National statistics

Family Resources Survey: background information and methodology

Updated 21 July 2023

1. Introduction

This background note accompanies the main Family Resources Survey 2021 to 2022 report.

The purpose of this report is to provide further contextual information to aid understanding of the statistics presented in the main report and detailed tables. It outlines points to note as well as strengths and limitations of the information presented in each section of the main report; alternative data sources; as well as changes to the survey this year compared to last year.

A detailed description of the Family Resources Survey (FRS) methodology, fieldwork operations, data processing and quality assurance are presented within the relevant sections in this report. These descriptions are intended to help users in the use and interpretation of FRS 2021 to 2022 data.

The FRS is a major study of income levels in the UK and has a prominence in the landscape of income data resources. For the collection of data in 2021 to 2022 it was recognised that there was still the need to maintain several coronavirus (COVID-19) related changes. However, the publication of FRS estimates and data for 2021 to 2022 was uninterrupted, allowing for a continued depth of insight on household circumstances.

The coronavirus (COVID-19) pandemic continued to affect the FRS to some degree, with some of these effects having a bearing on the survey results. Later sections discuss the changes to the survey questionnaire and fieldwork approaches, both of which continued to be different to a typical FRS year. Later sections also set out the difference in response rate, and composition of responses, achieved this year. There have also been changes to the production process of the survey results, particularly around the grossing methodology.

Editorial team

Alex Brandon-Bravo, Anna Britton, Claire Cameron, Annabel Connolly, Jake Lipscombe, Sheridan Lomas, Justyna Owen, Clive Warhurst

Feedback

If you have any comments or questions about any aspect of the FRS, or are interested in receiving information on consultations, planned changes, and advance notice of future releases, please contact:

FRS Team, Surveys Branch, Department for Work and Pensions

Email: team.frs@dwp.gov.uk

Acknowledgements

In a year of what were still challenging circumstances, we wish to give special acknowledgement to:

• all the respondents in households across the United Kingdom who agreed to and made the time to be interviewed

• interviewers from the Office for National Statistics, NatCen Social Research and the Northern Ireland Statistics and Research Agency who conducted and then collated the interviews

• all those who have contributed towards the Family Resources Survey 2021 to 2022 publication, through providing quality assurance and feedback

• our web support team

• the UK Data Service, who distribute our research data, as well as the ONS Secure Research Service

2. Background

The Family Resources Survey (FRS) is a continuous survey which collects information on the income and circumstances of individuals living in a representative sample of private households in the United Kingdom. The survey has been running in Great Britain since October 1992 and was extended to cover Northern Ireland in the survey year 2002 to 2003.

The primary objective of the FRS is to provide the Department for Work and Pensions (DWP) with information to inform the development, monitoring and evaluation of social welfare policy. Detailed information is collected on: respondents’ incomes from all sources including benefits, tax credits and pensions; housing tenure; caring needs and responsibilities; disability; expenditure on housing; education; childcare; family circumstances; child maintenance; household food security and food bank usage.

Microsimulation is central to DWP’s use of the data. Therefore, careful attention is paid to the accurate collection of information followed by meticulous data processing, editing, and quality assurance.

The FRS data is designated by the Office for Statistics Regulation (OSR) as National Statistics. The FRS provides the data for several other DWP National Statistics publications: Households Below Average Income, Pensioners’ Incomes Series. It also underpins the Official Statistics on Income-Related Benefits: Estimates of Take-up and Children in Low Income Families: local area statistics.

The survey contains information used by other government departments, particularly for tax and benefit policy analysis by His Majesty’s Revenue and Customs and His Majesty’s Treasury. The survey is also used extensively by academics and research institutes for social research purposes.

Status and development

These statistics underwent a full assessment against the Code of Practice for Statistics in 2011 and were confirmed as National Statistics in November 2012 by the Office for Statistics Regulation.

The OSR published its Review of Income-based Poverty Statistics on 19 May 2021. We took several recommendations on board, which have been maintained in this year’s publication. The following developments address some of these recommendations:

• DWP and ONS continue to work together to take account of the ongoing impacts of the coronavirus (COVID-19) pandemic on data collection methods. As virtually all interviewing continued to be via telephone, attention was focused on minimising the impacts of telephone interviewing on the quality of the achieved sample data

• DWP has had greater dialogue with expert users of the FRS-based statistics in response to COVID-19 related data issues. DWP expect this to continue in the future as part of its long-term work programme. In both the technical advice provided alongside the publication, and in the supporting documentation provided to the UK Data Service (UKDS) with the dataset, we have been clear on the strengths and limitations, appropriate uses and quality of the statistics and data, being transparent in how this may differ from past years. A Technical User Guide for the FRS is available on the UKDS to support microdata users

• we have further reviewed our publication tables and supporting guidance to ensure accessibility to users. For compliance with The Public Sector Bodies (Websites and Mobile Applications) Accessibility Regulations 2018, some formatting in the ODS tables, such as merged cells, has been avoided. For more information, please see the accessibility statement specific to DWP’s statistical releases

• the Glossary section of this Background Information & Methodology report has been updated to provide clearer and more up-to-date information on devolved benefits and geographical reporting layers

Recent improvements to the FRS include:

• the most substantial development this year has been the introduction of FRS data to the Stat-Xplore online tool. Stat-Xplore can be used to create bespoke tables and statistics, across a very wide range of FRS variables. Several years of data, going back to the 2002 to 2003 survey year, have been released

• the content of the FRS publication continues to evolve in response to user needs, with the addition of new material for emerging areas of policy interest. New information on food bank usage accompanies the Household Food Security chapter, for the 2021 to 2022 publication

• we have reviewed the presentation and content of the Care tables to ensure consistency across categories and improve clarity for users. We have also introduced a Sandwich Carer category for the first time, in some of the Care chapter tables

• the construction of Income and State Support tables has been reviewed and revisions have been applied to the denominator for income-related components, such that it now correctly includes income from Universal Credit

• auditing of processing methodology is a regular feature, and subsequent changes to imputation methodology and the construction of derived variables, such as components of income-related variables and adjusting the reported income bands to address the need for more granularity at both ends of the income distribution, have led to improvements in the quality of statistics

• other new questions and variables are added each year, as necessary to reflect changes in policy, such as benefit changes specific to some areas of the UK, and in different policy areas. This enables related policy analysis to be conducted

Users have been informed in advance of substantial changes to the FRS publication. Please see the DWP Statistical Work Programme and the FRS Release Strategy for more details.

3. Uses of FRS data

The FRS is used extensively both within and outside DWP. The main uses are:

Households Below Average Income (HBAI)

The HBAI publication uses household disposable incomes, adjusted for household size and composition, as a proxy for material living standards or, more precisely, for the level of consumption of goods and services that people could attain given the disposable income of the household in which they live.

Pensioners’ Incomes Series

The HBAI dataset is used in the Pensioners’ Incomes Series (PI), the Department’s analysis of trends in components and levels of pensioners’ incomes.

Income-Related Benefits: Estimates of Take-Up

Take-up figures are based on a combination of administrative and survey data. The FRS provides information about people’s circumstances, which is used to estimate numbers of people who are not claiming benefits to which they may be entitled.

Separated Families Statistics

Experimental statistics relating to separated families and their child maintenance arrangements.

DWP Policy Simulation Model and other policy analysis

DWP’s Policy Simulation Model (PSM) is used extensively for the development and costing of policy options. FRS responses are uprated to current prices, benefits and earnings levels and can be calibrated to the DWP Departmental Report forecasts of benefit caseload. Using FRS data has made it possible to model some aspects of the benefit system which could not be done previously, for example allowances for childcare costs. In addition to their use in formal modelling, FRS data play a vital role in the analysis of patterns of benefit receipt for policy monitoring and evaluation, and benefit forecasting.

Other government departments and the wider research community

The survey is widely used by other government departments, including His Majesty’s Treasury, His Majesty’s Revenue and Customs, the Department for Environment, Food and Rural Affairs and others.

The Department for Communities Northern Ireland uses the FRS to produce similar reports to those from DWP, which are focused on Northern Ireland statistics.

In-depth analysis of FRS-based HBAI data for Scotland can be found in the Scottish Government report on Poverty and income inequality statistics in Scotland.

In-depth analysis of relative income poverty in Wales, based on HBAI data, can be found on the Relative income poverty page of the Welsh Government website, which also has links to material deprivation analysis.

The Office for National Statistics produces small area model-based income estimates. These are the official estimates of annual household income at the middle layer super output area (MSOA) level in England and Wales. The estimates are produced using a combination of survey data from the Family Resources Survey and previously published data from the 2011 Census plus several administrative data sources.

The Race Disparity Unit has published a series of summaries of data on their ‘Ethnicity Facts and Figures’ website since June 2019. Ethnicity Facts and Figures provides information about the different experiences of people from a variety of ethnic backgrounds. It gathers data collected by government in one place, making it available to the public, specialists and charities. The FRS contributes data on state support, that is, receipt by ethnicity and type of benefit.

The FRS Household Food Security data is used as a UK indicator of Sustainable Development Goals. This relates to Indicator 2.1.2 – Prevalence of moderate or severe food insecurity in the population, based on the Food Insecurity Experience Scale.

Researchers and analysts outside government can also access the data through a variety of ways:

• the UK Data Service

• from 2023, the ONS Secure Research Service

• detailed statistics and tables can be self-generated using the department’s Stat-Xplore online tool for a wide range of FRS variables and across the years FYE 2003 to FYE 2022

4. Points to note

Impact of coronavirus (COVID-19) pandemic

FRS 2021 to 2022 is an important data resource, which presents several insights into UK household incomes as the nation continued to adjust to the effects of the coronavirus (COVID-19) pandemic.

For FYE 2022, we have undertaken extensive analysis and are content that levels of bias in the data, resulting from the mode change, are lower than FYE 2021 and are having less influence on the statistics. However, there remain areas where caution is advised when making comparisons with previous years and when interpreting larger changes. Each of the sections below provides further insight into the challenges to the results which were still affected by the pandemic.

The first section outlines several factors which continued to affect some topic areas in the survey. Subsequent sections then step through each chapter, to discuss the specific effects. This is contextual detail, which aims to show the strengths and limitations of the survey. It is intended to aid users in their interpretation of FRS 2021 to 2022 data.

The data in this report are from interviews conducted between April 2021 and March 2022. The FRS achieved sample this year returned towards that expected in a normal survey year (around 20,000 households), with well over 16,000 households interviewed in 2021 to 2022 (an increase from 10,000 in 2020 to 2021). However, it should be noted that the 2021 to 2022 sample was unbalanced across the two halves of the survey year, with an achieved sample of over 6,000 households for the period from April to September, and 10,000 for the period from October to March. This reflects the introduction of the pre-planned boost to the issued sample in England and Wales from 1st October 2021.

Whilst the whole of the previous 2020 to 2021 fieldwork year was affected by the coronavirus (COVID-19) pandemic, there continued to be effects of the pandemic in 2021 to 2022 which might have also affected the survey results this year.

Most of these stem from the data collection method and the distribution of characteristics among respondents. All of these represent ways in which the 2021 to 2022 survey is different to other survey years, both prior to the coronavirus (COVID-19) pandemic and compared to the first year of the pandemic in 2020 to 2021.

The key points to consider are:

• changes in individual behaviours and circumstances

• changes in the methods used to contact survey participants and response

• a change in the mode, between telephone interviewing and face-to-face

Changes in individual behaviours and circumstances

It is not possible to outline all issues that affected either participation in the FRS itself, or how differently circumstances were reported this year, compared to prior to the coronavirus (COVID-19) pandemic and compared to the main year of the pandemic in 2020 to 2021.

However, from a social research perspective, the pandemic had ongoing effects upon the UK labour market and household circumstances, and also our ability to measure them. This is because our measurement relies on data collected from a survey of households.

Many of the societal changes that took place during 2020 to 2021, such as an increase in home-schooling and social distancing were not present after July 2021, therefore did not materially affect survey participation this year. However, other factors, such as the continuation of the increase in remote working, effects on people’s health, and differences in attitudes regarding ‘normal’ social interaction, may have continued to influence the amount and type of data that the survey has been able to collect.

Family formation is the key building block of FRS results. The coronavirus (COVID-19) restrictions that led to people choosing to form support bubbles with close friends and family saw a reduction in house sharing amongst unrelated adults and an increase in multi-generational households (as adult children moved in with their parents or parents moved in with their children) in 2020 to 2021, compared with previous years.

FRS data for 2021 to 2022 shows that, compared to 2020 to 2021: the percentage of adults living with cohabiting adults has only decreased marginally (14.7% to 14.4%); the percentage of adults living with unrelated (not cohabiting) people has increased marginally (2.9% to 3.0%); and the number of households containing multiple generations has continued to increase (16.2% to 16.3%). Household composition has therefore largely remained as it was in 2020 to 2021, rather than reverting to the pre-pandemic situation.

It is also the case that the government response to the pandemic had a significant effect in supporting incomes, and on the UK labour market: government interventions allowed for the furloughing of workers until September 2021, which affects both reported incomes and other variables such as working hours.

This factor sits alongside wider labour market and business developments, whereby some businesses ceased operations, and many others altered their working practices. Remaining domestic COVID-19 measures in England, such as the legal requirement to self-isolate and the £500 isolation payment for people on low incomes who are required to self-isolate, were not lifted until 24 February 2022. However, COVID-19 provisions for statutory sick pay continued for a further month.

Whilst some change from year to year can be expected as a result of real-world changes in household circumstances, the longer lasting effects of the coronavirus (COVID-19) pandemic would likely have prevented some households from taking part in the survey who would otherwise have done so (e.g. those with caring responsibilities or ill health may have been less inclined to respond).

Changes in the methods used to contact survey participants and response

In 2021 to 2022 survey fieldwork operations continued to take account of the effect of government guidance relating to the coronavirus (COVID-19) pandemic and the restrictions placed on visiting others’ homes. The telephone interview approach, that was adopted during 2020 to 2021 survey year, continued throughout 2021 to 2022, with interviewers conducting the interview while working from their own homes.

The only face-to-face contact was interviewers making ‘knock-to-nudge’ visits at the addresses for which they had not received any contact details, to request and encourage their participation in the study. Knock-to-Nudge was used across the whole of Great Britain for all of the survey year, and in Northern Ireland from July 2021.

Full details of how engagement with respondents evolved during 2021 to 2022 is given in the Methodology section below. Response rates were also affected by difficulties in recruiting and training interviewers.

Change in the mode, to telephone interviewing

Government restrictions that were originally introduced in mid-March 2020 in response to the coronavirus (COVID-19) pandemic, forced a change in FRS processes to allow data collection by telephone. These changes remained in place for the whole of the 2020 to 2021 survey year. Although restrictions were gradually lifted both ahead of and during the 2021 to 2022 survey year, it remained the case that virtually all interviews were conducted by phone, with less than five per cent conducted by the usual method of face-to-face in the home, by the end of the survey year.

Ordinarily such changes would not be made without thorough testing to examine the effect on the data collected. In the time available, this was not possible. It is therefore difficult to quantify precisely how much the survey results have changed, because of the change to telephone interviewing, as distinct from real-world changes, in both the 2020 to 2021 and 2021 to 2022 surveys.

In broad terms, DWP’s assessment is that, where data has been collected, it is not materially different to what would have been collected from the same respondent face-to-face. This is because the FRS is almost wholly a survey of factual information rather than attitudinal information. Wherever possible, as part of the quality assurance process, results have been compared both with other data sources, and to previous FRS years.

