Our last post took a deep dive into Ofsted data, looking not only at how inspection types and ratings have changed over the years, but also at the ways in which various topics have waxed and waned in Ofsted's written reports. One thing evident from that analysis is that some topics come up much more often for primary schools than for secondary schools, and vice versa. It also suggests a further question: how do these topics vary between schools with high and low Ofsted ratings? This post explores both issues.
The underlying data used here is that the same as that for our previous analysis: inspection outcomes and reports published between January 2006 and early January 2023 (which in effect means inspections conducted up to around November 2022, or even very early December 2022, though some earlier inspections are likely to be missing, especially where their publication was delayed by challenge or revision). This analysis includes only mainstream state schools and only inspections that resulted in an overall rating. Because this analysis involves looking separately at primary and secondary schools, the small number of all-through schools have been omitted. A total of almost 60,000 reports were analysed, of which just under 2,800 were from inspections conducted since September 2021 (ie, under the current inspection framework).
Figure 1 shows the relative frequencies of terms between primary and secondary school inspection reports. A score of 100 means that the term appeared only in primary school reports, while a score of -100 means that it appeared only in secondary school reports. A score of zero means that it appeared with equal frequency across both phases.
Note that these scores don't say anything about the absolute frequency of terms across all schools. For example, 'punctuation' and 'reading' show similar degrees of skew towards primary schools, but over the years the latter has tended to be mentioned much more often in Ofsted reports than the former (see our previous analysis for details). To put it another way, the fact that a school is primary doesn't necessarily mean that these topics are likely to come up, but if they do come up then the school is very likely to be primary. This perspective becomes more interesting when we look at school ratings, as we will below.
Across the whole period from 2006 to 2022, the topics that skewed most towards primary schools were somewhat predictable: 'Early Years', 'phonics and so on. Perhaps less obvious was the tendency to talk a bit more about 'headteachers' and 'parents'.
As shown in our previous analysis, the topics mentioned by Ofsted have changed over time. Looking at the period over which the current Ofsted inspection framework has applied, from September 2021 onwards, provides a smaller sample size (making statistical anomalies more likely), but also gives a more focused view on current Ofsted practice. Among primary schools, the big risers are 'mathematics', 'computing', 'British values' and 'governors'. Again, this doesn't mean that these topics became more frequent overall (though some of them did), it means that their frequencies in primary school reports went up relative to those in secondary school reports.
Figure 1: Relative frequencies of terms in Ofsted reports – primary versus secondary schools
Notes:'Discipline' and 'off-rolling' omitted because they were too infrequently mentioned for meaningful analysis. Where appropriate, topics were identified by matching selected synonyms (eg, 'maths' and 'numeracy' as well as 'mathematics').
Looking again at the whole period from 2006 to 2022, some of the topics that skewed towards secondary schools were unsurprising: 'EBacc', 'destinations', 'sixth form' and 'careers'. Others seem to speak to the particular challenges of educating teenagers: 'absences', 'attendance', and 'exclusions'. It's also interesting to see the secondary skew of 'COVID-19' mentions. Terms that have become more secondary-oriented since September 2021 include 'behaviour', 'spiritual, moral and cultural development' and 'personal development'.
The good, the bad and the topic frequency analysis
Another interesting way to explore topic frequencies is to compare them against Ofsted's ratings. In other words, which topics does Ofsted tend to mention in reports about high-performing schools and how do these compare with the things they talk about in reports about underperforming schools? Given the differences between primary and secondary schools described above, it makes sense to look at each phase separately.
Figure 2 shows the relative frequencies of topics for 'Outstanding' or 'Good' primary schools (positive values) and 'Requires Improvement'1 or 'Inadequate' primary schools (negative values). Note that the horizontal axis here runs only from -50 to 50, not all the way from -100 to 100 as in Figure 1. The metrics are the same, but the values are smaller because the differences between better and worse schools within any given phase turn out to be less than the differences between schools in different phases. This probably shouldn't come as a surprise.
From 2006 to 2022, schools deemed good were more likely to see mentions of terms like 'art', 'culture', 'languages' and (slightly bizarrely) 'COVID-19'. Those deemed poor were more likely to have references to 'absences', 'exclusions', 'assessment' and 'teaching'. Since September 2021, the differences have grown starker (ie, the numbers are bigger), though this may be in part due to the smaller sample size. Among the biggest risers for schools deemed good were 'attitude', 'English as an Additional Language' (EAL) and 'science'. Terms that have cropped up more often than before for underperforming schools included 'British values', 'English and maths' and 'curriculum'.
Figure 2: Relative frequencies of topics in primary school Ofsted reports – Outstanding/Good versus RI/Inadequate
Notes:'Destinations', 'discipline', 'EBacc' and 'off-rolling' omitted because they were too infrequently mentioned for meaningful analysis. Where appropriate, topics were identified by matching selected synonyms (eg, 'maths' and 'numeracy' as well as 'mathematics').
Figure 3 shows a similar analysis for secondary schools. Since 2006, reports for schools deemed good tended to mention 'wellbeing', 'culture', 'EBacc' and (there it is again) 'COVID-19'. Those for schools deemed underperforming tended to mention topics such as 'absences', 'Alternative Provision', 'Pupil Premium', 'disadvantage', 'behaviour' and 'exclusions'.
Under the current inspection framework, the differences once again got bigger. The highest climbers for good schools included 'destinations', 'art', 'technology', 'English and maths' and 'science'. Terms that grew in relative frequency among schools deemed poor included 'British values', 'computing' and 'English as an Additional Language' (EAL).
Figure 3: Relative frequencies of topics in secondary school Ofsted reports – Outstanding/Good versus RI/Inadequate
Notes:'Discipline', 'Early Years', 'off-rolling', 'phonics' and 'punctuation' omitted because they were too infrequently mentioned for meaningful analysis. Where appropriate, topics were identified by matching selected synonyms (eg, 'maths' and 'numeracy' as well as 'mathematics').
Putting in a good word
Are these results of consequence or merely a curiosity? A bit of both, perhaps. We would urge against taking them too seriously, but at the same time they tell us something real about Ofsted's past and present preoccupations. On a practical level, if your inspectors are taking an overt interest in school culture, pupil wellbeing or art – or, more recently, in science and technology – then, at least statistically speaking, you would appear to be in good shape. On the other hand, if they're unduly concerned about absences, exclusions or British values2 then you might need to prepare that most British of all attributes, a stiff upper lip.
The 'Requires Improvement' group also includes schools assigned the now-defunct 'Satisfactory' rating.
'British values' were defined in 2014 by the Department for Education as "democracy, the rule of law, individual liberty, and mutual respect and tolerance of those with different faiths and beliefs". Outside Britain these are usually known as universal liberal values.
This is the SchoolDash blog, where we write about some of our projects and other things that spark our interest.
Blue links take you to other parts of the page, or to other web pages. Red links control the figures to show you what we're talking about.