Did you know? The A-Level Mathematics formula booklet contains formulas that are not part of every student’s examinable syllabus. A JC1 H1 Mathematics pupil can open MF27, see a long list of sophisticated results and mistakenly believe every line must be learnt. The booklet is shared across several A-Level Mathematics subjects. The H1 syllabus determines the actual scope.
For a family comparing JC1 H1 Mathematics tutors in Bukit Timah, this is an important test of tutoring quality. A capable tutor should show how to use the provided formulas, distinguish them from topics not required in H1, and make the student explain why a formula applies to a particular question. Remembering where the formula is printed is not the same as selecting it correctly.
Quick answer: the official MF27 List of Formulae and Results is provided in the Mathematics examinations listed by SEAB, including H1 8865. It includes useful binomial distribution, sample variance and regression formulas. But MF27 also contains material used by other Mathematics syllabuses. Match every formula to the official H1 content outline and the problem’s assumptions before using it.
Quick reading route: what MF27 is · selection routine · binomial formulas · sample variance · regression · FAQs.
What MF27 is and is not
The SEAB 2027 A-Level syllabus directory links MF27 as the list of formulae and results for H1 Mathematics 8865, H2 Mathematics 9758 and H2 Further Mathematics 9649. The booklet’s own cover also notes broader A-Level Mathematics use.
That means the formula list is not a syllabus checklist. It is a common reference containing results useful across several levels and courses. H1 content is separately defined by the 2027 H1 Mathematics 8865 syllabus.
For example, MF27 displays a section on Maclaurin series, certain trigonometric results, vectors and numerical methods. Those topics should not be assumed to be examinable in H1 simply because they appear in the shared booklet. Check what the H1 syllabus lists as included or excluded.
The list also contains formulas relevant to H1, including the binomial distribution’s probability, mean and variance; the usual unbiased sample variance estimate; and calculations for correlation and regression. A sensible student learns to recognise the relevant section quickly without treating unrelated sections as compulsory revision.
- Formula booklet: a reference made available in the exam.
- Syllabus: the official description of what can be assessed at H1 level.
- Question instruction: the specific task, including any requirement for working, precision or a graph sketch.
- Mathematical model: the assumptions that make a formula appropriate.
- Graphing calculator: a separate tool for evaluating numerical probabilities, functions and other calculations as allowed.
- Student’s explanation: the reasoning that connects the model to the question’s requested answer.
What the official H1 examination says
The 2027 H1 Mathematics syllabus 8865 confirms that candidates are given a list of formulae and results. It also states that an approved graphing calculator without computer algebra system is expected, and that unsupported GC answers are generally allowed unless the question says otherwise.
Where unsupported calculator answers are not allowed, candidates must show mathematical steps using mathematical notation rather than just the commands entered. The syllabus also instructs candidates to sketch graphs when they use graphs to find solutions.
The published 2027 paper has one compulsory three-hour, 100-mark assessment, split into 40 marks of Pure Mathematics and 60 marks of Probability and Statistics. An integrated real-world application question worth at least 12 marks may connect topics.
This article refers to the named 2027 examination. A student entering JC1 in one year may sit their A-Levels in another, so check the syllabus and formula-list arrangements for the actual examination year rather than treating a past document as permanent.
A four-step formula selection routine
- Define the requested quantity. Is the question asking for a probability, a count, an average, a variance, a regression prediction or something else?
- Identify the model and assumptions. For binomial, check fixed independent trials and constant p; for regression, identify which variable is predicted.
- Choose a syllabus-appropriate formula. Refer to the booklet only after the model is justified and match the symbols to the question.
- Substitute, calculate and interpret. Check whether the result’s size, sign, units and precision agree with the story.
This routine prevents the classic error of finding a formula that looks related and forcing the story into it. For example, the presence of the word ‘mean’ does not automatically mean the unbiased sample variance formula is appropriate; the question may concern the theoretical mean of a binomial count instead.
Similarly, a question containing paired data does not by itself say whether the student should calculate a correlation coefficient, a regression line predicting y from x, or a regression line predicting x from y. Read the requested output first.
Worked example one: the binomial formulas
MF27 contains the binomial probability formula for X ~ B(n,p), together with mean np and variance np(1 − p). These are genuinely relevant to the 2027 H1 syllabus’s binomial content.
