Learning Lens was built first for the system I knew best — HGIOS4, the self-evaluation framework that every school in Scotland uses to understand itself. The quality indicators, the evaluative language, the six-point scale from Unsatisfactory to Excellent — these were the structures familiar from years in Scottish schools. Ofsted was not the starting point; Scottish self-evaluation was, because that was the problem the product needed to solve first.
Then schools in England started asking whether it would work for them.
The answer was not straightforward, and working through it taught me more about both frameworks than years of using either one alone. When you have to encode an evaluation system into software — when you have to define, precisely, what each quality indicator means in terms of observable evidence — you are forced to confront the assumptions that sit beneath the language. Assumptions that are invisible when you are working within a single system, because the system is the water you swim in. It was only when I had to build for both simultaneously that the differences — and the surprising similarities — became clear.
This is a post about what I learned. It is also, as it happens, a post that arrives at a moment when both systems are being reformed at the same time, which has not happened before and which makes the comparison more than academic.
What is happening right now
In Scotland, His Majesty's Inspectorate of Education has formally separated from Education Scotland under the Education (Scotland) Act 2025. This is a structural change that has been debated for years — the argument being that the body responsible for supporting school improvement should not also be the body responsible for inspecting it. The separation is now law. HMIE is developing a new inspection framework for schools, with a draft expected in 2026. A public consultation — "School Inspections are Changing: Shape What's Next" — closed in November 2025 and its findings will inform the new framework. HGIOS4, the self-evaluation toolkit that has shaped how Scottish schools think about quality since 2015, may be significantly revised or replaced.
In England, Ofsted's new inspection framework went live in November 2025 — the most significant reform since the Education Inspection Framework was introduced in 2019. Single-word overall grades have been abolished. They have been replaced by report cards with a five-point scale — Urgent Improvement, Needs Improvement, Expected Standard, Strong Standard, Exceptional — applied across six evaluation areas. Inclusion is now a standalone judgement for the first time. The "best fit" model, which allowed inspectors to average across descriptors when making a judgement, has been replaced by "secure fit" — meaning every descriptor within a grade must be met for that grade to be awarded. The Ruth Perry tragedy, and the public reckoning it prompted about the psychological impact of high-stakes single-word judgements on school leaders, was the explicit catalyst for these reforms.
Both systems are, simultaneously, asking the same question: how should a country evaluate the quality of its schools? They are arriving at different answers, built on different assumptions, shaped by different histories. But the underlying problem — that the evidence schools produce about their own quality is not as good as it needs to be — is shared.
The philosophical divide
The most fundamental difference between the Scottish and English systems is not the quality indicators, the grading scales, or the inspection methodology. It is the starting assumption about whose job evaluation is.
HGIOS4 begins with self-evaluation. The framework is designed, explicitly, as a tool for schools to use on themselves. The quality indicators — 1.1 (Self-evaluation for self-improvement), 1.3 (Leadership of change), 2.3 (Learning, teaching and assessment), 3.1 (Ensuring wellbeing, equality and inclusion), 3.2 (Raising attainment and achievement), among others — are framed as questions a school asks about itself. How good is our leadership of change? How good is our learning, teaching and assessment? The school produces the evidence. The school makes the evaluative judgement. The inspector validates that judgement — or, where the evidence does not support it, challenges it. But the starting point is the school's own account of itself.
Ofsted begins with inspection. The framework is designed as a tool for inspectors to use on schools. The six evaluation areas — curriculum and teaching, achievement, inclusion, attendance and behaviour, personal development and wellbeing, leadership and governance — are framed as things the inspector assesses. The inspector gathers evidence, applies the grade descriptors, and makes the judgement. The school provides information and context, but the evaluative authority rests with the inspection team.
Both have defensible rationales. The Scottish model trusts schools to know themselves and positions the inspector as a critical friend who tests the quality of that self-knowledge. The English model recognises that self-evaluation can be self-serving and positions the inspector as an independent guarantor of standards. Each model compensates for a risk the other accepts. Scotland's model risks schools telling themselves a more flattering story than the evidence supports. England's model risks inspectors making high-stakes judgements on the basis of a two-day visit that may not capture the school's full picture. Both risks are real.
