The Value of Information, as a Means to an End or an End Itself

This blog was written by Courntey Tolmie (R4D) and originally published for the Transparency and Accountability Initiative in September 2013.

If you ask any partner in any low- or middle-income country whether absenteeism or drug stockouts or quality of services are a problem, the answer you are likely to receive (with little hesitation) is ‘yes’. That response may be based on extensive quantitative analyses, but more likely it is just common knowledge.

This raises the question of what is the value of gathering information about service delivery and spending problems. If everyone knows that the problem already exists, then why do CSOs spend time and resources monitoring whether teachers show up to class or the satisfaction of a community with its health service or the quality of supplies available for social services?

We asked these questions of fifteen CSOs working on transparency and accountability worldwide. Specifically: Did you gather primary data from facilities or civil servants? And why gather this data? Why not just start with advocacy to improve accountability?

Was gathering primary data necessary?

Only one of the fifteen organizations that we spoke with chose not to gather primary data on service delivery and spending problems — and they had interesting reasons for making this decision to which we will return. The fourteen organizations that did collect primary data all reported that they felt it was necessary before beginning their projects, and that they still held this opinion after completing their work.

But their reasons for gathering primary data were quite diverse:

• Data was not available at the local level (India).
• The problems may be commonly known, but they are not quantified. And policymakers wanted hard data to back advocacy (Uganda).
• Everyone knows the problems, but there is not reliable information about the reasons for them (Argentina).
• Data may be available, but no one believes what is being put out by the government (India).
• There is information on regulations, but not on implementation (Indonesia).

This wide range of reasons for collecting data suggests two things. First, collecting primary data is at least perceived as a necessary part of any successful transparency and accountability intervention. And second, different ‘transparency problems’ lend themselves to different types of data collection and different theories of change. Even the same service delivery problem that lends itself to the same type of intervention (say, an absenteeism study) suggests different triggers, opportunities, and roadblocks for translating transparency into accountability — something critical to consider before proposing a one-size-fits-all solution to a common development problem.

What about existing data?

Despite these problems with government data, all of the partners also reported that they used secondary data. In some cases, the goal was triangulation — comparing what the government reported with what seemed to be happening on the ground. In other cases, the partners needed some initial information regarding disbursement channels or school rosters before primary data could be useful — or even feasible to collect.

How does all of this work with a transparent government? A closed government? A country with a Freedom of Information Act (FOIA)? With a sample of just fifteen organizations, it is hard to make any generalizations about access to information in different environments. But a few themes did emerge which practitioners and researchers and donors should consider when designing and undertaking these types of interventions.

First, we sometimes think about transparency as just the availability of information. But most of the organizations we spoke to referred (directly or implicitly) to the issues of push transparency versus pull transparency. Not one of the organizations we spoke to found that all of the data that they needed from the government had been pushed into the public sphere. Instead, CSOs had to use their contacts in government or employ FOIA requests to pull the information they needed from the government. While this may seem like a minor point, it is worth noting that these organizations are all credible organizations working at the national level. A community-based organization or a citizen group would face many more challenges in getting data, even in an open political environment.

Second, even in countries with Freedom of Information legislation, many partners chose either to use other channels to get access to government data, or to drop their requests altogether. We worked with one organization in southern Africa that employed a FOIA request at the beginning of a grant period, and eighteen months later was still waiting for a response. Organizations cited delays in data, questionable quality of data, and other issues as reasons why they needed to employ different strategies in their transparency and accountability work. These issues affect the assumptions and actions built into a CSO’s theory of change.

Then again, sometimes problems with Freedom of Information legislation compel an organization to shift its goals and its intended impact entirely. Of the fifteen organizations we interviewed, one decided not to use any primary data. This Indian CSO had seen in many past projects that community-based organizations did not have access to the information they needed to advocate for better spending on health and education. And so they decided to focus on one question — can communities get basic information on social sector spending and services to be able to better advocate for improvements? Without collecting any firsthand data, they documented what information they were able to get — and what they could not gain access to. The result of the project was valuable information about where the system was broken and trigger points where accountability work could begin to improve access to information.