Suppose a hypothetical model has n = 5 independent trials and success probability p = 0.3. The random variable X counts successes. Thus X ~ B(5,0.3).
The expected success count is E(X) = np = 5(0.3) = 1.5. The variance is Var(X) = np(1 − p) = 5(0.3)(0.7) = 1.05.
Those two formulas are easy to locate. But they do not by themselves answer a different question such as ‘What is the probability of exactly two successes?’ That event needs the binomial probability expression.
For exactly two, write P(X = 2) = 5C2 × 0.3² × 0.7³ = 10 × 0.09 × 0.343 = 0.3087.
Why the combination term is there
A set of five trials can produce two successes in multiple positions. The combination 5C2 counts the ten ways to choose which two trial positions are successes. The other factors give the chance of one specified sequence under the independent constant-p model.
A formula sheet can display 5C2 in general notation, but the student must know what its count represents. If the problem involves dependent draws without replacement from a small bag, the binomial model’s conditions may fail even though the same-looking symbols appear elsewhere.
The word ‘exactly’ changes which booklet formula you need
If the problem says ‘exactly two’, compute P(X = 2). If it says ‘at most two’, add P(X = 0), P(X = 1) and P(X = 2), or use the appropriate cumulative function. If it says ‘at least two’, use 1 − P(X ≤ 1) when convenient.
The booklet does not interpret those English phrases. Our binomial distribution tutorial and graphing calculator guide explain how formula choice and cumulative-probability functions cooperate.
A strong student should be able to describe in words which counts are allowed before calculating the result. This protects against using a correct formula with an incorrect boundary.
How to use the binomial mean and variance in reverse
The formulas can also be used backwards. Suppose a binomial variable has mean 6 and variance 4.2. From np = 6 and np(1 − p) = 4.2, divide the latter by the former to obtain 1 − p = 0.7.
Thus p = 0.3 and n = 6/0.3 = 20, giving X ~ B(20,0.3). Check that twenty is a valid integer trial count and that 20(0.3)(0.7) = 4.2.
The binomial parameter reconstruction guide develops this method, including conditions under which the supplied theoretical mean and variance cannot describe a binomial distribution. The formula list is useful because it supplies the starting relationships, but the algebra and validation belong to the learner.
Worked example two: unbiased sample variance
MF27 contains a formula for the usual unbiased sample variance estimate. Its useful form is s² = [Σx² − (Σx)²/n]/(n − 1) for n greater than one.
Suppose five invented observations are 4, 6, 8, 10 and 12. Their sum is Σx = 40 and their sum of squares is Σx² = 16 + 36 + 64 + 100 + 144 = 360.
- Record n = 5, Σx = 40 and Σx² = 360.
- Compute (Σx)²/n = 40²/5 = 1600/5 = 320.
- Subtract: Σx² − (Σx)²/n = 360 − 320 = 40.
- Divide by n − 1 = 4: s² = 40/4 = 10.
- If asked for sample standard deviation, take the positive square root, approximately 3.162.
Notice that Σx² means square each observation before summing. It is not the same as (Σx)², which here equals 1600. Mistaking the two symbols changes the answer enormously.
The n − 1 denominator is part of estimating the unknown population variance from a sample under the usual conditions. It should not be applied blindly whenever a set of numbers appears; identify what the question wants to estimate.
A shifted-data variance question
The 2027 H1 syllabus explicitly includes unbiased estimates from summarised totals expressed as Σ(x − a) and Σ(x − a)², where a is a convenient reference value.
Take five synthetic values 9, 11, 14, 15 and 16, and subtract a = 10 from every observation. The shifted values are −1, 1, 4, 5 and 6. Their sum is 15 and squared sum is 79.
The mean of shifted values is 15/5 = 3; therefore the original mean is 10 + 3 = 13. The corrected squared-deviation sum is 79 − 15²/5 = 79 − 45 = 34, and the unbiased sample variance is 34/4 = 8.5.
A fixed shift leaves sample variance unchanged. That is why the same variance formula works using the shifted observations. The summarised sample variance guide includes more detailed explanations of this link.
The formula sheet does not select the appropriate variance
An H1 question may involve the theoretical variance of a binomial variable, the variance of a normal random variable or the unbiased variance estimate from sample observations. These are not the same task.
