Annual population survey (APS) QMI

• Statistical designation: official statistics in development

• Survey name: Annual Population Survey (APS)

• Data collection: quarterly/annually

• Frequency: quarterly – Quarter 1 (Jan to Mar), Quarter 2 (Apr to June), Quarter 3 (June to July), and Quarter 4 (Oct to Dec)

• How compiled: longitudinal sample survey

• Geographic coverage: UK

• Related publications: Labour market overview, UK bulletins

• Last revised: 6 February 2026

This quality and methodology information report contains information on the quality characteristics of the data (including the European Statistical System's five dimensions of quality) as well as the methods used to create it.

The information in this report will help you to:

• understand the strengths and limitations of the data

• learn about existing uses and users of the data

• understand the methods used to create the data

• decide suitable uses for the data

• reduce the risk of misusing data

• The Annual Population Survey (APS) is a continuous household survey covering the UK, with the aim of providing estimates of main social and labour market variables between censuses, down to a local-area level.

• The APS is not a standalone survey; it uses combined data collected from two waves of the main Labour Force Survey (LFS) and data collected on local sample boosts (for more information on sampling, see Volume 10: Analysis of data collected by the Labour Force Survey: Which dataset should I use? of our LFS user guidance).

• The APS covers multiple topics, including the labour market, housing, ethnicity, religion, health, wellbeing, smoking, sexual orientation, and education.

• We publish quarterly APS datasets that include 12 months of survey data for overlapping annual periods (January to December, April to March, July to June, and October to September); the current achieved sample size is approximately 80,000 UK households and 175,000 individuals per year, as of our October 2024 to September 2025 dataset.

• We provide APS quarterly datasets to government departments; they are available to approved researchers via our Secure Research Service (SRS) and the UK Data Archive, Essex University, which also provides non-disclosive data for public access.

• The ongoing challenges with response rates, levels, and weighting approach mean that both LFS- and APS-based labour market statistics will be considered official statistics in development until further review; estimates of change should be treated with additional caution because of the increased volatility of LFS and APS estimates (for more information, see our Labour market in the regions of the UK statistical bulletins).

Overview

The purpose of the Annual Population Survey (APS) is to provide information on important social and socio-economic variables at more granular levels than the Labour Force Survey (LFS). Published APS statistics enable monitoring of estimates between censuses for a range of policy purposes and provide local-area information for labour market estimates.

The LFS and the APS are the data sources recommended for employment-related statistics, such as estimates of the number of people in employment or unemployed. They are also a unique source of comprehensive, coherent information about economic inactivity, as they separate information about people who want a job and those who do not.

Annual Population Survey datasets

Household datasets

The LFS (and APS) is a household survey that aims to collect information for all eligible people in a sampled household. The LFS (and APS) can therefore be used to conduct household- and family-level analysis.

The APS Household dataset is produced for January to December periods only. It is used to carry out household or family analysis at a more detailed level than is possible using the LFS Household datasets.

The APS Household dataset achieved a sample of 185,000 individuals and 84,000 households for the January to December 2024 dataset.

Three-year pooled datasets

The APS three-year pooled dataset is designed to allow more robust analysis at lower-level geographies. Lower-level analysis is not always possible using the single year APS dataset, especially for certain topics whose achieved sample size is smaller.

This three-year dataset will contain a sample size of around 320,000 individuals and 150,000 households (from the January 2022 to December 2024 dataset). It will largely only include variables that appear in all three years that are covered.

Any analysis produced from the pooled dataset should be treated solely as point-in-time estimates. We do not recommend using the pooled datasets for any time series analysis.

Two-year longitudinal datasets

LFS respondents are interviewed once every quarter over a 13-month period and five times in total. APS boost respondents are interviewed once every year for four years and four times in total. This allows certain longitudinal analysis to be performed on the data.

Longitudinal analysis generally focuses on employment-related variables and on information about respondents who are working age. For example, we can analyse the number of people who have moved from employment to unemployment by comparing two data points that are one year apart (captured in a two-year dataset).

