This post from Savage Minds, now several weeks old, has an interesting discussion of trying to get ethnographic research approved by institutional review boards. The comments are a particularly good read. At Berkeley we certainly have our share of issues with IRB, but at this point I think they're more about institutional process than about ethnography in particular. I find that this is one of those frustrating situations where I can see both sides of the story. On the one hand, the University needs to protect itself from liability – its interest is certainly in protecting research participants, but perhaps moreso avoiding the consequences of our country's litigous ways. And who can blame them? The world is full of irresponsible researchers.

But on the other hand, many of the requirements and restrictions of IRB review and informed consent are so cumbersome and innapropriate for ethnographic practice that, even when researchers are diligent about the review process, they inevitably get tossed out the window once approval is in hand and research begins. IRBs aren't doing anyone any favors by requiring social scientists to abide by practices that, though they are more general now, are based on the needs of biomedical and psychological research. They might actually be making things more dangerous for participants because they are encouraging researchers to design protocols that they can't possibly follow.

During a recent fight with Berkeley's CPHS, which, thankfully, turned out well, I basically said to my representative, as I explained how ethnography works (again): 'How could it be that you don't have a policy for dealing with this?' I never got a response, but I think it's more critical than ever, especially now that many folks in the IT & design worlds are adopting qualitative methods. Here are a bunch of folks who are not used to thinking about the ethical obligations to research participants in the way that dedicated social scientists are, if only because the focus of their education has been about devices and not people (although this is changing). The ones I encounter mean well, but are often surprised that human subjects protection is something they need to deal with. My feeling is that must choose not to deal with it. Another CPHS representative at Berkeley said that he is positive that most people don't submit protocols for their research, because if they did, the office would be so swamped in proposals they couldn't get in the front door.