Policy paper

Proposed changes - CIN and CLA outcomes statistics

Updated 11 March 2021

This was published under the 2019 to 2022 Johnson Conservative government

Applies to England

What will happen next

The Department for Education (DfE) asked users to submit feedback on these proposals between 24 September 2020 and 29 October 2020.

The feedback is now being analysed and will be published at a later date at Children in need and looked after children statistics: proposed changes.

What we’re asking you to feedback on

We are requesting user feedback on proposals to change the annual children in children in need (CIN) and children looked after by local authorities (CLA) outcomes statistics based on both the commitment made at the end of the review of children in need to improve the annual CIN and CLA outcomes publications and on previous user feedback.

This includes recommendations to improve the consistency of definitions and the range of children reported on. For example, we currently do not report outcomes on the whole CIN cohort and we only report on children looked after for at least 12 months at 31 March – the impact on outcomes for those looked after for shorter durations is not currently reported on.

This request for user feedback on proposed changes provides DfE with the opportunity to check with users whether our proposals meet these aims. As part of the review of these National Statistics, an ad-hoc statistics release: outcomes of children in need including looked after children has been published. This ad-hoc publication includes changes to current cohorts, as well as additional cohorts. As part of this process, we are also reviewing the outcomes we publish and the outcomes of previously looked after children official statistics.

We propose producing a single publication, bringing together the CIN and CLA outcomes statistics into one place. We plan to publish a 4-year time series for each outcome at national level where possible and also provide local authority level data.

In addition to our proposed changes to the content of the future publication, we are also asking you to give your feedback on using Explore Education Statistics (EES), the new way in which we publish DfE statistics. The ad-hoc release was published using EES and will be used for future CIN, CLA and outcomes statistics.

Changes to cohorts

Background

Currently we publish statistics on:

• CIN at 31 March excluding looked after children but including looked after children that are on a child protection plan (CPP)
• children on a child protection plan at 31 March (for absence only)
• children looked after continuously for at least 12 months at 31 March
• children who were previously looked after and left care through adoption, special guardianship order or child arrangements order (for key stage 2 and key stage 4 only)
• for attainment measures we publish a non-looked after children cohort for comparison purposes

We’ve previously received feedback that the cohorts are not clear and there is also a group of children looked after that are not captured in the above cohorts (those children looked after for less than 12 months at 31 March that are not on a child protection plan).

Proposal and rationale

We propose to include the cohorts from the ad-hoc publication in the new publication (CINP is ‘children in need and other plans’ defined below):

• CIN at 31 March (all CIN at 31 March without any cohorts removed)
• CINP at 31 March (all CIN at 31 March excluding children looked after and children on a child protection plan)
• CPP at 31 March (all CPP at 31 March excluding children looked after)
• CLA at 31 March
• CLA for at least 12 months at 31 March
• CLA for less than 12 months at 31 March

---

• CIN at any point in the year (all CIN at any point in the year without any cohorts removed)
• CINP at any point in the year (all CIN at any point in the year excluding children looked after and children on a child protection plan)
• CPP at any point in the year (all CPP at any point in the year excluding children looked after)
• CLA at any point in the year

---

• Ever CIN – last 6 years
• Ever CLA – last 6 years

We propose publishing the majority of these cohorts as underlying data at national level but will focus on specific cohorts in the commentary and at local authority level (see proposal on changes to breakdowns). We are interested in your views on publishing these cohorts at national level.

In addition, we are proposing a new methodology for matching data based on that used in the ad-hoc publication, which means we would not be including a non-looked after children cohort for attainment but instead will make comparisons with all children using the relevant national level figures released in DfE’s statistics publications. These publications can have different coverage but historic comparisons have shown that the differences in results between the non-looked after children group and all children are small.

Information on children who were previously looked after and left care through adoption, special guardianship order or child arrangements order will continue to be published, see proposal on changes to outcomes).

Questions

You may find it helpful to consider these questions when providing feedback:

• do you think that the proposed changes to the definition of the CIN cohort and the inclusion of new cohorts will be an improvement on the existing statistics?
• if we were to reduce the number of cohorts we publish from the above list, which of the cohorts would be of most use to you at national level?
• will the removal of the non-looked after children cohort for attainment at national level cause any issues?

Changes to breakdowns

Background

We currently publish special education needs breakdowns at national level for attainment at key stage 2 and key stage 4. We also publish gender breakdowns for attainment at key stage 1, key stage 2 and key stage 4. For the headline measures, we currently publish local authority breakdowns (these do not include special educational needs or gender breakdowns).

Proposal and rationale

It is important for us to understand which cohorts are most useful and beneficial for our local authority figures. We propose to only publish local authority breakdowns for a number of specific cohorts and we are suggesting the following cohorts:

• CIN at 31 March (all CIN at 31 March without any cohorts removed)
• CINP at 31 March (all CIN at 31 March excluding children looked after and children on a child protection plan)
• CPP at 31 March (all CPP at 31 March excluding children looked after)
• CLA for at least 12 months at 31 March
• CLA for less than 12 months at 31 March

If we publish cohorts at 31 March, this removes duplicates where a child has moved local authority during the year. However, if we publish ‘at any point in the year’ and ‘Ever – last 6 years’ cohorts at local authority level, the same child can appear in more than one local authority. At national level this makes little difference but it may have more of an impact on local authority figures. In addition, the numbers in the ‘CLA for less than 12 months at 31 March’ cohort may be too small to be published at local authority level for all outcomes.

We propose to continue publishing special education needs for key stage 2 and key stage 4 at national level but to remove the gender breakdowns for key stage 1, key stage 2 and key stage 4 in order to improve efficiency - the gender breakdowns are considered secondary analysis.

In addition, we are proposing a new methodology for age filters based on that used in the ad-hoc publication, which means that the same age filters will be used for all cohorts. In the current publications there are some small variations between the cohorts detailed in the footnotes due to historic differences in methodologies.

Questions

You may find it helpful to consider these questions when providing feedback:

• what are your views on the proposed cohorts for local authority level data? Which cohorts would you consider to be the most useful at local authority level?
• if we were to publish ‘at any point in the year’ or ‘Ever – last 6 years’ cohorts at local authority level, would a child appearing in all of the local authorities they have been in during the year be acceptable for outcomes statistics?
• will the removal of gender breakdowns for attainment cause any significant issues? If it does, what would be the impact of removing the gender breakdown?

Changes to outcomes

Background

The following changes are due to coronavirus (COVID-19) and are expected to be temporary:

• there is no key stage 1 data available for 2020
• there is no key stage 2 data available for 2020
• key stage 4 data availability for 2020 will be confirmed in due course

In the ad-hoc publication we have included for the first time data on the early years foundation stage profile (EYFSP). There is no EYFSP data available for 2020 due to coronavirus (COVID-19). Furthermore, the EYFSP assessment is changing, with early adopters taking up the new profile in 2021 ahead of a statutory roll out in 2022. As this process progresses, we’ll review how best to incorporate EYFSP data into future outcomes publications.

Proposal and rationale

To improve the efficiency of the publication by focusing on the most useful statistics, we propose that the following longer term changes are made (key stage 2 and key stage 4 proposals to be introduced when data is next available):

• special education needs – see 2019 national tables:
  • remove primary type of special educational needs (only published for CLA at national level), as this is considered secondary analysis (see table 2b in the 2019 national tables)
• at key stage 4 - see 2019 national tables:
  • remove or reduce the number of EBACC headline measures (only published at national level for CLA but at national and local authority level for CIN), as numbers are too small to publish at local authority level for some of the proposed cohorts and generally other measures are considered a better gauge (see table 5a in the 2019 national tables)
  • remove Progress 8 school type breakdown (only published at national level), as this breakdown does not appear to be widely used (see table 5d in the 2019 national tables)
  • remove Progress 8 length of most recent period of care (only published at national level), as the general result from this breakdown does not appear to change year on year (see Table 5e in the 2019 national tables)
• school type – see ad-hoc statistics release: outcomes of children in need including looked after children:
  • add school type – (see underlying data in the ad-hoc publication)
• include statistics on previously looked after children (official statistics) – see 2019 additional tables in outcomes for children looked after by local authorities (31 March 2019):
  • continue to include key stage 2 and key stage 4 (only published at national level) – see tables A1, A2, B1, B2 and B3 in the 2019 additional tables.
  • add exclusions at national level

We did not include progress measures at key stage 2 in the ad-hoc publication but we do intend to continue publishing progress measures at key stage 2 when data is next available. In addition, we did not include destinations measures in the ad-hoc but we intend to continue publishing this at national level.

Questions

You may find it helpful to consider these questions when providing feedback:

• will there be any significant issues if we remove primary type of SEN statistics?
• will there be any significant issues if we make the proposed changes at key stage 4?
• do you find the data on school type included in the ad-hoc publication useful and would it be beneficial for us to include in future? If so, would it be beneficial to provide this data at national level and local authority level?
• will it be beneficial to add in exclusions data for previously looked after children?

Contact

If you have any questions about this request, email cla.stats@education.gov.uk.