If X ~ B(5,0.3), use np(1 − p) = 1.05. If X ~ N(70,100), the population variance is already 100, because the second parameter is σ². If five sample observations are supplied and the question asks for an unbiased estimate, use the sample variance calculation with n − 1.
The repeated word ‘variance’ is not enough to decide which formula applies. The random variable, given information and target quantity determine the correct choice.
Worked example three: correlation and regression
MF27 also contains formulas involving the estimated product moment correlation coefficient and a regression line. The 2027 H1 syllabus includes using scatter diagrams, interpreting r, calculating least-squares regression and making responsible predictions.
Consider the hypothetical paired dataset x = 1, 2, 3, 4, 5 and y = 3, 5, 4, 8, 10. These numbers are made up for teaching and do not measure actual student performance.
The mean x-value is x̄ = 3, and the mean y-value is ȳ = 6. The sum of squared deviations of x is Sxx = 10, the corresponding sum for y is Syy = 34, and the paired-deviation sum Sxy = 17.
The product moment correlation coefficient is r = 17/√(10 × 34) ≈ 0.922, indicating a strong positive linear association for these five illustrative pairs.
For predicting y from x, the least-squares slope is b = Sxy/Sxx = 17/10 = 1.7. The intercept is a = ȳ − bx̄ = 6 − 1.7(3) = 0.9. Thus ŷ = 0.9 + 1.7x.
At x = 4.5, the fitted line predicts ŷ = 0.9 + 1.7(4.5) = 8.55. This is interpolation because 4.5 falls inside the observed predictor range 1 through 5.
What MF27 cannot decide for you
The booklet supplies algebraic relationships, but the student must decide which variable is being predicted, whether a linear model looks plausible and whether a proposed new x-value is within the observed range.
The same paired data can be used to predict x from y, but the least-squares regression of x on y is not generally obtained by rearranging the line of y on x. The direction matters because the two regressions minimise different residuals.
Our correlation and regression guide uses the same small dataset to explain these ideas. A graphing calculator can fit the line, but it does not guarantee that a prediction outside the data range will be reliable.
Why seeing a formula in MF27 does not make it H1 examinable
MF27 includes material for several syllabuses. The presence of a Maclaurin expansion, vectors or the Poisson and geometric distributions in the shared booklet does not establish that every such topic is listed in 8865.
Look at the actual H1 2027 content outline: its named probability models are binomial and normal distribution, together with sampling and the other specified statistics topics. For the 2027 H1 syllabus, do not promote the shared booklet’s additional discrete models into compulsory H1 chapters without checking the content requirements.
This distinction also affects tuition quality. A tutor who regularly teaches unfamiliar H2-only content to a new H1 learner should explain its clear educational purpose. Otherwise it may distract from the assessed H1 skills still needing repair.
A practical syllabus-first rule
Make a two-column study note. In the first column list topics explicitly included in the student’s H1 syllabus. In the second identify formula-list pages that support those topics. Do not reverse the process by assuming that every formula printed implies an H1 learning requirement.
If the paper for the student’s eventual examination year is revised, update the note using the newly applicable official documents. This prevents a guide created for an earlier cohort from becoming an unintended source of incorrect exam preparation.
The role of binomial expansion in the shared booklet
The MF27 booklet contains a binomial expansion identity as part of its broader mathematics content. The word ‘binomial’ there refers to expanding expressions like (a+b)n, which is not the same task as modelling a count with B(n,p).
These two ideas use combinations, but they have different purposes. An expansion expresses powers of a sum; a binomial probability model counts successes across independent yes-or-no trials with a fixed success probability.
The 2027 H1 syllabus includes permutation and combination notation for counting and B(n,p) for probability. A student should not assume a broad algebraic binomial-expansion chapter is required merely because its identity appears on the shared booklet.
The role of the graphing calculator
A formula booklet and a GC solve different problems. The booklet helps retrieve an identity or formula, while the calculator evaluates suitable numerical expressions and distributions or displays graphs.
For binomial probabilities, the calculator can obtain a cumulative value without manually summing every count, but the learner must still specify n, p and the correct event. For normal questions, the learner must identify the mean, variance, standard deviation and the intended tail.
For a regression, the calculator can process paired data but may return a plausible equation for the wrong column order if the pairs were entered incorrectly. A scatter plot and sensible estimate of slope direction should come first.
