Updated following the release of the 2025 statistical release
First issue with the release of the 2024 statistical release
This page is a guide to the stability measures for children looked after in England statistics published by the Department for Education. It sets out information on:
This release brings statistics on all three measures of stability together. The methodology will be updated with each new statistical release.
The statistics in the stability measures for children looked after in England publication are derived from information collected in the children looked after (CLA) SSDA903 return and the school census.
The DfE has collected the CLA SSDA903 return from local authorities for each full 12 month period since 1992. The latest collection covers the year from 1 April 2024 to 31 March 2025. There have been a number of significant changes to the collection since 1992. More information on the SSDA903 return is available in the CLA return: guide to submitting data (opens in new tab). The collection provides the latest information on children looked after by local authorities and care leavers. A more detailed timeline of changes to the SSDA903 is in the main CLA publication methodology. The SSDA903 data is collected in a longitudinal database, with one record for every episode of care. Local authorities update the database every year, including making amendments to previous years’ records where there have been changes.
The school census collects individual school characteristics and pupil records 3 times a year. In England, all of the following schools are required to provide the information collected in this census to the Department for Education (DfE):
Maintained nursery
Primary
Middle-deemed primary
Middle-deemed secondary
Secondary
All-through
Special schools (including non-maintained special schools)
Pupil referral units or alternative provision schools (PRU or AP)
Academy schools (including free schools, University Technical Colleges (UTCs) and studio schools)
City Technology Colleges (CTCs).
Further information on the school census can be found in the guide to completing the school census (opens in new tab).
For school stability, data is derived from matching the SSDA903 collection to the school census. The main matching key is the child’s unique pupil number (opens in new tab) (UPN), which identifies each pupil attending a state funded school in England. Allocated on first entry to a school, the UPN is generated using a nationally specified formula and is expected to remain with a pupil throughout their school career regardless of any change in school or local authority. Local authorities are required to return UPNs as part of the SSDA903 return. For more information on matching using UPN, see the methodology document for the ‘Outcomes for children in need, including children looked after’ statistical release. Once matched, children whose date of birth differed between the SSDA903 collection and the school census were removed.
Placement stability analysis was previously published in the Children looked after in England including adoptions publication. This included extended placement stability analysis at national level (including breakdowns by child characteristics) in the 2022, 2023, 2024 and 2025 publications. The statistics in the 2025 publication are also published here. These focus on the following social care groups:
For the first four of these social care groups, the figures are broken down by first placement during the year and corresponding legal status. However, for a small number of children (up to 50 in each year), their placement on 31 March has been used instead. The majority of these children were in respite care or a temporary placement for their first placement.
For CLA aged under 16 on 31 March who had been looked after continuously for at least 2.5 years and who were living with the same carer for at least 2 years, note the methodology has changed compared to figures published in the 2020 publication of Children looked after in England including adoptions. The figures now a. exclude children who started a period of care under an agreed series of short-term placements which later became continuous and b. include children in fostering placements with relative(s) or friend(s) or other foster carers who is/are also an approved adopter(s) – fostering for adoption/concurrent planning (placement codes U2 and U5) as a placement change.
The number of placements during the year (or previous 2 years) includes a child's first placement during the year (on 1 April, or from the date they first started care during the year if they were not in care on 1 April), and any subsequent placements involving a change in carer.
High placement stability is defined as a looked after child experiencing 3 or more placements during the year (or previous 2 years) ending 31 March.