The change in fieldwork approach also affected the composition of the FRS achieved sample. Last year, in the 2020 to 2021 sample, there were substantially more owner-occupied households and fewer renters and a skew toward older respondents (aged 65 or over), fewer households with children and fewer respondents educated to below degree level, than in 2019 to 2020. This year, 2021 to 2022, the achieved sample was closer to 2019 to 2020 in its composition, yet has still not returned to pre-pandemic proportions.

Later sections discuss the effects and limitations that this and other compositional changes bring to the results. Some of the effects have been mitigated by a change to the survey grossing regime (and the later section on grossing outlines the steps taken to improve representativeness). The move to telephone interviewing is likely to have changed the profile of non-responders to the FRS. Those who provided their telephone contact details may have had certain attributes (higher civic engagement or more available time) which we cannot directly adjust for when weighting the sample.

Income and state support

The FRS continued to capture information about people on the Coronavirus Job Retention Scheme (CJRS), until its closure in September 2021, through a set of interview questions. Employees who were out on furlough during this period were classified as employed, but temporarily away from work. This will mean that workers on furlough still counted towards the number of people in employment (or the employment rate). The classification of Economic Status remains the same as in previous years, with anyone on furlough in the Benefit Unit being classified as an employee.

Any income received through the Coronavirus Job Retention Scheme (CJRS) or any reported grants from the Self-Employment Income Support Scheme (SEISS) during FYE 2022 are treated the same way as in FYE 2021. The calculation of ‘income from employment’ uses wages which are treated as being wholly income from employment irrespective of any CJRS payments that the respondent’s employer was receiving in respect of their employment (that is, treated as wages as opposed to state support).

The calculation of earnings uses actual pay (GRWAGE) over usual pay (UGROSS) for people on furlough. This aligns to the Annual Survey of Hours and Earnings (ASHE) employee earnings methodology, which uses actual payments made to the employee from company payrolls. The calculation of self-employed income and therefore total individual income does not directly include grants reported by the respondent received from SEISS. More detail of how SEISS grants are treated in self-employed income is given in the Self Employment section.

The decision to treat both CJRS and SEISS in this way follows our discussions with our Expert Advisory Group.

Income from Dividends have been included in the source of income for the first time in the 2021 to 2022 release. New variables, generated from new questions asked about directors’ dividends, DIVIDGRO (Gross dividend) and DIVIDNET (Net dividend) have been added to the ADULT table and incorporated into ‘Individual Income’ [INDINC] and ‘Net individual Income’ [NINDINC] respectively. Neither of these variables feature as distinct items within the publication tables, but the positive effect upon income tables is that a greater level of precision has been added to the calculation of these input variables, and the resultant summation of income variables.

We have also presented ‘Income from Universal Credit (UC)’ as its own distinct category as a source of income. This has allowed for more detailed breakdowns within Income and State Support tables. This change is because the volume of UC cases is now sufficiently high to allow for meaningful analysis. Also, revisions have been applied to the denominator of overall income where results relate to components of gross household income, to accommodate accurately any income from UC. Note that this affects percentage estimates in the publication data tables only; it has not changed the aggregation of the main income variables in either the FRS tables of pounds of income, or the released dataset (sum of household income and its tributary variables).

All income figures are presented gross of tax, national insurance and before other deductions from wages except where noted.

It is thought that household surveys underestimate income from both self employment and investment income. We rely on respondent recall of very detailed financial information across a comprehensive range of income sources. Some of these are hard for respondents to recall. The FRS interviewers ask respondents to check pay slips, tax returns and other financial paperwork at the time of the interview. This helps to improve the reliability of what respondents report they earn.

The FRS captures detailed information on benefit receipt. In most cases this is analysed at a benefit unit (family) level because income-related benefits are paid to families as a whole rather than being separately assessed for each individual. Some respondents do not know or do not have the necessary information to answer specific questions about individual benefits which makes it difficult to collect accurate information. State Benefits on the Family Resources Survey (WP115)

Relative to administrative records, the FRS under-reports numbers on benefit (caseload). See Methodology Tables M_6a and M_6b for a comparison of (i) numbers on benefit (caseload) and (ii) the average £ per week received, showing any differences between DWP administrative data and the numbers implied by the survey results. However, one of the strengths of the FRS is that it collects many personal and family characteristics which are not available from administrative sources. This means that the FRS can be used to analyse income and benefit receipt in ways which are not possible from administrative sources alone.

Tenure

The difference between the distribution of households across council tax bands in Great Britain captured in the survey, compared to administrative data has been minimised by the grossing regime applied. It should be noted that the grossing regime applied in 2021 to 2022 was more effective in enabling a representative dataset of English and Scottish households, than Wales, where sample sizes were relatively smaller. See Methodology Table M_3.

As presented in the FRS, the “social renting sector” is a combination of the categories “Rented from Council” and “Rented from a Housing Association”. These categories are combined because some housing association tenants may misreport that they are council tenants. For instance, where their home used to be owned by the council and although ownership has now transferred to a housing association, the tenant may still think that their landlord is the council (local authority).

FRS outcomes are similar in composition to the English Housing Survey (EHS), but it should be noted that the EHS also switched from face-to-face interviews to telephone, so the ability to contact interviewees would have been similar to that of FRS interviewers.

Disability

For those above State Pension age, the reduction seen between FYE 2020 and FYE 2021 may have been a consequence of the change in mode from face-to-face to telephone interviewing, as fewer participants reported impairments in hearing, memory or vision. There is still evidence of a mode effect in the FYE 2022 disability sample, with visual, hearing, and memory impairments again under-represented compared to before the coronavirus (COVID-19) pandemic. This largely affects the pensioner disability sample. However, we are reassured that growth in other impairment types is broadly similar to external sources showing changes in the composition of the disabled population since the pandemic. Outcomes for disabled people in the UK - Office for National Statistics (ons.gov.uk)

In FYE 2021 it was considered that the increase in working-age levels of disability reported in the FRS was a reflection of “real-world” change; that it may have been because of measures which limited movement outside of the home and restrictions imposed on how people were able to socialise.

The situation was different in FYE 2022, with non-essential retail and all outdoor venues reopened from April 2021; by 19 July 2021 most legal limits on social contact were removed and the final closed sectors of the economy reopened. The COVID-19 guidance to work remotely ceased to apply after 26 January 2022. However, for disabled working-age adults, those reporting a mental health impairment has increased further from 42% to 44% in 2021 to 2022; the most prevalent impairment type among this age group. This may suggest that some effects are longer lasting, which can also be seen in The employment of disabled people 2022; the ONS publication sourced from data captured in the APS and LFS.

It has been recognised that in this survey year the FRS may be reporting a wider disability employment gap when compared with other sources. The FRS should be looked at in conjunction with both the Annual Population Survey (APS) and the Labour Force Survey (LFS). This reports a disability employment gap of 26.8 percentage points, compared to the FRS gap of 31 percentage points (table 4_7). However, it should be noted that the FRS reports for working-age adults only, whereas the APS reports on all adults.

The APS aims to provide more detailed analysis of key labour market indicators (employment, unemployment, and economic inactivity) for sub-groups of the population including disability. The latest analysis considers long-term trends as well as the more recent impacts of the coronavirus (COVID-19) pandemic. The APS is not a stand-alone survey: it uses data combined from two waves of the main Labour Force Survey (LFS), alongside a local sample boost. The APS is a recommended source for employment statistics for smaller groups of the population. The LFS is the source recommended for employment-related statistics, such as estimates of the number of people in employment and is the key source for trend data on different measures of disability employment. The latest releases can be found in the Alternative Data Sources section of this report.

The FRS does not record information on individuals in nursing or retirement homes. This means that figures relating to elderly people may not be representative of the UK population, as many elderly people may have moved into homes where they can receive more frequent help. Therefore, it is likely that disability figures and impairments among all older people are higher than estimated from the FRS.

The way in which disabled people have been identified in the FRS has changed over time. From FYE 2003 statistics were based on responses to questions about barriers across several areas of life; figures from FYE 2005 to FYE 2012 are based on those reporting barriers across nine areas of life.

From the 2012 to 2013 survey year, a person is considered to have a disability if they regard themselves as having a long-standing illness, disability or impairment which causes substantial difficulty with day-to-day activities. This updated definition is consistent with the core definition of disability under the Equality Act 2010, and complies with harmonised standards for social surveys published in August 2011.

An impairment is different to a medical condition. It looks at the functions that a person either cannot perform or has difficulty performing because of their health condition. For example, glaucoma is a medical condition but being unable to see or being partially sighted is an impairment.

Some people classified as disabled and having rights under the Equality Act 2010 are not captured by this definition, such as people with a long-standing illness or disability which is not currently affecting their day-to-day activities. More information is available from the GSS Policy Store.

Care

Those receiving care

FRS respondents are asked if they receive care from anyone. This includes both professional help – paid-for care from the local authority, health professionals or domestic staff – but it also includes informal care. This is any care where their carer is not doing it as a paid job; it can be for many, or only a few hours a week, and can take several different forms. The survey is intentionally not prescriptive about what counts as care; it could, for example, include going shopping for someone, or helping them with paperwork.

The FRS does not record information on individuals in nursing or retirement homes. This means that figures relating to elderly people may not be representative of the UK population, as many elderly people may have moved into homes where they can receive more frequent help. Therefore, it is likely that care provision and receipt amongst older people is higher than estimated from the FRS.

Where a care recipient is receiving care at least once a week, further questions are asked to determine the care provider and the length of time spent providing care by each carer.

Informal carers

FRS respondents are also asked if they provide care to someone else, on an informal basis. That person could be living with them, in their household, or they could live somewhere else (outside the household).

From this publication, all respondents in receipt of Carer’s Allowance are now counted as informal carers, regardless of their independent responses to questions on care provision. This only resulted in minor adjustments and has not materially affected the published estimates for numbers of care recipients.

New to this publication’s data tables is the classification of some informal carers as ‘Sandwich Carers’. In the FRS data a Sandwich Carer is defined as someone (aged 16-70), caring for a child in their household and for an adult relative. The definition used in this analysis can be found in the Glossary and is adapted mainly from the ONS definition, which is taken from Understanding Society. Improvements have been made to define a more precise definition of Sandwich Carer, clearly specifying: that the child within the household of the carer is dependent on them or receiving care from them; that the adult receiving care is any adult relative; and that children can potentially be Sandwich Carers.

Following a review of the data tables in this chapter, figures published this year are now calculated differently to correct for observed inconsistencies. Breakdowns showing proportions of informal carers looking after care recipients have been revised, to show proportions on a consistent basis with the accompanying totals.

As a result of these changes, figures published this year have contributed to some notable differences compared to previous years of published FRS results, such as the apparent increase in the proportion of informal carers caring for a parent inside their household, which rose to nine per cent (would have been eight per cent without the methodology change), and the apparent increase in the proportion of informal carers caring for their child inside their household, which rose to 19% (would have been 15% with the previous methodology). As a result of these changes, figures published this year are not directly comparable to those published in previous years.

Table 5_7 has been redesigned to form two new tables, now 5_7a and 5_7b, and they include some additional breakdowns. The relevant charts have also been updated. These new breakdowns were previously only available to users of the published FRS dataset; they were not previously included in the publication. Comparisons against the previous methodology are still available from FRS data published on Stat Xplore.

This year we have presented ‘Income from Universal Credit (UC)’ as its own category as a source of income. This has allowed for more detailed breakdowns within Care Tables 5_5 and 5_10. This change is because the volume of UC cases is now sufficiently high to allow for meaningful analysis.

Pension participation

The Coronavirus Job Retention Scheme (CJRS), until its closure at the end of September 2021, covered wages, but employers were still responsible for paying National Insurance contributions and the minimum employer pension contribution of three per cent of qualifying wages required under automatic enrolment. Salaries of all furloughed workers, full-time and part-time, remained pensionable.

The FRS pension participation data tables present estimates for both ‘all adults’ and ‘working-age adults only’. Those over State Pension age are often excluded from analysis of pension participation in other publications, although they could continue to work and participate in pension schemes. The ‘all adults’ category allows data for this group to be represented and provides continuity across all chapters within the FRS.

Employer-sponsored pensions comprise any company or occupational pension scheme run by an employer, including group personal pensions and group stakeholder pensions.

Individual pensions include individual stakeholder pensions and retirement annuity contracts as well as personal pensions.

Although the numbers are relatively small, self-employed people can contribute to an employer-sponsored pension scheme, for a variety of reasons. Doctors and dentists in private practice can be members of an occupational scheme. People who have recently become self-employed can continue to contribute to their previous employer scheme and people whose main job is self-employed, may work part-time as an employee and contribute to an employer scheme. These circumstances are captured within the FRS tables under the ‘Self Employed – Other’ category.

Savings and investments

The FRS does not capture information on non-liquid assets. Physical wealth and pensions accruing are not included in FRS estimates. The survey also does not capture detailed information on expenditure (except for housing costs). Therefore, it is not possible to show how households are coping financially, in terms of income versus outgoings.

However, the FRS does capture information on liquid financial assets, referred to in the survey as ‘savings and investments’. Estimates for savings and investments should be treated with caution, as they are likely to be under-estimates, since respondents often inaccurately report their account details.

The process of gathering information on savings and investments maintained the same methodology as that introduced in 2020 to 2021:

• respondents who have at least one account are asked, as a benefit unit, to say which of several £ bands their total level of savings and investments are in

• benefit units that report between £1,500 and less than £30,000) (35% of applicable benefit units) are then asked, for each of their accounts and assets, how much each is worth and how much interest they accrue. The total level of savings and investments is then calculated using this set of reported values

• benefit units with reported savings and investments outside those limits – below £1,500 or above £30,000 (65% of applicable benefit units) – are only asked how much interest each account and asset accrue. These respondents are also asked to estimate the value of all their current accounts and basic bank accounts combined

Self employment

It is difficult to calculate current-year income for the self-employed. In line with international standards, the FRS calculates self-employed income from the profit data for a previous tax year or regular self-employment income over the past twelve months. Whilst this provides less of an issue when incomes are broadly stable, this was more of a challenge in both FYE 2021 and FYE 2022, given the changes in self-employed incomes caused by the coronavirus (COVID-19) pandemic. The statistics for these years suggest a materially lower number of self-employed people when compared to survey years prior to the pandemic. This is consistent with trends in other sources.

The survey has also had to adapt to several forms of government assistance for the self-employed:

• for those claiming Universal Credit (UC), the government determined that, from 6 April 2020 until 31st July 2021, the Minimum Income Floor would be temporarily relaxed. Self-employed people claiming UC would thereafter have their UC calculation based on their submitted earnings and not the Minimum Income Floor. It follows that where FRS incomes include UC, that UC payment will include the additional amount

• the government introduced the Self-Employment Income Support Scheme (SEISS) to help the self-employed who were affected by the coronavirus (COVID-19) pandemic. Although the FRS specifically asked about receipt of SEISS grants from June 2020 to the end of the scheme in September 2021, self-employed income amounts reported in the FRS do not include the grants reported by the respondent received from SEISS. This means that household and individual income amounts do not directly include grants reported by the respondent received from SEISS

The calculation of self-employed income and then total individual income does not include any grants reported by the respondent received from the Self-Employment Income Support Scheme (SEISS). FRS calculates self-employed income from the profit data for a previous tax year.

