Outcomes for disabled people in the UK methodology, 2021 update

This methodological article provides the detail around the definitions and data sources used in the Outcomes for disabled people in the UK: 2021 release. It also discusses how a number of data sources used in this release were affected by the coronavirus (COVID-19) pandemic, and what the Office for National Statistics (ONS) had done to address this.

For this release, different datasets are used to produce estimates on a number of areas of life, covering different geographies. The Annual Population Survey covers the UK and produces estimates on education, employment, housing and well-being. The Crime Survey for England and Wales produces estimates on crime. The Community Life Survey covers England and produces estimates on social participation and loneliness

Annual Population Survey based estimates

There is significant public and policy interest in how the coronavirus (COVID-19) pandemic has affected employment, population and migration. At the same time, the coronavirus pandemic has both affected our ability to collect data, including the need to pause face-to-face surveys, and the population projections to which the survey responses are weighted.

Because of the coronavirus pandemic all face-to-face interviewing for the Labour Force Survey (LFS) was suspended and replaced with telephone interviewing. This changed the profile of responders to the survey. The LFS, which feeds into the Annual Population Survey (APS), is one of the main data sources used in this release. In addition, the population estimates used as part of the LFS and APS statistics, predate the coronavirus pandemic and so do not show the demographic and structural impacts.

Initially to counter this, we introduced tenure weighting, which addressed some of the issues, but we acknowledged further work was needed and we committed to introducing further reweighting to improve LFS and APS estimates. Later, using Real Time Information (RTI) data from HM Revenue and Customs' payroll tax system, we created new population weights for the survey to better reflect changes to international migration. We also introduced a non-response bias adjustment to better reflect the socio-economic makeup of the population.

Revised weights have been applied to LFS and APS data from January to March 2020 onwards. Earlier data will still use weights consistent with the Office for National Statistics' (ONS) population estimates and projections. Estimates for education, employment, housing and well-being have been weighted according to the updated methodology.

When the recent weighting methodology for the LFS was applied, there was a small error in the implementation of RTI growth rates. This led to a slight overestimation of the non-UK population by approximately 0.5%. The impact on LFS economic-activity estimates at national level is mostly below 0.1%, and the impact on rates is less than 0.02 percentage points.

Crime Survey for England and Wales based estimates

The Crime Survey for England and Wales (CSEW) is the main source for crime statistics in England and Wales. It was suspended in March 2020 because of the coronavirus pandemic. A telephone-operated version of the CSEW (TCSEW) was specifically designed to allow us to continue measuring crime during this period, while face-to-face interviewing is not possible. All survey data published alongside police recorded crime up to the year ending March 2020 were based on the face-to-face CSEW and largely unaffected by the coronavirus pandemic. All data since have been published using data gathered via the TCSEW.

On 20 May 2020, the ONS launched an interim TCSEW based on a reduced set of questions usually operated via the regular face-to-face survey. Because of ethical guidelines, questions from the more sensitive self-completion modules of the face-to-face CSEW are not included in the telephone-operated survey. As a consequence we have not been able to produce an update to our usual crime indicators, such as sexual assault or domestic abuse in this release and have instead included additional data from March 2020.

Community Life Survey-based estimates

The fieldwork for all four quarters of the 2020 to 2021 community life survey (CLS) took place during the coronavirus pandemic. The effects of the coronavirus pandemic should be taken into consideration when interpreting the results of this report.

Disability-related definitions

Disability

To define disability in this publication we refer to the Government Statistical Service (GSS) harmonised "core" definition. This identifies "disabled" as a person who has a physical or mental health condition or illness that has lasted or is expected to last 12 months or more, that reduces their ability to carry out day-to-day activities.

The GSS definition is designed to reflect the definitions that appear in legal terms in the Disability Discrimination Act 1995 (DDA) and the subsequent Equality Act 2010.

The respondent is asked the GSS-harmonised questions in the survey, meaning that disability status is self-reported.

Impairment

For all analysis using the Annual Population Survey (APS), that is, for employment, education, housing and well-being, impairment is defined as any physical or mental health conditions or illnesses lasting or expected to last 12 months or more. Respondents were presented with a list of impairments and then asked to select all and subsequently their “main health problem”.

The commentary in this release refers to the main health problem. Analysis is limited to those who are also defined as disabled and does not explore where disabled people experienced more than one impairment. For further details see Volume 3: Detail of Labour Force Survey variables.

