League tables divide opinion. For some they support our core purpose of securing the best outcomes for our pupils, providing in the process the transparency, accountability and feedback that all organisations need to sustain improvement – sunlight is the best disinfectant, and all that. But for their critics, league tables say little about what really matters in schools; not only do they fail to capture the complexity of life in our classrooms, they distort our behaviour and encourage teachers and leaders to make decisions based on what looks best in league tables rather than what’s best for our pupils.
Over the following posts I want to see how league tables have evolved since they were introduced in 1992, before taking a detailed look at Progress 8, which has become the headline figure for secondary schools. We’ll finish by suggesting how schools might respond to the ongoing flux of league tables.
League tables have long been one of two ‘eyes’ of accountability, with the other being Ofsted. With signs that Ofsted want to focus more on the inputs of education than the outputs, it’s all the more important that the gaze of league tables is fixed on the things that matter.
First, a quick point on language. ‘League tables’ is a bit of a misnomer. What we’re really referring to is the DfE’s Compare School Performance service which provides open access to a wide range of measures from any state school in the country. When we search for a particular school we see key measures (see example of Paddington Academy below) and then more specific measures for that school, and we can compare a group of schools, or all schools, by any measure we want.
Part 1: Cat and Mouse
Given that league tables were introduced to provide transparency and to invite public scrutiny, it’s fitting that all league tables published since their inception in 1992 are available online. Naturally I started my search at the school I attended – Cantell School in Southampton – where in 1992, the year before I joined, 32% of pupils gained 5 A*-C grades in their GCSEs (not necessarily including English and Maths). The national average that year was 38.3, and once again this was 5 A*-C grades in any subject – it wasn’t until 2006 that English and Maths had to be included. By the time this headline measure was phased out in 2015, 64.9% of pupils left school with this basic benchmark. Our profession is good at self-flagellation, and a week rarely goes by without another article bemoaning the state of our schools, but things were a lot worse just a generation ago. We can of course debate whether league tables have caused this improvement, or simply revealed it.*
The most striking finding from a year-by-year check of league tables is how much they’ve changed over time. I’ve identified some key changes here (all quotations below are from DFE’s guidance document accompanying each year’s league tables):
- 1992-1996: 4 simple measures captured in league tables %5A-C, %1A-C, %5A-G and %1A-G
- 1997: GNVQs combined with GCSEs for the first time this year on the basis of ‘broad equivalencies’ e.g. intermediate GNVQ equivalent to 4 GCSEs at A*-C.
- 1998: the GCSE/GNVQ average point score per 15 year old is introduced. “This provides a fuller picture of the GCSE and GNVQ achievements of pupils of all abilities. The average point score is calculated by dividing the total GCSE/GNVQ points achieved by all 15 year olds by the number of 15 year olds”.
- Introduction of KS2-KS3 value added measure and KS3 to KS4 value added measure
- Average capped point score based on pupils’ best eight results introduced to deter school from entering pupils for an excessive amount of qualifications
- 2004: Introduction of KS2-KS4 value-added measure.
- Contextual Value Added (CVA) introduced: “CVA takes into account the varying starting points of each pupil’s KS2 test results, and also adjusts for factors which are outside a school’s control (such as gender, mobility and levels of deprivation) that have been observed to impact on pupils results.”
- 5A*-C including English and Maths included for the first time. This had a significant impact on the league table position of some schools, as reported by this BBC article “The effects on some schools have been dramatic. One school went from a score of 82% passing the equivalent of five A*-Cs to just 16% when maths and English were included.”
- 2009: Progress figure for English & Maths included for first time
- 2010: “The percentage of pupils who have met the new English Baccalaureate requirements reported for the first time this year”
- CVA measure scrapped, replaced by VA measure (best 8 with bonus for E&M), including separate VA measure for each EBacc subject
- Figures for high, middle and low attainers introduced for first time
- Headline figures published with and without equivalences
- In-school gaps published for first time, revealing the gap between the GCSE outcomes of each school’s FSM pupils and its non-FSM pupils
- iGCSEs included
- Pupil premium reported (in place of the gap analysis above)
- Destination measures introduced
- Gender breakdown introduced
- Progress gaps now revealed alongside attainment gaps
- ‘Similar schools’ table introduced
- No early entry – only first entry counts for EBacc subjects
- Wolf reforms lead to the “removal of around 3,000 qualifications from performance measures; adjustment of the point scores of non-GSCEs and the restriction on the number of non-GSCE qualifications that count to two per pupil”.
- Last year of 5A*-C with English and Maths
- Early entry policy now counts for all subject areas
- Progress 8 score published for schools who opted in
- 2016: Progress 8 introduced for all schools, so the headline measures are now:
- Progress 8
- Attainment 8
- English and Maths at C+
- EBacc (entering and achieving)
- Staying in education
- iGCSEs no longer count
- 1-9 grades replace A*-G in English and Maths
- ECDL no longer counts
- New measure for EBacc (average points score)
- 1-9 grades replace A*-G in most subjects
This dizzying array of changes (and there are plenty more I didn’t include here) reveals one of the challenges of league tables: they don’t tend to remain stable for long enough to guide school improvement in a meaningful and sustainable way. Given the fact that league tables are a retrospective check on the outcome of 5 years of schooling, you would think that stability would be built into their design so that schools can gradually work towards known metrics. Instead, the frequency of reform has left some schools lurching from one change to another.
