There is a sentence that every tutor, every school and every course hears at some point, usually a week or two in.
“I’m not sure this is working for me.”
Start with the words rather than the worry, because the words contain a mistake.
Tuition does not work for anybody. It is not a treatment. Nothing is being done to the student that could take effect or fail to. What a tutor supplies is a set of ideas, methods and corrections. Those only turn into a grade when the student picks them up and uses them — attempts the work, applies the method to their own writing, brings the confusion back to the tutor, and does something with the feedback when it comes.
So the honest version of the sentence is not “is this working for me”. It is: am I using this?
That is not a way of shifting blame onto a fourteen-year-old. It is what the evidence says. In one of the best-known studies of tutoring, researchers stopped the tutors from doing the thing everyone assumes tutoring is — explaining and giving feedback — and had them prompt the students instead. In the authors’ own word, the result was “surprising”. The students learned just as effectively. What the researchers credited was not the teaching they had removed. It was what the students then did: they built more of the understanding themselves, and they took more control of their own learning.[3]
Which means a family can be paying for excellent teaching and getting very little, and the cause is not the teaching. It is also why the second mistake matters so much.
Because having asked the question, most people reach for the wrong instrument to answer it. How do the lessons feel? That is the one measurement known to point the wrong way.
That is not an opinion. It has been measured, and the result is one of the more uncomfortable findings in education research.
The experiment
In 2019 a group of physicists at Harvard ran a study designed to separate two things that usually get muddled together: how much students actually learn, and how much they think they have learned.[1]
The design is the part that makes it hard to argue with. They took a single fifteen-week introductory physics course. For the first eleven weeks every student was taught the same way. Then, in week twelve, the class was split at random. One half carried on with lectures from an experienced and highly rated instructor. The other half was taught actively, which meant the students had to attempt problems, get things wrong, and be corrected as they went. The content was identical. The handouts were identical. The instructor made no effort to persuade anybody that either method was better.
Then they measured two things: what the students had actually learned, tested; and what the students believed they had learned, asked.
The two results pointed in opposite directions. The students in the active classes learned more. The students in the active classes felt they had learned less.
The researchers’ own summary of it is blunt: students in the active classroom “learn more, but they feel like they learn less”.[1]
I should be careful here, because it matters. That study was run on university physics students, not on fourteen-year-olds writing English essays. I am not going to claim it transfers exactly. What it establishes is the mechanism, and the mechanism is not about physics: effort feels like failure, and fluency feels like understanding. Sitting comfortably while someone explains something clearly feels excellent. Being made to attempt it yourself, badly, feels like proof that you cannot do it.
Why this matters for a parent making a decision
Now put that finding next to the ordinary way families judge a new course.
The researchers spelled out the consequence themselves. Trying to evaluate teaching on the basis of how much students think they are learning, they warn, “could inadvertently promote inferior (passive) pedagogical methods”.[1]
Read that slowly, because it is the whole article. If you judge teaching by how good it feels, you will systematically choose the teaching that works less well. Not occasionally. Systematically. The comfortable lesson wins the survey and loses the exam.
So when a student says “I don’t know if this is working”, after two weeks of being made to plan properly, being marked strictly, and being told to write the paragraph again — that report is exactly what the research predicts effective teaching to produce. It is not evidence that the teaching is failing. It is not evidence that it is succeeding either. It is not evidence.
The second problem: two weeks is not a measurement
There is a separate reason an early verdict cannot be trusted, and this one is simpler.
Learning is a durable change. What anybody can watch in a lesson is performance — how well the student did that morning. Those are not the same thing, and a major review of the evidence puts it plainly: what we can observe during teaching is performance, “which is often an unreliable index” of whether learning has taken place.[2] The same review notes that improvements in performance can fail to produce learning at all, and that some teaching choices push the two in opposite directions.[2]
There is also a plainer point that has nothing to do with research. One measurement is not a trend. If a student writes one essay and it is marked, you have a single number. A single number has no direction. You cannot tell from it whether anything is moving.
The smallest thing that has a direction is a completed cycle: the student writes something, it is marked against the examiner’s criteria, the feedback names what to change, the student redrafts it, and the redraft is marked again. Only at the end of that does a second number exist. Two numbers make a line. One number makes a feeling.
That is why a week cannot answer the question. Not because the answer is inconvenient, but because in a week the second number does not exist yet.
What we owe you in return
An argument like this can be used badly. The limit matters as much as the argument.
“You cannot judge it yet” is a reasonable thing for a teacher to say in week two. It becomes an excuse the moment it is used to avoid ever being judged. If we ask a family to wait for the first full cycle before drawing conclusions, then we owe them the evidence the moment that cycle closes — the marked piece, the feedback, the redraft, the new mark, and an honest reading of the difference between them.
That is a fair exchange. Hold any teaching to it, including ours. Ask for the criteria. Ask what changed between the first mark and the second. If nothing changed, that is a real answer and you are entitled to it. What you should not accept is a grade with no working, from us or from anyone.
The phrase itself
One last thing, about the words rather than the evidence.
“Does it work for me?” describes a student as somebody things are done to. A course arrives, and either it works on them or it does not, in the way a medicine either agrees with you or does not.
Learning does not behave like that. Nobody has ever been taught to write well by being present while writing was explained. The student who improves is the one who attempts the paragraph, gets it wrong, listens to why, and writes it again — and then does that fifty more times. The work is theirs.The teaching organises it, sequences it, marks it honestly and makes the next attempt better targeted than the last. It cannot do it for them.
So the more useful question is not whether the course is working. It is: am I working, and is my work getting better? The first question has no answer in week two. The second one does, and the student already knows it.
If you would like an honest starting measurement — which specific parts of your child’s writing are costing the marks, judged against the examiner’s criteria — our free assessment takes about five minutes.
—
Sources
- Deslauriers, L., McCarty, L. S., Miller, K., Callaghan, K., & Kestin, G. (2019). Measuring actual learning versus feeling of learning in response to being actively engaged in the classroom. Proceedings of the National Academy of Sciences, 116(39), 19251–19257. Randomised, identical course materials and handouts. A single 15-week introductory physics course at Harvard: the first 11 weeks taught identically to all students, then in week 12 the class randomly assigned to active or passive instruction. Students in the active classrooms learned more but reported lower perception of learning; the authors warn that evaluating instruction by students’ perceptions “could inadvertently promote inferior (passive) pedagogical methods”. https://www.pnas.org/doi/10.1073/pnas.1821936116
- Soderstrom, N. C., & Bjork, R. A. (2015). Learning Versus Performance: An Integrative Review. Perspectives on Psychological Science, 10(2), 176–199. “During the instruction or training process, however, what we can observe and measure is performance, which is often an unreliable index of whether the relatively long-term changes that constitute learning have taken place.” https://journals.sagepub.com/doi/abs/10.1177/1745691615569000
- Chi, M. T. H., Siler, S. A., Jeong, H., Yamauchi, T. and Hausmann, R. G. (2001). Learning from human tutoring. Cognitive Science, 25(4), 471–533. Tutors were suppressed from giving explanations and feedback and encouraged to prompt the students instead; the authors report that “surprisingly, students learned just as effectively”, and attribute that learning to the students’ own construction and to “their greater effort to take control of their own learning”. https://doi.org/10.1207/s15516709cog2504_1
