IGCSE Private Tutor in Dubai

After three sessions with an IGCSE private tutor in Dubai you should hold four things. A marked piece of your child's work with annotations that reference the mark scheme. A named error, stated specifically enough to count. A number: how often that error occurred across the work seen so far. And a sentence about what happens next, in which the error is the subject rather than the child's attitude.

If those four things exist, the tuition is auditable. You do not need to understand chemistry to look at a sheet that says "unbalanced ionic equations: 7 of 10 in week one, 4 of 10 in week three" and know what is going on. If they do not exist, you are buying company. Pleasant company, often, and sometimes useful for confidence, but not something that can be shown to have moved marks.

The reason tutors avoid this is that it creates accountability in both directions. A tracked count that does not fall after four weeks is an argument the tutor has to answer. Most will prefer to report that your child is improving in confidence, which may even be true, and which cannot be checked. Our position is that the count gets shown to the parent whether it moved or not, and a flat count triggers a change in approach rather than a reassurance.

A note on what three sessions is not enough for. It is not enough to move a grade, and anyone presenting it that way is overselling. Genuine content gaps need six to ten sessions per significant topic. Three sessions is enough to produce a diagnosis you can trust and to see whether your child actually talks to this tutor, which are the two things that should be settled before you commit money to a block.

The single most common reason a Dubai IGCSE candidate who knows the material scores below their understanding is that they answer a different question from the one the command word asked. This is not a vague risk. It is countable, it is visible on any marked script, and it is the cheapest category of mark to recover because no new content has to be learned.

Command words are specified. "State" or "give" wants a short factual answer with no reasoning; writing a paragraph wastes time without adding marks. "Describe" wants what happens, in order. "Explain" wants why, and a description scores nothing against an explain mark scheme no matter how accurate it is. "Suggest" signals an unfamiliar context and is marked on reasoning rather than recall, which is why candidates who leave it blank are leaving the most forgiving marks on the paper. "Calculate" wants working, because the method marks are awarded before the answer is reached.

The session technique is simple and it works on fifteen-year-olds. Put the student's own answer on one side of the screen and the published mark scheme on the other. Ask them to award themselves the marks. Nearly every student is stricter than the examiner on content and far more generous than the examiner on form. Watching themselves fail to award a mark they thought they had earned does more in ten minutes than a term of being told to read the question carefully.

This is also where board differences bite. A Cambridge mark scheme and a Pearson mark scheme for the same topic credit different wording, and the number of marks per question differs. A tutor marking Pearson work against a Cambridge scheme produces a document that looks like evidence and is not. Establishing the board before the first marking exercise is not administrative tidiness; it determines whether the evidence means anything.

A first session that teaches content is a wasted session. Nobody yet knows which content to teach. The first hour reads evidence, and the most useful evidence is the most recent mock with the marks on it. If the script is not available, we set a past paper and mark it ourselves before any teaching happens, because a plan built on a report card grade is a plan built on a summary of a summary.

The reading happens in a fixed order. First, total marks lost and where on the paper they sit, because a cluster at the end is a timing problem rather than a knowledge problem. Second, each lost mark gets a category: technique, process, timing or content. Third, the categories get a mark total. Fourth, the totals get ranked. The resulting list is the plan, in order, and it frequently surprises the family because the topic they were worried about is rarely the one costing the most marks.

A worked example from a typical 0580 Extended Paper 4 script. Fourteen marks lost: five to rounding and unit errors spread across the paper, four to two blank questions in the final section, three to a genuinely unlearned topic, and two to method marks forfeited by writing only answers. The family had asked for help with the unlearned topic, which is worth three marks. Process and timing together are worth nine, and both respond faster. Starting with the unlearned topic would have felt responsive and returned the least.

The output of the first session is written, goes to the parent, and is short: the ranked category list with mark totals, the single error we will track first, a target count, and the fortnight by which we expect it to move. Everything after that is execution against a document you can check. If a tutor cannot produce that document after reading a script, the diagnosis did not happen.

A term of tuition should leave a document behind. Not a folder of worksheets, a ledger: what was marked each week, what the tracked count was, and what changed. The point of writing it down is that memory is kind to tuition. Everyone remembers the session where something clicked. Nobody remembers the four weeks where a count sat still.

