We like facts. Anyone who follows our work in the Global Development and Population Program knows that many of our grant portfolios are heavily weighted toward collecting, analyzing, and using information. Our largest grants, just as examples, are to support policy research organizations in South Asia, Latin America, and Africa, and to fund the collection and sharing of information about the quality and impact of education and health services.

We like facts because we think they can help people who are in decision-making positions—government officials, politicians, donors—understand the nature and magnitude of social and economic problems that are beyond what they can observe first-hand. And we like evidence about how governments are responding to those problems because we think citizens have a right to be informed so that they can take corrective action, while those in government use the evidence to improve public programs and services. We also like facts because we like to be surprised, to have our assumptions and beliefs challenged. And the “what the world really looks like” information is often quite surprising indeed.

Liking facts and supporting organizations that share this focus does not mean we (or those organizations) are naïve. We know that facts are constructed within a social context; we know, for example, that it takes a concerted effort to overcome gender and other types of bias that are embedded in how data are collected. We also know the challenges of connecting data and evidence to policy debates, whether in politically charged environments like the United States or in countries where “science” is abused as a tool to promote wrongheaded policies. (Look no further than Ugandan President Museveni’s remarks upon signing the terrible anti-gay bill into law.) Like many others in the “research to policy” arena, we understand that chances for policies to be affected by science are few and far between. Those with cold, hard facts have to invite themselves to the party rather than waiting for politicians to seek their wisdom.

One way to find those opportunities to affect policy is to deconstruct the levels of “policy” and find the levels in any particular context where even a tiny little opening exists. That opening may already exist within the public sector, or it may be created through citizen awareness and action.

At the top level, we can think about “big P” policies like the creation of a new cash transfer program, the removal of school fees, the commitment to disclose information about public budgets and contracts, and other actions that might appear in a newspaper headline. Others might argue, but I think the main role facts play at this level is as part of the advocacy artillery, selectively used to back up a pre-established pro or con position.

Below that level, however, is a cascade of implementation decisions, taken variously by Ministers, technical personnel in ministries, operational managers, and front-line workers. Regardless of whether governments function in a centralized, decentralized, or chaotic way, choices made at each of these levels—“small p” policies and practices—can be better when facts are marshalled to support decision-making. They can also be better when citizens become more knowledgeable about what decisions have been taken and what their impact is. For a cash transfer program, for example, the nitty-gritty design and implementation choices—for whom, how often, with what conditions (if any), through what type of banking system, and so on—are amenable to the knowledge accrued from similar programs elsewhere, as well as from tailored information-gathering, studies, and periodic evaluation. Information about how the targeting was supposed to occur, how efficiently the public sector is delivering on its promises, and who is responsible for taking action, when made available to citizen and civil society organizations, can play an important role in fostering greater accountability.

While some combination of social values and raw political ambition may drive virtually all “big P” policies, the levels underneath have a lot to do with what happens after the headlines fade. And those levels are often far more open to being influenced, directly and indirectly, by empirical information.

If we were only reading the headlines, it would be easy to become cynical about the notion that research is ever used in making policy. But when we see how many of the organizations we support are successfully finding a low profile but high impact route, we are happy to be able to be part of the story. And that’s a fact.