What building for both taught me is that the practical consequence of this philosophical difference is significant. In the Scottish system, the school needs evidence that it can present in its own voice, using the evaluative language of the framework, structured as a self-evaluation narrative. The evidence must be rich enough to withstand external scrutiny but the school controls how it is framed. In the English system, the school needs evidence that will be legible to an inspector who arrives with a defined set of criteria and a limited amount of time. The evidence must be immediately accessible, clearly mapped to the evaluation areas, and structured in a way that an external reader can navigate quickly.
The tool that serves both systems has to do both things. It has to produce evidence that a Scottish school can use to write its own self-evaluation narrative — with the evaluative quantifier language (almost all, the majority of, a minority of) that HGIOS4 expects. And it has to produce evidence that an English school can present to an inspection team — mapped to the six evaluation areas, structured for rapid access, with the Ofsted-specific quantifier language (the vast majority, many, some, few) that the framework uses. Same observation data. Different evaluative languages. Different audiences. Different purposes.
What the evaluative language reveals
The evaluative quantifier language is where the differences become most revealing, because it is the point at which a philosophical position becomes a sentence on a page.
HGIOS4 uses a six-point scale with specific quantifier terms: "all" (100%), "almost all" (91–99%), "most" (75–90%), "the majority of" (50–74%), "a minority of" (15–49%), "a few" (less than 15%). These quantifiers are used to describe the extent to which a quality indicator is met — "almost all learners are motivated and engaged" means something precise, even if the precision is a range rather than a number. The language is designed for the school to use about itself. It assumes that the school is generating the evidence and making the evaluative judgement, and that the language provides a consistent vocabulary for expressing that judgement.
Ofsted's quantifiers are different: "the vast majority" (roughly equivalent to "almost all"), "many" (roughly "most"), "some" (roughly "a minority of"), "few" (roughly "a few"). The language is designed for inspectors to use about schools. The terms are less precisely defined than the HGIOS4 equivalents — Ofsted does not publish percentage bands — and they carry a subtly different tone. "The vast majority" is more emphatic than "almost all." "Some" is more neutral than "a minority of." The language reflects the different relationship: an external body describing what it observed, rather than a school describing what it knows about itself.
When I built the evaluative language engine in Learning Lens — the system that generates framework-specific prose from observation data — I had to map these two vocabularies onto the same underlying evidence. The same observation data that produces "In almost all lessons observed, learners demonstrated metacognitive awareness" for a Scottish school produces "In the vast majority of lessons, pupils demonstrated metacognitive awareness" for an English one. Same data. Same classrooms. Different words. Different frameworks. Different assumptions about who is speaking and who is listening.
The mapping forced me to confront ambiguities that are invisible when you work within a single framework. Where exactly does "most" end and "almost all" begin in HGIOS4? Where does "many" shade into "the vast majority" in Ofsted's vocabulary? These thresholds matter. A school that crosses from "most" to "almost all" has crossed from Good to Very Good in HGIOS4 terms. The quantifier is the evaluative judgement, expressed as a word rather than a number. Getting the threshold wrong means getting the self-evaluation wrong, which means presenting an inaccurate picture to an inspector — with consequences that are real and potentially serious.
Where they converge
For all the differences in philosophy and language, the two frameworks converge on what matters. Both care about the quality of learning and teaching. Both care about outcomes for all learners, with particular attention to the most disadvantaged. Both care about the effectiveness of leadership. Both care about wellbeing — HGIOS4 through its SHANARRI-aligned QI 3.1, Ofsted through its new standalone inclusion and personal development judgements. Both expect schools to have evidence — not impressions, not assertions, but structured evidence — that demonstrates the quality of what they do.
The convergence is most striking in the area that Learning Lens is designed to serve: classroom-level evidence. Both frameworks expect schools to know what is happening in their classrooms. HGIOS4 QI 2.3 asks: how good is our learning, teaching and assessment? Ofsted's curriculum and teaching evaluation area asks: is the quality of education consistently good across the school? Both questions require the same thing — systematic evidence from classrooms, gathered through observation, analysed for patterns, connected to what the school knows about its pupils and their outcomes.
