Main findings: state-funded schools inspections and outcomes as at 31 December 2019
Published 26 March 2020
Applies to England
This release includes:
- schools’ most recent inspections and outcomes as at 31 December 2019
- provisional data for inspections completed between 1 September and 31 December 2019
- revised data for inspections completed between 1 September 2018 and 31 August 2019
Inspections from September 2019 onwards were carried out under the new education inspection framework. This framework puts more focus on assessing whether a school has a well-designed and ambitious curriculum, and whether it has the same high ambition for all pupils. These are the first official statistics reporting on inspections carried out under the new framework.
Introduction
This academic year so far, 77% of schools have been judged good or outstanding for overall effectiveness.
Figure 1: Overall effectiveness of schools inspected this academic year, by phase (includes outcome for section 8 inspections that did not convert)[footnote 1]
Under both the current and previous inspection frameworks, the outcomes for effectiveness of leadership and management were similar to the outcomes for overall effectiveness.
Schools at their most recent inspection
Eighty-six per cent of all schools are good or outstanding. The percentage of schools judged good or outstanding has remained the same since August 2018. In this same period, the percentage of schools judged outstanding has declined from 21% to 19% while the percentage of schools judged good has increased from 65% to 67%.
Figure 2: Overall effectiveness of state-funded schools at their most recent inspection, by year
More primary schools are good or outstanding than secondary schools: 88% compared with 76%. Additionally, 98% of nursery schools, (no change from last year), 91% of special schools, (compared with 92% last year), and 84% of pupil referral units (PRUs), (compared with 83% last year) are good or outstanding.
Inspections between 1 September 2019 and 31 December 2019
This academic year so far, 77% of schools have been judged good or outstanding for overall effectiveness.
To date in this academic year, we have carried out:
- 853 full inspections, including section 8 inspections of good and non-exempt outstanding schools and monitoring visits that became a full inspection
- 773 were inspections of good and non-exempt outstanding schools that did not convert to a full inspection
This year, 77% of schools have been judged good or outstanding for overall effectiveness, a decrease from 80% last year. The percentage judged good has decreased by 1 percentage point compared with last year, and the percentage judged outstanding has decreased by 3 percentage points.
Figure 3: Overall effectiveness of schools inspected in 2018/19 and 2019/20[footnote 1]
The schools inspected this year have a slightly weaker grade profile than those inspected last year, with 75% being previously good or outstanding this year compared with 78% last year. This may be contributing to the slightly lower proportion of schools judged good or outstanding this year compared with last year.
Figure 4: Previous overall effectiveness of schools inspected in 2018/19 and 2019/20
Inspections of schools previously judged to be outstanding
The percentage of exempt schools that have remained outstanding this year has increased slightly to 17%, compared with 16% last year.
Since 1 September 2019, we have inspected 122 previously outstanding primary and secondary schools that are exempt from routine inspection, compared with 392 exempt schools that were inspected last year.
The percentage of exempt schools that have remained outstanding this year has increased slightly to 17%, compared with 16% last year. This may be partly due to changes in the way exempt schools are selected for inspection. This now includes taking into account the length of time since last inspection.
Inspections of schools previously judged to be good
We have carried out 1,036 section 5 and section 8 inspections of previously good schools this year. Of these, 84% remained good, 2% became outstanding, 12% now require improvement and 2% became inadequate.
Of these, 304 had a planned section 5 inspection and 728 had a section 8 inspection, of which 7 (1%) converted to a full inspection. A further 4 schools had a ‘no formal designation’ visit, for instance to investigate a complaint, and this then became a full inspection.
Of the 304 planned full inspections of good schools, 72 were inspected because a previous short inspection had indicated that the school may decline at its next full inspection. Of these, 60% declined to requires improvement or inadequate. A further 68 full inspections were because a previous short inspection had indicated that the school might improve at a full inspection. Of these, 34% improved to outstanding. The remaining 164 schools had full inspections because a risk assessment of the school had identified concerns or because the school had undergone significant change. Of these, 54% declined to requires improvement or inadequate.
Of the 721 section 8 inspections of good schools that did not convert, 78% confirmed that the school remains good and did not require further inspection, similar to 77% last year. A further 32 (4%) confirmed the school remains good but indicated it might improve at a full inspection. These schools will have a full inspection next, usually within 1 to 2 years. Another 124 (17%) also confirmed the school remains good but indicated it might decline at a full inspection. These schools will also have a full inspection next, within the same timescales.
Inspections of schools previously judged to require improvement
This year, 56% of schools inspected improved from requires improvement to good, compared with 64% last year.
We have carried out 329 full inspections this year of schools that were previously judged to require improvement. This year, 56% of schools inspected improved from requires improvement to good, compared with 64% last year. A further 38% remained requires improvement, an increase from 27% last year.