From January 2020 onwards, the list of impairments in the APS changed because of the addition of "Autism (including Autism Spectrum Condition, Asperger's syndrome)" as a possible response option. To include the most detailed list, the analysis of impairment types is based on a selection of APS data collected after January 2020 that incorporate this change. This meant in our last release we had to use a half-year dataset (January to June 2020 data). In this release we now have the first full-year period since changes were made. To enable the changes, both periods have been reweighted to the UK population following the latest methodological development to account for the impact of coronavirus (COVID-19) on data collection. As the response options changed in January 2020, comparison with previous years is not recommended.  

The impairments or condition categories compared in this release relate to the categories within the question in the survey. The exception is speech impediment, which has been grouped with the "other" category because of low sample size.

For all analysis using the Crime Survey for England and Wales (CSEW), the GSS harmonised principle for impairment is used.

An impairment refers to health conditions affecting:

• vision (for example, blindness or partial sight)

• hearing (for example, deafness or partial hearing)

• mobility (for example, walking short distances or climbing stairs)

• dexterity (for example, lifting and carrying objects or using a keyboard)

• learning, understanding or concentrating

• memory

• mental health

• stamina, breathing or fatigue

• socially or behaviourally (for example, associated with autism, attention deficit disorder or Asperger's syndrome)

• other (please specify)

Respondents were asked to select all impairments that applied to them.

Severity

Disabled people whose ability to carry out day-to-day activities is self-reported as "limited a lot" or "limited a little" by their impairment. Respondents were asked: "Does your condition or illness reduce your ability to carry out day-to-day activities?" with the responses, "yes, a lot" and "yes, a little" being taken to indicate severity of disability.

Outcome-related definitions

Anti-social behaviour

Anti-social behaviour (ASB) is any behaviour that causes people nuisance, annoyance, alarm or distress. This can include behaviours that are aggressive or intimidating or that impact upon a person's quality of life.

More details about these, and the new questions on experience of ASB asked for the first time in the year ending March 2012 CSEW, can be found in section 5.7 of the User guide to crime statistics for England and Wales.

Civic participation

Civic participation refers to engagement in democratic processes, both in person and online, including contacting a local official (such as a local councillor or MP), signing a petition, or attending a public rally (this excludes voting).

Civic consultation

Civic consultation refers to taking part in a consultation about local services or problems in the local area through completing a questionnaire, attending a public meeting, or being involved in a face-to-face or online group.

Civic activism

Civic activism refers to involvement in activities in the local community, such as being a local councillor, school governor, volunteer special constable or magistrate (for those aged 18 years and over). Civic action also includes involvement (in person or online) in decision-making groups in the local area, for example, a group making decisions about local health or education services, a tenants' decision-making group, or a group set up to tackle local crime problems or to regenerate the local area.

Employment

Employment measures the number of people in paid work or those who had a job that they were temporarily away from (for example, because they were on holiday or off sick). Employment differs from the number of jobs because some people have more than one job. The employment rate is the proportion of people aged 16 to 64 years who are in paid work.

The Guide to labour market statistics contains a glossary of other terms used in this release.

Formal volunteering

Formal volunteering refers to giving unpaid help through groups, clubs or organisations.

Highest qualification

Highest qualification applies to all respondents aged 21 to 64 years not enrolled on a course.

Housing situation

Conventional analysis of housing tenure explores outcomes at the household level. The analysis presented here uses a different approach, defining housing situation for the person level. The article reviews the housing situation of disabled people. It incorporates both tenure (the legal arrangements under which a person lives in a property, for example, owner occupier, social-rented housing or private-rented housing), responsibility for the property, and the relationship to the household reference person (HRP), who is the person legally responsible for the household. Where more than one person has indicated they are responsible for the property the HRP is allocated based on highest income and age.

For the first three categories (owner occupier, social-rented housing, private-rented housing) the respondent has either indicated they have responsibility for the property (including the HRP themselves) or the relationship to the HRP is either spouse, cohabitee, civil partner or same sex cohabitee.

The housing situation of the individual is defined as:

• owner occupier – includes owned outright, buying with mortgage or loan, or part rent and part mortgage

• social-rented housing – includes renting from local authority, council, Scottish homes, housing association, charitable trust or local housing company

• private-rented housing – includes renting from employing organisation or individual employer, relative of household member, another private landlord, or another organisation

• living with parents – this category ignores tenure (except to exclude rent-free or squatting) and looks at the relationship to the HRP (all those with one of the following relationships to the HRP are included in this category: child, stepchild, foster child, grandchild or child-in-law; this includes adult children)

• other – this category ignores tenure, except to include rent-free or squatting and looks at the relationship to the HRP (all those with one of the following relationships to the HRP are included in this category: parent, stepparent, foster parent, parent-in-law, grandparent, brother or sister, stepbrother or stepsister, foster brother or sister, brother- or sister-in-law, other relation, other non-relative, or undefined)

This is a small change from the methodology used in previous releases where tenure and relationship to the head of household (HOH) was used. In the latest data tables accompanying this article, estimates for earlier years have also been calculated using the new methodology.

Informal volunteering

Informal volunteering refers to giving unpaid help as an individual to people who are not a relative.

Loneliness

Those who feel lonely "often or always" refers to those who when asked: "How often do you feel lonely?" selected the answer "often or always" from the list of responses: "often or always", "sometimes", "occasionally", "hardly ever" and "never". These responses have been grouped together into an "other" category. This analysis is available in the datasets.

Social action

Social action refers to involvement with issues affecting the local area by doing things like setting up a new service or amenity, stopping the closure of a service or amenity, stopping something happening in the local area, running a local service on a voluntary basis, or helping to organise a community event.

Standard Occupational Classification 2010

For details of the general nature of qualifications, training and experience for occupations, see Table 2 of SOC 2010 Volume 1: structure and descriptions of unit groups.

Statistical significance

Any changes or differences mentioned in this release are "statistically significant". The statistical significance of differences noted within the release are determined based on non-overlapping confidence intervals.

Well-being

Personal well-being measures ask people to evaluate, on a scale of 0 to 10, how satisfied they are with their life overall, whether they feel they have meaning and purpose in their life, and about their emotions (happiness and anxiety) during a particular period.

Annual Population Survey (APS)

The Annual Population Survey (APS) is an annual survey based on data collected in Wave 1 and Wave 5 of the Labour Force Survey (LFS), combined with an annual local area boost sample run in England, Wales and Scotland. The APS dataset contains approximately 200,000 individuals.

The survey does not cover communal establishments, except for NHS staff accommodation. Those living in student halls of residence or boarding school are included as part of their family household.

The APS datasets are produced for four different overlapping 12-month periods: January to December, April to March, July to June and October to September. Analysis for this publication was conducted from July 2020 to June 2021.

Analysis conducted using the APS has been restricted to those aged 16 to 64 years in the case of employment, housing and well-being analysis. This is because the survey does not collect data for those aged under 16 years, and the disability variable is not robust for those aged 65 years and over. Disability status is only collected for people aged 65 years and older at their first contact, resulting in less data for this population. The weighting used does not account for the reduced sample size for this age group, making the data not fully representative of the population.

Education analysis using the APS has been further restricted to those aged 21 to 64 years to only include those who are likely to have finished their education. Those currently enrolled on a course are also removed from the analysis.

Well-being, because of the nature of the topic, is not collected by proxy. This further reduces the sample size used in the analysis by approximately 100,000 APS respondents.

The age restriction of the survey means we are not capturing the employment status, housing situation, well-being or highest education attainment of respondents aged 65 years and over.

Community Life Survey

The Community Life Survey (CLS) is a household self-completion survey of approximately 10,000 people aged 16 years and over in England. The survey can be completed either in a paper or online format; the question regarding disability status is asked online only. Analysis was conducted in the year ending March 2019.

The loneliness and social participation analysis are restricted to people aged 16 years and older for England only. This is because the CLS does not collect data for those aged under 16 years or the other countries in the UK.

Crime Survey for England and Wales

The analysis presented in this release uses data from the Crime Survey for England and Wales (CSEW). This is a face-to-face victimisation survey. It asks approximately 35,000 respondents, aged 16 years and over residing in households in England and Wales, about their experiences of crime in the 12 months prior to the interview.

The survey also asks for opinions on different crime-related issues such as the police and criminal justice system and perceptions of crime and anti-social behaviour. Individuals residing in group residences (for example, care homes or halls of residence) or other institutions are not included in the CSEW sample.