The HMIE thematic inspection published in March 2025 found that "self-evaluation processes are not always rigorous enough" across Scottish schools, and that there was "notable variability in the consistency and quality of support provided to schools." Ofsted's move to "secure fit" — requiring every descriptor to be met for a grade — was driven by a similar diagnosis: that the "best fit" model allowed too much variation in how evidence was interpreted. Both systems are responding to the same problem: that the evidence base for school evaluation is not consistently good enough.
This is the problem that observation systems were supposed to solve, and that most observation systems have failed to solve — because they capture impressions rather than evidence, because they produce paragraphs rather than data, because they are designed for compliance rather than for genuine self-evaluation. The Scottish school that writes a compelling QI 2.3 narrative without structured observation data to support it is vulnerable to the same challenge as the English school that presents well during an inspection but cannot demonstrate consistent quality across the curriculum. Both are relying on assertion where the framework demands evidence.
What the reforms might mean
Scotland's new inspection framework is not yet published, but the direction of travel is clear from the consultation and from the structural changes already enacted. HMIE's independence from Education Scotland signals a system that wants clearer separation between support and accountability — a move that could sharpen the scrutiny schools face while also creating space for Education Scotland to focus on curriculum improvement without the inspection function pulling in a different direction. The question for Scottish schools is whether the new framework will retain HGIOS4's self-evaluation-first philosophy or shift toward a more directive inspection model. The consultation asked about this directly, and the answer will shape what evidence schools need to produce and how they need to present it.
England's reformed framework is already live, and the early indications suggest that the shift to secure fit grading and standalone inclusion judgements is substantial. Schools that were comfortable under the best fit model — where a strong performance in one area could compensate for a weaker performance in another — will find that compensation is no longer available. Every evaluation area must meet the standard independently. This places a premium on consistency: not just good teaching in some classrooms, but good teaching in all of them. Not just inclusion as a policy, but inclusion as an observable, evidenced practice across the school.
Both reforms increase the demand for structured, specific, school-level evidence. Scotland's new framework, whatever form it takes, will ask schools to demonstrate that they know themselves well — and the evidence for that self-knowledge needs to be more rigorous than a collection of observation paragraphs. England's framework explicitly requires schools to evidence every descriptor within a grade, which means the evidence needs to be granular enough to map to specific criteria rather than offering a general impression of quality.
What I took from building for both
Building a tool that serves both frameworks forced me to think about what observation evidence is for — and the answer turns out to be the same on both sides of the border. It is for understanding what is happening in classrooms with enough specificity that the school can say something truthful about itself. The evaluative language differs. The inspection methodology differs. The relationship between the school and the inspector differs. But the underlying need — specific, structured, research-grounded evidence of teaching quality — is identical.
The qualifier taxonomy in Learning Lens is the same for Scottish and English schools. The 59 practices, the 11 categories, the descriptive sub-tags — these are grounded in research that does not stop at the border. Rosenshine's Principles are not more or less valid in Edinburgh than in Manchester. Hattie's effect sizes do not change when you cross the Tweed. What changes is the framework within which the evidence is interpreted — and that interpretation layer is what the evaluative language engine handles, translating the same observation data into the vocabulary that each framework expects.
What surprised me most was how much the process of building for both clarified my understanding of each. Working within a single framework, you absorb its assumptions without examining them. Working across two, you see each through the lens of the other — and the assumptions become visible. Scotland's trust in self-evaluation looks different when you see it next to Ofsted's assumption that external scrutiny is necessary. England's inspection model looks different when you see it next to Scotland's belief that schools should evaluate themselves first. Neither is wrong. Both are incomplete. The school that does this well — that genuinely understands its own practice, can evidence it specifically, and can present that evidence in whatever framework the system requires — is the school that will thrive regardless of which side of the border it sits on, and regardless of which reforms are coming next.
The frameworks are changing. The need for evidence is not.
Jamie Scobie writes from extensive experience in Scottish secondary education, including pastoral care, data for improvement, and school self-evaluation. This blog is an independent publication: he writes in a personal capacity as the creator of Learning Lens, writing about classroom observation, teaching evidence, and education policy. He speaks here only for himself and for Learning Lens, not for any employer or other organisation. He holds an MSt from Cambridge (Distinction) and a Masters from Stirling.