Sunday, July 23, 2006

Student Research

I thought I'd share my thoughts on student research since that's a topic that vexes IRBs, especially at universities. Most IRBs only review student research if it contributes to "generalizable knowledge" and, since most student research does not, most student research is not reviewed by IRBs. I believe that all student research, including undergraduate class projects, needs oversight to ensure that the rights and welfare of the subjects are adequately protected. I say this for two reasons. First, it doesn't matter to the subjects whether the person conducting the research is a student or a faculty member; their rights and welfare could still be at jeopardy. Second, the students lack of experience puts subjects at greater risk than faculty research.

Let me share an anecdote with you that illustrates my point. One day, when I was at Albany, a group of us were eating lunch on campus when a student came up to us and asked us if we'd answer some questions for a survey her class was conducting. Of course, I asked her if it had been approved by the IRB and who was her instructor (she had certainly come to the wrong table!). While she was standing there, another student, clearly from the same class, came over and asked her, "Do you know Joe over at the information desk? What until you hear what he said on the survey!" Now this was a fairly innocuous survey, but Joe's privacy had still be violated.

So, all student research needs ethical oversight. Can't the instructors provide that oversight since they are responsible for the research? I don't think so because they have a conflict of interest. In fact, they have two conflicts. The first conflict is a pedagogical one. They have an educational goal in assigning the research project and part of accomplishing that goal is keeping the students engaged. They might not want to limit the students' research for fear of discouraging them. The second conflict is temporal. They have an interest in getting the students done within a limited amount of time and would be disinclined to do anything which holds the student up. Like all human subjects research, all student research needs an independent, objective review.

I can already hear IRB people moaning how it's impossible for the IRB to take on that burden. Well, I can tell you that it doesn't have to been such a burden. This was our policy at Albany for the 20 years that I ran the IRB there and we were able to do it with a reasonable amount of work. First of all, we strongly encouraged all students to limit their research to exempt or expedited categories. We didn't require it, but told them that, if their research needed full review, they might not be able to complete the project on time. Second, we developed a three-tier system which helped facilitate the process. For projects where all of the students did the same thing or picked from a list of set projects, we had the instructor submit a protocol which detailed the project. This could be done before the class began. For projects where the students where designing their own research, either individually or in groups, we developed a one-page form for students to fill out which described their research. They submitted those along with any instruments they were using to the instructor who submitted them in an "umbrella" protocol for the whole class. Students conducting research which did meet the criterion of "generalizablity" or whose research did not qualify for exemption or expedited review had to submit their own protocols. I reviewed all of the student research and, even for classes in the second category with 75 or 100 students, I could review them in a few minutes. When reviewing those packets, we didn't hold up the whole class if some students needed more review. So, the burden is primarily on the expedited reviewer(s) and isn't a great burden at all.

Now, the good news is that, since student research doesn't meet the definition of research in the federal regulations, the detailed requirements of the regulations don't apply and institutions have a lot more flexibility in how they oversee student research. So, it doesn't have to be the IRB that does the review. An institution could use some sort of departmental review (as long as the reviewers are knowledgeable about human subjects issues) or, me favorite idea, a Student IRB. A Student IRB could be made up of advanced students, trained in human research protections, with an advisor from the IRB or the IRB office. This would be true peer review and would give the students on the IRB good experience and something valuable to put on their resumes. Actually, I think student reviewers would be tougher than the regular IRB. The University of Minnesota has a Student IRB, but I don't think it reviews class projects.

So, in conclusion, I think all student research needs ethical oversight. IRBs can provide this oversight without an undue burden, but institutions have other options for providing this oversight.