Data Governance
Intersectional approaches are vital in African data governance because overlapping identities (e.g., gender, age, location) create exacerbated barriers. Inclusivity in data systems is a human rights imperative, especially for Africa’s “under-sampled” communities. Policy must reflect the continent’s diverse realities, consider its colonial legacy, avoid importing non-African frameworks, and align with standards set by bodies like the African Commission on Human and Peoples’ Rights.
Policy Development in Data Governance without Intersectionality
While intersectionality is widely recognised as important for addressing social inequalities, translating it into policy practice remains difficult. Scholars note that intersectionality’s conceptual complexity and relatively recent adoption in policy settings present ongoing challenges for decision makers.9
The following section outlines specific systemic barriers that have been cited as making adopting an intersectional approach difficult.
Nubian Identity in Kenya
The Nubian community in Kenya, descendants of Africans forced to migrate to Kenya by the British colonial administration over a century ago, has long faced exclusion from citizenship and its associated rights. Because of their ethnic and religious origins, Nubians are subject to a unique, burdensome vetting process to obtain the national ID card required for full participation in Kenyan life. Most are unable to secure this documentation, leaving them effectively stateless.
Without an ID card, Nubians cannot equally access public education, health care, formal employment, land ownership, or even vote. The government’s refusal to recognise their property rights in Kibera, where they have lived for generations, further entrenches their marginalisation, as does the lack of public services provided to their community. The African Commission on Human and Peoples’ Rights found that the State’s practices violated the African Charter, citing discrimination, arbitrary deprivation of nationality, and the prohibition of statelessness.10
The Digital Divide for Youth with Disabilities in KwaZulu-Natal, South Africa
A recent study in KwaZulu-Natal, South Africa, surveyed over 3,000 people with disabilities and found that the digital divide is especially severe for young people with disabilities, particularly those in rural areas and those with non-visible or psychosocial disabilities. Only 41% could use digital devices independently, and 43% had no access at all. Barriers included inaccessible devices, high costs, limited training, and persistent stigma. The study also found that many students with disabilities remain excluded from education and digital services due to a lack of assistive technology, accessible infrastructure, and affordable connectivity.11
These intersecting barriers make it harder for young people with disabilities to participate in learning, work, and society, and their needs are often invisible in policy and data. In practice, this can mean being denied the ability to participate fully in education systems, decision-making, and development opportunities, especially in contexts where data systems don’t track the intersection of these identities.12
Policy Development with Intersectionality
What if the true strength of public policy lay not in uniformity, but in its ability to navigate the complex realities of diverse lives? In Africa, where histories, cultures, and identities are deeply layered, the task of crafting effective data and technology policies requires more than ticking the boxes of inclusion. It calls for a fundamentally different approach.
Viable Approaches to Intersectional Data Governance
As explored previously, intersectionality is multifaceted, meaning there is no single way to apply it. However, moving from theory to practice requires specific analytical frameworks that go beyond looking at identity categories in isolation. Instead, they present different perspectives to examine how these factors interact and compound one another, sometimes creating new forms of disadvantage.
For example, as noted in a recent data systems guide, when disability is recorded as a single category, data systems may overlook people who experience both non-visible disabilities and additional challenges linked to their age, place of residence, or migration background.13
Rather than treating characteristics like gender, age, disability, or geography as independent variables, intersectionality sees them as deeply intertwined. In policy development, this lens reveals how power, privilege, and oppression intersect, shaping who is counted and who benefits from change. In short, keeping intersectionality at the center of policy development changes:
- How we define problems (asking “who is missing or left out?”)
- How we design solutions (testing them for multiple, overlapping barriers)
- How we evaluate results (checking for unintended consequences across different identities)
To operationalise this, policymakers can adopt three complementary frameworks:
Multi-Dimensional Approach 17
Taking into consideration different layers of intersectionality. Unlike approaches that focus on a single strand of identity (e.g., only gender or only race), the multi-dimensional perspective recognizes that inequality operates at different levels simultaneously. Drawing on the framework by Jubany, Güell, and Davis (2011), this approach requires analyzing data governance across three distinct levels:
- The Micro Level (Individual): How individuals subjectively experience their multiple identities and potential discrimination.
- The Meso Level (Institutional/Community): How institutions (schools, hospitals, local government) and administrative practices either accommodate or exclude specific groups.
- The Macro Level (Structural): How broader laws, cultural norms, and national policies create the conditions for inequality.
How to apply a Multi-Dimensional approach
To implement a multi-dimensional strategy, data governance policies must connect these levels rather than treating them separately. Policymakers should ask:
- Micro: Does the data allow individuals to self-identify with multiple characteristics?
- Meso: Do administrative systems create barriers for specific combinations of identities?
- Macro: Do national privacy laws or data standards inadvertently harm specific subgroups identified at the micro level?