School stability analysis was previously published in the Children looked after in England including adoptions publication. This included extended school stability analysis at national level (including breakdowns by child characteristics) in the 2022 and 2023 publications. Since then, school stability analysis has been published in this release only. These focus on the following social care groups:
To create the main spine of children for whom we would track school moves, CLA on 31 March each year were matched to the spring census of the preceding year (for the one-year measure) or the preceding two years (for the two-year measure). For example CLA on 31 March 2025 were matched to the spring census 2024 for the one-year measure and to the spring census 2023 for the two-year measure*. This provides a starting point in time from which we can measure school moves. An exception to this is where a child was not aged 4 to 15 in the first of the academic years spanned by the one-year or two-year measures. In these cases, a child was matched to the autumn census in the academic year when their age at the beginning of the academic year was 4 (when they were due to start school). This is shown in the tables below for CLA on 31 March 2025:
One-year measure (spanning academic years 2023/24 and 2024/25):
| Age at the beginning of the 2023/24 academic year | One-year measure census starting point |
|---|---|
| Children aged 4 to 15 (mainly those in reception to year 11 in 2023/24) | Spring 2024 |
| Children aged 3 (mainly those starting reception in 2024/25 academic year) | Autumn 2024 |
Two-year measure (spanning academic years 2022/23, 2023/24 and 2024/25):
| Age at the beginning of the 2022/23 academic year | Two-year measure census starting point |
|---|---|
| Children aged 4 to 15 (mainly those in reception to year 11 in 2022/23) | Spring 2023 |
| Children aged 3 (mainly those starting reception in 2023/24 academic year) | Autumn 2023 |
| Children aged 2 (mainly those starting reception in 2024/25 academic year) | Autumn 2024 |
* This is a change from the methodology used for the 2022 publication where children were additionally required to match to the summer census of the following year. This improvement was made to improve the match rate. This had the benefit of increasing the number of children in the spine (by around 750 children in the 2022 figures). The impact on the main measures was minimal. Looking again at the national 2022 figures, the percentages of children with no moves or at least one move during the year were unchanged. The percentages of children with no mid-year moves or at least one mid-year move changed by less than one percentage point.
The majority of CLA on the spine were tracked across up to five school censuses (for the one-year measure) or eight school censuses (for the two-year measure), beginning with their census starting point (see tables above). So for the majority of CLA on 31 March 2025, school moves were tracked across the following censuses:
| One-year measure | Two-year measure |
|---|---|
2022/23 academic year | |
| Spring 2023 | |
| Summer 2023 | |
2023/24 academic year | |
| Autumn 2023 | |
| Spring 2024 | Spring 2024 |
| Summer 2024 | Summer 2024 |
2024/25 academic year | |
| Autumn 2024 | Autumn 2024 |
| Spring 2025 | Spring 2025 |
| Summer 2025 | Summer 2025 |
If a child was not aged 4 to 15 in all the academic years spanned by either the one-year or two-year measures, their moves were tracked only in the academic years during which they fell into the relevant age span. To illustrate this, the two diagrams below show the school moves included for the different age groups of CLA on 31 March 2025:
A school move was included from either 1 April, if the child was already CLA on that date, or from the date a child entered care for the first time during the year, even if they left care and re-entered during the year.
Some children have been excluded from the 2021 and 2022 measures as the summer census 2020 was not carried out, due to the COVID pandemic, as follows:
‘Any move’ is defined as a child moving school (or from no school to a school) in either the one-year or two-year period ending 31 March.
A ‘mid-year move’ is defined as a child moving school (or from no school to a school) in either the one-year or two-year period ending 31 March, excluding the period(s) 1 August to 30 September.
Expected transitional moves from one school phase to another (for example primary school to secondary school at the age of 11) are excluded, where the move happened between 1 August and 30 September. These are defined as those where the child either reached the maximum age of the school they were moving from, or the minimum age of the school they were moving to.
Where a school has academised (resulting in a change of a school’s Unique Reference Number), this move is also excluded.
Special Educational Need (SEN) and primary need - taken from the spring census of the relevant year, e.g. spring census 2023 for 2023 data. Where a child was not tracked in this census, for example children aged 15 in the 2021/22 academic year for the one-year measure for the 2023 data, it was taken from the most recent spring census that their moves were tracked in. There are a small number of children who were not in the relevant spring census, and therefore their SEN is unknown.
School phase – the phase of the school the child was attending prior to their first move during the year (or previous 2 years). A small number of children whose first phase was a state-funded nursery are included in the figures for state-funded primary schools.