The two-year APS longitudinal datasets have been produced for January to December periods from 2013; the 2012 dataset is linked with 2013, the 2013 dataset is linked with 2014, and so on. These datasets were introduced to supplement the LFS five-quarter longitudinal datasets because they had a relatively small sample in recent years. The APS two-year dataset has an achieved sample of approximately 23,000 individuals (from the January 2024 to December 2024 dataset).

Uses and users

Users of APS data often combine them with related data from other sources to provide an overall view of the state of the labour market. The Bank of England's Monetary Policy Committee is one of the most important users of this sort of assessment.

Other important users who receive APS data are HM Treasury and the Department for Work and Pensions. They are interested in a variety of indicators of the state of the labour market, including:

• the number of people in employment

• the number of hours worked

• the number of unemployed people (defined according to the International Labour Organisation (ILO))

These users often analyse these series by age groups, regions and gender.

Other data recipients include:

• the Department for Business and Trade 

• the Ministry of Housing, Communities and Local Government

• the Department for Education 

• Northern Ireland's Department for Economy

• the Department for Culture, Media and Sport 

• the Department for Transport 

• Northern Ireland's Department of Finance and Personnel

• the Department of Health and Social Care

• the UK Data Archive

• the Health and Safety Executive

• HM Revenue and Customs

• the Home Office 

• the Low Pay Commission 

• the Office for the Pay Review Bodies

• the Scottish Government and Scottish Executive 

• the Welsh Government 

At the international level, APS data are used by the Organisation for Economic Co-operation and Development (OECD) and the International Labour Organization (ILO).

Other data users include:

• local authorities

• the Trades Union Congress (TUC)

• the Employer's Association

• the Confederation of British Industry

• the Institute of Employment Studies

• the Institute for Public Policy Research

• the National Institute of Economic and Social Research

• the Policy Studies Institute

• the Institute for Fiscal Studies

• academic researchers

• the media

• the general public

Strengths and limitations

Strengths

• The APS has the largest sample size of any household survey.

• APS sampling errors are small, because the survey has a large sample size and a single-stage, random sample of addresses.

• The APS covers a large range of employment-related variables and non-employment-related variables, which allows for cross-linking analyses to be undertaken (for example, earnings linked with educational attainment).

Limitations

• The sample design provides no guarantee of adequate coverage of any domain, because the survey is not stratified by type of domain. 

• APS coverage does not include communal establishments, except for NHS housing (though the relevant sample frame has not been updated since its inception), and students in boarding schools and halls of residence.

• Members of the armed forces are only included if they live in private accommodation.

• Workers aged under 16 years are not included.

Recent improvements

Transformation of the Labour Force Survey

Following the 2014 National Statistics Quality Review of the LFS, we started a transformation programme to address the recommendations resulting from this review. After a series of large-scale quantitative tests, the Transformed Labour Force Survey (TLFS) was launched in March 2020, in response to the coronavirus (COVID-19) pandemic.

For more information on the design of this experimental survey, see our Transformed Labour Force Survey – user guidance. For further updates on the transformation and transition from the current LFS to the new TLFS see our the Labour Market Transformation – update on progress and plans article series.

We have recently faced challenges maintaining the quality and reliability of survey data. The COVID-19 pandemic accelerated long-term trends in declining response rates. In June 2025, we published our Survey Improvement and Enhancement Plan for Economic Statistics, reflecting the urgent need to address these challenges by outlining a clear roadmap to restore confidence, enhance data quality, and modernise systems.

This section describes the quality characteristics of the data and identifies issues that should be considered when using the statistics.

Relevance

Relevance is the degree to which the survey meets users' needs.

The purpose of the Annual Population Survey (APS) is to provide information on important social and socio-economic variables at a local level. The data enable monitoring of main estimates between censuses, for a range of policy purposes. They also provide local area information for labour market estimates.

The APS has the largest coverage of any household survey and enables the generation of statistics for small geographical areas. Sampling errors are smaller that other social survey designs, because the APS has a single-stage sample of addresses. For information about the survey's coverage constraints, please see the Strengths and limitation subsection of Section 4: Quality summary 

Accuracy and reliability

Accuracy and reliability are the degree of closeness between an estimate and the true value.