Receipt of the first three SEISS grants was treated as taxable income when calculating profits in FYE 2021 tax returns to His Majesty’s Revenue & Customs. As such, money received from the scheme will have been automatically included in income estimates for self-employed persons who reported their FYE 2021 profit data. Therefore amounts reported as SEISS grants are not added to FRS ‘self-employed income-related’ variables in the calculation of income, to avoid double counting.

The expected effect of SEISS in reporting levels and characteristics of self employment is that people will remain as self-employed but may class themselves as temporarily away from work and record no hours of employment. However, as under the terms of the scheme, they can continue to work or take on other employment, their economic status and number of hours worked may change during the scheme’s lifespan. This may affect the reporting of self-employed income.

The FRS asks a detailed set of questions to capture earnings from self employment, as described in the Glossary, at the end of this document. Income from dividends have been included in the source of self-employed income for the first time in the 2021 to 2022 release. New variables, generated from new questions asked about directors’ dividends, DIVIDGRO (Gross dividend) and DIVIDNET (Net dividend) have been added to the ADULT table and incorporated into ‘Individual Income’ [INDINC] and ‘Net individual Income’ [NINDINC] respectively. Neither of these variables feature as distinct items within the publication tables, but the positive effect upon income tables is that a greater level of precision has been added to the calculation of these input variables, and the resultant summation of income variables.

The FRS does not fully capture information on all types of income-in-kind accurately – for example, benefits of vehicles, computers and mobile phones purchased by the business – that are also for personal use. And these benefits are likely to be more important for the self-employed than for employees. Therefore, the FRS earnings measures are likely to underestimate the true monetary and other benefits of self employment. However, it is very difficult to quantify this.

Other benefits of self employment compared to employment are not captured, such as flexibility in working patterns, independence and flexibility in the way money is drawn from the business. The complexity of self-employment circumstances, with irregular income and benefits-in-kind coming from a range of sources, could also contribute to inaccuracy of information capture.

For self-employed individuals, net income figures are presented after any deductions which include, but are not limited to tax, national insurance and pension contributions. Where gross income figures are presented, these include all these elements.

One of the significant advantages of the FRS is that it has captured self employment in a consistent way over time. Therefore, the trends in self employment compared to employment are likely to be reasonably accurate.

The Labour Force Survey (LFS) is considered the definitive source where numbers participating in the labour market are concerned. It has previously been recognised that the FRS does undercount the number of people reporting self employment compared with the Labour Force Survey. This continues to be the case.

Both the LFS and the FRS saw a reduction of self-employed numbers in 2020 to 2021, compared to 2019 to 2020, and earlier years. However, in 2021 to 2022 LFS saw a further reduction (0.2 million decrease), whereas FRS saw an increase in self-employed people (0.3 million increase). The overall numbers of self-employed people reported by the two surveys are now closer than ever, with only a 0.3 million difference, (LFS 4.2 million; FRS 3.9 million).

In 2020 to 2021, both overall and for female respondents, the fall in the numbers of self-employed was larger for the FRS than in the LFS. The proportions by gender in 2021 to 2022 are once again consistent across the two surveys (as they were in years prior to 2020 to 2021).

The LFS has published a report into Understanding changes in self-employment in the UK: January 2019 to March 2022

Household food security

Since the introduction of questions on household food security in the 2019 to 2020 survey year the FRS continues to provide evidence on this subject. From April 2021, the FRS asked additional questions on food bank usage. Food banks can provide support other than food, such as financial advice or mental health support, but the FRS records “usage” as visits to a food bank for the purpose of obtaining emergency food supplies only.

Questions covering household food security and food bank usage are asked of the person in the household who knows the most about buying and preparing food. In common with the rest of the FRS, household food security questions focus on the period immediately before the interview (30 days).

For household food bank usage, questions are asked about two separate time periods: a lead-in question asks about usage within the 12 months prior to interview; and then households that report using a food bank in the last 12 months are asked about usage within the 30 days prior to the interview. The questions do not directly ask about the food bank usage needs of children, and it cannot be determined which individual or individuals the food parcels are for.

Caution is needed when comparing household food security status with 12-month food bank usage. The effect of household food security upon food bank usage, cannot be fully deduced because the former only asks about the household’s circumstances in the last 30 days.

There is no standard approach to either household food security or food bank usage. The questions used by the FRS are similar to those used by other public bodies in the UK, and also internationally, but there are some differences in their application via interview.

These statistics should be treated with caution when interpreting them:

• where a household is food insecure, information about the individual experiences of food insecurity within the household is not available. A young child’s experience in a food-insecure household may be very different from their parent’s, for example

• household food security statistics do not directly measure hunger. They instead explore the financial situation of households and how that affects their access to food. Only households with very low food security would anticipate substantive disruption to their food intake

• the statistics presented exclude shared households, such as a house shared by a group of professionals

With both measures of household food bank usage captured in the FRS, caution should be taken when interpreting figures and comparing with other sources. Most other sources captured usage from a single period, so may only be comparable to one FRS measure, if at all. It should also be noted that some sources measure individual food bank usage rather than household.

For further information see the Glossary section and the relevant publication tables.

Adjusting for inflation

Some figures in the main FRS report and the accompanying tables combine several years of income data. In these circumstances, uprating factors are used to adjust for inflation by bringing values from previous years into current price terms.

Since the 2014 to 2015 FRS, the Consumer Price Index (CPI) has been used to adjust for inflation. Read more information concerning this methodological change.

5. Policy changes for the year 2021 to 2022

Council Tax

The Department for Levelling Up, Housing and Communities estimated that the average Band D council tax set by local authorities in England for 2021 to 2022 increased by 4.4% from 2020 to 2021 levels.

In Wales, the average Band D council tax for 2021 to 2022 represented an increase of 3.8% from 2020 to 2021 levels.

In Scotland, the average Band D council tax for 2021 to 2022 has not increased from 2020 to 2021 levels.

In Northern Ireland, there were increases in rates (poundage) of no more than one per cent in some council areas, but in others the rates (poundage) remained as it was in 2020 to 2021.

National Living Wage

On 1 April 2021, the National Living Wage increased to £8.91 per hour for employees aged 23 years and above.

Employees aged under 23 years continued to receive the National Minimum Wage. On 1 April 2021, the National Minimum Wage increased to £8.36 per hour for those aged 21 to 22 years inclusive, £6.56 per hour for those aged 18 to 20 years inclusive and £4.62 per hour for those aged below 18 years (but over compulsory school leaving age). Additionally, the National Minimum Wage rose to £4.30 per hour for apprentices, both those aged below 19 years and those aged 19 years and above who were in the first year of their apprenticeship.

Universal Credit and Tax Credit uplift

From April 2021, the temporary £20 per week Universal Credit uplift continued until 6 October 2021 when it ended. For working households receiving Tax Credits, a one-off payment of £500 was announced by the government and made in April 2021.

Universal Credit removal of minimum income floor for self-employed people

Between 6 April 2020 and 31 July 2021, the government suspended the Minimum Income Floor (MIF) so that self-employed Universal Credit claimants with earned incomes below the MIF would not be treated as earning more than they actually had.

Reducing the Universal Credit taper rate

At the Autumn 2021 Budget, it was announced that the Universal Credit taper rate would be reduced from 63% to 55%. The taper is a reduction to a claimant’s Universal Credit based on their earned income. This change was implemented from 1 December 2021.

Up-rating

In April 2021:

• inflation-linked benefits and tax credits rose by 0.5% in line with the Consumer Prices Index (CPI)

• the Basic and New State Pension increased by 2.5% in line with the ‘triple lock’. The ‘triple lock’ ensured that in 2021 to 2022 both the Basic and New State Pension increased by the highest of the increase in earnings, price inflation as measured by the CPI or 2.5%. The Basic State Pension increased from £134.25 per week to £137.60 per week, a cash increase of £3.35 per week. The New State Pension increased from £175.20 per week to £179.60 per week, a cash increase of £4.40 per week

• the Standard Minimum Guarantee in Pension Credit increased by 1.9%. For those who were single, the Standard Minimum Guarantee in Pension Credit increased from £173.75 per week to £177.10 per week, a cash increase of £3.35 per week. For couples, this increased from £265.20 per week to £270.30 per week, a cash increase of £5.10 per week

• both the lower and higher Universal Credit Work Allowances rose broadly in line with the CPI

Pensioner TV licences

Eligibility for a free licence for all pensioners aged 75 years and older was limited to those in receipt of Pension Credit from August 2020. Those not in receipt of Pension Credit could continue to receive a free TV licence until their annual renewal. After 31 July 2021, only 75 year olds receiving Pension Credit will have been entitled to a free TV licence.

Rent and mortgage payments

The government announced, on 17 March 2020, that those struggling to pay their mortgage or rent because of the coronavirus (COVID-19) pandemic and landlords with buy-to-let mortgages whose tenants were unable to pay the rent were able to apply for a payment holiday. Payment holidays were extended until 31 July 2021.

Payment holidays could either last up to three months or up to six months. Extra support should have been given by lenders through tailored forbearance options for those who continued to face financial struggles once their payment holiday ended.

The introduction of the Coronavirus Act 2020 stopped both landlords and lenders from evicting those who occupied their properties. Additionally, there was a ban on repossessions from November 2020 until the end of May 2021.

Housing Support for private renters: in April 2021, Local Housing Allowance rates were maintained at the 30th percentile of market rents.

Self-Employment Income Support Scheme

The Self-Employment Income Support Scheme (SEISS) was introduced by the government to help those who were self-employed or a member of a partnership in the United Kingdom and lost income because of the coronavirus (COVID-19) pandemic.

The fourth round of the SEISS paid 80% of three months’ average trading profits of the claimant, up to £7,500 in total, and covered from February 2021 up to and including April 2021.

The fifth and final round of the SEISS was partly determined by the amount a claimant’s turnover had reduced in the year April 2020 to April 2021. The grant was worth either 80% of three months’ average trading profits, up to £7,500 in total, for claimants with a turnover reduction of 30% or above or 30% of three months’ average trading profits, up to £2,850 in total, for claimants with a turnover reduction below 30%. This round of the SEISS covered from May 2021 up to the end of September 2021.

‘Furlough’ - Coronavirus Job Retention Scheme

The Coronavirus Job Retention Scheme (CJRS), announced by the government in March 2020, was extended until 30 September 2021. Employers who were unable to keep their workforce due to the coronavirus (COVID-19) pandemic were able to put their employees on furlough and apply for a grant. During this scheme, government and employer contributions varied to ensure that employees received 80% of their monthly wage, up to £2,500 per month. At the employer’s own expense, they could top up their employees’ wages above this threshold.

’IR35’ private sector changes

Reforms to worker classification, from self-employed to employee, were implemented from April 2021.

ONS (Labour Force Survey) statistics on self employment show that in January to March 2021, prior to the implementation of the reforms, 155,000 workers (72.8%) who moved from self employment to employee status reclassified. In April to June 2021, this fell to 88,000 workers (62.2%) who reclassified. This fell again in January to March 2022, with 75,000 workers (53.5%) who reclassified.

Household Support Fund

The government announced, on 30 September 2021, that vulnerable households in England would be able to access a £500 million support fund with the aim of helping them with necessities over winter. This was in place from 6 October 2021 up to and including 31 March 2022. On 23 March 2022, an extension to 30 September 2022 to the Household Support Fund was announced with an additional £500 million of funding. To help directly the most vulnerable households, the Household Support Fund was distributed by councils in England through small payments. The aim was to ensure that daily needs such as food, clothing, and utilities of those in such households were met.

6. Alternative data sources

Income and state support

A Guide to Sources of Data on Earnings and Income

The Income and Earnings Interactive Tool where you can filter by government department and country of interest to find relevant statistics.

The Effects of Taxes and Benefits on Households

Living Costs and Food Survey

Cost of living insights tool

Wealth and Assets Survey

Income, spending and wealth: how do you compare? – joined-up data from the Wealth and Assets Survey (WAS) and the Living Costs and Food Survey (LCF) providing insight into the financial vulnerability of different households.

Annual Survey of Hours and Earnings

Labour Force Survey

Benefits statistics on Stat-Xplore

Households Below Average Income on Stat-Xplore

Pensioners’ Incomes Series on Stat-Xplore

Income Dynamics: Income movements and persistence of low incomes

ONS: explanation of incomes and earnings

Changing trends and recent shortages in the labour market, UK: 2016 to 2021

Tenure

English Housing Survey

Private Landlords Survey

English Private Landlords Survey

Index of Private Housing Rental Prices

Housing affordability in England and Wales: 2022

Rent affordability: Literature and evidence review: 2019

Landing page of the GSS housing and planning statistics interactive tools (civilservice.gov.uk)

Disability

Life Opportunities Survey

Outcomes for disabled people in the UK 2021

The employment of disabled people 2022

Labour market data for protected groups in Wales and the UK, April 2004 to March 2021

Disabled people in the labour market in Scotland 2019

Disability Employment Gap in Northern Ireland 2020

Revisions to medical condition ICD high level grouping codes for ESA and IB/SDA

Care

Census 2021 (Unpaid Care) dataset

Census 2021 (Unpaid care by age, sex and deprivation)

Health, Disability and Unpaid Care, Census 2021 in England and Wales

The geographic divide in general health, disability and unpaid care: Census 2021 - Office for National Statistics (ons.gov.uk)

Personal Social Services Adult Social Care Survey Series (England)

UK adult social care statistics

Health Survey for England Series

Scottish Health Survey Series

National Survey for Wales Series

Health Survey Northern Ireland Series

English Longitudinal Study of Ageing Wave 9, 2002-2019

Understanding Society data access

Sandwich Carers dataset

Pension participation

Occupational Pension Schemes Survey

Note that the collection and publication of the annual Occupational Pension Schemes Survey (OPSS) has ceased. A quarterly publication has superseded this.

Employers’ Pension Provision Survey

The Pensions Regulator – DC Trust: a presentation of scheme return data

The Pensions Regulator

DWP State Pension Statistics from November 2020 to May 2022

HMRC Pensions Tables

English Longitudinal Study of Ageing Wave 9: 2002-2019

Annual Survey of Hours and Earnings (pension tables)

Savings and investments

Annual savings statistics 2022: HMRC statistics on Individual Savings Accounts, Child Trust Funds and Help to Save accounts

Self employment

Trends in self employment in the UK

Labour Market overview UK (including breakdown of the self-employed)

Household food security

Food and You 2 - Wave 3, Food Standards Agency

Food and You 2 - Wave 4, Food Standards Agency

Food and You 2 - Wave 5, Food Standards Agency

ONS Opinions & Lifestyle Survey

Public opinions and social trends, Great Britain - Office for National Statistics

7. The FRS questionnaire

Relevant changes made to the FRS questionnaire in 2020 to 2021, due to the ongoing effects of the coronavirus (COVID-19) pandemic, were kept in place for the fieldwork year 2021 to 2022, to continue to collect information on government support initiatives. This is important to enable the statistics to reflect changes in the economy and society, and to inform policy. FRS interviews were conducted using Computer Assisted Telephone Interviewing (CATI) during 2021 to 2022.

The questionnaire is divided into three parts.

