The purpose of this report is to provide an overview of the complexity of measuring disability in Official Statistics. It provides a summary of how disability is currently measured across Central Statistics Office (CSO) surveys, and more generally, as it sets out clearly how disability is defined and conceptualised. The benefit of this report is that it serves as a single source of information on disability measurement in the CSO. This can help data users understand what specifically is measured and, importantly, what is not measured.
It is important for readers of this report to note the following:
As the CSO prepares for IDS data collection in 2028 there will be ongoing consideration of methodological choices and user needs to ensure the resulting statistics on disability are robust, comparable and policy relevant. The CSO and the National Disability Authority (NDA) have collaborated to develop this report on disability measurement, ahead of the IDS. As set out in the National Human Rights Strategy for Disabled People 2025 – 2030 (NHRSDP), the Department of Children, Disability and Equality (DCDE) have proposed running the IDS in 2028. Disability statistics are essential to compare the equality of opportunities and outcomes for disabled and non-disabled people, to inform policy development, and to meet national and international reporting obligations. The IDS is an opportunity to provide a more in-depth source of disability data.
The report outlines the standards required to produce high-quality statistical data, so that the data are valid and reliable. We also highlight some of the regulatory constraints that impact statistical comparability, and consistency across different outputs, as well as user relevance. This report highlights relevant frameworks, legislation and models that underpin how disability is defined and understood. In doing so, we pay particular attention to areas where these diverge to demonstrate the challenges involved when collecting disability information in surveys. This report therefore could be used by anyone developing, analysing and reporting on measures of disability data. It will help users consider the different meanings of the word disability, to understand the requirements that must be met to produce reliable data, and to ensure the accessibility of the measure to respondents. The report aims to articulate the challenge of achieving a balance between those three aspects.
Before disability can be measured statistically, we need to define and understand the underlying concept that is being measured. This is challenging as disability is one word with many meanings. These meanings can have a lot in common with each other, but they can also mean different things. For instance:
The picture is complicated further by the fact that people disagree about the precise definition for each of the above meanings of the word disability. For instance, people differ in the extent to which they believe disability to be caused by a person’s biology or psychology, by societal oppression, or by some interaction between each of these things.
The ability to categorise someone as disabled or non-disabled is fundamental to our ability to count how many disabled people live in Ireland. The ICF, developed by the WHO, recommends a biopsychosocial model for statistical disability measurement where impairments are understood in a healthcare context [1]. This is consistent with the more broadly applied social-relational model of disability. That is, both models understand disability as resulting from an interaction between a difficulty and a societal response to that difficulty. However, the social-relational model does not require an impairment to be understood using healthcare terminology. These types of models are the most widely adopted to inform disability measurement in Official Statistics across the EU and in CSO surveys.
To meet the standards required by the CSO to produce high-quality statistical data, the upcoming IDS will use the social-relational model, and the questionnaire content has been defined by stakeholders across the disability sector including DCDE, the NDA, people with lived experience, representatives across government, and academia. As such, a person can be categorised as disabled if they report an impairment (or difficulty) and if they also report that this has an impact on their ability to participate in daily life. Importantly, this impact on daily life can be attributed to an effect of the impairment itself (such as fatigue), the way in which society responds to this impairment (such as an inaccessibly designed building), or to both. Based on this report, we recommend a social-relational model of disability is used when designing survey questions and categorisation rules to best capture instances of disability that can result from societal barriers, from impairment effects or from a combination of the two.
Lastly, there is an opportunity for the upcoming IDS to support the CSO’s coordinating role, to bring increasing harmonisation to disability measurement. It is important that the NDA as chair of the Disability Research and Data Advisory Group (DRDAG), and other stakeholders support the CSO as chair of the National Data Infrastructure (NDI) group, to harmonise disability measurement through the promotion of disability statistical standards across the Irish Public and Civil Sector.