As the APS is a sample survey, it provides estimates of population characteristics rather than exact measures. In principle, many random samples could be drawn and each would give different results. This is because each sample would include different people who would give different answers to the questions asked. The spread of these results is the sampling variability.

Confidence intervals are used to present the sampling variability. For example, with a 95% confidence interval, it is expected that in 95% of survey samples, the resulting confidence interval will contain the true value that would be obtained by surveying the whole population. The statistical methodology used to calculate the Labour Force Survey (LFS) sampling variability is published in our LFS quality and methodology information (QMI).

Response rates for the LFS main sample are monitored and published in our quarterly LFS performance and quality monitoring reports. Response rates for the local area boost sample that feeds into the APS is monitored separately. Because of reduced sample sizes, there is potential for increased non-response bias; this has been accounted for in our weighting methodology. For more information, see the How we process the data subsection of Section 6: Methods used to produce the data.

Increased volatility of APS estimates, resulting from smaller achieved sample sizes, means that estimates of quarterly change should be treated with additional caution. We recommend using APS estimates as part of our suite of labour market indicators, alongside workforce jobs, Claimant Count data, and Pay As You Earn (PAYE) Real Time Information (RTI) estimates.

For information on how to use the APS for different purposes, and the weights applied to the datasets, see Volume 1: Background and methodology and Volume 10: Analysis of data collected by the Labour Force Survey: Which dataset should I use? in our our LFS user guidance.

Coherence and comparability

Coherence is the degree to which data that are derived from different sources or methods, but refer to the same topic, are similar. Comparability is the degree to which data can be compared over time and domain, for example, geographic level.

Labour Force Survey and Annual Population Survey

We published the first APS dataset for the period January to December 2004. Historically, the APS includes combined data from the quarterly LFS (Waves 1 and 5) and the local LFS (LLFS) boosts for England, Wales and Scotland (for more information, see Volume 6: APS user guide in our LFS user guidance). From 2026, the Scottish boost local sample was removed (for more information, see Section 6: Methods used to produce the data).

The main LFS produces rolling quarterly outputs each month. However, its main output is produced at the end of each calendar quarter (covering Quarter 1 (Jan to Mar), Quarter 2 (Apr to June), Quarter 3 (July to Sept), and Quarter 4 (Oct to Dec)). Before 2006, the main LFS and LLFS outputs were produced at the end of each seasonal quarter (December to February, March to May, June to August, September to November).

The APS is one of several data sources about the labour market. Some sources provide data that overlap with APS data on employment, unemployment and earnings. We have published guidance about the strengths and limitations of each source, in relation to these topics, and have indicated which source is the most appropriate for different purposes. While the APS and LFS use many of the same interviews, they have different achieved samples and results because they are based on different pools of data. We do not attempt to calibrate the results of one survey to the other. This means that the two surveys will have different non-response biases and therefore, will give different results for the same measure.

We recommend the LFS (and APS) for employment-related statistics, such as estimates of the number of people in employment or unemployed. The LFS (and APS) are also unique sources of comprehensive information about economic inactivity, where they provide information separately about people who want a job and those who do not. The appropriateness of each source depends on the balance between timeliness and precision, while considering the granularity of the estimate. The LFS gives more timely indicators, whereas the APS gives more precise estimates.

Other sources of labour market data

Our lead indicator of short-term changes in earnings is average weekly earnings (AWE). It is designed to capture monthly changes in the AWE of employees in Great Britain. AWE gives timely information on short-term trends in earnings. We publish these estimates in our Average weekly earnings in Great Britain bulletins. However, this measure lacks the detail available from our Annual Survey of Hours and Earnings (ASHE).

The ASHE is our main measure of earnings, in terms of structural statistics. It provides information about the levels, distribution and composition of earnings and hours paid for employees in most industries and occupations across the UK. It is also the lead source of estimates for the gender pay gap and the number of jobs paid below the National Minimum Wage or National Living Wage. We published our ASHE estimates in our Employee earnings in the UK bulletins.

HM Revenue and Customs' (HMRC's) data include information on people paid through the Pay As You Earn (PAYE) Real Time Information (RTI) system. We focus on median monthly pay in our Earnings and employment from PAYE RTI, UK bulletins. However, estimates of mean pay, and breakdowns by industry, age-band, various geographies, and pay distribution are also available.

