How do you measure a student’s progress in private tutoring?

A student can make progress without their average rising straight away. They may start their exercises without waiting for a hint, make fewer method errors, hand in their work more regularly or be able to explain a concept they used to apply mechanically. Conversely, a single good mark does not prove that a skill has been lastingly acquired.

The useful question is therefore not only “are the marks going up?” but what can the student now do with less help, more consistency and greater precision? It is also a good criterion for judging a digital learning tool: a dashboard is only valuable if it helps the tutor, the student or the parent make a better teaching decision, which is what the 5 useful pieces of information to pass on to parents after a private tutoring session make possible.

1. Defining a starting point before looking for progress

You can only measure change if you know where the student is starting from. The first session or the first few weeks should therefore produce a sufficiently precise snapshot of the initial level.

This snapshot does not need to be an exhaustive test. Above all, it should answer a few questions:

• what knowledge is already in place?
• which tasks does the student succeed at without help?
• at what point do they ask for a hint?
• which errors keep coming back?
• how do they organise their work?
• what do the latest school assessments show?

The underlying idea is close to formative assessment. The Éduscol (the French Ministry of Education’s resource portal for teachers) guide to assessing learning at lycée (upper secondary school) distinguishes diagnostic assessment, which identifies the level and needs; formative assessment, used during learning to help the student see where they stand and progress; and summative assessment, which certifies a level of mastery at the end of a period of learning. Although the guide concerns the school setting, this distinction is particularly useful in private tutoring.

Choosing a few observable skills

Avoid a baseline such as “average level in maths” or “lacks method”. Prefer elements that can be compared over time:

Starting indicator	Possible observation
Understanding	Explains the rule correctly but cannot yet use it in a problem.
Independence	Needs a hint to start three exercises out of four.
Errors	Regularly forgets a specific step in the same type of calculation.
Consistency	The work set is prepared irregularly.
School results	The latest assessments show difficulties concentrated on certain skills.

Follow-up then becomes much simpler: you just need to see whether these observations change on comparable tasks.

Digital tools should remember, not invent the diagnosis

A digital tool is useful for keeping the date, the topic, the difficulties observed and the work planned, notably the homework to set between two private tutoring sessions. It prevents information from disappearing into messages or the tutor’s memory.

On the other hand, no interface can seriously conclude that a student “has mastered functions” just because a box has been ticked. The data has to come from a real task: an exercise, an oral explanation, a test paper, a quiz or a problem. The software organises the information; the tutor interprets what the performance means.

2. Tracking understanding before looking only at the average

The mark is an important indicator because it reflects the assessments actually faced at school. But it aggregates many factors: the difficulty of the test, the marking scheme, the chapter, time management, careless mistakes and the skills assessed. It does not always say why the student has progressed or slipped back.

To track understanding, the tutor can use short checks during or after the session:

• have the student explain a concept in their own words;
• ask for an example and a counter-example;
• have them solve a similar exercise without a model;
• change the wording slightly to check transfer;
• ask the student to identify their own error.

In its resources on tutoring, Stanford’s National Student Support Accelerator recommends aligning formative assessments with session objectives and using the information gathered to adjust subsequent teaching. These standards concern structured tutoring programmes, not French private tutoring specifically, but the mechanism remains relevant: measuring is only worthwhile if the information changes the next action.

Testing the same skill in a different form

A student may succeed because they recognise the exercise they saw five minutes earlier. To check understanding, you sometimes need to change the surface of the problem without changing the skill.

In maths, after a standard equation, the tutor can set a short problem that leads to the same equation. In French, after working on the introduction to a commentary, they can change the text while keeping the same criteria. In English, they can ask the student to reuse the grammatical structure in a new sentence.

Progress becomes more credible when the skill holds up under this change of context.

3. Measuring independence and the quality of methods

One of the most interesting indicators in private tutoring is the amount of help needed. The tutor is there to guide; the risk is therefore confusing the student’s performance with the performance of the student when supported.