• the first part is the household schedule which is addressed to one person in the household (usually the household reference person, although other members are encouraged to be present) and mainly asks household level information, such as relationships of individuals to each other, tenure and housing costs

• next is the individual schedule which is addressed to each adult in turn and asks questions about employment, benefits, pensions, investments, and other income. Information on children in the household is collected by proxy from a responsible adult

• a final section asks the value of investments (by type) for respondents with savings between a lower and an upper pound limit

Interviewers new to the FRS are briefed on the questionnaire and an annual re-briefing is given to all interviewers on changes to the questionnaire.

For 2021 to 2022 additional training was provided to interviewers on key aspects of the questionnaire to be aware of, with respect to telephone interviewing. This included how to ask questions which were designed to reference in-person showcards, including:

• adapting the question wording to avoid mentioning the showcard

• for most questions, simply reading aloud all responses as listed on the showcard

• some further specific guidance for certain questions with unusually long showcards

Recommended approaches for collecting information from two-person benefit units were also covered, ideally speaking to both respondents at once, using for example a speakerphone or two handsets.

Interviewers who have worked on the survey for some time also completed a written field report each year, describing their experiences with specific parts of the questionnaire and commenting on how changes were received in the field.

Prior to the start of fieldwork, DWP consults FRS users and draws up a list of possible questionnaire changes. Users are asked to identify individual questions or sections which were no longer of interest. The FRS questionnaire is lengthy and demanding and a key concern is, where possible, to reduce (or at least not increase) its length, so as not to overburden respondents or interviewers.

As part of the process of agreeing annual changes, suggestions from contractors are also considered, as well as those arising from an evaluation of feedback from interviewers. Any changes to the questionnaire are checked for consistency with the harmonised standards for social surveys across government.

Questionnaire changes

The main aim of the annual questionnaire consultation is to improve our understanding of respondent circumstances, which should in turn improve data quality (editing) decisions taken by the FRS Project team. New questions are added with the expectation that they will produce useful data that can be delivered to users through additional variables and used to support future policy analysis.

Whilst the resulting variables will be released in the FRS 2021 to 2022 dataset, they have not all been added to the main FRS publication on GOV.UK; nor to the set of FRS tables available from the Department’s Stat Xplore.

Whilst the questionnaire was largely unchanged from the previous 2020 to 2021 survey year, several important changes were made:

Food bank usage

Due to the significant increase in food bank usage over the past decade, it is important to collect information to increase understanding of household food security and the circumstances of people using food banks. New questions were added at the end of the household food security block to establish whether the respondent had used a food bank, firstly in the last 12 months, and then in the last 30 days.

Internet access

A new question was added to identify how many people can access digital services online, and whether they would need help in doing so. Many public services are now online, including DWP’s, so it is important to the department to have this insight.

Doctors and dentists

Changes were made to the question ‘Nature’. The revised response options now record doctors and dentists separately, because their pay and working arrangements differ to most of the working population. New questions were added after Nature to collect better information about the doctor’s exact role (NHS, versus private or teaching; hospital versus GP; and partner or otherwise) with dentists identified as either partners or employees in their practice.

Directors and dividends

Directors of limited (or PLC) companies have complex incomes, that can involve taking dividends, alongside receiving salaries and/or drawing from the company’s profits. To get as complete a picture as possible of a company director’s entire income stream, new questions were added: These included further changes to the question ‘Nature’, with a new category recording whether the respondent was a ‘Director, Managing Director, Chief Executive of a limited company, or PLC’; and then questions on any dividends received.

Discretionary Housing Payments (DHPs)

DHPs are either one-off or regular payments given by a Local Authority for a range of reasons to people who need more help with housing costs. New questions have been added to collect information on whether people renting their accommodation have received a Discretionary Housing Payment (DHP) and how much, and how regularly they received the payment.

Flexible working

A new question was added to collect information about the different types of flexible working arrangements that people may have. This question will help users understand how particular forms of work, including zero-hours contracts and “gig” work, affect the standard of living of people involved in such work.

Pension freedoms

Three new questions were added, asked of respondents who reported taking a lump sum or withdrawal from their pension. This will allow users to gain a better understanding of how people are accessing their pension and whether they are using their pension freedoms, which allow them to withdraw tax-free lump sums from their pension.

A new question was also added to ask the age at which the pension was first accessed (received). As pension freedoms now allow people to access their defined contribution (DC) pension pots from age 55, this information will be crucial to understanding when people are taking their pension. This question will give an insight into whether there are certain ages or types of people who are more likely to take a lump sum and whether there are certain circumstances which mean the decision to take a lump sum is made.

Renewable energy installation and income (Northern Ireland only)

With a new Energy Strategy being developed within Northern Ireland, a new set of questions was added to the NI questionnaire to identify households with renewable energy installations, and then whether they received any income from the installations.

Other changes

Numerous minor updates and changes to the questionnaire were made in response to feedback from interviewers on the operation of the questionnaire. Changes also stemmed from categories or definitions which were new for the 2021 to 2022 survey year. These included changes in relation to areas of policy overseen by the devolved administrations.

As in every survey year, a small number of removals were made, of questions which were either no longer relevant, or which were answered by too small a pool of respondents to yield useful information.

8. Population and sample selection

The FRS sample is designed to be representative of private households in the United Kingdom.

The sampling frame in Great Britain

The Great Britain FRS sample is drawn from the Royal Mail’s small users Postcode Address File (PAF). The small users PAF is limited to addresses which receive, on average, fewer than 50 items of post per day and which are not flagged with Royal Mail’s “organisation code”. The sampling frame therefore excludes people living in institutions, for example nursing homes, halls of residence, barracks or prisons, and homeless people living rough or in bed and breakfast accommodation.

An updated version of this list is obtained twice a year. By using only the small-user delivery points most large institutions and businesses are excluded from the sample. Small-user delivery points which are flagged as small business addresses are also excluded. However, some small businesses and other ineligible addresses remain on the sampling frame. If sampled, they are recorded as ineligible once the interviewer verifies that no private household lives there.

Sample design in Great Britain

The Great Britain FRS uses a stratified clustered probability sample design. The survey samples 1,417 postcode sectors, from around 9,200 in Great Britain, with a probability of selection that is proportional to size. Each postcode sector is known as a Primary Sampling Unit (PSU). The PSUs are stratified by 27 regions and three other variables, described below, derived from the 2011 Census of Population. Stratifying ensures that the proportions of the sample falling into each group reflect those of the population.

Within each region the postcode sectors are ranked and grouped into eight equal bands using the proportion of households where the household reference person (HRP) is in National Statistics Socio-Economic Classification (NS-SEC) 1 to 3. Within each of these eight bands, the PSUs are ranked by the proportion of economically active adults aged 16-74 and formed into two further bands, resulting in sixteen bands for each region. These are then ranked according to the proportion of economically active men aged 16-74 who are unemployed. This set of stratifiers is chosen to have maximum effectiveness on the accuracy of two key variables: household income and housing costs. The table below summarises the stratification variables. Within each PSU a sample of addresses is selected.

FRS sample stratification variables for Great Britain

Regions	19 in England (including Metropolitan vs non-Metropolitan split; 4 in London) 2 in Wales 6 in Scotland
The proportion of households where the HRP is in NS-SEC 1 to 3	8 equal bands
The proportion of economically active adults aged 16-74	2 equal bands
The proportion of economically active men aged 16-74 who are unemployed	Sorted within above bands

Each year, half of the PSUs are retained from the previous year’s sample, but with new addresses chosen; for the other half of the sample, a fresh selection of PSUs is made (which in turn will be retained for the following year). This is to improve comparability between years.

Within each PSU a sample of addresses was selected. In FYE 2022, the number of addresses selected within each PSU varied according to location and month. This was because in the 2021 to 2022 survey year a substantial sample boost was introduced in October. Between April – September 2021, 28 addresses were selected per PSU for most regions, in Scotland 38 addresses were selected. During the period October 2021 to March 2022, addresses per PSU varied by region (from 44 to 84) because the boost was targeted at certain areas.

The total Great Britain set sample size in FYE 2022 was 58,764 addresses. Each address had approximately a 2-in-967 chance of being included in the survey. For England each address had approximately a 1-in-529 chance of inclusion in the survey. In Wales each address had approximately a 1-in-471 chance of inclusion in the survey. In Scotland each address had approximately a 1-in-288 chance of inclusion in the survey.

The sampling frame in Northern Ireland

The sampling frame employed on the Northern Ireland FRS is the NISRA Address Register (NAR). The NAR is developed within NISRA and is primarily based on the Land and Property Services (LPS) Pointer database, the most comprehensive and authoritative address database in Northern Ireland, with approximately 752,000 address records available for selection.

The sample design in Northern Ireland

A systematic random sample is selected for the Northern Ireland FRS from the NISRA Address Register. Addresses are sorted by district council and ward, so the sample is effectively stratified geographically. Multi households are not selected in Northern Ireland.

In Northern Ireland the sample size and sample approach was the same as in the previous survey year, FYE 2021. A systematic random sample of 4,080 addresses was selected for the 2021 to 2022 Northern Ireland FRS from the NISRA Address Register. Addresses are sorted by district council and ward, so the sample is effectively stratified geographically. Each address had approximately a 1-in-158 chance of being selected for the survey.

9. Data collection

FRS 2021 to 2022 survey fieldwork operations continued to take account of the effect of government guidance relating to the coronavirus (COVID-19) pandemic and the restrictions placed on visiting others’ homes. The telephone interview approach that was adopted during 2020 to 2021 survey year continued throughout 2021 to 2022, with interviewers conducting the interview while working from their own homes.

Use of “Knock to Nudge”, where interviewers visited addresses for which they had not received any contact details, continued throughout the year in Great Britain and was introduced to Northern Ireland in July 2021. This approach supported overall response rates. Overall, the fieldwork approach was more stable in 2021 to 2022 than in 2020 to 2021, but conditions remained challenging.

Data collection in Great Britain

A consortium consisting of the Office for National Statistics (ONS) and NatCen Social Research conducts fieldwork for the FRS in Great Britain on behalf of the Department for Work and Pensions (DWP).

Each month the PSUs are systematically divided between the two organisations and then assigned to interviewers. If more than one household receives mail at an address a single household is interviewed.

FRS 2021 to 2022 survey fieldwork operations continued to take account of the effect of government guidance relating to the coronavirus (COVID-19) pandemic and any restrictions placed on visiting others’ homes. The telephone interview approach that was adopted during the 2020 to 2021 survey year continued throughout 2021 to 2022, with interviewers conducting virtually all interviews remotely.

Respondents were offered various options to provide their telephone numbers, including an online portal and a freephone number. Telephone numbers for sampled addresses were also obtained where possible to supplement those provided directly by respondents. This included telematching sourced via ‘ReAD’ and telephone numbers sourced by matching the FRS sample with telephone numbers held on DWP’s internal databases and other administrative systems.

During this time the main face-to-face contact with respondents was via doorstep contact whereby the interviewer visited the sampled addresses, as would be usual in face-to-face interviewing, to talk to the residents and encourage participation in the study. This approach is referred to as “knock to nudge”.

There was also a small-scale iterative trial where interviewers volunteered to trial face-to-face (FTF) interviewing between August-December (though was paused mid-December due to the Omicron variant). The FTF-led approach was introduced in March and so saw an increase in FTF interviews, with 12% of interviews conducted face-to-face in March; this only amounted to four per cent of all interviews across the year.

Both ONS and NatCen sent reminder letters to respondents although these were administered differently.  All ONS cases receive two letters as standard: one sent out by NatCen’s printers and then another directly from their individual interviewer. ONS interviewers had the option to send out a further “chaser letter” and were instructed to do this for all cases where they had not obtained a phone number. All NatCen cases receive two centrally despatched letters, one advance and one reminder sent a week later.

Data collection in Northern Ireland

In Northern Ireland the sampling and fieldwork for the survey are carried out by the Central Survey Unit at the Northern Ireland Statistics and Research Agency. The responsibilities for programming the survey questionnaire, making annual modifications, initial data processing and data delivery are retained within ONS and NatCen.

As in Great Britain, survey fieldwork operations in NI continued to take account of the impact of guidance relating to the coronavirus (COVID-19) pandemic and the restrictions placed on visiting others’ homes. The telephone interview approach that was adopted during the previous survey year continued throughout the whole of the 2021 to 2022 survey year, with interviewers conducting the interview whilst working from their own homes.

To facilitate interviewers making contact with respondents, participants were asked, in the advance materials, to express their willingness to participate in the survey in the first instance, by phone or online. For the first quarter of the 2021 to 2022 survey year NISRA also sent an ‘interviewer follow-up’ letter to all sampled addresses approximately one week after the initial advance letters were issued. This follow-up letter included the name and direct contact details for the interviewer assigned to that address, as well as details for the online portal and office phone numbers as included on the advance letter. This follow-up letter was discontinued from July 2021 once interviewers were permitted to make face-to-face visits at their sampled addresses.

From July 2021 through to March 2022 interviewers were permitted to visit sampled addresses, as would be usual in face-to-face interviewing, to apply a ‘knock-to-nudge’ method of talking to the residents to encourage participation in the study. During this time the only face-to-face contact with respondents was through this doorstep contact. This has significantly increased participation in the survey compared with response rate achieved in 2020 to 2021 (and the first quarter of 2021 to 2022).

10. Response

A household is defined as fully co-operating when it meets the requirement of interviewing all adults aged 16 and over, except those aged 16 to 19 who are classed as dependent children and there are fewer than 13 ‘don’t know’ or ‘refusal’ answers to monetary amount questions in the benefit unit schedule (i.e. excluding the assets section of the questionnaire).

Proxy interviews are accepted when a household member is unavailable for interview. In 2021 to 2022, for those households classed as fully co-operating, proxy responses were obtained for 29% of adults. While this is higher than in previous years it could be a reflection of using telephone interviews (an increase that was observed in 2020 to 2021), with the additional effect of the removal of restrictions upon people leaving the house, such that the likelihood of other people in the household not being available at the time of interview had increased.

It should be noted that all data shown in the main body of this publication refer only to fully co-operating households.

Methodology Table M_1 summarises the household response.

The UK-wide sample chosen for 2021 to 2022 consisted of 62,844 households. In total 16,376 households UK-wide fully co-operated (26%), 515 partially co-operated (one per cent) and 20,040 refused to proceed with the interview (32%), either refusing when contacting the office, refusing when contacted by the interviewer, or being away during the fieldwork period. The interviewer was unable to contact 25,913 households (41%), which is a decrease from 2020 to 2021 when it was 59%. However, it should be noted that these figures cannot be compared to 2019 to 2020 and earlier years, as it was not possible to remove ineligible addresses from this figure.

It should be noted that these percentages are not comparable to survey years 2019 to 2020 and earlier. This is due to the switch to telephone interviewing, which removed the possibility of determining the eligibility of households (typically 10% ineligible). Therefore, for 2020 to 2021 and 2021 to 2022, the whole set sample has been used as the denominator for all response rate calculations.

Response rates are calculated as follows:

‘The number of fully co-operating households multiplied’ by 100 divided by ‘The number of households contained in the sample’

The overall response rate for the FRS in 2021 to 2022 was 26%.

The following chart compares response rates in NI with those in Great Britain.

