Disability data in developing countries: opportunities to support inclusion

Data can be a powerful tool for understanding the challenges and opportunities faced by people with disabilities in developing countries, and for improving their welfare and access to relevant services. High-quality disability data, when accessible and used effectively, can help communities and their advocates, policymakers and local officials better understand and prioritise interventions that benefit people with disabilities. However, it is unclear what data is currently available to these stakeholders, and how it could be improved to better support the inclusion of people with disabilities.

That’s why Development Initiatives (DI) is working with Sightsavers as part of the Inclusion Works consortium . The consortium's work is funded by UK aid, and aims to create and test innovative approaches to improve the long-term economic empowerment and inclusion of people with disabilities in Bangladesh, Kenya, Nigeria and Uganda.

What we’ve done

DI’s role is to improve the quality and use of disability data in these four target countries. To establish the project baseline, we conducted an initial mapping of the disability data that exists – which involved identifying official sources of disability data, i.e. censuses, national studies and surveys; assessing the quality of the data those sources provide; and presenting analysis of disability prevalence for working-age persons and employment rates of those with disabilities from publicly available microdata (individual response data in surveys and censuses).

We’ve conducted some preliminary scoping around this objective, and we’ve already begun to uncover trends in the four target countries that will guide our work going forward.

How we did it

We identified sources of official disability data from national and international statistical websites. The most recent national household and population censuses were also considered, as were other surveys listed on the International Household Survey Network . We then assessed official data sources against criteria to determine their data quality and usefulness to stakeholders.

We scored these official sources over five categories of data quality, with each domain containing individual scoring factors that contribute to an overall ‘data quality index’ (see Table 1). Evaluating the quality of a source’s disability data framework relied on both an assessment of the delivery practice [i] and the measured outputs from disability-related survey questions. [ii]

Having identified the highest quality national data sources with publicly available microdata, DI produced disability prevalence estimates for the working age populations, and employment rates of those with disabilities.

What we found

In total, we identified 18 official sources of disability data across the 4 target countries. Uganda produced the most consistent estimates of disability prevalence and were largely assessed to be of high quality (as most of the sources used best practice questions, as well as providing high levels of disaggregation and geographic coverage). However, we identified significant challenges in the official data sources in Bangladesh, Kenya and Nigeria, which would benefit from a consistent approach to how disability is measured. Better application of best practice in question delivery, for instance, would improve both the consistency and quality of the data available.

We conducted disability and employment analysis of the highest quality surveys with available microdata. In Bangladesh, women with a disability were significantly less likely to be employed than men; however, in Kenya, Nigeria and Uganda, women with disabilities had higher employment rates than men with disabilities. Prevalence of people with disabilities in working age populations was found to be lower than the national average in each target country; the most common domains of disability among working age populations were physical or mobility impairments, while the least common were self-care and cognitive impairments.

In each target country the employment rate of persons with disabilities was significantly lower than the national employment rate. Employment rates of people with disabilities varied widely, depending on the type of impairment, with those with cognitive or self-care impairments having the lowest employment rates.

Next steps

Building on our findings so far, we will engage with national disabled people’s organisations (DPOs) to assess their data needs. The process will entail tailored data surveys and key informant interviews to evaluate what data DPOs require to improve effectively the welfare and opportunities for people with disabilities.

We’ll also conduct a disability data audit, which will expand the range of the preliminary data scoping to include unofficial and academic sources not already captured. This audit will comprehensively identify what disability data is currently available for each of the target countries, who produces it, where it is stored, and who is currently using it – giving us a more complete picture of the data that is available and how it could be improved.

Notes

[i] ‘Delivery practice’ refers to the way questions about disabilities are presented – for example, best practice for delivery does not ask for respondents to self-report that they are disabled, and does not include the word “disability” in questions, but rather assesses functioning.

[ii] Measured outputs are the range and depth of functioning which is assessed: across how many domains (types) of impairment, and the severity of impact on the respondent.

This blog post was updated on 10 October 2019.

Photo: Martin Kharumwa/ Sightsavers