Example in the context of Data Governance
In the context of data governance, a multi-dimensional intersectional approach means designing data systems and policies that can capture, analyse, and respond to the many, overlapping ways people can be marginalised.
Space-Based Approach 18
Treating “place and context” as part of the analysis. Inequality is rarely geographically neutral. A Space-Based approach treats “place”19 not just as a backdrop, but as an active component of disadvantage.
This method pushes analysts to examine how power, infrastructure, and inclusion are distributed unevenly across different environments, such as urban centers, rural border zones, informal settlements, or conflict-affected regions.
How to Apply This Approach | The Spatial Audit
- The Assumption Test: What infrastructure does this policy assume exists everywhere?
- The Visibility Test: Who becomes visible in these specific spaces, and who disappears?
- The Agency Test: Who has the power to define “success” in this location?
Example in the context of Data Governance
A digital ID rollout may “work” in capital cities (registration centers, documents, connectivity) but systematically fail in informal settlements or rural border zones, so the policy’s spatial assumptions become exclusion mechanisms.
Policy Process Approach 20
Developed by Bishwakarma, Hunt, and Zajicek,21 this systematic approach integrates intersectionality throughout the typical policy cycle. It operates on the premise that both governmental and non-governmental organizations must incorporate intersectionality at every stage of policymaking, not just at the end.
Instead of treating inclusion as a retrospective checklist, this method demands asking intersectional questions at four phases: Agenda Setting, Formulation, Implementation, and Evaluation.
How to apply this approach
The Goal: Identify whose problems are prioritized.
Intersectional Question: Is the problem experienced differently by different groups? Who is defining the “crisis” or “need”? (e.g., Is “safety” defined by property owners or by those living in informal settlements?)
The Goal: Construct inclusive solutions.
Intersectional Question: Who is at the drafting table? Are we relying on data proxies that fail to capture specific populations (e.g., using “household” surveys that miss unhoused populations)?
The Goal: Ensure equitable access in practice.
Intersectional Question: Do the delivery mechanisms (digital portals, physical offices) create barriers for intersectional groups (e.g., women with limited mobility in rural areas)?
The Goal: Measure impact across identities.
Intersectional Question: Are we measuring outcomes for specific intersecting groups, or only reporting aggregate success?
Example in the context of Data Governance
Consider a government launching an “Open Data Portal.” In the Standard Approach, the portal is simply launched, and success is measured by “total hits.” While the aggregate numbers might appear high, they hide a critical failure: the data is mostly being accessed by those living in urban areas with high-speed internet, leaving the most vulnerable populations out. In contrast, applying a Policy Process Approach transforms the initiative entirely. Before a single line of code is written, for instance, civil society groups representing rural women are consulted to define what data they actually need to hold local officials accountable. Recognising the intersection of poverty and geography, the team designs the interface specifically for low-bandwidth mobile devices rather than high-end desktops. Finally, success is no longer centered around total traffic; instead, metrics specifically track usage rates from marginalised regions. If usage in those areas is low, it is flagged as a system failure rather than a lack of interest, triggering future improvements.
REFERENCES
1. Bishwakarma, R., Hunt, J., & Zajicek, A., “Intersectionality in the Policy Process,” 2023.
From theory to practice
To move from theory to practice, it helps to see how intersectional thinking changes outcomes on the ground. The following two case studies illustrate how change makers across the continent of Africa have begun integrating intersectionality into real-world data and policy solutions.
Collection of Data through an Intersectional Lens
When Morocco set out to update national statistics on violence against women and girls (VAWG), intersectionality was woven directly into the survey process. Rather than limiting the data to prevalence rates, government actors worked hand-in-hand with civil society organizations to broaden the scope. They used new survey modules to look at how violence affects not only survivors but also their families, and the costs they bear.
To gather this nuanced data, CSOs specialising in gender-based violence played a leading role in training enumerators. These trainers equipped data collectors with methods for asking sensitive questions, like helping interviewees recall traumatic experiences in a safe and respectful manner, applying ethical protocols, and providing direct referrals to support services. During data collection itself, women’s networks and advocacy groups were present as “listeners,” working alongside official survey teams. Their involvement did not just safeguard the wellbeing of respondents, it improved the quality of responses, as enumerators could introduce questions about violence more carefully and pick up on subtle cues.
By involving local actors throughout the process, the survey design better reflected the lived realities of diverse groups of women in Morocco. The resulting dataset captured layers of experience that standard surveys often overlook, giving policymakers a more grounded foundation for designing support services and prevention programmes. This collaborative model shows how approaching data collection with an intersectional approach, linking gender, context, and social support systems, can produce data that is richer and more relevant for real-world decisions.
Example adapted from Open Data Watch & Data2X (2023)15
Digital Pathways for Young Women in Kenya (Ajira Digital)
Kenya’s Ajira Digital Program was developed to expand opportunities for young people in the digital economy and to address the low representation of women in STEM and technology spaces. Through AjiraForShe, the programme focuses on the different barriers that shape women’s participation, including caregiving demands, cultural expectations, financial constraints, and limits in access to digital tools or safe learning environments.
