Update on the harmonisation of disability data collection in UK surveys (part 2).

Abstract

Background: This article reports on the field testing work undertaken, leading to the proposals for a primary harmonised standard for disability data inputs: designed to derive statistical measures of disability from social surveys using a face-to-face mode of data capture. These proposals were submitted to the National Statistics Harmonisation Group (NSHG) for approval in December 2010. The proposed primary harmonised standard data inputs are designed to meet the data needs arising from the equality legislation introduced in 2010 and improve international comparability: by better meeting the definitions for measures of longstanding illness and disability derived from the European Union's Statistics on Income and Living Conditions (EU-SILC).

Methods: Field testing was conducted during two discrete periods: between October 2009 and February 2010 for stage 1 testing, and between May and July 2010 for stage 2 testing. The Office for National Statistics (ONS) Opinions Survey acted as the data collection vehicle and the design was a split sample in each case. In stage 1, question suite variants were compared, together with the Family Resources Survey (FRS) Disability Discrimination Act module running on the Opinions Survey in equivalent months. The second stage also applied a split sample to compare the revised proposed harmonised questions with the question on disability planned for the 2011 Census. Derivations of disability to meet measures required under the national Equality Act legislation and the European Union-Statistics on Income and Living Conditions were applied to field-test estimates to assess coherence with contemporaneous external sources.

Results: The field test results found good comparability with estimates of disability used in the FRS publications and with the field tested 2011 Census disability question.In stage 1 testing, the measure of disability derived from the version 1 suite was closer to the FRS measure of disability used in FRS publications than that derived from version 2's questions. In stage 2, the proximity of the measure derived from the proposed harmonised questions to the FRS measure of disability improved, with a 0.5 percentage point difference.The stage 2 measure of disability was found to be consistent with the EU-SILC measure of disability in 2009 for the UK published on the Eurostat website. Furthermore, the harmonised questions produced a similar estimate of disability to the 2011 Census question, also field tested concurrently with the proposed harmonised questions.

Conclusion: The harmonised questions tested provide appropriate data inputs to measure persistent illness, impairment and disability, consistent with the components of disability used in the disablement process and the International Classification of Functioning Disability and Health. Their implementation across social survey data sources using a face-to-face mode of data capture will enhance the consistency of statistical measurement and their relevance to the data requirements embodied in national Equality legislation and the EU-SILC European regulation.List of Tables, 33.