You can track a very simple scale:

1. the tutor demonstrates the method;
2. the student succeeds with several guiding questions;
3. a single hint is enough;
4. the student chooses the strategy alone;
5. they check their own result and can explain their choice.

This is not an official scale. It is a practical tool for seeing whether the support needed is decreasing.

Recent resources from the Education Endowment Foundation on metacognition and self-regulation describe independence through the ability to plan, monitor and evaluate one’s own learning, skills that a study skills session at lycée works on. They recommend explicit teaching of strategies, first modelled and guided, then gradually used independently.

Measuring decisions, not just answers

Two students can reach the same correct result with different levels of independence. The first applies the method only after the question “which formula do you need?”. The second identifies the type of problem, chooses the formula and then checks the result.

The final result is identical; the learning progress is not.

A few independence indicators are easy to track:

• number of hints needed;
• ability to start alone;
• ability to choose between several methods;
• ability to check their answer;
• ability to explain why a strategy works;
• ability to go back over an error before the tutor corrects it.

A digital tool can keep these observations session after session. It should not artificially turn them into an “independence score” presented as scientific if no validated method justifies it.

4. Analysing recurring errors rather than counting all errors

Making fewer errors can signal progress, but not all errors carry the same weight. A one-off calculation slip and a repeated conceptual confusion should not be counted in the same way.

The most useful thing is to track errors that come back despite correction:

• the same confusion of vocabulary;
• the same step forgotten;
• a poor choice of method in several contexts;
• a recurring sign error;
• a systematic lack of justification;
• reading instructions too quickly.

The National Student Support Accelerator recommends precisely that tutoring programmes examine the student’s actual work, not just quantitative data, in order to identify accuracy errors, misunderstandings and difficulties encountered on tasks. This analysis can then inform the content and structure of future sessions.

Creating an error → correction → check loop

An error has only really been “worked on” if you then check that it no longer comes back in a similar situation.

Step	Example
Observe	The student regularly mixes up image and preimage.
Understand	They know the definitions but misread the question.
Correct	Systematically restate what is known and what is being looked for.
Check	Set a new exercise the following week without a reminder.

The sign of progress is not just “the error has been explained”. It is “the student then spots or avoids it without the tutor reminding them of the rule”.

5. Tracking homework and consistency without confusing quantity and quality

Between two sessions, independent work provides information that the session itself does not. A student may succeed with their tutor and be unable to redo anything alone three days later. Prepared work therefore makes it possible to test the continuity of learning.

Useful indicators are not just “homework done / not done”. You can look at:

• was the task started without an adult stepping in?
• did the student see it through to the end?
• did they identify where they got stuck?
• did they use the answers before or after trying?
• do the errors from the previous session reappear?

The EEF summary on homework finds a positive average impact of work done at home, particularly in secondary school, but highlights wide variation depending on practice. The quality of the work, the clarity of its objective and its link to learning matter more than simply piling up tasks.

Consistency is context, not a universal explanation

If the planned work is never done, the tutor needs to know, because it limits what can be consolidated between sessions. But you should avoid automatically translating “homework not done” as “lack of motivation”.

Was the instruction clear? Reasonable? Did the student have the prerequisites? Did they know where to find the material? A digital record can show that the problem keeps recurring; it does not on its own explain the cause.

6. Combining quantitative data and qualitative observations

A solid measurement uses several families of indicators. This is also the logic of Stanford’s Tutoring Quality Standards, which provide for tracking individual progress over time with indicators of academic growth but also adaptive indicators, such as engagement or confidence.

Type of data	Examples	What it brings
Results	Marks, quiz scores, proportion of correct answers	A comparable numerical measure when the tasks allow it
Understanding	Oral explanation, transfer to a new exercise	Shows whether knowledge goes beyond immediate memorisation
Independence	Number of hints, ability to start and check alone	Shows how dependence on the tutor is changing
Errors	Confusions that disappear or keep recurring	Helps target the next sessions
Consistency	Work prepared, attendance, continuity	Gives the context in which to interpret results
Session dynamics	Engagement, preparation, observed understanding	Provides an additional signal to be read over time

Avoiding the falsely scientific dashboard

Digital tools make it very easy to create charts, percentages and scores. A piece of data does not automatically become reliable just because it is visualised.

