Labour Force Survey performance and quality monitoring report: April to June 2026

The achieved sample size, excluding imputed cases, for the UK Labour Force Survey (LFS) was 62,630 individuals in 28,655 households during Quarter 2 (Apr to June 2026). This represents a decrease of 8.9% in household interviews and a decrease of 9.0% in achieved person interviews, compared with Quarter 1 (Jan to Mar) 2026.

The dataset size, including imputed households, for the UK LFS was 81,193 individuals in 36,267 households during Quarter 2 2026. This represents an increase of 1.4% in individual cases and an increase of 1.2% in household cases, compared with Quarter 1 2026.

LFS data collection during Quarter 2 2026 was affected by an operational issue with how telephone interviewers’ time was allocated. This issue affected the data collection of the survey and directly affected response rates by causing a reduction in those excluding imputed cases. More information can be found in Section 2: Summary of quality and our recent blog post.

In Quarter 2 2026, the main response rates were as follows:

• the total response rate for Great Britain excluding imputed cases was 22.8% (Table 2); this is down 2.3 percentage points on the previous quarter

• the response rate for Great Britain excluding imputed cases was 33.6% in Wave 1 and 17.3% in Wave 5 (Figure 3); this compares with 33.4% and 20.7%, respectively, in the previous quarter

• the total response rate for Great Britain including imputed cases was 29.2% (Table 3); this is up 0.4 percentage points on the previous quarter

• the total response rate for the UK excluding imputed cases was 23.4% (Table 4); this is down 2.1 percentage points on the previous quarter

• the total response rate for the UK including imputed cases was 29.6% (Table 5); this is up 0.5 percentage point on the previous quarter

• non-contacts made up 16.9% of unproductive cases (Figure 4); this is up 1.6 percentage points on the previous quarter

• other unproductive cases included circumstantial refusals (5.1%, down 0.1 percentage points from the previous quarter), outright refusals (65.7%, up 2.9 percentage points on the previous quarter) and other refusals, which made up the remainder

• the English region with the highest accumulated response rate across the five waves was the South West (34.8%) (Table 6); the lowest was West Midlands Metropolitan Council (20.2%)

• the overall proxy response rate was 36.4% (Table 7); the highest proxy response rates were in the 16 to 17 years age group (95%), in males (40.5%), and in the combined total proxy response rate of all ethnic groups other than the White ethnic group (42.8%)

• the average income response rate was 82.3% (Table 8)

• the data on attrition rates show (in percentage change terms) that those who drop out of the survey are overrepresented in the 20 to 29 years age group, in employees, in households with five people, and in the Greater Manchester region (Table 9)

Respondent characteristics proportions for age, tenure, and index of multiple deprivation (IMD) remained stable in Quarter 2 2026. More information can be found in Section 5: Respondent characteristics.

More information about the LFS and technical definitions can be found in our LFS quality and methodology information (QMI) and our LFS user guidance.

We introduced several collection measures in Quarter 4 (Oct to Dec) 2023 and the Labour Force Survey (LFS) sample was increased from January 2024. Together, these two actions have incrementally led to increases in the achieved sample and subsequently, the quality of the data. The five-wave structure of the LFS means that some changes can take up to 15 months to fully feed through into survey estimates. Quarter 1 (Jan to Mar) 2025 was the first period of data that reflected the full impact of the January 2024 sample increase. More information can be found in our Labour Market transformation – update on progress and plans: July 2025 article and our LFS quality update: May 2025 article.

We introduced a coding practice during and after the coronavirus (COVID-19) pandemic that involved coding cases that were unproductive in two consecutive waves as headquarter (HQ) refusals. This prevented these cases from being attempted again in the following wave. This measure was put in place to better manage interviewer workloads while resources were stretched. This practice was discontinued in December 2025. For more information see Section 6: Quarterly changes and updates.

LFS data collection during Quarter 2 (Apr to June) 2026 was affected by an operational issue with how telephone interviewers’ time was allocated, as first reported in our blog post in June 2026. This affected collection of survey data during May and early June. A mistake in operational systems at this time led to more telephone interviewers being allocated to the Transformed Labour Force Survey (TLFS) than intended. The impact of this was that LFS telephone interviewing was underresourced, and households were therefore contacted fewer times than they would have been under normal circumstances.

Telephone interviewing on LFS is primarily used for Waves 2 to 5 (following up cases who were interviewed at the previous wave). Interviewing of respondents who did not participate at the previous wave is generally undertaken face-to-face. The impact of the operational issue is therefore evident within the Quarter 2 2026 response figures presented within this report for Waves 2 to 5, while Wave 1 is unaffected.

If a respondent from one quarter is not interviewed in the consecutive quarter, their responses from the earlier quarter may be carried forward (for one quarter only), which is referred to as “imputation”. This process is detailed in Section 7: Understanding response rates. This imputation approach largely mitigated this quarter’s data collection issue within the final dataset. This means that while a drop can be observed in the size reported for the achieved sample excluding imputed cases, the achieved sample including imputed cases continued to increase quarter-on-quarter.