The LFS and APS also collect information on the earnings of employees. While the earnings data are known to be underestimated in the LFS/APS, they are a useful source of breakdowns not available in our other sources (for example, by ethnicity and education). LFS/APS data on individuals' earnings are of a lower quality than AWE, ASHE or RTI data. This is because LFS/APS data are self-reported by employees.

You can read more about our data sources on earnings in our Comparison of labour market data sources methodology and our Income and earnings statistics guide.

Accessibility and clarity

Accessibility is the ease with which users are able to access the data, also reflecting the format in which the data are available and the availability of supporting information. Clarity refers to the quality and sufficiency of the release details, illustrations and accompanying advice.

The UK Data Service at Essex University provides free access to the various Labour Force Survey (LFS) and Annual Population Survey (APS) datasets. They can be contacted through the UK Data Archive website.

Tables are available publicly on NOMIS, which uses the APS for estimates of employment, unemployment and inactivity rates, down to unitary authority and parliamentary constituency area. Estimates of educational qualifications are also available on NOMIS.

Our Social Survey Data Advice and Relations team provides advice, APS data, and some ad hoc analysis of data for a fee. They can be contacted by email at socialsurveys@ons.gov.uk.

We regularly publish labour market data, including data from the LFS and APS, including narrative text, tables and charts. The data are widely available, generally free of charge, through a range of media. Our Labour market overview, UK bulletins and time series data included within the releases are available. Our labour market helpline can be contacted by email at labour.market@ons.gov.uk or by telephone on +44 1633 455400.

Our recommended format for accessible content is a combination of HTML web pages for narrative, charts and graphs and usable formats like CSV and Excel for data. Our website also offers users the option to download the narrative in PDF format. In some instances other software may be used, or may be available on request. Available formats for content published on our website but not produced by the Office for National Statistics (ONS), or referenced on the ONS website but stored elsewhere, may vary. For further information, please email the Data Advise and Relations teams at socialsurveys@ons.gov.uk.

More information about all available APS microdata can be found in Volume 6: APS user guide of our LFS user guidance.

Timeliness and punctuality 

Timeliness refers to the lapse of time between data collection and data delivery. Punctuality refers to the gap between planned and actual data delivery dates.

The APS Person data are disseminated quarterly, with each dataset covering 12 months' data. Final data are released three months after the close of the survey period to which they relate. The APS Household microdata are released at the same time as our Workless households for regions across the UK bulletins.

For more details on related releases, the GOV.UK release calendar provides 12 months' advance notice of release dates. If there are any changes to the pre-announced release schedule, public attention will be drawn to the change and the reasons for the change will be explained fully at the same time, as set out in the Code of Practice for Official Statistics.

Concepts and definitions

Concepts and definitions describe the legislation governing the output and a description of the classifications used in the output.

Economic activity

Economically active people are those aged 16 years and over who are either in employment or unemployed (according to the International Labour Organisation (ILO) definition of unemployed). This group of people are those who are active in the labour force.

Economic activity rate

The economic activity rate is the number of people who are economically active, presented as a percentage of the total population.

Economically inactive

Economically inactive people are neither in employment nor unemployed (based on the ILO measure). This group includes people who are caring for their family or are retired, and those aged under 16 years.

Employment

Employment includes those people who are aged 16 years or over who:

• did some paid work in the reference week (whether as an employee or self-employed)

• had a job that they were temporarily away from (for example, on holiday)

• are on government-supported training and employment programmes

• are doing unpaid family work (that is, working in a family business)

Employment rate

The number of people in employment, presented as a percentage of the population in that age group.

International Labour Organisation unemployment

The ILO measure of unemployment refers to people without a job who were able to start work in the two weeks following their Annual Population Survey (APS) interview, and who had either looked for work in the four weeks before interview or were waiting to start a job they had already obtained. This measure is different to the Claimant Count measure of unemployment, which is based solely on eligibility for benefits (that is, Jobseekers' Allowance).

The ILO measure of unemployment is generally accepted to be a more comprehensive measure and is usually higher than the Claimant Count for any given period, especially for women.