The number of school moves during the year (or previous 2 years).
High school instability is defined as a looked after child experiencing at least one mid-year school move during the year (or previous 2 years) ending 31 March - a move at any point in the year excluding the period(s) 1 August to 30 September.
Social worker data for CLA on 31 March was collected by DfE for the first time during the 2024 SSDA903 data collection. Statistics on the social worker stability of CLA were previously published in the ‘Stability Index’ by the Children's Commissioner's Office (opens in new tab) (CCO). Their most recent publication containing social worker stability was the 2019 Stability Index (opens in new tab), relating to the stability of CLA on 31 March 2018. In the 2022 release of the Children looked after in England including adoptions publication, DfE published social worker stability data collected by CCO for CLA on 31 March 2019 and 2020 on their behalf. Further details on this can be found in the social worker section of the methodology document (opens in new tab) relating to this publication.
These statistics focus on the following social care groups:
Figures have also been published for the reason a child is assigned a social worker for both the one-year and two-year measures.
Information on social worker stability was collected for the first time in 2024 via the SSDA903 return. For the 2025 data collection, for each CLA on 31 March 2025 (excluding those solely looked after on respite care), local authorities were required to submit information on every primary social worker episode for the two-year period between 1 April 2023 to 31 March 2025 – including any social worker episodes in periods within this timescale where the child was not looked after. This included the following information:
A number of automated data validation checks are carried out at the point of data entry to identify and remove issues such as:
Any record which fails the validation checks is highlighted and must be corrected. An explanation of each validation check and guidance on how corrections can be made are documented in the list of validation checks available on the DfE children looked after collection website (opens in new tab).
Some aggregate return level checks are performed in the system to highlight large changes in return level data compared to the previous year. These are there as prompts for local authorities to check their data. Any outstanding aggregate checks need to have explanatory commentary added to the system before the local authority can sign off their return.
At the end of the data collection a ‘snapshot’ of the social worker data was taken. This snapshot table is checked to make sure:
The snapshot table then went through a process of data cleaning. This included:
In the 2025 collection, for CLA on 31 March 2025, social worker episodes between 1 April 2023 and 31 March 2025 were submitted by local authorities. For a subset of these children, those who were also CLA on 31 March 2024, social worker episodes between 1 April 2023 and 31 March 2024 had already been submitted in the 2024 collection. Therefore, flags were added to the cleaned dataset to indicate which social worker episodes should be used for each child for each measure, depending on whether the child was CLA on 31 March 2024 only, CLA on 31 March 2025 only, or CLA on 31 March in both years. Where social worker episodes were submitted in both data collections, data from the most recent collection was used, as shown in the table below:
One-year measure | Two-year measure | ||
|---|---|---|---|
| 2024 (social worker episodes between 1 April 2023 and 31 March 2024) | 2025 (social worker episodes between 1 April 2024 and 31 March 2025) | 2025 (social worker episodes between 1 April 2023 and 31 March 2025) | |
| CLA on 31 March 2024 only | 2024 collection | N/A | N/A |
| CLA on 31 March 2024 and 2025* | 2025 collection | 2025 collection | 2025 collection |
| CLA on 31 March 2025 only | N/A | 2025 collection | 2025 collection |
*Where a child was CLA on both 31 March 2024 and 2025, a small number of local authorities did not re-submit social worker episodes between 1 April 2023 and 31 March 2025 during the 2025 data collection, but only those between 1 April 2024 and 31 March 2025. Therefore, in the creation of flags for the 2024 one-year measure and the 2025 two-year measure, social worker episodes provided for these children during the 2024 data collection, covering the period 1 April 2023 to 31 March 2024, were used instead. These local authorities were:
The cleaned dataset was then used to produce the data for this statistical release.
Barnet local authority were unable to complete the social worker data return in the 2024 data collection, citing the reason that their system had not been upgraded in time to collect the relevant information. Therefore, they are excluded from the 2024 one-year measure.