The shape below is what a maths or science term typically looks like. Weeks one to two diagnose and set the baseline. Weeks three to six attack the highest-value error category, usually process or technique, because those return marks fastest. Weeks seven to nine move to content gaps, which are slower. Weeks ten to twelve run full timed papers, which test whether the earlier gains survive under pressure. They often do not at first, which is exactly why that block exists.

What a parent does with the ledger is ask one question every fortnight: did the number move. If it did, continue. If it did not, the conversation is about why, and the honest answers are a narrow set. The error was misdiagnosed. The independent work between sessions is not happening. The tutor's explanation is not landing with this particular child. Or the gap is genuinely content-shaped and needs more weeks than estimated. Each of those has a different response, and none of them is "keep going and hope".

The ledger also solves a problem families do not anticipate: tutor changes. Tutors move cities, change jobs, and occasionally turn out to be the wrong fit in week five. When the record exists, the next tutor starts from the diagnosis rather than redoing it, which saves two to three sessions. We keep the record with the family's file and hand it over on request, and we delete marked work when asked.

A predicted grade is a professional judgement, not a measurement, and it is produced under constraints parents rarely see. Dubai schools generate them from internal assessment data, set placement and teacher judgement, usually in a short window and for a whole cohort at once. They are frequently accurate. When they are not, the family needs to know what evidence sits underneath before challenging anything.

The question that cuts through is about raw marks. A mock grade is a raw mark converted using some set of grade boundaries, and boundaries are set per series, per paper, after the exam. A mock marked against last year's boundaries produces an indicative grade, not a prediction. So the useful request to a head of department is not "can the prediction change", it is "what raw mark did my child score, out of what total, and which series boundaries were applied". That is a factual question with a factual answer.

Raw marks also make progress legible in a way grades do not. A child who moves from 94 to 112 out of 130 has gained eighteen marks and may still be sitting in the same grade band, which looks like no progress on a report card and is in fact substantial. The reverse happens too: a one-mark move can cross a boundary and flatter a term of work that achieved very little. Tracking marks protects a family from both illusions.

For forecast grades there is a hard date in the calendar, and it belongs to whichever series your child is in. A UAE candidate sits June or November, so for a June candidate the sequence is an entry closing in the spring, a forecast submitted before the series opens, and papers in June; your examinations officer can give you the exact submission date for that series. The implication is the same either way: the conversation about a prediction belongs in the autumn term, while there is still a term of evidence left to gather. By the time the forecast is submitted, the evidence that would have changed it needed to already exist.

Some IGCSE subjects carry a component the student produces outside an exam hall. Wherever that exists, there is a line, and the line is not negotiable. A tutor may teach the skills the component needs, explain the assessment criteria in plain language, discuss the student's chosen approach, question their reasoning, and respond to a draft the student wrote. A tutor may not produce assessed material, rewrite a draft into a submittable state, or supply sentences for a student to hand in.

The practical test we use is whether the feedback could be read aloud to the student's teacher without anybody being uncomfortable. "Your second paragraph asserts a conclusion your data does not support; which figure would you need to show to support it?" passes. Handing over a corrected paragraph does not. The first builds the skill the criteria are actually assessing. The second transfers the mark to the tutor and exposes the student to a malpractice finding they, not the tutor, will carry.

This market has a specific problem that makes the boundary worth stating loudly. One site competing for these searches sells completed Internal Assessments outright at a published price. Families encountering that offer sometimes assume it is normal and that everyone does a version of it quietly. They should not. It is an academic-integrity breach, it is the kind of thing boards investigate, and the student is the one who faces the consequence.

Our process makes the boundary auditable rather than merely asserted. Feedback is logged, so the trail of advice can be read in order. Sessions on a coursework component are scheduled against the school's internal deadlines so the work is produced over time rather than assembled at the end. And the boundary is written down and sent to the parent before the first draft session, because the moment to agree it is before anyone is under deadline pressure.