International Labour Organisation unemployment rate

The percentage of economically active people who are unemployed, according to the ILO measure. This refers to those aged 16 years and over, or those of working age.

For the range of international classifications used, please see Volume 5: Labour Force Survey (LFS) classifications in our LFS user guidance.

Geography

Additional geography variables were introduced to the LFS and APS after the 2011 Census. This is covered in detail in Volume 3: Details of LFS variables in our LFS user guidance.

Users should note that datasets from after January to December 2025 will not be suitable for lower-level analysis of associated Scottish geographies. This is because of changes regarding the Scottish boost sample, which are discussed in Section 6: Methods used to produce the data.

Why you can trust our data

The ONS is the UK's largest independent producer of statistics and is the UK's National Statistics Institute. We outline how data are collected, secured, and used in the publication of our statistics in our Data Policies and Information Charter. We treat the data that we hold with respect, keeping it secure and confidential, and we use statistical methods that are professional, ethical and transparent.

We apply statistical disclosure control methodology to LFS/APS data. This ensures that information attributable to an individual is not disclosed in any publication, and that confidentiality of respondents is protected in datasets. The Code of Practice for Official Statistics, and specifically Principle 4: Manage data responsibly, sets out the principles for protecting data from being disclosed. The Statistics and Registration Service Act 2007 includes data confidentiality regulations that apply to the ONS. More information about disclosure control for survey microdata is available in our Policy for social survey microdata.

Main data sources

The Annual Population Survey (APS) combines data from four quarters of our main Labour Force Survey (LFS) using Wave 1 and Wave 5 data with rolling-year data from the English and Welsh local Labour Force Survey (LLFS) or "boost" sample. The achieved sample size is approximately 80,000 households (or 175,000 respondents) on each annual APS dataset (as of the September 2024 to October 2025 dataset).

Labour Force Survey main sample

The LFS main quarterly sample of private households includes five waves. Each cohort is interviewed in five successive quarters – in any one quarter, one cohort will receive their first interview (Wave 1), another cohort will receive their second interview (Wave 2), and so on.

More information about the methodology, sampling and data collection of the LFS and APS can be found in Volume 1: Background and methodology of our LFS user guidance and our LFS quality and methodology information (QMI). For information about the achieved sample and response rates for the LFS main sample, see our LFS performance and quality monitoring reports.

The APS dataset is created by taking Wave 1 and Wave 5 from four successive quarters to obtain an annually representative sample. Over the four successive quarters, Wave 1 and Wave 5 will never contain the same households, which means responses from any household are not included more than once in the dataset.

Boost samples

The APS sample has historically been increased by three annual boost samples. Respondents to the annual boosts are interviewed four times at yearly intervals and approximately one quarter of the sample is replaced each year.

The first sample boost began in England in survey year ending 2001. It is combined with the LFS data from respondents in England to form the English local LFS (LLFS). The second sample boost began in Wales in survey year ending 2002. It is combined with the LFS data from respondents in Wales to form the Welsh LLFS. The third sample boost began in Scotland in survey year ending 2004. It was combined with the LFS data from respondents in Scotland to form the Scottish LLFS.

Changes to the Scottish boost sample

The Scottish boost sample has historically been funded by the Scottish Government. They decided to discontinue funding for the sample from Quarter 4 (Oct to Dec) 2025 to explore alternative sources that may better suit users' needs. This decision is explained in detail in the Scottish Government's Changes to Labour Market Statistics in Scotland: What You Need to Know article.

Quarter 1 (Jan to Mar) 2026 was the first quarter without the Scottish boost sample. The Office for National Statistics (ONS) maintained collection of the Scottish boost sample for Quarter 4 (Oct to Dec) 2025 to allow safe decommissioning from 2026. As a result, only the English LLFS and Welsh LLFS are now collected and added to the APS dataset.

The Scottish boost sample for the October 2024 to September 2025 dataset (11,320 individuals) accounts for 55.4% of all Scottish data and 6.4% of the total data in the sample. From Quarter 1 2026, no further Wave 1 Scottish boost samples will be drawn and no further Wave 2 to Wave 4 Scottish boost cases will be reinterviewed. The Scottish boost cases remaining in the datasets will reduce by 25% each quarter until the end of 2026. The first dataset with no Scottish boost cases remaining will be January to December 2026 (Figure 1). The main sample for Scotland will remain the same.