The graphing calculator accuracy guide shows why a screen result can be mathematically valid yet answer the wrong wording or be hidden by a poor viewing window.
How to use the booklet in a long mixed examination
The first task in a mixed question is not ‘Find a page number’. It is to identify the model and requested quantity. Once the learner knows it is a binomial count, an unbiased sample variance or a regression prediction, the relevant formula becomes much easier to locate.
Use the booklet to reduce unnecessary memory load, but avoid interrupting every elementary algebra step to search pages. Frequent flipping can consume time when a student has not internalised the meaning of the most-used relationships.
A practice strategy is to work a short mixed set using the official booklet, record which sections are actually needed, then repeat with changed questions. This builds a natural retrieval map rather than a memorised sequence of page numbers.
What is not automatically printed for every problem
A formula list cannot define a random variable in the context of a given experiment. It cannot select which events count as ‘at least two’ or justify a survey’s sample selection. It cannot tell a learner whether a curve is above or below another without interpreting the actual function.
Those decisions are central to H1 Mathematics because the official assessment objectives include selecting a relevant strategy, forming mathematical models and interpreting results. A reference page helps with formulas; it does not replace mathematical thinking.
A strong tutor asks the student to explain the method before searching the formula. After the calculation, the student should be able to tell the story behind the result, including limits of the model.
A five-minute formula audit
- Open the actual MF27 booklet, not a photograph of an unrelated formula sheet.
- Find the binomial distribution’s probability, mean and variance entries.
- Locate the usual unbiased sample variance expression and identify n, Σx and Σx².
- Find the product moment correlation and regression formulas; note which variables are being predicted.
- Identify one booklet section that is not named in the current H1 syllabus and explain why its presence does not create an examination requirement.
- Choose one short H1 question and state which, if any, booklet relationship is actually helpful.
- After solving, verify the numerical result and its interpretation independently.
The aim is not to memorise every symbol in the shared booklet. It is to use a small number of relevant entries confidently while respecting the syllabus boundary.
An example of a wrong shortcut and its repair
Suppose a question asks for the unbiased sample variance of 4, 6, 8, 10 and 12. A learner sees the binomial mean-and-variance table on MF27 and tries to apply np(1 − p). But no trial count or success probability is defined. That is a model error.
The correct approach is to recognise five sample observations and compute the sample variance estimate, obtaining s² = 10. A good tutor’s repair is not to insist ‘Use page 6 instead of page 5’, but to explain why the target quantity is an estimator from data rather than the theoretical variance of a binomial count.
Now reverse the story: if X ~ B(5,0.3), using the sample variance estimator on the numbers 5 and 0.3 makes no sense. The required theoretical variance is np(1 − p) = 1.05. The formulas answer different questions despite sharing statistical vocabulary.
A hypothetical 90-minute tutorial plan
Begin with ten minutes identifying the relevant H1 syllabus sections and comparing them with the shared formula booklet. Let the student point to three useful entries and one entry that belongs to a different syllabus context.
Use twenty minutes to solve a binomial question and an inverse parameter problem, explaining each symbol. Then spend twenty minutes on sample variance using raw and summarised data.
The following twenty minutes can compare scatter plots, r and regression predictions. Close with ten minutes of graphing calculator checking and ten minutes of an unseen mixed question where the tutor does not reveal the appropriate booklet section.
This is an illustrative lesson design. Real tutoring should follow the JC’s taught topics and identify what a particular student can perform independently. The main goal is method selection, not the number of formula pages visited.
The established Bukit Timah tuition format
The immutable eduKateSG small-group reference describes three-pupil tutorials in a typical 90-minute weekly format near Sixth Avenue MRT, at 8 Fourth Avenue, Singapore 268674. It serves as a teaching-format reference, not a promise that a particular JC1 H1 Mathematics group is presently available.
In a three-student setting, one learner may recognise the formula but not the probability event, another may confuse Σx² with (Σx)², and a third may fit a regression without identifying the response variable. The tutor should give each learner an appropriately different checking task.
Parents should confirm the tutor’s current H1 syllabus familiarity, fees, availability and group fit. Bringing a real marked question can reveal whether the student chose the wrong formula because of vocabulary, conceptual knowledge or a rushed calculation. See consultation details.
Seven days to become comfortable with MF27
- Day 1: read the official syllabus outline and mark which MF27 sections support the taught topics.