The increased level of imputation within the final dataset was estimated to have had minimal impact on the headline estimates of employment, unemployment and inactivity rates. However, the impact on average actual weekly hours was found to be slightly larger. The operational issue will affect the longitudinal dataset, which covers two quarters and does not include imputed cases. The longitudinal dataset will therefore be smaller, which may lead to increased variability in the estimates. More detail will be available in our LFS quality update article, publishing in August 2026.

We reweighted the LFS estimates from Quarter 1 2019 onwards, alongside the changes to the sample and data collection. The reweighted LFS estimates incorporate information on the size and composition of the UK population, based on our 2022 mid-year estimates. For England, Wales and Northern Ireland, they are projected forward using scaling factors from our 2021-based national population projections published in January 2024. For Scotland, they are projected forward using scaling factors from 2020-based national population projections published in January 2023. More information can be found in our Impact of reweighting on Labour Force Survey key indicators: December 2024 article. A further exercise to reweight data all the way back to 2011, using the latest available population estimates and projections, is underway. Because of the scale of this project, revised estimates are due to be published over a period of around two years, with the first revised estimates expected to be available in late 2026.

We continue to advise users to apply caution when interpreting short-term changes observed in the estimates from the data. We suggest users read our commentary on our LFS-based statistics, which will consider ongoing quality challenges.

The total response rate is the aggregate response rate for the quarter across all waves, based on all eligible, in-scope households. For a full definition of response categories and the method used to calculate response rates, see Section 7: Understanding response rates.

We monitor the distribution of the sample across a range of demographic characteristics on a quarterly basis to help identify potential biases in the data. We have seen changes to the make-up of the Labour Force Survey (LFS) sample since the onset of the coronavirus (COVID-19) pandemic in March 2020. The following charts show the trends observed across all waves of the LFS in age, tenure, and index of multiple deprivation (IMD) quintile since Quarter 1 (Jan to Mar) 2019.

This section details changes that may have had an impact on response rates and data included in the charts and tables in Section 5: Respondent characteristics. Details of historical operational changes can be found in Volume 1: Background and methodology of our Labour Force Survey user guidance.

Quarter 2 2026 questionnaire changes

There were no questionnaire changes in Quarter 2 (Apr to June) 2026.

Methodological changes

In response to fluctuation in response rates since the onset of the coronavirus (COVID-19) pandemic, we made several adjustments to the size of the LFS Wave 1 sample in recent years. These adjustments included:

• doubling the Wave 1 sample in May 2020
• reducing the Wave 1 sample to 160% of pre-pandemic levels in April 2021
• reducing the Wave 1 sample to 150% of pre-pandemic levels in October 2021
• increasing the Wave 1 sample to 155% of pre-pandemic levels in April 2022
• reducing the Wave 1 sample to pre-pandemic levels in July 2023
• increasing the Wave 1 sample to 155% of pre-pandemic levels (25,800 addresses in total per quarter) in January 2024

Northern Ireland Statistics and Research Agency (NISRA) are responsible for the LFS data collection in Northern Ireland. NISRA reduced the Wave 1 issued sample in Northern Ireland by 20% (from 1,287 to 1,040 addresses per quarter) in August 2024.

Fieldwork and operational changes

Since the start of the pandemic a coding practice was put in place. This coding practice involved coding cases that were unproductive in two consecutive waves as HQ refusals so that these would not be attempted again in the following wave. These cases were less likely to result in an interview compared to those that had achieved a productive outcome in previous waves, and this helped mitigate increasing workloads for interviewers whilst capacity was stretched.

This measure was paused in October 2023 when LFS recovery measures were introduced. This resulted in an observed decrease in refusal rates in Waves 2 to 5. To improve our fieldwork operations, this practice was reintroduced in July 2024 after the Wave 1 sample increase led to an increase in unproductive cases in follow-up waves.

With the implementation of the Survey Improvement and Enhancement Plan interviewer capacity in the field has been significantly increasing. As such, more cases could be worked in 2025 than in 2024 which has reduced the need for this coding practice. Consequently, the practice was discontinued in December 2025, and only Survey Enquiry Line (SEL) should now record HQ refusals.

This change in coding practice is reflected in non-response data, including the composition of non-response meaning that the increase in outright refusals in the last two calendar quarters is largely offset by a decrease in the refusals to HQ (see Fig 4).

The Labour Force Survey (LFS) is managed by the Social Surveys directorate of the Office for National Statistics (ONS) in Great Britain, and by the Central Survey Unit of the Northern Ireland Statistics and Research Agency (NISRA) in Northern Ireland. Response rates for the UK are calculated by combining the achieved samples for Great Britain and Northern Ireland.

More information and technical definitions can be found in our LFS quality and methodology information (QMI) and Volume 1: Background and methodology of our Labour Force Survey user guidance.