Response rates by month, Great Britain and Northern Ireland, 2021 to 2022

When respondents refuse to participate in the FRS, interviewers record up to three reasons for refusal. The most common reasons for refusal in 2021 to 2022 are shown below:

Reasons for refusal to participate in the FRS, Great Britain, 2021 to 2022

Reason for refusal	Percentage of households
Couldn’t be bothered	27
Invasion of privacy	16
Genuinely too busy	14
Don’t believe in surveys	14
Concerns about confidentiality	11
Disliked survey of income	8
Personal problems	6
Anti-government	3
Temporarily too busy	3
Late contact – insufficient field time	1
Bad experience with previous surveys	1
Total number who gave a reason for refusal	12,470
Total number of refusals	16,397

People can give up to three reasons for not responding, so percentages do not sum to 100%. The percentage shown is of the households where someone gave a reason for refusal.

Refusals include those residents of an address that make contact to refuse to participate and residents who refuse to participate in the survey when contacted by the interviewer, either by telephone or on the doorstep. Those who aren’t available to proceed with the interview for other reasons, such as being away throughout the fieldwork period, are not counted in “Total number of refusals” shown here.

Methodology Table M_2 shows response rates broken down by region. All regions saw an increase in response in 2021 to 2022, compared with response in 2020 to 2021 (23%). COVID-19 restrictions continued to be in place during some periods of the year; this had an effect on data collection. It had a particular effect during the times of renewed government measures in December 2021, such as the response to the rise in the Omicron variant, which saw a return to the “work from home” rule.

Response rates by region showed that the East had the highest response rate in England, where 29% of all selected households responded fully. London had the lowest response rate where 18% of selected households fully co-operated.

Non-response

The lower the response rate to a survey, the greater the likelihood that those who responded are significantly unlike those who did not, and so the greater the risk of systematic bias in the survey results.

For a United Kingdom survey of the size and complexity of the FRS, the total non-response rate typically seen of around 50% is not considered unreasonable. However, given the effects of coronavirus (COVID-19) and the subsequent difficulties in interviewer recruitment, which have had impacts upon data collection, the non-response rate for the 2021 to 2022 survey year was 74%.

Any information that can be obtained about non-respondents is useful both in terms of future attempts to improve the overall response rate and potentially in improving the weighting of the sample results.

Non-response form analysis

Direct information about the non-responding households is valuable, although difficult to obtain.

In a normal survey year, some non-responding households who are not willing to take part in the full survey are willing to provide basic information by completing a non-response form. A detailed analysis of these forms is usually conducted to monitor characteristics of non-respondents and trends in non-response. However, due to the switch to telephone interviewing and other changes to field procedures introduced because of the coronavirus (COVID-19) pandemic, it was not possible for interviewers to record full non-response information during 2021 to 2022.

As interviewers did not visit all the sampled properties, they were not always able to record features such as barriers to entry that are relevant to non-response. They were also not able to record any observable characteristics of non-responding households such as the age and sex of non-responders.

FRS non-response and Council Tax band

Comparisons were made between the achieved sample of FRS responses in Great Britain, and administrative data for FYE 2022 on the number of households within each Council Tax band.

Methodology Table M_3 shows that the achieved (ungrossed) FRS sample has a smaller proportion of households in the lower Council Tax bands (A and B) than the administrative data, and a higher proportion of households in higher Council Tax bands.

Table M_3 also shows the extent to which the FRS grossing regime controls for this bias in the achieved sample, effectively correcting it to be closer to the proportions seen on the administrative data.

Composition of final sample, by mode of contact, 2021 to 2022

Length of interview

The telephone interview approach that was adopted during 2020 to 2021 survey year continued throughout 2021 to 2022. The length of each fully co-operating interview is recorded by the questionnaire program. In 2021 to 2022 the median interview length for Great Britain was 53 minutes, but the time varied according to the size of household and its circumstances.

The distribution of interview lengths in Great Britain is shown below, with full data in Methodology Table M_7. The timings exclude interviewer time spent preparing for and completing administration tasks after the interview. They are based on completed audit data from 14,616 fully productive ONS and NatCen interviews.

Distribution of FRS interview lengths, 2021 to 2022, Great Britain

Respondent burden

The Code of Practice for Statistics states that producers of statistics should consider the burden on survey respondents. The FRS can measure the burden placed on respondents by using measured interview times for 14,616 full interviews, in Great Britain.

Respondent burden in Great Britain is calculated as Number of responses multiplied by median interview time.

The median interview time for these 14,616 interviews was 53.1 minutes. Therefore, the respondent burden for the FRS in 2021 to 2022 was 776,110 minutes [539 days].

Consultation of documentation

Interviewers encourage respondents to consult documentation at all stages of the interview to ensure that the answers provided are as accurate as possible. For some items, whether certain documents are consulted or not is recorded on the questionnaire. This assists FRS users in assessing the accuracy of the data.

It should be noted that, due to the predominant use of telephone interviewing in 2021 to 2022, the consultation rates reported below may be less reliable than for face-to-face interviewing, as the interviewer was not able to observe directly whether documents were being checked.

• employees have consulted their latest payslip for 30% of jobs they have reported. Of all employees, 97% reported having one job only and three per cent reported having more than one job

• employees did not have a payslip to consult for four per cent of jobs they reported; 21% could not consult a payslip because their payslips were only received electronically

• sixty-two per cent of all reported benefit and payable Tax Credit receipt involved consultation of documentation (that is, a letter from DWP or HM Revenue and Customs, or a bank statement)

• sixty per cent of households in Great Britain consulted a Council Tax bill or statement in answering questions on their Council Tax payments

11. Validation, editing, conversion and imputation

In addition to unit non-response, where a household does not participate, a problem inherent in all large surveys is item non-response. This occurs when a household agrees to give an interview, but either does not know the answer to certain questions or refuses to answer them. This does not prevent them being classified as fully co-operating households because there is enough known data to be of good use to the analyst (although see the first paragraph of the Response section above for information about non-response to monetary questions).

The fact that the FRS allows missing values in the data collection can create problems for users, so missing values are imputed where appropriate. The policy is that for variables that are components of key derived variables, such as total household income and housing costs, and areas key to the work of DWP, such as benefit receipt, there should be no missing information in the final data.

In addition to imputation, prior to publication, FRS data is put through several stages of validation and editing. This ensures the final data presented to the public are as accurate as possible. The stages in the validation, editing, conversion and imputation process are laid out below:

Stage one – the interview

One of the benefits of interviewing using CATI is that in-built checks can be made at the interview stage. This helps to check respondents’ responses and that interviewers do not make keying errors. There are checks to ensure that amounts are within a valid range and cross-checks which make sure that an answer does not contradict a previous response. However, it is not possible to check all potential inconsistencies, as this would slow down the interview to an unacceptable degree, and there are also capacity constraints on interviewer notes. FRS interviewers can override most checks if the answers are confirmed as accurate with respondents.

Stage two – post-interview checks

Once an interview has taken place, data is returned to ONS, NatCen, or NISRA. At this stage, editing takes place, based on any notes made by interviewers. Notes are made by the interviewer when a warning has been overridden, for example, where an amount is outside the expected range, but the respondent has documentation to prove it is correct. Office-based staff make editing decisions based on these notes. Other edits taking place at this stage are checking amounts of fixed-rate benefits and, where possible, separating multiple benefit payments into their constituent parts, such as separating Disability Living Allowance into the Care and Mobility components.

Stage three – data conversion

Before further validation, FRS data is converted from CATI format into SAS-readable tables. Using DWP specifications, SAS-readable tables are created by ONS, with each table displaying information from different parts of the questionnaire. Both DWP and ONS then carry out validation checks on key input and output variables to ensure that the data have converted correctly to the new format. Checks include ensuring that the number of adults and children recorded is correct, and that records are internally consistent.

Stage four – state support validation

Information on benefit and tax credit receipt is one of the key areas of the FRS, and it is very important that this section is thoroughly validated and cleaned.

It is not appropriate to use the imputation methods outlined in stages five and six (below) for benefits data so instead a separate procedure of validation and editing is used. The following types of validation were carried out for 2021 to 2022 FRS data:

Missing values

For cases where a respondent had answered ‘yes’ to whether they are in receipt of a particular benefit, but did not give the amount received, we impute using linked data where possible, depending on the benefit. For benefits such as Universal Credit, where the rate could vary greatly depending on the circumstances of the respondent, we replace all reported amounts with linked amounts, because of the difficulty of making individual benefit assessments.

However, for benefits with no access to suitable administrative data, such as the Armed Forces Compensation Scheme, or where fewer rates apply, a more general method has been used and an imputation decision has been made, based upon all the available evidence available about that person’s circumstances.

Near-zero amounts

Where benefit amounts are recorded as near-zero, the case is examined individually, and an edit decision is made.

Multiple benefits

Any combined benefit amounts (for example where State Pension is paid with Attendance Allowance) are edited by carrying out benefit entitlement assessments on individual cases, while preserving the reported total wherever possible.

Validation reports

Computer programs are run to carry out a final check for benefit entitlement and to output any cases that look unreasonable. All cases detected because of this validation exercise are individually checked and edited where necessary.

Stage five – other pre-imputation cleaning

In preparation for imputing missing values, data is made as clean as possible. This involves edits and checks of the following nature:

Weekly amounts

In the FRS, most monetary amounts are converted to a weekly equivalent. To calculate this, respondents are usually asked the amount, then the length of time this amount covered. The latter is known as a “period code”. Period codes are used in conjunction with amounts to derive weekly figures for all receipts and payments. Some variables, such as interest on savings accounts, refer to the amount paid in the whole of the past year. These are also converted to a weekly amount.

Sometimes the period code relates to a lump sum or a one-off payment. In these cases, the corresponding value does not automatically convert to a weekly amount. For the data to be consistent across the survey, edits are applied to convert most lump sums and one-off payments to weekly amounts. In the same way, where period codes are recorded as ‘don’t know’ or ‘refused’, these are imputed so that the corresponding amount can be converted to a weekly value in the final dataset.

Near-zero amounts

It is not possible for interviewers to enter zero amounts where it is inappropriate to do so. For example, in response to a question on receipt of benefit, a zero amount will result in a warning message being displayed. Some interviewers try to avoid this message by recording near-zero amounts. As a result, all near-zero values are examined and a decision taken as to whether the value is genuine or whether the value should be treated as missing.

Outliers

Statistical reports of the data are produced to show those cases where an amount was greater than four standard deviations from the mean. For the seven largest values over this limit, the individual record is examined and where necessary (but only if a value looks unrealistic), the case is edited. The outliers remaining in the dataset are verified by examining other relevant data. Compared with earlier FRS years, only a small number of these edits are now carried out, because of the many range checks in the computerised questionnaire.

Credibility checks

Checks are carried out for the internal consistency of certain variables. For example, one check on mortgage payments ensures that payments to the mortgage from outside the household are not greater than the mortgage payment itself. Such cases are examined and edited where necessary.

Stage six – imputation

The responses to some questions are much more likely to have missing values than others. For example, it is very unlikely that a respondent will refuse to give or will not know their age or marital status; whereas it is much more likely that they will not be able to provide precise information on the amount of interest received from their investments.

Two areas where missing values are a problem are (1) income from self employment and (2) income from investments. Results in the tables provided in this publication include imputed values. Elsewhere however, values are left to remain as missing in some variables (such as hours of care).

Methodology Table M_4 illustrates the extent of missing values. Of the 10.4 million set values in the 2021 to 2022 FRS dataset, two per cent were originally recorded as either ‘don’t know’ or ‘refused’. Out of 155,729 missing values, approximately 98% were then imputed.

Closing down routes

As with any questionnaire, a typical feature of the FRS is a gatekeeper question positioned at the top of a sequence of questions, at which a particular response will open the rest of the sequence. If the gatekeeper question is answered as ‘don’t know’ or ‘refused’ then the whole sequence is skipped.

A missing gatekeeper variable could be imputed such that a further series of answers would be expected. However, these answers will not appear because a whole new route has been opened. For example, if the amount of rent is missing for a record and has since been imputed, any further questions about rent would not have been asked. From the post-imputed dataset, it will appear that these questions should have been asked because a value is present for rent.

For this reason, where the gatekeeper question has been skipped the onward routes should be closed. In most cases, gatekeeper variables are of the ‘yes or no’ type. If missing, these would be imputed to ‘no’, on the basis that if a respondent does not know whether an item is received or paid, then it is not.

Hot-decking

This process looks at characteristics within a record containing a missing value to be imputed and matches it up to another record with similar characteristics for which the variable is not missing. It then takes the known variable and copies it to the missing case. For example, when imputing the Council Tax Band of a household, the number of bedrooms, type of accommodation and region are used to search for a case with a similar record. This method ensures that imputed solutions are realistic and allows for a wide range of outcomes which maintain variability in the data.

Algorithms

These are used to impute missing values for certain variables, for example variables relating to mortgages. The algorithms range from very simple calculations to more sophisticated models, based on observed relationships within the data and individual characteristics, such as age and gender.

‘Mop-up’ imputation

This is achieved by running a general validation report of all variables and looking at those cases where missing values are still present. At this stage, variables are examined on a case-by-case basis to decide what to impute. Credibility checks are re-run to identify any inconsistencies in the data caused by imputation, and further edits are applied where necessary.

All imputations, by each of the methods above, are applied to the un-imputed dataset by a transaction database. This ensures auditability in that it is always possible to reproduce the original data.

Points to note with imputed data:

• whilst several processes are used to impute missing values, it should be remembered that they represent only a very small proportion (typically two per cent) of the dataset

• imputation will have a greater effect on the distribution of original data for variables that have a higher proportion of non-response, as proportions of imputed data will be higher

• as mentioned above, in certain situations, imputed values will be followed by ‘skipped’ values. It was decided in some cases that it was better to impute the top of a route only, and not large amounts of onward data. For a small proportion of imputations it is not possible to close down a route. These cases are followed by ‘skipped’ responses (where a value might otherwise be expected)

Stage seven – derived variables

Derived variables (DVs) are those which are not created by the original interview, but instead are made by combining information, both within the survey and from other sources.

They are created at the FRS user’s request. Their main purpose is to make it easier for users to carry out analysis and to ensure consistent definitions are used in all FRS analyses. For example, INDINC is a DV which sums all components of income to find an individual’s total income. This is possible because of the various sources collected by the survey. As new information is collected in the survey, the relevant DVs are updated as necessary.

12. Grossing

The grossing regime applied in FYE 2021 was adapted to try to control for the larger impacts of the coronavirus (COVID-19) pandemic upon the achieved sample. Whilst the previous FRS grossing regime brought estimates close to the age and tenure profile of the UK population, it retained a disproportionate number of working-age respondents who had been educated to at least first-degree level. It was important to adjust for this bias because income levels are strongly correlated with the level of education achieved.

Therefore, additional grossing controls were introduced to rebalance the educational levels of those in the sample. After adding in a control total for working-age adults with degree level education, there was still a bias towards younger adults with degrees so the grossing control was split into two: working-age adults aged 16 to 45 with a degree and working-aged adults over 45 with a degree.

When preparing the FYE 2022 estimates, we found, albeit to a lesser degree than in FYE 2021, that the weighted sample included a disproportionate number of respondents with education levels at first-degree level or above, and too few below degree level. Therefore, we maintained the use of educational grossing controls, using annual growth in the degree population measured in the Annual Population Survey (APS).

There has, however, been no change to the overall population basis for the estimates. These remain the population in private households, as estimated by ONS. ONS has not adjusted these figures in the light of coronavirus (COVID-19).

