School census expansion 2023: young carers
Blog School census / January 19, 2023
Summary of the change
From today, January 19 2023, the DfE will add a “young carer” label to the termly School Census pupil-level data collection.
Section 17ZA of the Children Act 1989 says a young carer is a person aged under 18 who provides or intends to provide care for another person (which isn’t to a contract or voluntary work). The labels will default to not declared (which itself could be misleading compared with ‘not a young carer’) and enable a record to be edited by the school to indicate a child identified as a young carer by the parent/ guardian, or by the school, but not self-declared.
While the change has been warmly welcomed for a range of reasons by young carers’ advocacy and support organisations, we have a number of practical concerns with how it will be done.
The incomplete scope of the settings in the school census
The geographical scope of this data item is the same as the scope of rest of the school census and applies to state funded and non-maintained special schools in England.
- maintained nursery primary
- middle-deemed primary
- middle-deemed secondary
- secondary
- all-through
- special schools (including non-maintained special schools)
- pupil referral units / alternative provision (PRU / AP) (*compulsory new change in 2023)
- academies (including free schools, university technical colleges (UTCs) and studio schools)
- city technology colleges (CTCs)
Considerations and concerns
1. Accuracy
The new value is being seen as a way to identify at national level how many young carers are in England’s education system. It won’t do that for both coverage and completion rate reasons, and there is a risk that data analysis and the media may accept an undercount and a misleading figure once it is collected which will be worse than today because “the data” will be used as-is, rather than challenged.
-
- Not all state educated children attend mainstream education in primary or secondary schools, or the other settings where the School Census collects pupil level data (as above). Changes in the 2023 census appear to include the placements of the children attending Alternative Provision (AP) settings for the first time, previously only counted in the AP Census, but others such as Further Education appear to remain out of scope.**
- Clarification is needed which groups of children may not be counted using the new young carer label and when in the course of the year i.e previously those in Alternative Provision would have been omitted or double counted in both the AP and the School Census but the parallel change on AP placement may cover this.
- Analysis of the risk that the count will be wrong and estimations of by how many in which groups should be done to compensate for an undercount and a misleading figure once it is collected, and plans put in place to rectify any gaps and to account for it in communications or any correlations at the time of publication.
2. Completion rate
It appears there is no incentive for schools who do not already know who the young carers are in their community to identify them; simply to add more data into the census collection. Even those schools who do already know them and may or may not support them may see little value in additional national data curation without any attached benefit for the child or the school setting.
3. Consensual data processing (not to be confused with the data protection law ‘consent’ basis)
The codes suggesting identification may be made by the school or the parent/guardian but not the child themselves. The option (S) might also suggest that schools could add this onto pupil’s record without telling schools that they have an obligation to inform families (a) the label will be put on a child’s record and (b) how it will be given away to commercial companies among other third parties and (c) of their rights.
This must be changed and all settings that may assign the code on a child’s records should know that they must inform the family and child; schools need urgent guidance on what happens if families disagree or challenge the label, and what can or cannot be recorded as a result. It is likely that the data could only lawfully be shared with parental consent.
Children should also be empowered to be able to self-declare and have due weight given to their voice and rights, in accordance with the principle of UNCRC Article 12.
Impact Assessment of the data processing must assess whether there is not only (a) an unnecessary and disproportionate interference with the students’ right to privacy, enshrined in Article 16 of the UN Convention on the Rights of the Child (UNCRC) and Article 8 of the European Convention on Human Rights (ECHR); but (b) that of the parent since inferences are made about them (potentially health inferences which are special category data) though the application of the “young carer” label.
[additional comment Jan 31, 2023 after receipt of information from the DfE] The Data Protection Impact Assessment wrongly states the label is self-declared (para 16, page 16). The RFC states that the data fields expressly omit this, and instead it provides a field to be ‘identified by the school’. This, where similar guidance was given in the past led to ethnicity being assigned by schools and parents and children (a) not told and (b) having it imposed. The DPIA also fails to assess the shared-aspect of this data about the child creating inference about the parent/guardian.]
About the National Pupil Database
Data collected in the school census is added to the The National Pupil Database (“NPD”). It is an administrative ‘melting pot’. A single dataset created by joining up different data collections controlled by the Department for Education (“DfE”). The NPD contains data (child-level and school-level) on all pupils in state schools in England and after collection, is retained indefinitely.
The National Pupil Database is “one of the richest data resources about education in the world” according to the Data Protection Impact Assessment carried out in 2019, at that time holding over 21 million individuals’ named records, increasing by around 1 million each year.
In October 2020 the ICO published the high level Executive Summary of their findings from a compulsory audit of the DfE stating that of 139 findings over 60% were urgent or high priority. To date there has been no publication what exactly these are, what has changed and what remains outstanding to bring the Department practice into compliance with the law. In November 2022, the UK Data Protection Authority, the ICO, called data practices at the Department “woeful.”
Next steps
We believe that there should be wider discussion on this change and include:
- Data accuracy: identify and include the full range of where children should be identified in datasets;
- Data quality: identify barriers to process and needed actions in educational settings staff capability, change and training, because while the fields may be in the dataset it does not mean that the data entered will be any better than it is today;
- Data completeness: in particular, consider Further Education, improve guidance on who to count and double counting in AP/PRU and (even if small numbers) a potential for gaps at borders where children are schooled and live in different Local Authorities;
- Data rights
- Stop until this has been done lawfully and safely and in full cooperation with families;
- Fair processing: Clarity in guidance that families must always be asked / informed by default must become realised, this is a legal obligation since parents’ own and potentially sensitive health data are infered by the use of a young carer label;
- Respect for and publishing a route for Subject Access Requests and corrections where mistakes are made on an historic or current record;
- Identify the termly mid-census year process by which any young carer whose status changes (to no longer be a young carer or becomes one) is amended or the data risks being inaccurate (and unlawful)
- Consensual data processing; before any new release of national pupil data at pupil level the young carer label should be excluded to protect confidentiality and the individual from potential stigma for life not knowing “who knows” and unable to restrict its use by choice eg entering Higher Education;
- Protection from misuse at national regional and local levels, from non-educational purposes such as immigration enforcement, police use, identification of fraud and error e.g at the DWP, and an opt-out of all indirect and commercial re-uses of the pupil records.
** We are seeking clarification of this scope (especially regards Further Education) at the time of writing. (Jan. 19, 2022 — confirmation is in the FOI response of Jan. 31, 2023)
Our legal team is in correspondence with the Department for Education and next steps in our action are imminent. The UK government approach to public administrative data handling and the new ICO approach to its regulatory enforcement, fail to meet the test of their data handling by public authorities: there should be no surprises. There is no effective remedy at scale. And the upcoming changes to UK data protection law, will further tip the balance of power between the people whose personal lives these data are about, even more strongly in favour of government and commercialisation.