Method of calculating response rates

The response rate is how many interviews were achieved as a proportion of those eligible for the survey. The formula used is:

Response rate (RR) equals full response (FR) plus partial response (PR), divided by FR plus PR plus outright refusal (OR) plus circumstantial refusal (CR) plus refusal to headquarters (RHQ) plus non-contact (NC) plus refusal to re-interview (RRI) (for only Waves 2 to 5).

The calculation of response rates should take the panel design into account, because the LFS is a panel survey. Households may refuse further participation at any of the five quarterly visits they are due to receive.

Households that refuse further participation are not revisited at the next quarter, but they remain part of the eligible sample. The response rate for households is the ratio of the number of households responding at the current wave to the sum of the number of eligible households found at the same wave at the sampled addresses, plus any households that have refused outright to participate at a previous wave. Outright refusals are distinct from circumstantial refusals and non-contacts, which are revisited at the next wave. Outright refusals may occur either when the interviewer calls or asks for permission to re-call in three months.

This simple model does not take account of situations like net addition or subtraction of eligible households at sampled addresses at waves after Wave 1, for example by a net increase or decrease in occupation of household spaces. However, the definition of household spaces is fluid and does not provide a firm basis for response rate calculations. LFS response rates that take the panel design into account are based on the simple model.

Definitions of response outcome categories

Full response

A household in which each household member has answered all applicable questions.

Partial response

A household in which questions were not completed because someone refused to be interviewed, refused part way through the questionnaire, or refused to let someone else answer on his or her behalf. However, at least one question block must have been completed. If only part of the information has been collected for a one-person household, it is coded as a refusal or non-contact.

Outright refusal

A household that refuses to respond to the survey and the interviewer feels that there is no chance of an interview at the current wave or in any future wave.

Circumstantial refusal

A household where the respondent refuses to respond because of a temporary circumstance (for example, going on holiday or being too busy during the field period). A circumstantial refusal enables an interviewer to call back at the next wave.

Refusal to HQ

A household that contacts headquarters to refuse to participate in the survey in response to the advance letter.

Non-contact

When an address is occupied, but where it has not been possible to contact any member of the household in the field period.

Method of calculating income response rates

The income question is asked at Waves 1 and 5 only. The subset of respondents who are eligible for these questions includes individuals aged 16 to 69 years who are in employment in the reference week. The percentage response rates for the income questions are based on all eligible, in-scope respondents at Wave 1 and all eligible, in-scope respondents at Wave 5. The total response rate is the aggregate response rate for income for the quarter (Wave 1 and Wave 5), based on all eligible, in-scope respondents.

Proxy response

The LFS interviewers must complete fieldwork to a tight timetable and interview as many of the sampled households as possible. This leaves limited time for re-calls. LFS interviewers try to interview every adult (aged 16 years and over) in each sampled household. However, when a household member is unavailable for interview, interviewers accept information by proxy from another responsible adult in the household. The proxy respondents are normally people living with a partner on behalf of their partner, and parents on behalf of their adult offspring who live with them.

Attrition

Attrition is the term applied to respondents who begin the survey but then drop out. These respondents tend to have different characteristics to those who remain in all waves of the survey. This can result in attrition bias. For example, if respondents in a particular age group have a higher tendency to drop out (attrition rate) than respondents in other age groups, then they will be underrepresented in subsequent waves of the survey and in estimates.

Item non-response

Item non-response refers to the situation where a respondent fails to answer a specific question or item during the survey, even though they have completed other parts of the survey. This can occur for various reasons, such as the respondent not knowing the answer, choosing to skip a sensitive question, or misunderstanding the question. This type of non-response is distinct from unit non-response.

Unit non-response

Unit non-response refers to individuals in sampled households not participating in the survey. This can occur for a variety of reasons, such as the respondent refusing the survey, including circumstantial refusals and outright refusals, or being unable to reach or contact the selected respondent during the field period.

Imputed cases

If a respondent from one quarter is not available in the next consecutive quarter, then their responses from the earlier quarter may be carried forward (for one quarter only). This is referred to as “imputation”. Response rates excluding imputed cases are the best measure of true response, as these are calculated based solely on actual responders as a proportion of eligible cases for the current quarter. Imputed cases are included in analytical datasets to provide a greater dataset size. We also report response rates including imputed cases in this methodology to show the extent of imputation for the current quarter. More information about imputation can be found in Section 12 of our Volume 1: Background and methodology of our Labour Force Survey user guidance.

Timeliness refers to the lapse of time between publication and the period to which the data refer. Punctuality refers to the time lag between the actual and planned dates of publication.

To the Office for National Statistics Labour Market and Households Division

• The scheduled delivery date for file is 17 July 2026.
• The achieved delivery date for file was 17 July 2026.
• The time lag between achieved delivery date and the end of the reference period was 12 days.

Data file for other users

The scheduled availability date for regional public and government normal release user files is 18 August 2026.