The grossing regime in 2021 to 2022 has also been adapted to control for the differential level of response seen through the year: This reflects the introduction of the planned boost to the issued sample from 1st October 2021. Following this boost (in England and Wales) in the second half of the year, the achieved sample (for the UK as a whole) was approximately 10,000 for the period from October to March. This was significantly higher than the 6,000 achieved for the period from April to September. We accounted for this step change by introducing a biannual grossing control for Great Britain so there were equal numbers of private households from each half of the survey year in the weighted sample.

In Northern Ireland, changes to the approach of contacting respondents in July 2021 meant that the achieved sample increased markedly partway through the year. This is not normally a feature of the FRS achieved sample, with response normally spread relatively equally over each twelve-month run of fieldwork. We introduced a quarterly household grossing control to balance their sample across the year.

See the section on Sample Design for more information.

The FRS publication presents tabulations where the percentages refer to sample estimates grossed-up to apply to the whole population.

Grossing-up is the term given to the process of applying factors to sample data so that they yield estimates for the overall population. The simplest grossing system would be a single factor e.g. the number of households in the population divided by the number in the achieved sample. However, surveys are normally grossed by a more complex set of grossing factors that attempt to correct for differential non-response, at the same time as they scale up sample estimates.

The system used to calculate grossing factors for the FRS divides the sample into different groups. The groups are designed to reflect differences in response rates among different types of household. The FRS stratified sample structure is designed to minimise differential non-response in the achieved sample. Grossing is then designed to account for residual differential non-response.

They have also been chosen with the aims of DWP analysis in mind. The population estimates for these groups, obtained from official data sources, provide control totals. The grossing factors are then calculated so that the FRS produces population estimates that are as close as possible to the control totals. As an example, a grossed FRS count of the number of men aged 35 to 39 would be consistent with the ONS population estimates of the same group.

In developing the grossing regime careful consideration has been given to the combination of control totals, and the way age ranges, Council Tax Bands and so on, are grouped together. The aim has been to strike a balance so that the grossing system will provide, where possible, accurate estimates in different dimensions without significantly increasing variances.

Some adjustments are made to the original control total sources, so that definitions match those in the FRS. For example, an adjustment is made to the demographic data to exclude people whose residence is not a private household. It is also the case that some control totals have to be adjusted to correspond to the FRS survey year which runs from April to March.

A software package called CALMAR, provided by the French National Statistics Institute, is used to reconcile control variables at different levels and estimate their joint population. This software makes the final weighted sample distributions match the population distributions through a process known as calibration weighting. It should be noted that if a few cases are associated with very small or very large grossing factors, grossed estimates will have relatively wide confidence intervals.

A review of the FRS grossing methodology was carried out by the ONS Methodological Advisory Service in 2013.

A number of relatively minor methodological improvements were made as a result, with the grossing calculations updated to use 2011 Census data at that point. Further details on the methodological changes.

Both Great Britain and Northern Ireland data use the same CALMAR software to reconcile control variables at different levels and estimate their joint population. There are minor differences between the methods used to gross the Northern Ireland sample as compared with the Great Britain sample:

• local taxes in Northern Ireland are collected through the rates system, so Council Tax Band is not applicable as a control variable

• Northern Ireland housing data is based largely on small-sample surveys. It is not desirable to introduce the variance of one survey into another by using it to compute control totals; therefore tenure type is not used as a control variable

Details of the control variables used in the grossing methodology for Great Britain and Northern Ireland are shown below.

Grossing regime for Great Britain 2021 to 2022

Control variables used to generate grossing factors for private households

Variable	Groupings	Source of data
Individuals (Age, sex and Region)	Male children: 0-9, 10-19 Male adults: 16-24, 25-29, 30-34, 35-39, 40-44, 45-49, 50-59, 60-64, 65-74, 75-79, 80+ Female children: 0-9, 10-19 Female adults: 16-24, 25-29, 30-34, 35-39, 40-44, 45-49, 50-59, 60-69, 70-74, 75-79, 80+ Each grouping is further broken down by region: North East, North West, Yorkshire and the Humber, East Midlands, West Midlands, East, London, South East, South West, Scotland and Wales	Office for National Statistics (ONS) Mid-year private household population estimates, DWP estimates using data derived from ONS and HMRC
Benefit units (with children)	England and Wales (combined), Scotland	HMRC Child Benefit data
Benefit units (with children	Lone parents: Male, female	Labour Force Survey estimates
Households (Tenure type)	Local Authority or Housing Association renters, private renters, owner occupiers	Department for Levelling Up, Housing and Communities (DLUHC)
Households (Council Tax Band)	A and Not Valued Separately, B, C-D, E-H (and band I in Wales only)	Valuation Office, Scottish Government
Households (Region)	North East, North West, Yorkshire and the Humber, East Midlands, West Midlands, East, London, South East, South West, Scotland and Wales	ONS (England), Welsh Government (Wales), Scottish Government (Scotland)
Households by half-year of interview	April 2021 to September 2021 and October 2021 to March 2022	See Households by region /tenure/council tax band above
Working-age adults with degrees	Working-age adults aged 16 to 44 and working-age adults aged over 44	Annual Population Survey

Grossing regime for Northern Ireland, 2021 to 2022

Control variables used to generate grossing factors for private households

Variable	Groupings	Source of data
Individuals (Age and sex)	Male children: 0-9, 10-19 Male adults: 16-24, 25-29, 30-34, 35-39, 40-44, 45-49, 50-59, 60-64, 65-74, 75-79, 80+ Female children: 0-9, 10-19 Female adults: 16-24, 25-29, 30-34, 35-39, 40-44, 45-49,50-59, 60-69, 70-74, 75-79, 80+	Office for National Statistics (ONS)
Benefit units (with children)	Lone parents	Department for Communities Northern Ireland (DfCNI) estimates
Households		Northern Ireland Statistics and Research Agency (NISRA)
Households by quarter of interview	Each quarter of April 2021 to March 2022
Working-age adults with degrees	Working-age adults aged 16 to 44 and working-age adults aged over 44	Annual Population Survey

13. Reliability of estimates

All survey estimates have a sampling error attached to them, approximated from the variability of the observations in the sample. From this, a margin of error (confidence interval) is estimated. It is this confidence interval, rather than the estimate itself, that is used to make statements about the likely ‘true’ value in the population; specifically, to state the probability that the true value will be found between the upper and lower limits of the confidence interval. In general, a 95% confidence interval is calculated within which there is a 95% chance that the true value of the population is found. A small margin of error will result in a narrow interval, and a more precise estimate of where the true value lies.

The achieved sample in FYE 2022 (16,376 households) is meaningfully higher than that achieved in FYE 2021 (10,007), but still somewhat short of the sample size achieved in the years prior to the COVID-19 pandemic (typically 19,000-20,000). In general, this means that the uncertainty on this year’s survey estimates is materially smaller than last year, but still slightly larger than in the years preceding the pandemic.

The FRS sample in Great Britain, as described earlier, is selected using a stratified design, based on addresses clustered within postcode sectors. As a result, FRS sampling error is not just dependent on the variability between sample units (households, benefit units, individuals), but also on the variability between postcode sectors. For example, if a sample characteristic is distributed differently by postcode sector (i.e. is clustered) the representativeness of a given sample is reduced, and the variability of repeated samples would be greater overall than would occur in a simple random sample of the same size.

Therefore, the correct mathematical accounting for clustering causes the complex (actual) sampling error to be greater than a sampling error calculated under the assumption of simple random sampling. Stratification attempts to account pre-emptively for some of the variation between clusters using information at a cluster level known prior to the survey, and so its effect is to reduce the sampling error, relative to what it would otherwise have been. Clustering is not used in Northern Ireland, but households are contacted using systematic random sampling, which has a similar effect to stratification.

Following the survey fieldwork, far more is known about those sampled. For certain characteristics, this information can be compared to known population totals and households weighted, to ensure proportionate representation of those otherwise disproportionately sampled (see the Grossing section of this report). This can be thought of as increasing the representativeness of the sample, or reducing the variability that would be observed if making repeated samples. The effect is therefore similar to stratification and reduces the sampling error. For this reason, this process is also called post-stratification.

As ever, standard errors and confidence intervals will vary from survey estimate to survey estimate; as in previous publications, standard errors for a selection of survey estimates are set out. However, it should be noted that the standard errors in this year’s FRS publication are not directly comparable with those of recent years: for this publication, a new methodology has been adopted. Standard errors, design factors, and confidence intervals on estimates are now calculated using a bootstrap resampling method that accounts for the complex survey design and post-stratification weighting as fully as possible.

This change has been made because the previous method did not fully take into account a number of relevant factors. The FRS, like many household surveys, employs a complex sampling design. Standard techniques which do not account for these complexities will therefore yield imperfect estimates of the extent of uncertainty. The previous method did take into account the variation ‘between’ clusters and the number of clusters.

In contrast, the bootstrap methodology can be refined so that it accounts as fully as possible for the relevant complexities of the FRS sampling design. Essentially, the key features of the sampling design accounted for are stratified sampling, clustered sampling, the fact that the sample is weighted ex post using external information about the characteristics of the population, and non-independence of the samples in consecutive years (relevant for statistics such as year-on-year changes in incomes).

Whereas, the previous method didn’t take account of the variance ‘within’ clusters, the new method more accurately assesses of the true uncertainty in the survey estimates, because it takes into account “all sources” of uncertainty in the survey design. The new method also takes into account the adjustments that are made to adjust for uncertainty, such as weighting ex post (grossing), in the same way as the method employed with HBAI standard errors. This change therefore has the added benefit that it brings harmonisation between the FRS and the HBAI standard errors.

A perfect method for calculating variability in survey estimates would be to have performed the survey many times, independently, and measure the range of estimates observed from so doing. As the survey fieldwork has of course been performed only once, the real-world method of estimating uncertainty is as follows:

• uncertainty is approximated by treating the achieved sample as if it were the population and repeatedly drawing sub-samples at random from that achieved sample

• this process is referred to as ‘resampling’ and the result is a series of ‘resamples’. Resamples are drawn so as to mimic the original sampling methodology and replicate its effects on the reliability of any results. This means that stratification and clustering information and systematic random sampling processes are used to replicate the original FRS household selection process

• households can be selected multiple times, and the unequal probability of selection of households is accounted for. Each FRS resample is smaller than the original sample size by about half. Once households are selected, each FRS household is assigned a grossing factor in a process identical to the full sample. Altogether, this produces a series of alternative samples from which to calculate a series of alternative estimates

The variability in these alternative estimates is used to quantify the uncertainty in the original estimate in two ways:

• the standard deviation of these alternative estimates is the approximate standard error of the original estimate

• the series of estimates produced are ranked by size, and the 2.5th and 97.5th percentiles extracted. These results define the lower and upper confidence limits of the approximate 95% confidence interval around the original estimate

The size of the actual standard error relative to the standard error calculated under the assumption of simple random sampling is represented by the design factor, which is calculated as the ratio of the two. Where the standard errors are the same, the design factor equals one, implying that there is no loss of precision associated with the use of a complex sample design with post-stratification.

In many cases, the design factor will be greater than one, implying that FRS estimates are less precise than those of a simple random sample of the same size. Conversely a design factor of less than one implies the FRS estimate is more precise than would be obtained from a simple random sample.

Standard errors

Methodology Tables SE_1 to SE_9 provide standard errors, design factors and confidence limits for a selection of variables from this year’s survey.

Example of how to interpret figures in this table

Example: Uncertainty measures for household composition, table SE_1

Table SE_1 shows that 72% of households did not contain any children.

The standard error is estimated as 0.1 percentage points. This is the final estimate after rounding.

The design factor for this variable is 0.3. That is, the effect of using a complex survey design and post-stratification, rather than a simple random sample, is a reduction in uncertainty of 70%, when using standard error as the measure of uncertainty. In contrast, a design factor of 1.5 would have denoted an increase in such uncertainty of 50%. Among smaller groups, such larger design factors are not uncommon.

The 95% confidence interval is given as 72.2% to 72.6%. That is, if sampling error is the sole source of error, there is a 95% chance that the true percentage of households without children lies within this range.

See the methodology paper for information on estimating variance and confidence intervals in special circumstances, for example where the occurrences of a response in the sample are very small.

In addition to sampling errors, consideration should also be given to non-sampling errors. Sampling errors arise through the process of random sampling and the influence of chance. Non-sampling errors arise from the introduction of some systematic bias in the sample compared with the population it is supposed to represent.

As well as response bias, such biases include inappropriate definition of the population; misleading questions; data input errors; data handling problems; or any other factor that might lead to the survey results systematically misrepresenting the population. There is no simple control or measurement for such non-sampling errors, although the risk can be minimised through careful application of the appropriate survey techniques from the questionnaire and sample design stages through to analysis of results.

14. Linking FRS data to administrative data

In line with the DWP Digital data strategy the department is committed to transforming its surveys by linking administrative data from the full range of available sources (for example, from other parts of government). Furthermore, as a national statistic, and in line with the Code of Practice for Statistics (Value V4.1) DWP looks to improve the FRS, year on year.

The introduction of GDPR in 2018 has provided an alternative to consent as the legal basis for linking. We can now link all respondents on the basis that the processing is necessary for the department to carry out its functions as a public body (GDPR Article 6(1)(e)).

This change, together with improvements in our linking methodology, means that we can now link at least 95% of FRS respondents to their administrative records. This has opened the potential to realise a range of financial, data quality, respondent and user benefits through the integration of administrative data into the survey.

The FRS Administrative Data Transformation Project was established to investigate the potential of all administrative data sources and realise the benefits. This is in the wider context of DWP Digital’s Data Strategy Developing a data strategy and delivering it - Data in government (blog.gov.uk) for optimising use and re-use of administrative data, the UK Statistics Authority’s Strategy for data linking Joining Up Data for Better Statistics – Office for Statistics Regulation (statisticsauthority.gov.uk) and the Office for National Statistics move towards a ‘survey and administrative data integration’ approach to meeting government information needs Data collection transformation - Office for National Statistics (ons.gov.uk)

It should also be noted that development aligns with several OSR recommendations in their review of income-based poverty statistics, on the use of integrated survey and administrative data. These were:

• the strategic recommendation that innovation is needed for the statistics to deliver their full potential and serve the public good; opportunities for data linkage should be maximised and data gaps should be addressed, building on work already underway in the GSS to explore the use of administrative data and its integration with social surveys

• DWP and ONS, building on existing work to explore the feasibility and potential of social survey and administrative data integration, should explore whether integration can help improve the timeliness and robustness of income-based poverty statistics

• DWP and ONS should prioritise work to address under-reporting at the bottom end of the income distribution; they should consider a multifaceted approach to solving this problem, such as data linkage and making greater use of administrative data

Development in this area has been taken forward in the 2020 to 2021, 2021 to 2022 and 2022 to 2023 survey years. Please see the DWP Statistical Work Programme for more details. Our existing long-term work programme developing integrated survey-administrative datasets see section 2.5 of the Work Programme will meet the aims of these objectives in the future.

Comparisons of survey and administrative data

We make comparisons of FRS survey and administrative data in several ways. These are outlined in the following tables:

Please see Methodology tables M_6a and M_6b, for a summary of how FRS benefit caseloads and amounts compare with DWP administrative data.

Methodology Table M_6a

This compares the grossed number of benefit recipients in the FRS 2021 to 2022 data, with the total caseload on benefit from administrative data sources. For all benefits, and as in most previous years, the FRS numbers in receipt are below those seen in administrative data. The difference varies by benefit.