A full list of the strengths and limitations of the Crime Survey for England and Wales can be found in the Crime in England and Wales QMI.

Causality

The analysis conducted is for the purpose of comparing the outcomes of disabled and non-disabled people. The analysis describes differences in these two populations but does not explore the cause of this difference. Further analysis, which is outside the scope of this article, is required to make judgements on causality.

Crime

The survey has a large nationally representative sample survey that provides a good measure of long-term crime trends for the offences and the population it covers (that is, those resident in households). However, small sample sizes observed for anti-social behaviour data has resulted in greater variability surrounding the estimates for the disabled population, limiting the conclusions that can be drawn from this analysis.

The survey covers crime not reported to the police. It is not affected by changes in police recording-practice and is a reliable measure of long-term trends.

A full list of the strengths and limitations of the Crime Survey for England and Wales can be found in the Quality and methodology section of the Crime in England and Wales statistical release.

Education as an outcome

This analysis accounts for all people aged 21 to 64 years who have a self-reported disability at the time the survey was conducted. However, prevalence of disability is highest among older people and survey respondents may or may not have had a disability during the time of their education. Factors such as age and disability onset may have a confounding influence on the data reported.

People with different disabilities have different potential to attain qualifications. This analysis accounts for the highest level of achievement, but is unable to reflect if a disabled person has met their potential.

Age was restricted to those aged 21 years and over as most people will not have been able to complete a degree level qualification before this age. Additionally, participants aged 21 years and over who were enrolled in education were excluded.

The severity section of the education analysis has been revised since the 2020 release of this publication. All estimates, including those using data collected in earlier years, have been updated. In the previous releases, disabled people with lot and a little activity-limitation were treated as part of the same population for estimation. However, they are now treated as separate populations, in line with estimation for other outcomes in this release. Therefore datasets from previous years cannot be compared, and comparisons should only be made within the February 2022 dataset.

Data on the educational outcomes for disabled children are limited. Education-outcomes data tend to report on Special Education Needs or Additional Learning Needs, which do not cover the same population as disabled children. In the absence of these data, we have used Annual Population Survey (APS) data to understand how attainment differs for disabled and non-disabled people. We are exploring how the existing evidence gap of educational experiences and outcomes for disabled people could be addressed.

Housing situation

The analysis has been carried out at the person level, meaning we can determine the living situation of the individual rather than just the housing tenure of a household that includes (or does not include) a disabled person. This is seen in particular in the category "living with parents" where we have focused on the relationship to the homeowner rather than tenure. This allows us to explore differences in disabled and non-disabled people living with parents. Please see the Glossary for the definition of each housing category.

Students living away from home or those in boarding school are included in the household. This is particularly important when considering the numbers living with parents.

Impairments

Analysis by impairment with APS data is based on the "main impairment" as reported by the respondent. People often experience more than one impairment, but this analysis does not account for co-morbidities or the cumulative impact of living with more than one impairment simultaneously.

Loneliness

Loneliness proportions were calculated using the Community Life Survey, which is used in England only.

Social participation

This article focuses on participation in two domains of social participation: civic engagement and social action and volunteering. In each domain, participation refers to any involvement in the 12 months prior to completing the survey. In some areas (such as formal and informal volunteering), the Community Life Survey also records participation in the four weeks prior to completing the survey – these data could be used to assess more regular participation in these areas. However, to be consistent across our analysis, we have looked only at involvement in the previous 12 months.

Uncertainty and quality

The results in this release are survey-based estimates, so they are subject to a level of uncertainty as they are based on a sample rather than the whole population. Confidence intervals are provided around every estimate and give an indication of the range in which the true population value is likely to fall. The estimates in this release are supported with confidence intervals at the 95% level. This means that, if we repeated the sample, we would expect the true population value to fall within the lower and upper bounds of the interval 95% of the time (that is, 19 times out of 20).

Well-being

The Personal well-being in the UK Quality and Methodology Information report contains important information on the strengths and limitations and uses of the data, as well as how outputs are created and the quality and accuracy of those outputs. For more information on personal well-being, please see the Personal well-being user guidance and Harmonised principles of personal well-being.

All analysis of well-being measures in this release compares mean scores between groups. Another method of comparing well-being scores between groups is to look at the proportions of people who are within certain thresholds that indicate well-being levels.