Sunday thoughts: Is it time to take a look at Ofsted reform?

I turn this week, slightly reluctantly, to Ofsted.

Not really to talk about Ruth Perry. It’s an utterly tragic case and I can’t imagine I have anything useful to say about it beyond the sheer human heartbreak for all involved.

There’s a reason that they say hard cases make bad law. Turns out that a petition to abolish Ofsted that’s being associated with her was started this time last year by an organisation that didn’t like Ofsted anyway and still doesn’t. And turns out that people and groups that do like Ofsted, still do. All in all, I suspect we won’t gain much from the debate through linking it to this tragedy, on either side.

But it is a wider debate, and it has been running for most of the 30 years that Ofsted has been in existence. And so while I hate the fact that the current discussion is being framed through human agony, it is worth thinking about what some of the issues are.

I do start, and its worth saying this, from the perspective that Ofsted is a good thing, and is needed in the system. Almost everyone says that, of course. Almost everyone who campaigns for change doesn’t just say abolish Ofsted, or abolish other forms of accountability. But a more honest debate would, I think, allow some people to say that external accountability, if it must exist at all, should come at best second to a system whose primary job should be to create a sense of shared mission of improvement, and — implicitly — that the necessity of accountability comes well below the inherent trust that should be held in professionals to run their services. I don’t hold to that line, but some people do, and we’d have a better debate if it were more honest. Equally, those of us who do believe explicitly in accountability more or less in its current form should acknowledge that there’s a strong possibility that we are wrong to a greater or lesser degree, and that the current approach is doing more harm than good. “What would it take for you to change your mind?”, is often a good question to pose to someone (including me) in issues such as this.

Part of the issue with an Ofsted judgement in its current form, as Sam explained very well the other day, is that it’s so high stakes precisely because it is now the only form of accountability at school level in the system. This is a relatively recent development. There used to be two main forms of accountability for schools — exam grades (at 11 or 16) and Ofsted. If a school fell below a certain threshold of exam passes, there were sanctions (such schools were often known, colloquially, as ‘failing schools’ and ‘coasting schools’). And, as now, an Ofsted grade of Inadequate, or Requires Improvement, also led to a triggering of some intervention.

In 2018 and 2019, Damian Hinds as Secretary of State did something which may in retrospect be incredibly significant for his quite short tenure. Following his speech to NAHT on accountability, he consulted on whether the formal consequence architecture of floor standards and other exam based metrics for triggering intervention should be abolished. He also published a short (and pretty good) statement on principles of clear and simple accountability.

At the time, this proposed change made a huge amount of sense. One of the major issues was that school leaders didn’t know whether the Regional Schools Commissioners (as were) or Ofsted were the ones making judgments on school performance. (It perhaps didn’t help that the individuals at the top of those two organisations also didn’t agree on who it was that was making judgements). The double system was creating a lot of workload. And it was duplicative, because a lot of Ofsted reports tracked school performance anyway — so that schools that had better exam results, or were in richer areas, were much more likely to get Outstanding judgements.

In May 2019, the consultation response showed, perhaps unsurprisingly, a lot of support for this premise. So the changes were made — floor targets and coasting school metrics were abandoned from Autumn 2019, and a new Ofsted framework also came in at the same time that deliberately focussed more on a rounded judgement as to how well schools were performing, including a focus on the curriculum. Exam factories, it was briefed out, would no longer automatically be garlanded by the inspectorate.

And it has indeed done that. There is now a ‘broken link’ between exam scores (whether measured in absolute terms or P8) and Ofsted scores.

Problem solved?

Or not.

I recall thinking at the time that the 2018 and 2019 changes made a lot of sense, for the reasons mentioned above. But now, I’m not so sure. There’s too many instances — and Mo Ismail of Star Academies regularly tweets them out — where schools with either low exam results, or negative Progress 8 scores, are getting Good or better judgements. I’m sorry, but I just don’t accept this. This is partly an issue of language, and upper case versus lower case adjectives, but a school that is removing value, regardless of its circumstances, simply cannot be a good school. And — with some caveats — a school with incredibly low exam results is highly unlikely to be a good school either. As the Schools Week piece shows, 10% of all Ofsted inspections in 2022/23 to date with a Good rating at time of writing had Progress 8 of -0.5 or lower. I want to have a balanced approach to judgements, and I don’t want it to simply be the case that high ranking schools and rich schools automatically get the top grades, but the current outcome isn’t right either.

And the second problem is that because we don’t have two sets of balancing judgements, everything comes down to Ofsted. It is now a zero failure system, and there’s no redundancy or fail safe in it. And given what we know, because humans are humans, about questions of reliability and validity of inspections, this strikes me as overly concentrating risk in a metric that might not be able to bear the weight of it.

We don’t solve this, incidentally, by renaming Ofsted as a school report card, or changing grade descriptors. When it is the single point of accountability, however framed, then it will remain high stakes. There’s also a risk in England that if we did abolish all grades from Ofsted, and didn’t reintroduce some form of accountability measurement based on exams, that there wouldn’t be any clear point of intervention from a government level to a school that was underperforming.

What could we do about it? I’m not sure a simple reintroduced de minimis floor level works either. Depending on whether you set the bar, you either have a situation where the state has to intervene in more schools than it can legitimately tackle, or a bar that is set low to avoid that with the permanent worry that some schools which are not good are nevertheless above the minimum expected. Plus an annually refreshed floor standard simply leads to ongoing pressure about results, even more so than now. And thirdly, in a high stakes system, we risk also putting too much pressure on the validity of exam grades in a binary above / below floor target based on data alone.

But in principle, some form of dual key system — such that Ofsted reports and exam data (using perhaps raw data and Progress 8, or a three year rolling average of exam scores), would seem right to reconsider. You could, for example, have a pretty low floor standard based on raw grades or very negative Progress 8, plus a continuation of intervention based on Ofsted Inadequate or double RI — and schools who are in both of these categories are the ones intervened on. Ofsted would also be used, potentially again with Progress 8, to identify schools that are what used to be called coasting. There’s bound to be lots more options as well.

The time is right to think again about this. We are about to have, in under a year, a new HMCI. We may well have a new framework accompanying them. We may have a new government in 18 months (and we’re bound to have a new Secretary of State, regardless). If we can change all these things together, as we did in 2019, then this will reduce sector confusion. So now is the time to consult on what changes might need to be made. We could start from the Hinds piece on principles of a good accountability system, and consider how best we can address, on behalf of children and taxpayers, poor schools, while removing some pressure of judgements from those who are performing well under a myriad of different circumstances.

Because in short, a set of changes in 2018 and 2019 that was aimed at simplicity and clarity, has now led to an acute agglomeration of all pressure and consequences and sanctions to one point within it. A regulatory system can’t bear, I don’t think, that level of single concentration — especially when questions of human reliability and validity are baked within it.

We need to somehow separate out the measures of accountability once again. Yes, this may at times lead to tension between the measures. And yes, it may add further issues for leaders to consider. But the Enterprise of the current accountability system, with more and more piled on to one inspectorate, simply cannae take it any more, captain.