The extent of this lurching is not insignificant. In 2016 there were more than 300,000 entries for iGCSE, the final year that they were included in league tables. In 2017 the figure had fallen to 110,000, creating one of the more unusual examples of our divided school system as independent schools retained the iGCSE while state schools took flight. Independent schools, of course, do not feature in league tables.
The rise and fall of the BCS Level 2 European Computer Driving Licence Certificate in IT Application Skills (ECDL) is another example of the influence of league tables on schools’ behaviour. ECDL entries increased from 26,000 in June 2015 to 117,000 the following year, and rose again in 2017 as the ECDL was promoted as an accessible qualification that can be quickly delivered to boost the open bucket of Progress 8. Figures for 2018 are not yet released but between January and March 2018 there were just 2,800 entries for ECDL compared to 37,650 in the same quarter in 2017. Did schools suddenly decide that their pupils no longer needed this certificate in computer skills? Of course not – this qualification fell out of favour the moment it ceased to count in league tables.
As I was searching for the entry figures above I found these questions on web forums for parents and students:
- Mumsnet, March 2016: “Is ECDL worth having? I’m sure it is in its own right. But is it equivalent to taking Physics, History, Art or French GCSE?”
- Student Room, February 2017: “I have asked many people on whether or not ECDL counts as a GCSE and they all give different answers. Does anybody with the ECDL qualification know? Say you get 5 GCSE’s and ECDL, would a job/uni/sixthform count that as 6 GCSEs? Thank you”
“My school calls it a GCSE” is the depressing reply of two respondents to the question posed by a student.
Since the ECDL has been discounted, some school leaders have searched for the next quick win, which no doubt will itself be discounted in time. And thus continues the bizarre dance of cat and mouse that has played out between the DFE and schools since league tables were introduced. Take this example from an IPPR report in 2003:
“The Government’s decision to give intermediate GNVQs an equivalency rating of four A*-C GCSEs has led to a surge of schools taking advantage of what is seen as easy league table success. Thomas Telford School in Shropshire, for example, has embraced GNVQs from its inception. Yet now, all of their pupils take at least one GNVQ and some leave with a total equivalent of nineteen GCSEs contributing in large part to their outstanding league table performance.”
One final example of the responsiveness of schools to changes in performance measures: in 2014 when the government decided that only first entries would count for EBacc subjects, the number of early entries for GCSE Maths fell from 169,000 in 2013 to 31,000 in 2014 – a profound overnight change to the delivery of Maths at thousands of schools.
It’s difficult to defend a system which appears to result in such obviously tactical behaviour by so many schools, and it is pupils, particularly poorer pupils at schools vulnerable to weak league table performance, who are caught up in this not-so-merry dance between schools and the DFE. We see this in the fact that so many independent schools continue to enter pupils for the iGCSE, while state schools abandoned the iGCSE when it ceased to count. Pupils in independent schools therefore benefit from teachers who have taught the same qualification for several years, while state schools gradually familiarise themselves with new specifications. In my experience it is disadvantaged pupils, even the high attainers among them, who are more likely than their more affluent peers to take vocational qualifications such as BTEC Sport or BTEC Business, which again might be because they tend to find themselves in schools desperately seeking any possible league table advantage.
Does it have to be like this? One school in York gained an outstanding Ofsted judgement in 2017 with a report which praised leaders for putting the needs of pupils ‘above all else’. The report continues: “The curriculum reflects leaders’ integrity because it is designed to match pupils’ needs and aspirations regardless of performance table measures. Leaders have made provision for almost all pupils to study a modern foreign language because research tells them that pupils will develop valuable skills for their future.” Sure enough, the league tables show that 83% of this school’s Year 11 pupils took a language GCSE in 2017, with a progress score for language of -0.43 – sharply at odds with the school’s progress score in other subjects (e.g. science 0.48, humanities 0.63).
250 miles south, in another cathedral city, one school gained a good Ofsted judgement in 2018 despite a Progress 8 score of -0.71. The reason, as I understand it, is that the school continued to enter pupils for iGCSE in English as they believed this was in the pupils’ best interests, even though the league table performance of the school would suffer as a result (the P8 score for English here is -2.91).
Perhaps these two schools, which have flourished despite taking decisions that have hampered their league table standing, prove that the bark of the league table is worse than its bite? What would happen if more schools did their own thing rather than dance to the DFE’s tune?
More on this in our next post
*After checking my own school I went on to check the schools that I work with at United Learning. Just look at these improvements between 1992 and now:
Obviously this is just a handful of schools among the 3000 or so secondaries across the country, but let’s be grateful that the woeful outcomes evident in the 1992 column hardly exist anymore.