How often should we measure progress, and what should we measure?

The best measure answers the decision staff need to make. Testing too often can take time from teaching; testing too rarely may mean an unhelpful approach continues for weeks.

Match the measure to the review decision

For a taught spelling pattern, a short check with unfamiliar but comparable words may show whether learning transfers. For independent starts, observe the first action and adult prompts in similar lessons. A broad attainment score may be too insensitive for either immediate question.

Choose a feasible interval that leaves enough teaching between checks and gives time to act on the information. Keep task difficulty and support comparable and note important changes. Look at several observations rather than one unusually good day. Record accuracy, independence or quality according to the goal, and stop collecting information that does not inform a decision.

Look for the barrier

Decide what change would count as meaningful: accuracy, fluency, independence, task quality or use in class. Choose a short repeatable check and an appropriate interval for the skill. Record enough context to interpret unusual scores.

Measurement frequency should follow the kind of learning. A short daily skill practice might use a brief probe regularly; a complex classroom outcome may need work samples over several lessons. Decide what the number represents before making a chart. Correct responses, accuracy, independence, adult prompts and the pupil's own account answer different questions. Record enough to see a trend without turning a ten-minute teaching session into ten minutes of paperwork. Establish a starting point using an equivalent task.

What to try in school

For a practised micro-skill, a brief regular probe may be helpful. For a broader outcome such as writing a coherent paragraph, a comparable work sample every few weeks may tell you more. Plot or list results where staff can see a pattern rather than relying on one better session.

Choose one direct measure of the taught skill and one check that matters in class. Keep the conditions reasonably comparable: instructions, task length, item difficulty and help available. A graph of correct answers is difficult to interpret if the adult sometimes supplies half the responses. Note absences, changes in teaching or major events beside the series. Agree a review point in advance and keep the pupil aware of progress in an encouraging, non-competitive way. The data should prompt a teaching decision, not exist for its own sake.

Measurement should be proportionate to the decision. When staff are deciding whether to change a teaching method, a few comparable samples may be enough. When a pupil is gaining a very small component skill, closer tracking can reveal whether practice helps. Avoid measuring only speed for skills where accuracy, meaning or confidence matters. Plan who collects the data and who interprets it; the TA should not be left to make a major change alone because a line on the graph flattened.

Be transparent about what a probe can and cannot show. A quick one-minute check may be sensitive to a narrow skill, but it will not explain whether the pupil can use that skill in a longer piece of work. A pupil may also become tired, anxious or distracted during timing; ask about their experience and select an appropriate method. Compare several points rather than celebrating an unusually high score or reacting to one low score. When the graph changes, look back at teaching and context before attributing the line to the pupil alone.

Try this: For a new spelling pattern, check a few untaught example words at the same time each week and note both accuracy and whether the pupil uses the pattern in writing.

A worked example

A TA teaches a Year 5 pupil to recall a small set of multiplication facts. They make a brief record of correct responses to equivalent sets during practice and the teacher samples how the pupil uses those facts in weekly mixed problems. Accuracy rises in the probe, but class problems still take a long time because the pupil cannot choose the required operation. Staff continue fact practice only if it remains useful, while teaching the problem-selection step separately. Both measures prevent a misleading all-or-nothing conclusion.

Review what changes

Discuss the trend with the teacher, TA and pupil. Change the plan if results are flat or the pupil can only succeed under the exact practice conditions.

Look for a pattern across several observations, not a straight line or an arbitrary percentage after one session. Check the number of practice opportunities actually delivered. If progress is erratic, inspect the task and how it was administered before attributing the variation to the pupil. When the skill is secure, reduce frequent probes and check retention or application after a delay. Tell parents and the pupil what changed in ordinary learning, not just the chart's final number.

What does the evidence tell us?

EEF implementation guidance supports monitoring whether a change is delivered and whether it benefits pupils. EEF SEND guidance emphasises reviewing individual support. Neither prescribes a single measurement schedule for every intervention; the right measure depends on the intended skill and the classroom decision.

Further reading

Use this resource to explore the wider evidence and adapt ideas to the pupil and school context.

More practical advice: Browse the advice.

Share this article

Know someone who would find this useful?