It should be noted that the source for State Pension administrative data has changed this year, to DWP State Pension Statistics. Data is not currently available from the previously used source on Stat Xplore, due to issues with obtaining “Get Your State Pension” (GYSP) data.

Methodology Table M_6b

This compares the average weekly receipt of state support in the FRS 2021 to 2022 data, with average receipt from administrative data sources. Some benefit types have not been included in this analysis because no directly comparable administrative data source is available.

Methodology Table M_8

This also compares FRS and administrative data, but linked at a record level, and shows receipt of DWP benefits for the 2021 to 2022 survey year across either or both of those sources. Percentages are on a post-grossing basis. Some benefits are better represented on the FRS than others: For example, 99% of adults in receipt of State Pension are represented on the FRS, while only 65% of those in receipt of Attendance Allowance are.

Percentage of adults shown in receipt of DWP benefits, FRS and administrative data, 2021 to 2022, Great Britain

15. Glossary

This glossary provides a brief explanation for each of the key terms used in the Family Resources Survey (FRS). Further details on these definitions, including full derivations of variables, are available on request from the FRS team.

Adult

All individuals who are aged 16 and over are classified as an adult, unless the individual is defined as a dependent child. All adults in the household are interviewed as part of the FRS.

Age

Respondent’s age at last birthday (at the time of the interview).

Automatic enrolment

Automatic enrolment requires all employers to enrol their eligible workers into a workplace pension scheme if they are not already in one. This enrolment also commits the employer to make contributions into the employee’s pension. The staged timetable began in October 2012 for larger firms, with enrolment for all employers completed in 2018. To preserve individual responsibility for the decision to save, workers can opt out of the scheme. To be eligible for automatic enrolment, the jobholder must be at least 22 years old, under State Pension age, earn above the earnings threshold for automatic enrolment, and work or usually work in the UK.

However, those not eligible for automatic enrolment may be entitled to opt in. Those people now defined as self-employed could have been a member of an employer scheme, from auto-enrolment, but are entitled to remain in their auto-enrolled scheme and make their own contributions. Likewise, someone who is now an employee, who was previously self-employed can have employer contributions to their previous scheme.

For more information see this pensions guide

Benefit unit or family

A benefit unit may consist of: a single adult, or a married or cohabiting couple, plus any dependent children. Same-sex partners (civil partners and cohabitees) have been included in the same benefit unit since January 2006. Where a total for a benefit unit is presented (such as total benefit unit income) this includes income from adults plus any income from children.

There are various types of benefit unit:

• pensioner couple: Benefit units headed by a couple where the head of the benefit unit is over State Pension age. Note that this differs from definitions used in the Households Below Average Income, Income Dynamics and Pensioners’ Incomes Series reports. These publications define a benefit unit as a pensioner couple if either the head of the benefit unit or their partner is over State Pension age

• pensioner couple, married or civil partnered: Benefit units headed by a couple where the head of the benefit unit is over State Pension age and the couple are either married or in a civil partnership

• pensioner couple, cohabiting: Benefit units headed by a couple where the head of the benefit unit is over State Pension age, and the couple are neither married nor in a civil partnership

• single male pensioner: Benefit units headed by a single male adult over State Pension age

• single female pensioner: Benefit units headed by a single female adult over State Pension age

• couple with children: Benefit units containing two adults, headed by a non-pensioner, with dependent children

• couple with children, married or civil partnered: Benefit units containing two adults, headed by a non-pensioner, with dependent children and the couple are either married or in a civil partnership

• couple with children, cohabiting: Benefit units containing two adults, headed by a non-pensioner, with dependent children and the couple are neither married nor in a civil partnership

• couple without children: Benefit units containing two adults, headed by a non-pensioner, with no dependent children

• couple without children, married or civil partnered: Benefit units containing two adults, headed by a non-pensioner, with no dependent children and the couple are either married or in a civil partnership

• couple without children, cohabiting: Benefit units containing two adults, headed by a non-pensioner, with no dependent children and the couple are neither married nor in a civil partnership

• single with children: Benefit units containing a single adult (male or female), headed by a non-pensioner, with dependent children

• single male without children: Benefit units containing a single male adult, headed by a non-pensioner, with no dependent children

• single female without children: Benefit units containing a single female adult, headed by a non-pensioner, with no dependent children

Benefits

Financial support from the government. Most of these benefits are administered by DWP. The major exceptions are Housing Benefit and Council Tax Reduction, which are administered by local authorities.

Child Benefit is administered by HM Revenue and Customs. HMRC also administer Tax Credits. These are not treated as benefits, but both Tax Credits and benefits are included in the term State Support. Tax Credits will ultimately be superseded by Universal Credit.

Benefits are often divided into income-related benefits and non-income-related benefits. In assessing entitlement to the former, the claimant’s income and savings will be checked against the rules of the benefit. In contrast, eligibility for non-income-related benefits is dependent on the claimant’s circumstances (a recent bereavement, for example), rather than their income and savings.

Some benefits have sample sizes which are too small to be presented separately in this publication. A list of the main state benefits shown in “state support” tables in this publication can be found in the table below:

United Kingdom benefits

UK income-related benefits
Council Tax Reduction
Employment and Support Allowance (income-related element)
Housing Benefit
Income Support
Pension Credit
Universal Credit

UK non-income-related benefits
Attendance Allowance
State Pension
Child Benefit
Carer’s Allowance
Disability Living Allowance (both mobility and care components)
Personal Independence Payment (Daily Living and Mobility components)
Employment and Support Allowance (contributory element)

Devolved benefits

The Scotland Act 2016 gave Scottish Parliament powers over a number of social security benefits effectively transferring policy ownership from DWP to the Scottish Government. In September 2018, Carer’s Allowance became the first of these benefits to have policy ownership transferred from DWP to Scottish Ministers. On 1 April 2020, executive competence transferred to the Scottish Government for all remaining disability and industrial injuries benefits due to be devolved.

Child Disability Payment will steadily replace child DLA, with the first new claims accepted in 2021 to 2022; however, many existing child DLA claims will continue until 2025.

More information on this can be found in the DWP benefits statistical summary background information note.

Disability-related benefits

‘Disability-related benefits’ is the term used to describe all benefits paid on the grounds of disability. These are:

• Personal Independence Payment
• Disability Living Allowance
• Severe Disablement Allowance
• Attendance Allowance
• Armed Forces Compensation Scheme
• Industrial Injuries Disablement Benefit *Northern Ireland Disability Rate Rebate

Child

A dependent child is defined as an individual aged under 16. A person will also be defined as a child if they are 16 to 19 years old and they are:

• not married nor in a civil partnership nor living with a partner; and

• living with parents (or a responsible adult); and

• in full-time non-advanced education or in unwaged government training

Child Benefit

This is a non-income related benefit in terms of eligibility, but remains taxable in households where one adult is earning more than £50,000 per year.

Council Tax

The tax is based on a set of bands that a property’s value falls into and is evaluated accordingly by each council. Its headline rate is based on two adults per household.

Disability

The definition of disability used in this publication is consistent with the core definition of disability under the Equality Act 2010. A person is considered to have a disability if they “have a physical or mental impairment that has a ‘substantial’ and ‘long-term’ negative effect on their ability to do normal daily activities”. Whereby ‘substantial’ means more than minor or trivial, and ‘long-term’ means 12 months or more. However, some individuals classified as disabled and having rights under the Equality Act 2010 are not captured by this definition:

• people with a long-standing illness or disability who would experience substantial difficulties without medication or treatment

• people who have been diagnosed with cancer, HIV infection or multiple sclerosis but who are not currently experiencing difficulties with their day-to-day activities

• people with progressive conditions, where the effect of the impairment does not yet impede their lives

• people who were disabled in the past but are no longer limited in their daily lives are still covered by the Act

This definition of disability differs from that used for Economic status.

Economic status

This classification follows the harmonised output category for economic status, based on respondents’ answers to the survey questions. All definitions conform to those of the International Labour Organization (ILO):

• employee: where respondents have an arrangement with an employer, whereby work is done in exchange for a wage or salary. This would include those doing unpaid work in a business that a relative owns

• self-employed: where respondents report regular working activities, which over time are responsible only to themselves (and not an employer). Various groups are classified as self-employed, including farmers, doctors in private practice and some builders, as well as anyone whose job is habitually done on a freelance basis (for example, journalists or musicians). The self-employed include anyone doing work for their own business, but which is currently unpaid

Several respondents have more than one job. The FRS identifies which of these is their ‘main job’. This is the job which the respondent says is the dominant activity. Where they cannot decide, the number of hours worked will determine which is the main job. This process of categorisation also applies to respondents who are employees in one job but self-employed in another; whilst the survey will capture information on both jobs, only one can be their main job.

• unemployed: Adults who are under State Pension age and not working, but are available and have been actively seeking work in the last four weeks; includes those who were waiting to take up a job already obtained and were to start in the next two weeks

• economically inactive: Individuals who are both out of work, and not seeking or not available to work. There are several sub-categories:

  • retired: individuals who are over State Pension age, or say they are now retired
  • student: individuals who have not completed their education
  • looking after family or home: working-age individuals who are looking after their family or their home
  • permanently sick or disabled: working-age individuals who have been sick, injured or disabled for longer than 28 weeks
  • temporarily sick or disabled: working-age individuals who have been sick, injured or disabled for less than 28 weeks. Note that the sick or disabled definitions are different to that used for Disability, as they are based on different questions that are only asked of working-age adults who are not working
  • other inactive: all respondents not already classified above

Employment status

This classification is equivalent to economic status but includes those in employment only.

Ethnic group

The ethnic group to which respondents consider that they belong. Ethnicity representation rates are now calculated from known declarations and exclude ‘choose not to declare’ and ‘unknown’.

Where respondents do volunteer their ethnicity, this is captured as one of 18 recognised groups. This is consistent with the harmonised principles for ethnicity, as set out by the Government Statistical Service, wherever social surveys are carried out.

• White

• Irish Traveller

• Mixed or Multiple ethnic groups

• Asian or Asian British

• Indian

• Pakistani

• Bangladeshi

• Chinese

• Arab

• Any other Asian background

• Black or African or Caribbean or Black British

• Other ethnic group

Sample sizes for ‘Gypsy, Traveller or Irish Traveller’ are small. In Northern Ireland, ‘Irish Traveller’ is included in ‘Other ethnic group’ whereas elsewhere ‘Gypsy or Irish Traveller’ is included in ‘White’.

Food bank usage

See Household food bank usage.

Food security

See Household food security.

Full-time education

Individuals registered as full-time at an educational establishment. Students on sandwich courses are coded as working, or studying, depending on their position at the time of interview.

Harmonised principles

The harmonised principles contain harmonised definitions, survey questions, standards for administrative data and standards for presentation. They have been developed by topic groups, after wide consultation with producers and customers across the GSS and beyond. Further information is available via the Government Statistical Service pages

Head of benefit unit

If the household reference person does not belong to the benefit unit, then the head of benefit unit is simply the first person from that benefit unit, in the order they were named in the interview. If the household reference person does belong to the benefit unit, they are also the head of that benefit unit.

Household

A household consists of one person living alone or a group of people (not necessarily related) living at the same address, who share cooking facilities and share a living room or sitting room or dining area. A household will consist of one or more benefit units. Where a total value for a household is presented, such as total household income, this includes income from adults plus any income from children.

Household food bank usage

Household food bank usage in the FRS refers only to visits to a food bank when emergency food supplies (food parcels) were obtained. This excludes visits to the food bank made only for other support (e.g. financial advice or mental health support). The FRS asks food bank usage questions relating to two time periods:

• usage within the 12 months prior to interview

• usage within the 30 days prior to interview

Only households that report using a food bank in the last 12 months are asked about 30-day usage.

Household food security

“Food security” as a concept is defined as “access by all people at all times to enough food for an active, healthy life”. Questions relate to the household’s experience in the 30 days immediately before the interview.

The questions are put to the person in each household who is best placed to answer about food shopping and preparation. These respondents are asked the first three questions, on whether they are concerned about:

• food running out before they had enough money to buy more

• the food they had bought not lasting, and not having money to buy more

• not being able to afford balanced meals

The possible answers are ‘often, ‘sometimes’ or ‘never’ true. If respondents say that all three statements are never true they will not be asked further questions on food security. If respondents answer that any of these statements are sometimes or often true they will be asked further questions on the extent of their food security.

Taking the responses together, a household ‘score’ for food security is then derived. This is a measure of whether households have sufficient food to facilitate active and healthy lifestyles. This measure has four classifications:

• High food security (score=0): The household has no problem, or anxiety about, consistently accessing adequate food

• Marginal food security (score= 1 or 2): The household had problems at times, or anxiety about, accessing adequate food, but the quality, variety, and quantity of their food intake were not substantially reduced

• Low food security (score = 3 to 5): The household reduced the quality, variety, and desirability of their diets, but the quantity of food intake and normal eating patterns were not substantially disrupted

• Very low food security (score = 6 to 10): At times during the last 30 days, eating patterns of one or more household members were disrupted and food intake reduced because the household lacked money and other resources for food

High and marginal food security households are considered to be “food secure”. Food secure households are considered to have sufficient, varied food to facilitate an active and healthy lifestyle. Conversely, low and very low food security households are considered to be “food insecure”. Food insecure households are where there is risk of, or lack of access to, sufficient, varied food.

The broad structure and sequence of the questions is the same as those used internationally. They are used within the UK (Food Standards Agency) and are also used by other countries, including the United States Department of Agriculture, enabling broad international comparability of the results.

Household reference person (HRP)

The highest income householder.

• In a single-adult household, the HRP is simply the sole householder (i.e. the person in whose name the accommodation is owned or rented)

• If there are two or more householders, the HRP is the householder with the highest personal income, taking all sources of income into account

• If there are two or more householders who have the same income, the HRP is the elder

Where we refer to ‘Head’ in tables relating to households, this is the HRP. The Head of benefit unit will not necessarily be the HRP.

Individual

An adult or child. Where ‘people’ are presented, this is all adults and children.

Informal carers

Individuals who provide any regular service or help to someone. That person can be inside or outside of their household, and might be sick, disabled or elderly; this description excludes those who give this service or help as part of a formal job.

Marital status

This is the person’s de facto marital status:

• Married or civil partnership: currently married or in a civil partnership, and not separated from spouse (excludes temporary absences)

• Cohabiting: not married nor in a civil partnership, but living as a couple

• Single: is not currently cohabiting and has never been married nor in a civil partnership

• Widowed: widowed and not currently cohabiting

• Separated: married or in a civil partnership, but separated from spouse and is not currently cohabiting

• Divorced or civil partnership dissolved: marriage or civil partnership legally dissolved

Non-advanced education

Non-advanced education for benefits purposes includes:

• ‘A’ levels, or similar qualifications (eg. the International Baccalaureate and Pre-U)

• T levels (introduced in September 2020)

• Scottish national qualifications at higher or advanced higher level

• NVQ at Level 3

• study programme in England

• national diploma

• ordinary national diploma

• national certificate of Edexcel

If the young person is studying for a course that is not classed as advanced education, the education is normally treated as non-advanced. Non-advanced education does not include university courses.

Pension

• Employer-sponsored pension: schemes that are set up and run by the employer.

• Occupational pension: an occupational pension scheme is an arrangement an employer makes to give their employees a pension when they retire. They are often referred to as ‘company pensions’. As of 2017 the Occupational Pension Schemes regulations brought restrictions on the Early Exit charges for those aged 55 and older, and are eligible to access the pension freedoms.