A number of local authorities continued to report difficulties in supplying accurate information on reason for social worker change for some or all of its social worker episodes. In the 2025 data collection, these local authorities were:
Where local authorities were unable to record a reason a child was assigned a social worker, the majority reported that they mainly used ‘Other reason'. There may also be instances of other reasons being used.
In the 2024 data collection, local authorities were asked to use ‘Other reason’ if the social worker episode was open prior to 1 April 2023 and the reason for the new social worker episode was unknown. In the 2025 data collection, local authorities were able to leave the reason field as null for this reason. These have been combined with the ‘Other reason’ category in the social worker episodes reason table in the release. In the 2025 data, the proportion of ‘Other reason’ which were nulls was 32 per cent for the one-year measure and 43% for the two-year measure.
The number of social workers assigned to a child during the year is defined as the number of unique social workers assigned to a child during the year ending (or previous 2 years ending) 31 March. For example, if a child had a social worker, moved to a different social worker, then back to the original social worker during the course of the year (or previous 2 years), then this would count as 2 social workers during the year (or previous 2 years). The count of social workers includes any social workers assigned to a child within the relevant timescale when the child was not looked after.
High social worker instability is defined as a looked after child experiencing 3 or more social workers during the year (or previous 2 years) ending 31 March.
Rounding and suppression is applied to the data. The National Statistics Code of Practice requires that reasonable steps should be taken to ensure that all published or disseminated statistics produced by the DfE protect confidentiality. The publications follow the DfE policy statement on confidentiality.
National and regional figures have been rounded to the nearest 10. Local authority figures are unrounded.
Where any number is shown as zero, the original figure submitted was zero.
The following symbols have been used in the releases (updated to align with GSS standards):
For percentages:
These statistics are badged as "official statistics in development". As the social worker data collection is new, we expect the quality of the data returns to improve over time as the collection becomes established.
‘Official statistics in development’ means statistics that are undergoing a development. They have been produced as far as possible in line with the Code of Practice for Statistics.
This can be broadly interpreted to mean that these statistics are:
Find out more about the standards we follow to produce these statistics through our Standards for official statistics published by DfE guidance (opens in new tab).
This statistical release is produced using a final version of the dataset. We do not plan to make any revisions to the publications. If we later discover that a revision is necessary, this will be made in accordance with the DfE statistical policy statement on revisions.
Children looked after in England including adoptions: Information on children looked after in England, including numbers of looked after children adopted, care leavers and looked after children who are missing. Data is taken from the annual SSDA903 data collection.
Outcomes for children in need, including children looked after by local authorities in England: a range of outcome measures at national and level authority level for children in need ?(show glossary term definition), including children looked after ?(show glossary term definition) by local authorities in England.
https://explore-education-statistics.service.gov.uk/find-statistics/children-looked-after-in-england-including-adoptions
2020 Stability index: information on the stability of children looked after on 31 March 2019 (opens in new tab) published by the Children's Commissioners Office.
Data is collected and published independently by each of the four countries in the UK. Although there are similarities between the data collected by the four nations, there are also differences which may be down to different policies and legislation and differing historical data collections.
Scotland: Statistics on children’s social work (including children looked after) in Scotland and statistics on the outcomes of looked after children in Scotland are available on the Scottish Government website (opens in new tab).
Wales: Statistics on children’s services (including children looked after) in Wales are available on the Welsh Government website (opens in new tab).
Northern Ireland: Statistics on children’s services (including children looked after) in Northern Ireland are available on the Department of Health website (opens in new tab).
If you have a specific enquiry about Stability measures for children looked after in England statistics and data:
If you have a media enquiry:
Telephone: 020 7783 8300
If you have a general enquiry about the Department for Education (DfE) or education:
Phone: 0370 000 2288
Opening times:
Monday to Friday from 9.30am to 5.30pm (closed on bank holidays)
Don't include personal or financial information like your National Insurance number or credit card details.