- Day 2: find the binomial mean and variance entries and explain what n and p mean.
- Day 3: solve a binomial probability event with the formula and check the cumulative event on the GC.
- Day 4: calculate an unbiased sample variance from five raw values and then from summary totals.
- Day 5: identify the correlation coefficient and regression line formulas in a simple dataset.
- Day 6: attempt three mixed unlabelled questions, selecting a formula only after naming the mathematical model.
- Day 7: complete a short timed set with the booklet and record which formula lookups were useful or unnecessary.
Use changed numbers on the final day. Recognition of yesterday’s worked example is not proof that the learner can choose the correct formula in a new story.
Six independent self-check questions
One: the binomial mean
For X ~ B(5,0.3), E(X) = 1.5. The result need not be a count that occurs in a single trial set.
Two: the binomial variance
For the same model, Var(X) = 5(0.3)(0.7) = 1.05. Do not confuse this with standard deviation √1.05.
Three: exactly two
P(X = 2) = 5C2(0.3)²(0.7)³ = 0.3087. The combination represents ten possible positions for the two successes.
Four: sample variance
For n = 5, Σx = 40 and Σx² = 360, the unbiased sample variance is [360 − 40²/5]/4 = 10.
Five: regression slope
For the illustrative dataset with Sxx = 10 and Sxy = 17, the least-squares slope for y on x is 1.7. It measures predicted y change per unit increase in x under the fitted linear model.
Six: syllabus scope
Does seeing a Poisson distribution formula in MF27 mean the 2027 H1 8865 syllabus automatically assesses Poisson distribution questions? No. MF27 serves multiple syllabuses; use the H1 content outline to determine what is examinable.
One last comparison: the same number can mean different things
Suppose a question contains the number five. It might mean five independent trials, five observations in a raw sample, five paired records in a regression table, or a cutoff measured in original units. The number has no mathematical role until the problem defines it. In a binomial model, n = 5 tells us there are five trial opportunities. In a sample variance question, n = 5 tells us the number of observations and leads to an n − 1 divisor of four for the usual unbiased estimate.
In a regression problem with five paired observations, each of those records contributes to the mean and deviation sums, but the task may be to predict y from x, not to compute a variance. The symbols on a formula page make sense only after the student labels the quantities in the actual data story.
This is a useful independent check during tuition: hand the learner three different questions containing the same number, then ask them to explain why three different formula-list entries are appropriate. Success means recognising the mathematical structure—not merely spotting a familiar number and searching for a matching expression.
Frequently asked questions
Will I be given the MF27 formula list in the H1 exam?
The official 2027 syllabus states candidates are provided with a list of formulae and results, and SEAB’s 2027 subject directory links MF27 for H1 Mathematics 8865.
Does everything on MF27 belong to H1 Mathematics?
No. The booklet is shared across several Mathematics syllabuses. The H1 syllabus, not the shared booklet, determines H1 examination scope.
Is the binomial formula on MF27?
Yes. The list includes the binomial distribution probability expression as well as the mean np and variance np(1 − p).
Is the unbiased sample variance formula available?
Yes. MF27 contains the usual sample variance estimate involving n, Σx and Σx². Students still need to identify the correct totals.
Are correlation and regression formulas included?
Yes, the booklet includes relevant correlation and regression relationships. The question determines which variable should be predicted and whether a linear fit is appropriate.
Can I use a Poisson formula because it is printed?
Not as evidence that Poisson is part of H1 8865. Confirm content from the syllabus; the formula sheet includes topics used in other Mathematics courses.
Can the booklet replace a graphing calculator?
No. The formula list is a reference to mathematical relationships. The approved non-CAS graphing calculator is a separate computational and graphical tool expected by the syllabus.
What should parents ask a tutor about MF27?
Ask the tutor to show a real H1 question, choose the relevant entry if needed, explain why it applies and check an independent changed problem afterwards.
One habit that makes the booklet genuinely useful
Name the mathematical job before locating the formula. Is it a count, a probability, an estimate of population variance or a prediction from paired observations? Once the learner knows the job, MF27 becomes a tool rather than a source of confusion.
Use the official MF27 formula list alongside the 2027 H1 Mathematics 8865 syllabus. For broader course planning, read our Bukit Timah H1 Mathematics overview and H1 8865 paper strategy guide. Verify the documents for the student’s eventual A-Level examination year.