There are 2 main types of occupational pension:

1. Defined-benefit (DB) schemes (also called salary-related pension or superannuation schemes). In a defined benefit scheme, the pension is based on the number of years you belong to the scheme and how much you earn. Your employer contributes to the scheme and trustees look after scheme members’ interests. Employees often have to pay contributions into the scheme on top of those made by the employer. Some schemes are ‘non-contributory’: The employee either makes no contributions, or makes a small contribution, typically 1%–2% of salary

2. Defined-contribution (DC) schemes (also called Money purchase schemes). A defined contribution scheme can be a personal pension arranged by the individual or a workplace pension arranged by the employer (such as NEST). Money is paid in by the individual or the employer over time and is then invested by the pension provider. The size of the pension available to take out when the individual retires depends on how much was paid in and the level of growth from the investments. With a defined contribution pension the individual can also decide how to take their money out

• Group personal pension: some employers who do not offer an occupational pension scheme may arrange for a third-party pension provider to offer employees a pension instead. The employer may have negotiated special terms with the provider, which means that administration charges are lower than those for individual personal pensions. Although sometimes still referred to as ‘company pensions’, they are not run by employers and should not be confused with occupational pensions, which have different tax, benefit and contribution rules

• Group stakeholder pension: like Group Personal Pensions, an employer can make an arrangement with a pension provider and offer their employees a Group Stakeholder Pension (see Stakeholder Pension)

• Personal pension: a pension provided through a contract between an individual and the pension provider. The pension which is produced will be based upon the level of contributions, investment returns and annuity rates; a personal pension can be either employer provided (see Group Personal Pension) or privately purchased (see Private pension)

• Private pension: includes occupational pensions (also known as employer-sponsored pensions) and personal pensions (including stakeholder pensions). People can have several different private pensions at once

• Stakeholder pension: enables those without earnings, such as non-earning partners, carers, pensioners and students, to pay into a pension scheme. Almost anybody up to the age of 75 may take out a stakeholder pension and it is not necessary to make regular contributions

For more information, see the GOV.UK pension guide.

Pension Credit

The qualifying age for Pension Credit has been increasing gradually in line with the increase in the State Pension age.

Region (also see Urban-Rural)

Regional classifications are based on the standard statistical geography of UK Regions: nine in England, and a single region for each of Wales, Scotland and Northern Ireland. Tables will also show statistics for the United Kingdom, Great Britain, and England as a whole.

Some split London into Inner and Outer where there is sufficient data to provide meaningful comparisons.

• Inner London boroughs: Camden, City of London, Greenwich, Hackney, Hammersmith and Fulham, Islington, Kensington and Chelsea, Lambeth, Lewisham, Southwark, Tower Hamlets, Wandsworth, Westminster

• Outer London boroughs: Barking and Dagenham, Barnet, Bexley, Brent, Bromley, Croydon, Ealing, Enfield, Haringey, Harrow, Havering, Hillingdon, Hounslow, Kingston upon Thames, Merton, Newham, Redbridge, Richmond upon Thames, Sutton, Waltham Forest

Sandwich carer

A sandwich carer is defined in the FRS as somebody who is aged 16 to 70 and:

• cares for a child within their household and/or has a child dependent on them within their household

• also cares for an adult relative

Savings

The total value of all liquid assets, including fixed-term investments. Pound amounts are informed by responses to questions on the value of assets or, in some cases, estimated from the interest on the savings. Note that banded savings do not include assets held by children in the benefit unit or household.

The FRS asks questions about all saving and investment products, including bank and building society accounts, and shares. These products go by many names. In this publication, the products are labelled as follows:

• Basic bank account: This type of account is similar to a current account. Payments can be received from other sources and it can pay bills by direct debit, but unlike a current account there are no overdraft facilities. Withdrawals can be made from cash machines and, in some cases, over the counter of the bank or building society itself

• Child Trust Funds (CTFs) have been replaced by Junior ISAs (JISAs) as the main tax-free savings account for children. See ISA

• Current account: This includes all accounts at both banks and building societies, which are used for day-to-day transactions; with a bank card. Overdraft facilities may be offered

• Company share schemes (profit sharing): Some companies provide extra rewards or bonuses to their employees depending on the profitability of the company. In publicly traded companies, this often takes the form of shares in the company. This label is given to any scheme which follows this general principle

• Credit union: A credit union is a financial co-operative similar in many respects to mainstream building societies. Its members both own and control the credit union, which is run solely for their benefit. All members of a specific credit union must share what is known as a “common bond” i.e. they must be connected in some way to the other members of that credit union. The members pool their savings into a single ‘pot’ from which loans can be made to members of the credit union. Members who have deposited money receive an annual dividend, while those to whom money is lent have to pay interest on the loan

• Endowment policy (not linked): An endowment policy taken out to repay a mortgage but no longer used to do so. This is where the mortgage has either been paid off or, more usually, converted to a different method of repayment. The respondent has decided to retain the endowment as an investment in its own right, even though it is no longer intended to repay the mortgage

• Informal assets: An informal asset is money that a respondent has given to someone else to look after or save on their behalf, or money that the respondent saves in cash

• ISA: An Individual Savings Account (ISA) pays interest on a tax-free basis

To be eligible for a Junior ISA, children must be under 18 and living in the UK. Junior ISAs are now included at the question ChSave. There are two types of Junior ISA. A child can have both types; a cash Junior ISA; a stocks & shares Junior ISA. There is a limit on annual payments into JISAs.

As with Child Trust Funds, the Junior ISA is a long-term savings account which can only be accessed by the child on their 18th birthday. The Junior ISA is then transferred to an Adult ISA so that the child can access their money.

• Investment trust: See Unit trusts.

• National savings bonds: All types of National Savings investments in this category are collected on the survey, except Easy Access and Investment accounts:

  • Fixed Rate Savings Bonds: replaced new issues of FIRST Option Bonds

  • National Savings Certificates: yield earnings in either a fixed or index-linked manner, for lump sum savings of £100 or more. Maximum earnings are obtained after five years and interest on investments is tax free

  • National Savings Income Bonds: minimum purchase is £2,000 and a maximum holding of £250,000; interest is paid monthly, and is gross of tax

  • Children’s Bonus Bonds: can be bought for any child aged under 16 as a five-year accumulating investment; interest is paid gross of tax

• NS&I savings accounts: The National Savings & Investments (NS&I) Investment Account and Direct Saver

• Other bank or building society account: Accounts belonging to adults recorded under categories “savings account, investment account or bond, any other account with bank building society, etc.”

• Post Office card account (POCA): This type of account can only be used to receive benefits and Tax Credit payments. Some other payments, such as Housing Benefit, occupational pensions, or wages cannot be paid into it. Payments can only be collected over the counter at a Post Office and will not incur any charges or accrue interest on money contained therein. Due to the limited capability to receive payments, these accounts are included or excluded in tables as noted

• Premium bonds: Investments which do not earn interest, but are entered in a monthly draw for tax-free cash prizes

• Stocks and shares: This includes all bonds, debentures and other securities which are usually traded on the financial markets. Bonds issued by the UK or foreign governments, or local authorities would also be recorded here. A share is a single unit of ownership in a company. ‘Stocks’ is the general term for various types of security issued by companies to raise financial support. If respondents are members of a shares club they will be included with those owning stocks and shares

• Unit trusts: A collectively managed investment in the financial markets, where investors buy ‘units’ of a fund, which invests in shares, stocks, Gilts, etc. Dividends are paid net of tax. The data presented for unit trusts also includes investment trusts, since these two assets are collected together in the FRS

• Any other type of asset: This is a catch-all category for the small numbers who own other types of financial asset. This includes Gilts (HM Government bonds) which raise money for the UK Government by offering a secure investment, usually over a fixed term, and usually with a set rate of interest although some are index-linked. Interest is paid half-yearly

The above products cover all types of savings. Some of them are grouped together in other ways in the tables:

• Direct payment account: A direct payment account is one that can accept electronic payment of benefits via BACS (the Banker’s Automated Clearing System). The types of accounts included in this grouping are:

• Current Account

• National Savings and Investments Savings Accounts

• Savings, investments etc.

• Basic Account

Where noted, Post Office Card Accounts are also included in this group.

Sources of income

• Wages and salaries: for a respondent currently working as an employee, income from wages and salaries is equal to: gross pay before any deductions, less any refunds of income tax, any motoring and mileage expenses, any refunds for items of household expenditure and any Statutory Sick Pay or Statutory Maternity Pay, plus bonuses received over the last 12 months (converted to a weekly amount) and any children’s earnings from part-time jobs

• Self-employed income: the total amount of income received from self-employment gross of tax and national insurance payments, based on profits (where the individual considers themselves as running a business) or on estimated drawings otherwise. Excludes any profits due to partners in the business. Any losses are recorded as such

• Self-employed respondents are asked questions on their most recent business accounts as submitted to HMRC: dates of the accounts, profit or loss figures, and amounts paid in tax and National Insurance

• They are then asked if they draw money from their business accounts for non-business purposes, such as for payments to themselves, personal spending, paying domestic bills etc. and how much this is per month on average. They are also asked if they receive other income from their business for personal use, for example cash in hand, and how much this is per month on average

• Those who do not keep annual business accounts and do not draw money for non-business purposes are asked for their income after paying for materials, equipment, goods etc. and whether they make tax and National Insurance payments on this amount

• Investments: Interest and dividends received on savings and investments. See Savings and investments for details of investments covered by the FRS

• Tax Credits: Income from Tax Credits

• State Pension plus any Pension Credit: for any adults who are over State Pension age, any State Pension plus any Pension Credit which is received; these benefits are shown together because of known problems with separating these amounts for pensioners

• Other pensions: payments received from pension schemes, including occupational, stakeholder or personal pension schemes; employee pensions for surviving spouses, annuity pensions, trusts and covenants

• Disability benefits: payments received from any of the benefits payable due to disability – see Benefits

• Universal Credit: (UC) is now the primary working-age benefit. Universal Credit replaces all the following state support: income-based Jobseeker’s Allowance, income-related Employment and Support Allowance, Income Support, Working Tax Credit, Child Tax Credit and Housing Benefit. It replaces these with a single, usually monthly payment, administered by DWP

• Other benefits: payments received from any of the other Benefits

• Other sources: payments from all other sources including, for example, baby-sitting, allowances from absent spouses including child maintenance, organisations, royalties, odd jobs, sub-tenants, educational grants, alimony and Healthy Start Vouchers

State Pension age

Since 6 April 2010, the State Pension age increased gradually for women and since December 2018 it increased for both men and women. The State Pension age for both men and women reached 66 by October 2020. During the whole of the 2021-2022 survey year State Pension age was 66.

Details of further planned changes to State Pension age

State support

An individual is in receipt of state support if they receive one or more benefits, or are being paid Tax Credits.

Tax Credits

Working Tax Credits and Child Tax Credits are paid by HMRC. Tax Credits are being phased out, as they are replaced by Universal Credit.

Tenure

This is the basis on which the head of household is resident in their dwelling. Types of renting or ownership as classified as follows:

• social renting: includes all cases where the landlord is either the local authority, or a housing association

• private renting: all cases where the property is rented from a private landlord, including those on a rent-free basis

Rent-free accommodation is any provided free by an employer or by an organisation to a self-employed respondent, provided that the normal activities of the tenant are to further the cause of the organisation (for example, Church of England clergy). Accommodation is not classed as rent-free if anyone, apart from an employer or organisation, is paying a rent or mortgage on a property on behalf of the respondent.

• Buying with a mortgage: includes local authority and housing association part-own-and-part-rent, and shared ownership arrangements

• Owned outright: households who pay neither rent, nor any mortgage or loan used to purchase the property. These households may have other loans secured on their property for which information is collected on the FRS. However, these payments are excluded from the costs of housing

Prior to 2008 to 2009, social renting was split into council and housing association groups. This division was removed because it was found to be unreliable. Comparison with administrative data showed that a significant number of housing association tenants wrongly reported that they were council tenants. Also, in 2008 to 2009, a split between furnished and unfurnished private renting was removed.

Universal Credit

Universal Credit (UC) is now the primary working-age benefit. Universal Credit replaces all the following state support:

• income-based Jobseeker’s Allowance
• income-related Employment and Support Allowance
• Income Support
• Working Tax Credit
• Child Tax Credit
• Housing Benefit

Most claimants will be of working-age, though claimants can be over State Pension age if their partner is still of working-age. UC supports those on low incomes with their housing and living costs, as well as child and childcare support where appropriate. It is not just for those who are out of work; it is also for those who are working, but whose earnings are low enough to qualify. Claimants must have capital of less than a set limit to be eligible.

UC completed its roll-out for new claims in Great Britain at the end of 2018, and is available for new claims throughout the UK. Legacy benefit claimants will continue to transfer to UC over a number of years.

The Universal Credit Official statistics provide the primary source of information about people and households on universal credit.

Urban rural classifications

England & Wales:

• The 2011 Rural-Urban Classification for Output Areas in England (and Wales)

The Classification defines areas as rural if they fall outside of settlements with more than 10,000 resident population. For the smallest geography areas, the classification assigns them to one of four urban (compressed to two for FRS use) or six rural categories:

• Rural: Hamlets and Isolated Dwellings in a sparse setting

• Rural: Hamlets and Isolated Dwellings

• Rural: Village in a sparse setting

• Rural: Village

• Rural: Town and Fringe in a sparse setting

• Rural: Town and Fringe

• Urban: City and Town in a sparse setting

• Urban: City and Town, including minor and major conurbations

Scotland

http://scottish-government-urban-rural-classification-2020-8-fold-map.pdf

Large urban areas	Settlements of over 125,000 people
Other urban areas	Settlements of 10,000 to 125,000 people
Accessible small towns	Settlements of between 3,000 and 10,000 people and within 30-minutes drive of a settlement of 10,000 or more
Remote small towns	Settlements of between 3,000 and 10,000 people and with a drive time of between 30 and 60-minutes to a settlement of 10,000 or more
Very remote small towns	Settlements of between 3,000 and 10,000 people and within a drive time of over 60-minutes to a settlement of 10,000 or more
Accessible rural	Settlements of less than 3,000, within 30-minutes drive to a settlement of 10,000 or more
Remote rural	Settlements of less than 3,000 people and with a drive time of between 30 and 60-minutes to a settlement of 10,000 or more
Very remote rural	Settlements of less than 3,000 people and with a drive time of over 60-minutes to a settlement of 10,000 or more

Northern Ireland:

Since the 2015 review of the ‘Statistical classification and delineation of settlements’, urban settlements in Northern Ireland are classified as those with a population greater than or equal to 5,000 people (Bands A-E).

Settlement (2015) Classification Bands	FRS categories
Band A: Belfast City	Belfast City
Band B: Derry City
Band C: Large Town, population greater than 18,000 people
Band D: Medium Town, population between 10,000 and 18,000 people	Other Urban
Band E: Small Town, population between 5,000 and 9,999 people
Band F: Intermediate Settlements, population between 2,500 and 4,999 people
Band G: Village, population between 1,000 and 2,499 people	Rural
Band H: Open Countryside and small villages with population less than 1,000 people

Working

All respondents whose employment status was employed or self-employed, irrespective of full-time or part-time working patterns.

Working-age

Adults (see Adult and Child) under State Pension age.