Figure 5: Grade movement for schools previously judged to require improvement, by year
The relationship between overall effectiveness and the other inspection judgements
Under both the current and previous inspection frameworks, the outcomes for effectiveness of leadership and management were similar to the outcomes for overall effectiveness.
When we carry out a full inspection, we make a judgement about the overall effectiveness of a school. Under the common inspection framework prior to September 2019, we also made judgements about the quality of teaching, learning and assessment, outcomes for pupils, personal development, behaviour and welfare and the effectiveness of leadership and management. Since September 2019, we now make judgements about the quality of education, behaviour and attitudes, personal development and the effectiveness of leadership and management.
Under both the current and previous inspection frameworks, the outcomes for effectiveness of leadership and management were similar to the outcomes for overall effectiveness. The other judgements under the current framework are not comparable with the previous framework. However, the outcomes for both behaviour and attitudes and personal development under the current framework are more positive than those for overall effectiveness. This was also the case under the previous inspection framework for the personal development, behaviour and welfare judgement.
Figure 6: Main inspection judgements for schools inspected this academic year (full inspections only)
As well as the judgements already mentioned, where applicable, we also make judgements about any early years and sixth-form provision.
Under the current framework, the percentage judged good or outstanding for early years provision is 77%, compared with 59% for the overall effectiveness of all schools inspected that have early years provision. The gap between these judgements has increased slightly when compared with inspections last year.
Figure 7: Overall effectiveness judgements compared with early years provision judgements for schools inspected this academic year
Under the current framework, the percentage judged good or outstanding for sixth-form provision is 77%, compared with 50% for the overall effectiveness of all schools inspected that have sixth-form provision. The gap between these judgements has widened by 11 percentage points compared with inspections last year. However, this is based on the outcomes of a small number of inspections (98).
Figure 8: Overall effectiveness judgements compared with sixth-form provision judgements for schools inspected this academic year
The outcomes of inspections by level of deprivation
There is very little difference between the percentage of schools that were judged outstanding in the most deprived and the least deprived groups of schools. There is a difference of 18 percentage points between the schools that were judged good in the most deprived group compared with the least deprived group: 66% compared with 84%.
In the least deprived group of schools, 87% are good or outstanding this year, compared with 69% in the most deprived groups of schools.
Figure 9: Overall effectiveness judgements by the income deprivation affecting children index (IDACI) quintile inspected this academic year[footnote 2]
Revisions to previous release
The provisional data in the previous release relates to inspections that took place between 1 April 2019 and 31 August 2019, with the reports published by 30 September 2019. Revised data for inspections in this period is provided in tables 1R and 2R of the Excel file accompanying this release: State-funded schools inspections and outcomes as at 31 December 2019, charts and tables.
This revised data includes an additional 19 full inspections that had not been published by 30 September 2019 but were published by 31 January 2020. These inspections resulted in the following number of additional judgements at each overall effectiveness grade:
- 0 outstanding
- 1 good
- 0 requires improvement
- 18 inadequate
Therefore, the proportion of all schools judged good between 1 September 2018 and 31 August 2019 changed from 54% to 53% and the proportion judged inadequate from 8% to 9%. These are the final figures for the year and therefore slightly different from the figures in HMCI’s Annual Report published in January 2020.
For primary schools, the proportion that require improvement changed from 29% to 28%.
For secondary schools, the proportion judged good changed from 43% to 42%.
For special schools, the proportion judged good changed from 42% to 41% and the proportion judged inadequate from 12% to 14%.
For PRUs, the proportion good changed from 51% to 50%, the proportion that require improvement from 19% to 18% and the proportion judged inadequate from 16% to 18%.
Revisions to data in this publication are published in line with Ofsted’s revisions policy for official statistics.
Notes
The purpose of these official statistics is to disseminate the data on school standards collected through Ofsted’s role as an inspectorate. They provide information about how the judgements of schools have changed over time and vary across different phases of education and different parts of the country.
This official statistics release reports on the outcomes of state-funded school inspections carried out under sections 5 or 8 of the Education Act 2005. These inspections were carried out between 1 September 2019 and 31 December 2019. This release also includes the most recent inspections and outcomes for all schools that have been inspected, as at 31 December 2019.
Throughout this release, the term ‘schools’ is used to cover all local authority maintained schools, state-funded academies and non-maintained special schools in England that Ofsted is required to inspect under section 5 of the Education Act 2005.
Inspections between September 2015 and August 2019 were carried out under Ofsted’s common inspection framework: common inspection framework.
From September 2019, inspections were carried out under the education inspection framework. education inspection framework.
You can find an explanation of the main uses of this data, further contextual information and the arrangements for quality assurance in the methodology report. This report provides information about strengths and limitations of the statistics.
