Learning loss research: Understanding progress in the 2020/21 academic year

Learning loss research: Understanding progress in the 2020/21 academic year

This research used data made available via the Office for National Statistics Secure Research Service, which is being expanded and improved with ADR UK funding.

Authors: Renaissance Learning, Education Policy Institute

Date: January 2021

Research summary

The education system has been severely disrupted by the coronavirus (Covid-19) pandemic. Analysis by the Education Policy Institute (EPI) and Renaissance Learning is helping to quantify the ‘learning loss’ of pupils in England and inform the necessary catch-up provision. This research has found several regional disparities in learning loss, with reduced in-person teaching widening the ‘disadvantage gap’ – the gap in school attainment between disadvantaged pupils and their peers. Up to 66% of the progress to close the disadvantage gap over the past decade is estimated to have been lost.

Using the Office for National Statistics (ONS) Secure Research Service (SRS), researchers analysed online assessment data provided by Renaissance Learning from over 5,000 schools linked with the National Pupil Database (NPD). Matching this data is providing insight into academic progress and is exposing socio-demographic inequalities experienced by pupils. This research has already informed a government funding programme to increase progress in certain schools. It has the potential to further inform education policies and help target interventions to support those who have missed out the most due to the pandemic.

Data used

The National Pupil Database is a collection of de-identified administrative datasets collected by the Department for Education, containing both pupil-level and school-level data of all pupils in English state schools. Only accessible through the Office for National Statistics Secure Research Service, the NPD includes data on pupils’ demographic characteristics, educational attainment, absences and exclusions and children in need or looked after children.

The NPD is an example of powerful administrative data, providing a longitudinal view of nearly the entire pupil population. This potential for in-depth investigations is increased through linkage with other datasets. For this research, the NPD was linked with online assessment data provided by Renaissance Learning, who are used by over 6,500 schools across the UK and Ireland typically for baseline assessments for reading and maths.

Methods used

Using an ‘ex-post direct’ measurement approach - which is a method looking at results after they have occurred to identify and predict future results - the researchers compared the 2020/21 cohort of pupils to previous years to estimate learning ‘lost’ due to the pandemic. The researchers developed regression models to predict pupils’ expected outcomes as though education was not affected by the pandemic. The models were built from previous years’ attainments and a range of contextual factors including data from academic years ending 2019 and 2020. The predictions were compared to actual outcomes, providing the estimated learning loss and identifying disparities in learning loss by region and pupil characteristics.

Research findings

As a result of the pandemic, the average national learning loss by October 2020 was found to be:

  • 3.6 months behind for mathematics
  • 1.8 months behind for reading

amongst pupils in primary schools.

The return to school in the autumn had a positive impact on learning loss and by December 2020, the average learning loss reduced:

  • from 3.6 months to 2.6 months for mathematics, and
  • from 1.8 months to 1.2 months for reading.

The national lockdown and restrictions to in-person learning at the start of 2021 was associated with further learning losses, though there was further recovery by the end of the school year.

Analysis found regional disparities in learning loss throughout the school year. Pupils in parts of the north and the midlands were amongst the worst affected. These regions experienced nearly double the average learning loss, which may reflect differences in learning provision through differing rates of attendance and local lockdowns or school closures.

Pupils from disadvantaged backgrounds were estimated to have the greatest learning loss. The relative learning loss for disadvantaged pupils was estimated to be the equivalent of losing around one third of the progress made over the past decade in closing the attainment gap in primary schools.

Research impact

The research has been cited in numerous discussions on education recovery and catch-up funding within the House of Parliament and House of Lords. It has also been referenced by the Schools Minister, chair of the Education Committee, the Shadow Secretary of State for Education and other prominent MPs.

It was also included in the rationale behind the Department for Education’s summer schools programme in 2021 which provided additional funding for certain schools to support incoming year 7 pupils to increase progress. The Education Endowment Foundations included the research in its ‘Best evidence on impact of Covid-19 on pupil attainment’.

Research outputs

Publications and reports

Blogs, news posts, and videos

About the ONS Secure Research Service

The ONS Secure Research Service (SRS) is an accredited trusted research environment, using the Five Safes Framework to provide secure access to de-identified, unpublished data. If you would like to discuss writing a future case study with us, please get in touch: srs.dev-impact@ons.gov.uk.  

Share this: