Is it time to move on from inclusive data?

Mothers in a listening group sharing as part of a citizen-generated data project on antimicrobial resistance in Kilifi, Kenya. Credit: Elphas Ngugi.

“We’ve been talking about ‘leaving no one behind’ and inclusive data for nearly a decade. Is it time to move beyond inclusion?”

The data for good community has talked about inclusive data as necessary to achieving the UN Sustainable Development Goals (SDGs) since their inception. It’s easy to think we’ve done enough, but the most recent SDG progress report paints a picture of why inclusion should still be on the agenda: Only 30 percent of countries are on track to end poverty (SDG 1) by 2030. Hunger has increased to levels not seen since 2005. Achieving gender equality is some 300 years away. At the same time, inequalities are at a record high and growing. Just 26 people have the same wealth as half of the world’s population. How will we address these inequities and others unless we have data to identify them? 

As we have been reflecting on what’s next for the Inclusive Data Charter, I’ve also found myself thinking about how inclusive data is analogous to governance. Good governance promotes equity and social justice by expanding representation and promoting equal access to opportunities. Likewise, inclusive data is a tool to identify and address inequalities and promote social justice. It helps reveal disparities, identify the needs of marginalized populations, and inform targeted interventions to ensure equitable development outcomes. 

In 2015, the data for the development community was characterized by broad optimism that innovation and data-driven development would unleash prosperity and opportunity. Better data would lead to better and more targeted services while disaggregation and filling data gaps would propel the Leave No One Behind agenda. Fast forward to 2023: with the power of hindsight and reflecting on what we now know, we need to think deeply about how we move ahead. In this post, I share some key considerations for what inclusive data should be about and how we can chart a path toward it.  

Inclusive data is not just about representation and data disaggregation.

The Leave No One Behind agenda has advanced efforts to identify inequality and discrimination through generating evidence and collecting and disaggregating data. But disaggregating data by sex, disability status, and other identities that affect people’s access to opportunity is only a first step; it’s not sufficient on its own. Disaggregation can’t improve the visibility of those who are excluded from data. It’s also impossible to disaggregate datasets by every relevant dimension, meaning some inequalities remain invisible. Whoever decides what data or what form of disaggregation is prioritized ultimately has power over which forms of disparities receive more attention over others. 

Inevitably, we are making choices about what to count, and how to count it. Inclusion is about working with communities and experts across the entire data value chain. It’s about taking an intersectional approach by analyzing how intersecting identities are captured in data and how societal power structures shape whether and how people are represented in data. It is about determining with people: what the data priorities are, how the data should be produced, who is best placed to produce the data, how the data is presented, what decisions are required from the findings, and how to implement those decisions. Inclusion cuts across the entire data value chain and thrives on effective collaboration. 

There’s no such thing as inclusive data without collaboration. 

Collaboration among stakeholders ensures exchange of data, knowledge, and expertise necessary for data to benefit all. Positioning inclusive data to cut across the entire data value chain means that no one stakeholder can do it alone. It can only succeed with effective collaboration. Effective collaboration starts with a comprehensive understanding of the data needs of a population group, geography, or community. It then moves to a process where the responsibility for data design, production, collection, analysis, and use does not sit only with a specific institution or stakeholder but is a shared responsibility. The entire process is co-designed, or co-created, with stakeholders responding to the needs of the people for whom the data is produced. This makes data fit for purpose. In the end, effective collaboration ensures that data gives people dignity and power, rather than seeing them as data subjects. 

This is why it’s also important to acknowledge the practical trade-offs of inclusive data. It goes without saying that the number of people who can directly participate in decisions is directly related to the time and monetary constraints associated with data production. My colleague Kate Richards wrote an excellent post with recommendations for balancing these trade-offs here. Her point is that our aim should be to move toward inclusive practices and not to see inclusive data as an end in itself. The practical challenges of including people in decisions that affect them shouldn’t deter us from taking steps to inject inclusive measures throughout the data value chain.

Trust is a building block to making inclusive data accessible. 

The Leave No One Behind agenda means that, where data does not exist, efforts should go toward finding ways to close the data gaps. To be useful, though, existing data should be both available and accessible. Data for public good should be freely available without unnecessary restrictions and barriers and should adhere to legal and ethical standards. This is far from the case in most countries. For example, the most recent Global Data Barometer measured the extent to which countries have governance frameworks to promote accessibility of data and found that, out of the 109 countries surveyed, only 17 had specific provisions relating to data. Regionally, the Middle East and North Africa region and sub-Saharan Africa had the lowest scores. 

Trust is a key ingredient to ensuring that data is useful and to making data available and accessible. Collecting data about or from communities that have faced historic marginalization requires establishing trust through collaboration. For decision-makers to use data, they must trust in its validity and reliability. Likewise, the public must be able to trust, not only the data, but also in the credibility of the data producers and in public institutions and decision makers to use their data in constructive, responsible, and transparent ways.

It’s time to move toward Inclusion 2.0.

Going beyond data disaggregation to inclusive data practices, seeking meaningful collaboration, and creating trustworthy, open data systems are concrete ways we can take the Leave No One Behind agenda to the next level. 

But how do we put these priorities into practice? As a first step, we must commit to truly inclusive data approaches. We must stand ready to look at our data production processes, starting from the data gaps and design stage and identifying the forms of exclusion and work towards producing the needed data. We must also be ready to examine existing data processes and find mechanisms to meaningfully include people, beyond only being respondents to data collection efforts. 

Lastly, inclusive approaches to data should be everybody’s aspiration—whether data is your key focus or not. In the data for development community, we’re accountable to people for how data activities affect lives. So, in answer to the question I proposed at the start of this post, we are ready to move forward with—not on from—the Leave No One Behind agenda.

Thanks for reading The Data Values Digest! Subscribe for free to receive new posts and support our work.

Related Articles

Responses

Your email address will not be published. Required fields are marked *