dialuk.info

Threshold — how disability is defined, counted and governed in the United Kingdom.

Counts

The continuous surveys

Entry in CountsSources named in full

Papers, forms and reference volumes on a desk
Smaller samples, more often, with confidence intervals that matter.

Photo: Lukas Blazek / Pexels

Running year-round instead of once a decade, the regular household surveys are the main source of disability prevalence figures — and their sample sizes set the limits of what can reliably be said.

Measurement in motion

The Annual Population Survey and the Family Resources Survey are the two instruments that generate most published disability prevalence data in the United Kingdom. Neither takes a single snapshot: both run continuously, accumulating interviews across the year before results are aggregated and published. The Annual Population Survey, a boost to the Labour Force Survey, draws roughly 320,000 individual interviews per year across Great Britain; the Family Resources Survey covers around 19,000 households in Great Britain, with a Northern Ireland equivalent running separately.

That difference in scale matters immediately. The Annual Population Survey is large enough to support breakdowns by region and, with care, by some intersecting characteristics. The Family Resources Survey, smaller, is the basis for benefit-unit analysis and income data, and its disability figures feed directly into DWP's annual Households Below Average Income release and the Family Resources Survey bulletin itself. The Office for National Statistics publishes disability figures from the Annual Population Survey in its Labour market overview series and in dedicated releases on the disability employment gap.

Papers, forms and reference volumes on a desk
Every published figure sits in a numbered table with a stated base population and reference period. The headline summarises that; it does not replace it.

Photo: Mikhail Nilov / Pexels

Both surveys use self-reported measures. Responses shift when question wording changes, which is why comparability over time is a persistent problem: a redesigned questionnaire breaks the series even if the underlying population is unchanged.

Grey archive boxes on a metal rack
Fig. 2Coverage, cadence and geography differ by instrument, so two totals for the same subject rarely describe the same population.

Because both are sample surveys rather than censuses, every figure they produce carries a confidence interval. For the overall prevalence rate — roughly a fifth of the working-age population under the Equality Act measure — the interval is narrow enough to be reliable. For smaller subgroups, particularly where disability intersects with ethnicity or geography, the interval widens to the point where year-on-year changes cannot be distinguished from sampling variation. The published bulletins carry these intervals; users who read only the headline rate are reading only part of the number.

More in Counts