How Ofsted misrepresents the able pupils issuePosted: June 14, 2013
Not long ago Michael Wilshaw joined in the assault on education academics and researchers telling them to “get out of their ivory towers”. It’s ironic therefore that Ofsted has now produced a report (“The most able students”) that is quite embarrassingly shoddy.
It begins by illustrating again that Ofsted is to all intents and purposes innumerate. It’s initial complaint is that less than half of pupils who get a Level 5 at age 11 go on to get A* or A in English and maths. Put on one side for a moment all that we know about the unreliability of Key Stage 2 tests and consider:
· Who said that all level 5’s should get A* or A? Level 5 covers a third of pupils and so a wide range of performance at 11. Is it surprising that they don’t all get the same GCSE result? Would we want them to? According to Ofsted’s own dashboard a Level 5 pupil makes the expected progress if he or she gets a GCSE B.
· Would GCSE be going its job of differentiating between the highest achievers and the not quite so high achievers (but still pretty good) if it just gave 25% or 30% of pupils a Grade A. Mr Gove actually wants finer distinctions amongst the highest GCSE performers.
· Has Ofsted not heard of “comparable outcomes?”. Ofqual has in effect taken us to a norm referenced system so more pupils can’t get a Grade A no matter how they perform.
In short there is no basis for concluding that pupils are under-achieving because they don’t all get A/A*. The system is designed to make sure they don’t.
The report then goes on to assert that selective schools do better in converting Level 5’s to A/A* than comprehensives. It does this with no attempt to analyse whether the population in selective schools is typical of all Level 5 pupils. In many areas it clearly isn’t and as Henry Stewart shows, 5a or 5b students are hugely much more likely to get an A that 5c’s. And these are likely in many cases to be the great bulk of pupils in grammar schools.[i]
The rest of the report comprises an extremely simplistic analysis of achievement by gender and ethnicity. The latter crudely compares white British and every category of non-white British lumped in together – not something likely to provide any very useful insights. Then there is a survey of the international evidence which takes a whole 17 lines. There is then a report on visits to 41 schools.
There is one section which begins to address some of the issues. It identifies the difference in performance between pupils on free meals and those not – not exactly an original finding but worthwhile nonetheless. But there is no attempt to discover whether there are places where that gap has been closed, and if so how. And of course there are, particularly in inner London comprehensives. But that wouldn’t fit the narrative.
There is also an interesting little section on why students don’t apply to Oxbridge or Russell Group – family, money, a fear of not fitting in (reinforced in one anecdote by actually meeting some Oxbridge students! But this goes nowhere, except to an exhortation to schools to do better in changing pupils’ minds.
One strong message is to reinforce setting. The report asserts that “It was evident in some of the schools visited that school leaders had responded to recent research findings about mixed ability teaching”. What research is not stated but it presumably didn’t include the comprehensive studies from the National Education Policy Center at the University of Colorado, which, building on a meta-analysis, which incorporated 500 studies and 14 previous meta-analyses , concluded that: “For several decades, researchers have documented the effects of tracking students into segregated classrooms according to perceived ability or achievement. Whether known as tracking, sorting, streaming, or ability grouping, an expansive body of literature conclusively shows tracking has been harmful, inequitable, and an unsupportable practice… and that tracking has minimal effects on learning outcomes and profound negative equity effects”.[ii]
The description of school visits is for the most part unexceptionable and tells us, surprise surprise, that some schools do this kind of thing better than others. But whether there are any patterns that would help us understand what is going on is left unsaid. There is no attempt to see whether there are issues of geography, whether size of school and sixth form makes a difference or whether schools under pressure to meet floor targets at A* to C behave differently. And incidentally no attempt to consider whether a sixth form college system gives any different results and whether the proliferation of new sixth forms recently has had any effect.
In short there are massive issues that are simply ignored or simplified beyond belief. The initial assertion is that “The most able students in non-selective secondary schools are not achieving as well as they should.” This is invalid because the statistical basis for it is simply wrong. The attempt to argue that grammar and independent schools do it better is equally statistically invalid and smacks more of an ideological agenda than genuine research.
What is never considered is that we have one of the most socially segregated school systems in the OECD. While that continues to be the case we are unlikely to see fairer outcomes.
Finally, we do have to ask ourselves whether this is the biggest problem we face. If Gove and Wilshaw have their way, the whole curriculum and testing regime will be geared to the needs of perhaps 20% of pupils. It needs to be asserted loud and clear that you are not a failure in life if you don’t go to the Russell Group and schools need to be preparing pupils for every kind of possible future.
[ii] Reported in the Education Journal 31st May 2013.