Comparing 14/20 on an easy test with 11/20 on a much harder test can give a false impression of decline. Calculating an average “focus” score over three sessions can also give artificial precision if the scale used has no stable definition.

Before displaying a curve, you should therefore ask:

1. are we always measuring the same thing?
2. are the tasks sufficiently comparable?
3. does the indicator change a teaching decision?
4. does the figure add anything to the raw observation?

If the answer is no, it is sometimes better to keep a short qualitative observation.

7. Taking stock every six weeks

For regular support, a review every six weeks is a practical benchmark, not a frequency imposed by research. The aim is simply to step back far enough to observe a trend without waiting for the end of a school term.

If sessions are weekly, a few weeks already produce several observations: tasks completed successfully, recurring errors, work prepared and any school assessments. The pace should nonetheless be adapted: intensive support before an exam may call for more frequent reviews; occasional support does not need a fixed cycle.

A progress review in five questions

1. What has now been mastered?
2. What remains shaky despite several sessions?
3. Does the student need less help?
4. Is the work between sessions regular and useful enough?
5. Should the current objective be kept, changed or scaled back?

The outcome of this review does not need to be a report several pages long. A “start / today / next priority” comparison is often enough.

Deciding after the review, not just noting

If errors are decreasing and the student is working more on their own, the difficulty can be gradually increased. If marks remain low but the method is becoming more solid, it may make sense to continue before concluding that the support is ineffective. If no change appears on the same indicators after several sessions, the diagnosis, the method or the format of the support needs to be re-examined.

The National Student Support Accelerator also recommends structured data review cycles to adapt tutoring. The important principle is not the “six weeks” frequency, but the regular loop of measure → interpret → act.

What digital tools genuinely improve

Digital follow-up is useful when it solves a concrete problem: information exists but disappears, stays scattered or is never compared with that of previous weeks.

A good tool can:

• keep the starting point;
• archive the observations from each session;
• attach the work to prepare to the right session;
• retrieve errors that have already been worked on;
• make comparison over several weeks easier;
• give the parent visibility without attending sessions.

On its own, it cannot:

• decide that a skill has been acquired;
• automatically explain the cause of a bad mark;
• distinguish with certainty between lack of work, misunderstanding and emotional difficulty;
• guarantee a rise in average marks;
• replace the tutor’s analysis.

How NOEO structures this follow-up without turning the student into a scoreboard

On NOEO, the tutor produces a session report after every session. This report can in particular record the topic, strengths, areas for improvement, the work to prepare and observations on focus, engagement, preparation or understanding.

The real value comes from the history: several reports make it possible to see whether the same difficulty keeps coming back, whether a priority changes or whether the level of help needed is decreasing. The work or exercises sent by the tutor after the session add to this picture by making it possible to check what the student can redo between two sessions.

The parent thus has visibility over the continuity of follow-up without needing to attend NOEO Live. The details of how it works are set out in the NOEO journey before, during and after the session.

Our article on what a useful session report should show parents goes into the quality of the information worth keeping. And when progress also concerns the student’s relationship with difficulty, the signs that show confidence gradually returning provide additional indicators that cannot be reduced to a mark.

Checklist: follow-up that really measures something

• A starting point is defined before comparing.
• The indicators are linked to the real objectives of the support.
• Understanding is checked on tasks, not just declared.
• Independence is tracked through the amount of help needed.
• Recurring errors are distinguished from one-off errors.
• Work between sessions is analysed for quality, not just quantity.
• Marks are kept but are not the only measure.
• Qualitative observations stay visible when a figure would be artificial.
• Comparisons cover comparable tasks wherever possible.
• A periodic review leads to a decision for the following sessions.