If cities are to ‘leave no one behind’, disaggregated data is invaluable
At the annual United Nations Statistical Commission meeting next week in New York, the commission will formally adopt the Cape Town Global Action Plan for Sustainable Development Data — a first-of-its-kind document that pledges member states to improve their data capacity at all levels.
Adopting the plan is a milestone in the U. N.’s “data revolution”, a call to arms for governments and the development community to strengthen the use of development data to achieve the new Sustainable Development Goals.
Yet the data revolution isn’t entirely about creating more data. More specifically, it’s about all of the ways to find the right data at the right time for the right purpose. In this, a core pillar of the data revolution is the call for more nuanced and granular data — more information from the ground, hopefully leading to tailored development initiatives that can double down on what is working and change track when something isn’t.
The formal term for this is “disaggregated” data, and it’s a key concept highlighted in the Cape Town plan, which was finalized in January. Disaggregated data includes information on populations based on characteristics such as age, sex, income, race, ethnicity, migratory status or disability. It also means breaking down raw data by geographic location — a key consideration for the SDGs, given the centrality that cities will be playing in making progress on these goals.
Indeed, the issue is rising to the fore as countries and cities have begun work to implement the SDGs. Those 17 broad goals sit upon nearly 170 specific targets — reducing maternal mortality by a certain amount, for instance, or providing safe access to public space. Advocates warn that raw national statistics on related initiatives don’t tell planners and other officials enough about what’s working and what isn’t. For that, they need disaggregated data.
“If we really want to ensure the full implementation of the [SDGs] agenda, we really need to improve the data availability and use on all population groups that are covered by and addressed specifically in the targets,” said Francesca Perucci, assistant director of the U. N. Statistics Division. “Not only that — even beyond the groups specifically addressed in the targets, [we need to] really focus on the ones that are normally not counted.”
Having sound disaggregated data available and properly analyzed allows policymakers to focus in on certain groups, devising policies and allocating resources to meet their needs. It also aids in more-detailed monitoring of progress on the SDGs.
But there is still a lot of work to be done in terms of data disaggregation, especially when it comes to cities. Several events this month will be focusing on this issue; by April, many are hoping for increased clarity on how to move forward, although the process will take far longer.
Disaggregation around urban development is a particularly important consideration when it comes to the “indicators” that city officials and others will have to report on to monitor progress on each of the SDG targets. “The data disaggregation work stream within the SDG indicator work is the most important one,” said Perucci.
“Give the demographic health survey to national policymakers, and they’ve got lots of stuff they can draw on. Give it to a city mayor or city civil servant, and it’s useless.”
International Institute for Environment and Development
Because many of these targets touch on issues that governments have not previously had to monitor, debate is continuing on how to define many of these indicators. The issue will receive significant attention at next week’s U. N. Statistical Commission meeting, where the commission will consider proposed refinements to around 10 indicators. That meeting also is expected to result in a draft resolution on the indicator framework.
It also will be the sole focus of an expert group’s meeting later in the month. There a group known as the Inter-Agency and Expert Group on SDG Indicators (IAEG) is slated to present its work plan on data disaggregation. Finally, a meeting of the World Council on City Data, in Dubai next week, also will address the role of city-level data in informing national policies, although this is outside the U. N. ambit.
All of this action underscores the concern that there are serious gaps in the SDG indicators. The United Nations has compiled available national data, which it uses to inform the annual “Progress Towards the Sustainable Development Goals” report, into an online tool, the SDG Indicators Global Database. According to that database, there is a serious paucity of disaggregated data available for the SDG indicators.
For many indicators, there is no disaggregated data available at all. This is the case, for instance, with Indicator 11.1.1, which addresses the proportion of the urban population living in slums, informal settlements or inadequate housing. The same goes for Indicator 7.1.1, on the proportion of population with access to electricity.
Other indicators have data that is disaggregated for just one or two variables. For instance, Indicator 6.1.1 — which looks at the proportion of the population using safely managed drinking water services — differentiates between rural and urban populations, but it doesn’t break those down into age groups or sex.
This means that, based on the available data for those indicators, policymakers or development researchers would not be able to identify or monitor the needs of specific demographics, such as girls of school-going age living in informal settlements, or the proportion of elderly people with access to electricity or safe drinking water.