Training hubs such as Madrasa-Tul-Falaah in Kibera and the Swahili Port Hub play an important role by offering free digital skills training in settings that reflect the cultural and social realities of the women who participate. These hubs work with community networks to reach women and girls who might not otherwise join formal training programmes. Participants learn practical skills in areas such as transcription, virtual assistance, content creation, and web design, while also gaining confidence to navigate online work.
One example shared in the programme’s documentation highlights a young woman who used AjiraForShe training to transition into paid digital work, support her household, and later mentor other women. Her story mirrors broader trends in which trainees report multiple sources of income, greater financial independence, and improved stability in their personal and professional lives.
This case shows how digital initiatives become stronger when they recognise that women’s experiences are shaped by intersecting identities such as gender, culture, location, and socioeconomic status. By working with these factors rather than treating women as a single category, the programme opens pathways for groups who are often overlooked in mainstream digital policy and skills development.
Adapted from two sources: “Transforming Lives Through Digital Empowerment,” 202422; and Pollicy’s Gender Data Futures Handbook (2025)23
6 Open Data Watch & Data2X. (2023). Integrating Intersectionality in Data Systems: A Practical Guide Across the Data Value Chain. https://opendatawatch.com/publications/integrating-intersectionality-in-data-systems/
7 Effoduh, J. O., Akpudo, U. E., & Kong, J. D. (2024). Toward a trustworthy and inclusive data governance policy for the use of artificial intelligence in Africa. Data & Policy, 6, e34. https://drive.google.com/file/d/15ahdi4jLHG2eBzEdmqGlGfQkt3AJEKIj/view
8 Effoduh, J. O., Akpudo, U. E., & Kong, J. D. (2024). Toward a trustworthy and inclusive data governance policy for the use of artificial intelligence in Africa. Data & Policy, 6, e34. https://drive.google.com/file/d/15ahdi4jLHG2eBzEdmqGlGfQkt3AJEKIj/view
9 Hankivsky, O., & Cormier, R. (2011). Intersectionality and public policy: Some lessons from existing models. Political Research Quarterly, 64(1), 217–229. https://www.jstor.org/stable/41058335
10 Open Society Justice Initiative. (2019). Nubian Community in Kenya v. Kenya. https://www.justiceinitiative.org/litigation/nubian-community-kenya-v-kenya
11 Buthelezi, S. P., Zondo, N. M., Nxumalo, L. T. M., & Vilakazi, M. (2024). Determining the digital divide among people with disabilities in KwaZulu-Natal. South African Journal of Information Management, 26(1), Article a1820. PDF Access
12 UNICEF. (2021). Children and young people with disabilities in West and Central Africa: Symposium report. United Nations Children’s Fund. PDF Access
13 Jubany, O., Güell, B., & Davis, R. (2011). Standing up to intersectional discrimination] A multi-dimensional approach to the case of Spain. Droit et cultures, 62, 197–217.
14 Inclusive Data Charter. (2022). Unpacking intersectional approaches to data: A white paper. Global Partnership for Sustainable Development Data.
15 Open Data Watch & Data2X. (2023). Integrating Intersectionality in Data Systems: A Practical Guide Across the Data Value Chain. https://opendatawatch.com/publications/integrating-intersectionality-in-data-systems/
16 Hankivsky, O. (2012). An intersectionality-based policy analysis framework.
17 Jubany, O., Güell, B., & Davis, R. (2011). Standing up to intersectional discrimination] A multi-dimensional approach to the case of Spain. Droit et cultures, 62, 197–217.
18 This approach outlined in this section is adapted from Hankivsky and Cormier’s discussion on intersectionality-informed policy models (Hankivsky & Cormier, 2011). Because these models were developed primarily in Western policy contexts, we supplement them with examples that render them more relevant to contexts on the African continent.
19 What do we mean by “Place”? A policy may assume people have (1) a fixed home address (2) a recognised neighborhood/village name in official records (3) a nearby registration office or clinic (4) reliable connectivity to use online services (5) transport to travel for in-person verification
20 The approach outlined in this section is adapted from Hankivsky and Cormier’s discussion on intersectionality-informed policy models (Hankivsky & Cormier, 2011). Because these models were developed primarily in Western policy contexts, we supplement them with examples that render them more relevant to contexts on the African continent.
21 Bishwakarma, Hunt, and Zajicek (2007) in (Hankivsky & Cormier, 2011).
22 Mastercard Foundation. (2024, December 13). Transforming lives through digital empowerment. https://mastercardfdn.org/en/articles/transforming-lives-through-digital-empowerment/
23 Pollicy. (2025). Gender Data Futures: A Handbook on Transformative Data Governance. https://pollicy.org/resource/gender-data-futures/