Figure 1: Existing Scottish boost sample cases will reduce by 25% every quarter after January to March 2026 because the boost was discontinued

Source: Annual Population Survey and Labour Force Survey from the Office for National Statistics

Notes:

1. Each block shows the Scottish boost cases present in each dataset.  
2. Q1 refers to Quarter 1 (Jan to Mar). 
3. Q2 refers to Quarter 2 (Apr to June). 
4. Q3 refers to Quarter 3 (July to Sept). 
5. Q4 refers to Quarter 4 (Oct to Dec).

Download this image Figure 1: Existing Scottish boost sample cases will reduce by 25% every quarter after January to March 2026 because the boost was discontinued

.png (75.5 kB)

How we process the data

APS datasets are weighted by using the most up-to-date official population data to reflect the size and composition of the general population. Weighting factors account for:

• the survey design, which does not include communal establishments

• the composition of the local population by age and sex

• non-response and tenure adjustments (the non-response model is only applied to data for Great Britain; the weights for other sample members are then adjusted to compensate for this)

Sample attrition is the term applied to respondents who begin the survey but subsequently drop out. These respondents tend to have different characteristics to those who remain in all waves of the survey, and therefore, this can result in sample attrition bias. The large sample size of the datasets helps to minimise sample attrition bias.

Every few years, all APS outputs (except APS Pooled datasets) are reweighted, based on the latest mid-year population and projection estimates available.

Imputation is carried out to minimise non-response bias in estimates, while simultaneously boosting the sample size. For person datasets, missing data are imputed by using roll-forward methods. For many variables, responses are rolled forward for Wave 5 only. This means that, if a respondent cannot be contacted for Wave 5, then the information that the respondent provided at their previous interview is used to impute a response for the current interview. This may give rise to error (for example, it is possible that the individual may have changed jobs).

To minimise non-response bias, we also collect proxy responses where there are other individuals in the household who can answer on behalf of the absent respondent. Responses from the annual boosts are not rolled forward because of the time elapsed since the response. For household datasets, missing data are imputed by using the "nearest neighbour" method.

How we quality assure and validate the data

Variable frequencies are compared with the previous period in each quarterly APS dataset. This identifies any significant discontinuities at an early stage. We investigate all discontinuities that we judge to be significant, to determine the reason for the discontinuity. For example, we assess whether a discontinuity is the product of questionnaire revision, a processing error, derived variable revision or error, or a real-world change. This process also ensures that the metadata associated with each variable are correct.

We check specific main-derived variables in detail by extracting the underlying variables and recalculating in another application, and then comparing the results with the values in the dataset. This ensures that the program used to calculate the derived variables is working correctly. We carry out structural integrity checks on the assembled annual dataset to ensure data are included from the relevant period, outcome and sample. We also perform time series checks on main variables. We carry out geographical consistency checks and all cases with missing geographical data are assigned missing values. We carry out post-weighting checks to compare APS totals with source population totals, and we investigate and correct any anomalies.

How we review and maintain the data processes

We conduct a number of engagements with both internal and external stakeholders as part of our review of the design, content and outputs of the LFS and APS. LFS production teams are represented on Operation Boards within our Social Surveys Directorate. They work with delivery teams to ensure the continued collection and production of LFS and APS microdata and the continued delivery of LFS and APS data for publication in the labour market and productivity outputs within our labour market portfolio.

We hold a monthly Management Board with stakeholders of LFS production, analysis and outputs of teams across the ONS. We also hold a quarterly User Group with stakeholders from across government to:

• ensure we are meeting their needs as users of these statistics

• give the opportunity for feedback on the performance and content of the survey

• consider any changes we need to make to the survey design

We also work closely with the ONS Secure Research Service team and the UK Data Service to ensure data are made available for all users, including academics, consultancies and commercial agencies, according to strict research purposes. We also ensure that any feedback from such analysts and organisations is considered as part of LFS and APS production.