Where percentages are quoted in this report, figures have been rounded and may not add to 100.
Glossary
Academies
Academies are publicly funded independent schools. Academies do not have to follow the national curriculum and can set their own term times. They still have to follow the same rules on admissions, special educational needs and exclusions as other state schools.
Academies get money directly from the government, not the LA. They are run by an academy trust, which employs the staff. Some academies have sponsors such as businesses, universities, other schools, faith groups or voluntary groups. Sponsors are responsible for improving the performance of their schools.
Academies include converter and sponsor-led academies, free schools, university technical colleges and studio schools.
Academy converters
Academy converters are schools that have opted to become academies. Most of these were good or outstanding LA-maintained schools before they became an academy.
Academy converters retain their latest inspection grade, even if the most recent inspection was of the predecessor school.
Sponsor-led academies
Sponsor-led academies are academies that have sponsors such as businesses, universities, other schools, faith groups or voluntary groups. Sponsors are responsible for improving the performance of their schools.
Free schools
Free schools are funded by the government and are not run by the LA. They are set up on a not-for-profit basis by charities, universities, and community and faith groups, among others.
Free schools are ‘all-ability’ schools, so cannot use academic selection processes like a grammar school. They can set their own pay and conditions for staff and change the length of school terms and the school day. Free schools do not have to follow the national curriculum.
LA-maintained schools
Maintained schools are funded by the government and run by the LA. They must follow the national curriculum.
Section 5 and section 8
Inspections are carried out under sections 5 and 8 of the Education Act 2005.
Ofsted is required to inspect at prescribed intervals all schools to which section 5 applies. The regulations set the interval for section 5 inspections ‘within 5 school years from the end of the school year in which the last inspection took place’. The exceptions to this requirement are schools that are exempt from section 5 inspection (known as ‘exempt schools’).
Section 8 enables Her Majesty’s Chief Inspector (HMCI) to carry out inspections for a range of purposes, including monitoring visits by HMI to schools that are in a category of concern following a section 5 inspection. HMI may also visit schools to aid HMCI in keeping the Secretary of State informed or to contribute to reports on, for example, the teaching in a curriculum subject or a particular aspect of the work of schools. Section 8 is also used to enable HMCI, if she has concerns, to carry out an inspection of those outstanding schools that are exempt from routine inspection under section 5.
Inspections carried out under section 8 include:
- section 8 inspections of schools judged to be good at their most recent section 5 inspection and those outstanding schools that are not exempt from section 5
- monitoring inspections of schools judged as requires improvement, having serious weaknesses or requiring special measures
- any inspection carried out in other circumstances if the inspection has no specific designation
- unannounced behaviour inspections
Section 8 inspections of schools judged to be good at their most recent section 5 inspection and those outstanding schools that are not exempt from section 5
These section 8 inspections are usually either a 1- or 2-day inspection and take place approximately every 4 years. The section 8 inspection determines whether the school continues to provide the same good or outstanding standard of education for pupils. The inspection does not result in individual graded judgements nor does it change the overall effectiveness grade of the school. If there are serious concerns, the section 8 inspection is converted to a full inspection, when inspectors will make the full set of graded judgements required under section 5.
From November 2017, some previously good schools are subject to a full section 5 inspection instead of a section 8 inspection if our risk assessment tells us that a section 8 inspection would be highly likely to convert. For example, this applies if a school underwent significant change, such as changing its age range, or if the quality of provision may have deteriorated significantly.
From January 2018, section 8 inspections are only converted to a full inspection if there are serious concerns. If a section 8 inspection does not convert, there are additional outcomes if the school remains either good or outstanding, but the next inspection will be a section 5 inspection.
Contacts
If you have any comments or feedback on this publication, please contact Louise Butler on 03000 131 457 or the schools data and analysis team if you are a member of the public. Send press enquiries to: [email protected].
Acknowledgements
Thanks to the following for their contribution to this statistical release: Clair Simpson, Matthew Spencer, James Jordan and Kim Chamberlain.
How to search
Click on Ctrl + F on a Windows laptop or Command + F on a Mac
This will open a search box in the top right-hand corner of the page. Type the word you are looking for in the search bar and press enter.
Your browser will highlight the word, usually in yellow, wherever it appears on the page. Press enter to move to the next place it appears.
How to print a copy of the guidance
Click on Ctrl + P on a Windows laptop or Command + P on a Mac
You can also choose to save as a PDF.
-
Includes outcomes of section 8 inspections that did not convert to a full inspection. ↩ ↩2
-
Based on the indices of multiple deprivation (IDACI) 2015. Each school is assigned a score based on the home postcodes of the pupils who attended the school at the time of the January 2019 school census. The schools are then placed into one of 5 deprivation bands. ↩