The Dubai market publishes rates unevenly, so here are the bands with their sources, as of October 2026. Budget online tutoring advertises from AED 45-150 an hour, with EdFlik publishing from AED 45-65 and Tutopiya from AED 33-52. A qualified subject tutor online sits at AED 150-300, with Unitors publishing IGCSE from AED 150 and IB from AED 200. In-home one-to-one visits run AED 180-500 inclusive of travel, per Bestimate at 180-350 and independent home tutors at 200-500.

Above that, IGCSE and A Level specialists are quoted at AED 200-450 and above, with TutorChase publishing AED 270, and premium named or examiner-badged tutors reach AED 500-650, with BartyED at AED 650. Monthly packages in this market include AED 799 for eight online sessions and AED 1,599-1,799 for four two-hour in-home sessions. Bestimate's monthly estimates for weekly one-hour sessions run AED 240-600 at the budget end, 600-1,200 mid-range and 1,000-2,000 plus for specialist exam-year work.

What the extra dirhams buy, in evidence terms, is mark-scheme literacy. A tutor at the bottom of the range usually knows the subject and not the scheme, which means they can explain content and cannot tell you that nine of fourteen lost marks were technique. That is a real distinction and it is most of the price gap. The upper bands exist because syllabus-specific marking experience is scarce, and scarcity in a small market prices accordingly.

A note on the in-home premium. The AED 180-500 band for visits is partly a travel bill, and travel is the one component of a tutoring fee that produces no teaching. IB Gram prices against the online bands rather than the in-home bands because nobody is driving anywhere. We do not publish a rate card on this page: the figure depends on subject scarcity, tier and session length, and the consultation after the free first lesson is where you get it. Any page quoting you a specific dirham figure as our price is not ours.

KHDA is responsible for enhancing and overseeing the quality of private education in Dubai, and it inspects private schools on a six-band scale running from Outstanding down to Very Weak. The report is public, it is detailed, and most families read only the headline band. The useful part sits further in, in the subject-level and phase-level judgements, because a school can be Very Good overall and have an inconsistent department in exactly the subject your child is struggling with.

That is the connection to tuition, and it is a practical one. If the inspection report flags attainment in a subject as acceptable while the rest of the school sits higher, a child in that subject is being taught in a department under scrutiny. Tuition in that subject is more likely to be worth the money than tuition in a subject where the department is strong and the child's problem is attitude or organisation.

The band also tells you something about how to read the school's own predicted grades. A department with inconsistent internal assessment produces predictions with wider error bars, which makes raw marks and external past-paper evidence more valuable. A strong department's prediction is usually well calibrated, and a parent who disputes it with a tutor's opinion rather than with marks will get nowhere, correctly.

One thing an inspection band is not is a statement about any individual child or any tutor. We name KHDA here because it is the regulator and because its reports are a free, public, well-evidenced source that families underuse. IB Gram is not approved, accredited, endorsed or licensed by KHDA, holds no permit from it, and operates no premises in the UAE. An operation with premises here would require a KHDA Educational Services Permit, which is a specific thing issued to training institutes, and we do not hold one or claim to.

An evidence-led arrangement has to be willing to produce an unwelcome answer, or it is not evidence-led. The count either moved or it did not, and when it has not moved after four to six weeks of consistent work there are only a few honest explanations. Confronting them is cheaper in week six than in month five.

The first is misdiagnosis. The error being tracked was not the one costing the marks, which happens most often when a family's stated concern overrode the ranked list. The fix is re-reading the script and re-ranking, and it costs one session. The second is independent work not happening between sessions, which no amount of tutoring hours compensates for; a child who does nothing for six days and sits a lesson on the seventh will plateau. That is a scheduling and expectation conversation, not a teaching one.

The third is personal fit, and it is more common than the industry admits. A tutor can be technically excellent and simply not land with a particular fifteen-year-old. The signal is a student who answers in monosyllables, does not bring questions, and performs better on independent work than in sessions. When we see that, we propose a different tutor and the ledger transfers, so nothing is restarted from zero.

The fourth answer is the one nobody advertises: sometimes the right decision is to stop. A child carrying nine subjects, a coursework backlog and three tuition slots a week is not short of teaching. Dropping a session and sleeping is occasionally the intervention with the highest return. We would rather say that and keep a family's trust than sell hours into a timetable that is already full. If a subject's marks are genuinely secure, the honest recommendation is to spend the money elsewhere or not at all.