had adverse impacts, and so defacto it involved humans in research that should have gone through the full IRB human subject oversight process.

belorn 5y ago

I wonder if IRB ruled in this way because of the assumption of algorithmic response for requests like DMCA take down notices. I can imagine that even for GDPR/CCPA requests, there is still no human involved for website like Google, facebook, youtube and other major sites that is primarily operated through automation. If there is no humans involved then there is no humans to have an adverse impact on. But as you said, researchers however must have suspected that responses would be made by humans or else the email would have included the fact that it was a study.

throw10920 5y ago

> Ethical guidelines on research exist to prevent an adverse impact on participants. Not relevant in this case, because (1) it's clearly not human research[1] and (2) all kinds of other clearly-non-human-experiments still have adverse effects on humans tangentially involved, so that's not unique to human research, either. When scientists were testing the first particle accelerators, they caused some people a lot of stress who were worried that they would destroy the world - does that mean that those tests were human experiments? (clearly not) [1] https://news.ycombinator.com/item?id=29656792 https://news.ycombinator.com/item?id=29656792

ineedasername 5y ago
I may be wrong, but I'm going to guess that you don't work with IRB's all that often. I do. I asked a colleague their thoughts on this and they were unequivocal in believing this should have been flagged as involving human subjects if the approving IRB had all of the details of the protocol. Their guess is that the proposal's presentation-- perhaps very innocently-- did not fully convey the details that would have resulted in oversight of the research as a human subject project. These are experts on the nuances of these laws. Of course, experts may disagree, so that reference is not definitive. But it is suggestive that dismissals from those here on HN that are arm-chairing this with no-- or minimal-- experience with this sort of research are not basing their opinions on a full understanding of the IRB & research ethics ecosystem. One or two IRB applications for a research project will not convey the understanding needed to evaluate research protocols under the laws & regulations involved. As for definitions of human subject, it seems like you are overlooking part of the regs you want to use to support your argument. Per the link in the comment you cite, it's a human subject research if the research obtains data "through intervention or interaction with the individual, and uses, studies, or analyzes the information" That was clearly part of the research in this case: It involved interaction with individuals. I'm not sure how you can overlook this part of the definition when making your determination. You should also be aware that regulatory laws do not stand alone: The government provides explanatory statements of interpretation and policy guideline. The grand-daddy of these in this case and a fundamental guiding document to anyone in an IRB is the Belmont Report in 1979 that guided the development of modern regulated IRB's. It is pretty clear: participants should "undertake activities freely and with awareness of possible adverse consequence" It's understandable that plenty of folks here on HN are not familiar with the body of clarifications and case-law that that guide interpretation of the law's statutes, but researchers & especially IRB members are supposed to know this stuff inside and out. Next: Deception to avoid advanced consent is allowed in only very limited circumstances, and is a significant red flag that an IRB needs to bring more scrutiny to the research protocols. I don't know how you (or the IRB if it was doing its job properly) can say that human subjects were not involved if the research protocol relied on the deception of human subjects to see their behaviors. All of which is somewhat besides the point: This entire research ethics process, independent of specific statutes, exists to prevent adverse impacts on human subjects. This research had adverse impacts, and so defacto it involved humans in research that should have gone through the full IRB human subject oversight process.