Posted college tuition has been on the rise in recent decades. There is a lot of debate about the the cause of rising sticker prices, but one factor that is often blamed is the increasing number of administrators. Less well appreciated are the factors causing the increasing number of administrators. One driver is the fact that more staff are needed to comply with increasing regulations and accountability imposed on universities from state and federal governments.
One example that affects my area of research most directly is human subjects research committees (or the so-called internal review board, IRB). Early in my career, if I wanted to survey food consumers or run an experiment in a grocery store, I just did it. I didn't have to ask permission or get approval from a university administrator. But, somewhere along the way, the federal government required universities wishing to receive federal monies to have projects approved by local IRBs. Now, all major universities have their own IRBs with various sized staffs and with faculty spending time chairing and serving on IRB committees (full disclosure: I've served as an alternative member of OSU's IRB for several years).
Some of the basic ideas behind IRB approval process are worthwhile: making sure people aren't being unduly coerced and are participating in research voluntarily, making sure research participants' information isn't being used in a way that embarrasses the participant or can be used against them in some way (i.e., protects confidentiality), and making sure the research doesn't generate undue risks for participants in research that are incommensurate with the benefits.
All that said, from a researcher's perspective, all this can be a real pain even for the most minor of surveys. Surveys that are anonymous are technically exempt from IRB approval, but the researcher doesn't have the ability to make that determination: they have to fill out a long form, send the research instrument, including information on participant recruiting, etc. to an IRB committee and wait for them make the determination (before all this, the researcher has to undergo a training on human subjects research and pass several tests). And wait you will. I've heard stories from colleagues having to wait several months for an IRB determination. And when you hear back, you often are asked to make changes to your research design that have little to do with the aforementioned purposes of the IRBs. If you're trying to do a survey on a current policy issue, you've now waited weeks or months for approval, and even if a project is approved, if you want to reword a question or add a new one to address evolving events, now you have to submit a change modification form that also has to be approved. Given these timing issues, it has become next to impossible, for example, to have graduate students do publishable quality surveys/experiments for class projects.
I've largely had positive experiences with IRBs (I've had a couple bad ones too), but one shouldn't mistake the cost this imposes on researchers, on the university, and ultimately the taxpayer and student. Whether the benefits of the system exceeds these costs is a question I've never seen seriously addressed.
Change is afoot. This is from an article by Richard Shweder and Richard Nisbett in the Chronicle of Higher Education back in March:
I suggest reading the whole thing. The authors provide some history of these programs and passionately convey their frustration with the present system. They also note that universities have till next year to figure out how to address the changes in federal regulations.
Another article in the New York Times is more critical of the changes and is less optimistic that real changes will occur for human subjects research. However, I'm beginning to hear rumblings at a few universities that they will not longer require prior approval from IRB for certain types of human subjects research. The end of the NYT article suggests what some of this is about: absent federal guidelines universities may still want to review research to reduce the risk of controversy, embarrassment, or lawsuits. Those are legitimate concerns but are likely to run up against issues of academic freedom and freedom of inquiry. And, they are concerns that are distinct from protecting participants of human subjects.