In an analysis of indicators “meant to be representative of the bare minimum of ‘leaving no one behind,’” researchers at the Center for Global Development (CGD), a Washington think tank, found that none of them have data disaggregated by income, race, ethnicity, migratory status or disability status.
And even the “better performing” indicators, those linked to food security and unemployment rates, at best have data disaggregated across two metrics, the researchers note.
In the eyes of many, that is a problem. The more policymakers can understand about the lived experience of citizens of varying sex, age, disability status and location, the more they can do to ensure the vulnerable are not left behind and tailor policy solutions for particular demographics.
“It’s really important to underscore the fact that the intersection of these different forms of disaggregation will then compound to be even more relevant when in combination,” said Megan O’Donnell, who co-authored the CGD analysis and now works for the ONE Campaign on gender issues.
“It’s all well and good to have an understanding of the differences between the urban and the rural,” O’Donnell said. “It will be much more effective, and even essential for good policymaking, to understand the divides between high-income and low-income populations within urban and rural populations.”
Data as ‘public good’
One of the major concerns around disaggregation is that much of the available data is based on national surveys, which don’t provide detailed local-level data for cities — even as cities are at the forefront of the sustainable development agenda.
“It’s really important to underscore the fact that the intersection of these different forms of disaggregation will then compound to be even more relevant when in combination.”
“If you go through the SDGs and all their suggested indicators, for urban areas a very high proportion are the responsibility of city and municipal government,” says David Satterthwaite, a senior fellow with the International Institute for Environment and Development’s human settlements group.
But, Satterthwaite points out, data collected at the national level doesn’t tell city authorities where the issues are in their specific territories or help them identify where to act at the street or ward level.
National sample surveys, like those done by national statistical offices in Africa, Asia and Latin America, and the Demographic and Health Surveys, are limited in terms of what they can produce in terms of disaggregated data, says Satterthwaite.
“Give the demographic health survey to national policymakers, and they’ve got lots of stuff they can draw on,” he said. “Give it to a city mayor or city civil servant, and it’s useless.”
In theory, censuses can provide data on housing, sanitation and water disaggregated to the street level. But typically, census authorities often don’t allow local governments access to the data they need — although this has changed in many Latin American countries, where census authorities provide detailed data to local governments, Satterthwaite says, quoting an official from Brazil who referred to census data as a “public good.”
As the United Nations and others continue to grapple with the issue of data disaggregation, it’s clear that improved data capacity will, in part, be contingent on partnerships between national statistics systems and other data producers. That includes those in local government, the private sector, academia and civil society.
There is a need to improve data at the local and national level, which means relying on administrative data and improving household surveys, along with improved reporting on municipal data, says the U. N.’s Perucci.
Using what’s there
Advocates such as Satterthwaite are calling for increased recognition of data needs at the local level. “All the people promoting the SDGs are going to have to think about how [the goals are] going to serve and support local action,” he said.
To do that, local governments can mesh locally collected data with census data from national authorities, like is being done in some of the better-resourced municipalities in Latin America, Satterthwaite explained.
To plug data gaps on informal settlements, authorities can make use of data collected by grass-roots federations of slum dwellers, such as those that are part of the umbrella organization Shack/Slum Dwellers International. The group does surveys on informal settlements to provide detailed data on populations that lack access to infrastructure and services, he says.
Improving the collection and use of disaggregated development data won’t come cheap, however. It’s going to take money and human resources.
“The key is to try use the data that are already available, and try to merge and join up the datasets that already exist,” says Perucci.
Additionally, it will be important for governments to re-address their priorities and make adjustments to current data tools, such as household surveys, to incorporate the new areas and groups that require data under the SDGs, she says.
To help tackle the issue of disaggregation, U. N. actors, national statistical offices, civil society groups, and the private sector need to come together and clearly lay out indicator definitions and priorities, said the ONE Campaign’s O’Donnell.
“Rather than being intimidated and overly daunted by this massive undertaking, prioritization needs to happen,” she said. “And that’s okay and necessary, so long as it’s carried out in a systematic way — in a well-defined, careful manner.”