Assessment is one of the most consequential tools a school has, and one of the most misunderstood. In British schools, it takes several distinct forms, each with a different purpose. Knowing the difference between them helps parents interpret the information schools share, ask better questions, and recognise the signs of a school that uses assessment well versus one that uses it carelessly.
Three Types of Assessment
Formative assessment happens continuously inside lessons. It includes questioning that checks whole-class understanding, short recall tasks at the start of a period, and feedback pupils can act on immediately. Its purpose is to guide teaching, not to rank pupils. A teacher who asks the class a question and listens carefully to the answers is conducting formative assessment. So is one who reviews a quick written task and adjusts the next twenty minutes of the lesson accordingly.
Summative assessment evaluates learning at a fixed point: a termly exam, a GCSEs paper, an internal test at the end of a unit. It produces a grade or a score intended to reflect what a pupil has learned. Summative assessments are most useful as indicators of curriculum effectiveness and as prompts for professional discussion. They become harmful when treated as the sole measure of a pupil or used punitively without context.
Standardised assessment, such as CAT4 or GL Progress Tests, measures performance against national norms. These tools offer statistical reliability and allow schools to identify unusual patterns, such as a pupil whose test results diverge sharply from teacher expectation. Their limitation is equally important: they cannot replace teacher judgement, and scores should be read as secondary data points, not verdicts.
Why Retrieval Practice Improves Learning
The research on what is sometimes called the testing effect is robust and consistent. Asking pupils to retrieve information from memory, rather than re-reading notes or passively reviewing material, significantly strengthens long-term retention. Spaced retrieval, returning to the same material across multiple sessions rather than concentrating all review in one block, reduces forgetting further. Low-stakes quizzes outperform highlighting, re-reading, and summarising as revision strategies.
The practical implication for schools is direct: brief, cumulative recall tasks built into daily teaching are more effective than a revision sprint before an exam. Schools that understand this build retrieval into the normal rhythm of lessons. Those that do not often see pupils who perform adequately during a unit but cannot recall the same material three months later.
The Stress Question
Exam pressure is real, and the debate around it is frequently framed in unhelpful terms. The question is not whether British schools test too much. It is whether testing is designed and sequenced in a way that makes high-stakes moments feel predictable rather than overwhelming.
Research is clear that moderate challenge improves performance. Harmful stress arises from a specific combination: poor preparation, unclear expectations, and high stakes arriving without warning. Schools that use spaced retrieval, communicate assessment criteria clearly, and limit the number of significant assessments in any given period tend to see lower anxiety at exam time. The problem is not GCSEs. It is inadequate preparation for them.
What Strong Schools Do Differently
Strong schools use assessment to serve the curriculum, not to drive it. Assessment reflects what has been taught rather than shaping what is taught next, which avoids the narrowing effect of teaching to the test. Internal exams are moderated: mark schemes are rigorous, different teachers review the same work, and grade boundaries are consistent. A pupil moving from one year group to the next should not encounter a grading system that operates on entirely different principles.
Weaker schools rely too heavily on a single data point, whether that is a CAT4 score, a termly ranking, or a predicted grade generated early in the year. They may also use assessment calendars that are too dense, producing data without acting on it. The measure of a good assessment system is not the quantity of data produced but what changes as a result.
ISJ uses a combination of GL standardised assessments, internal termly exams, and continuous formative checks across all year groups. The GL data is used to identify patterns and to contextualise teacher judgements, not to replace them. More detail on how assessment sits within the broader curriculum framework is on the curriculum page.
What Parents Should Look For and Ask About
When evaluating how a school handles assessment, the most revealing questions are practical. Ask how the school balances formative and summative approaches. Ask how internal exams are moderated and what standardised tools are used. Ask how the school communicates results and whether it helps parents interpret scores in context rather than in isolation.
Red flags worth noting: frequent tests with no clear rationale, idiosyncratic grading systems that change year to year, reliance on a single score for decisions about pupil progress, and a reluctance to explain how internal marking is quality-checked. Healthy indicators include a coherent assessment calendar, cumulative low-stakes retrieval embedded in teaching, transparent communication about what scores mean, and consistent expectations across subjects and year groups.
What is the difference between formative and summative assessment?
Formative assessment happens during teaching and is used to adapt instruction. Summative assessment evaluates learning at a specific point in time, such as a termly exam or a GCSE paper. Both are necessary, but they serve different purposes and should not be conflated.
Why does the British system place so much emphasis on exams?
External examinations such as GCSEs and A-Levels are valued for their reliability, comparability, and fairness. Anonymous marking, external examiners, and consistent grade boundaries reduce bias and provide a national standard that universities and employers recognise.
Are British exams too stressful?
Research shows that moderate exam pressure is normal and can improve performance. Stress becomes harmful when pupils lack preparation or face poorly sequenced assessments. Schools with strong curriculum sequencing and embedded retrieval practice typically report lower anxiety at high-stakes moments.
What are standardised tests and why are they used?
Standardised tests such as CAT4 and GL Progress Tests compare a child's performance with national norms. They are useful for identifying unusual patterns of progress and for providing a check on internal school data, but they should not be used in isolation from teacher judgement.
How reliable are internal school exams?
Internal exams are reliable when they are moderated, aligned to the curriculum, and supported by clear mark schemes. Without moderation, internal assessments can drift or inflate over time, making results difficult to interpret accurately.
Do British schools teach to the test?
Teaching to the test occurs when exam content drives curriculum content rather than the other way around. High-quality schools avoid this by ensuring that assessment reflects what has been taught through a well-sequenced curriculum. Pupils learn rich content, and assessments measure that learning.
Why does retrieval practice improve learning?
Retrieval strengthens memory by requiring pupils to recall information rather than re-expose themselves to it. Meta-analyses consistently show that spaced retrieval outperforms passive review strategies such as re-reading or highlighting as a method for building long-term knowledge.
What should parents look for in a school's assessment approach?
Look for a clear explanation of how data is used, moderation of internal assessments, a balanced assessment calendar, evidence of retrieval practice and spaced review, and transparent communication about what scores mean. Avoid schools that rely on a single metric or cannot explain how internal marking is quality-checked.
How should parents interpret standardised scores or predicted grades?
Scores should be read as indicators, not verdicts. Patterns over time matter more than any single data point. Predictions reflect probability based on available data and should be treated as starting points for a conversation, not fixed outcomes.