IRBs in development? Accountability mechanisms matter

Yesterday James Fearon asked an interesting question: “Why aren’t there IRBs for the development industry?” Academic researchers must answer to Institutional Review Boards (IRBs) on questions related to ethical research practice and treatment of human research subjects. Fearon wonders why there is no equivalent for development projects, which can have negative impacts on beneficiaries in the same way that academic research can. The question applies generally to a development organization’s work, and also specifically to the research that a development organization conducts as part of its projects (e.g. needs assessments, program evaluations, etc.) which would require IRB approval if it were conducted by an academic.

I’ve never dealt with an IRB, but Fearon got me thinking about accountability mechanisms.

I used to work for a company doing management/strategy consulting for universities. We spent a lot of time thinking about how to create accountability in those peculiar organizations. University leaders face an odd mix of constraints: external pressure from parents, alumni and legislators; internal pressure from employees who can never be fired (tenured professors); massive capital planning needs and ongoing operating expenses; and the need to at least break even while serving a social mission.

Within that environment, the primary incentive facing a professor is to conduct research, generate new knowledge, and publish papers. Note that this can all be achieved without any respect for the well-being or rights of human research subjects. Most researchers are guided by their own moral compass, but abuses like the Tuskegee syphilis experiment made it clear that a more formal check was needed. Hence the creation of the IRB, a mechanism to ensure that researchers consider the well-being of the subjects.

With that in mind, I’d like to re-phrase Fearon’s question:

Does the development industry need mechanisms to ensure that projects and programs respect the rights and well-being of the intended beneficiaries?

My equivocal answer: yes, and no, and also yes again. Let’s take each of those in turn.

  • Yes, in the sense that accountability mechanisms are needed. To demonstrate that principle, just look at Greg Mortenson’s Central Asia Institute. Good intentions are no substitute for shady accounting and a compromised board of directors. The accountability mechanisms within an organization should be structurally and strategically aligned to promote the mission.
  • But on the other hand, development organizations don’t need a separate accountability mechanism such as an IRB. Academic institutions use the IRB because no other institutional mechanism exists to ensure that researchers are concerned with a human subject’s well-being. Benefiting the subject just isn’t what a university is designed to do. The opposite is true of a development organization. The whole organization exists to promote the well-being and rights of the ultimate beneficiaries (or, if you don’t like the connotations of that term, partners). The strategic planning processes, staff hiring criteria, log-frames — all of these are designed to ensure that the organization’s activities benefit those being served. That’s the difference between being a research subject and a beneficiary.
  • At least, that’s the idea. In practice, the existing mechanisms could be so much better. Too often they are hijacked by fundraising concerns or donor-country politics. Or even worse, they simply fall victim to bad management.

Ultimately, I think Fearon was right when he wrote that development programs do face mechanisms, but that they are insufficient, especially in the face of the complicated political and social impacts of aid projects. I don’t think the answer is to introduce something equivalent or even analogous to IRBs. Even when it comes to research conducted as part of development programs, an IRB-type mechanism doesn’t make sense. The accountability mechanism of the IRB is appropriate for academic research because of the incentives that researchers face, not because of the activity of research itself. Development research conducted within an organization should face the same mechanisms as the organization’s other activities.

The big takeaway from all of this: Let’s keep talking about the accountability mechanisms and how they relate to organizational strategy, structure and mission. The development industry’s discourse focuses too heavily on technical solutions, while some of our biggest obstacles are managerial.

_______________

P.S. Despite my lack of experience with IRBs, I was once part of a research project at a university. It involved antimatter asymmetry. I couldn’t even tell you what that means anymore, but here’s the proof (second row, on the left).