Cambridge IGCSE Statistics (0479) teaches students how to collect data, choose meaningful summaries, interpret probability and evaluate conclusions drawn from numbers. It is a new Cambridge subject first examined in June 2027, with twelve syllabus topics and two equally weighted written papers. This guide explains exactly how the examination works, what students need to understand and how to prepare effectively.
The important skill is not merely applying a formula. A student can calculate an average correctly yet make a misleading claim because the sample was biased. Another can draw a scatter diagram but mistake correlation for proof of causation. Strong Statistics students learn to ask what was measured, how the data were obtained, which method is suitable and what the result actually supports.
Quick student reference: – Cambridge syllabus: IGCSE Statistics 0479, first examination 2027, June series. – Paper 1: 2 hours 15 minutes, 100 marks, 50%. – Paper 2: 2 hours 15 minutes, 100 marks, 50%. – Entry: both papers compulsory, grades **A*–G; questions in both may draw on all twelve topics. – Calculator: a permitted ordinary calculator should be used on both papers; algebraic and graphical calculators are not permitted under the 2027 syllabus. – Written method: essential working must be shown for full marks. – Availability: in 2027, only administrative zones 3 and 4, June series. From 2028, Cambridge lists zones 3, 4, 5 and 6, under a separate 2028–2030 syllabus. – Main topics:** data collection, displays, frequency, averages, spread, transformations, probability, rates, index numbers, relationships and time series.
Start with the correct subject code and exam year
The Cambridge IGCSE Statistics qualification is 0479. It is not another name for Mathematics 0580, International Mathematics 0607 or a statistics chapter in an O Level paper. Those courses contain relevant mathematics but have different content and examination arrangements.
Cambridge introduced Statistics 0479 as a separate syllabus for first examination in 2027. Its 2027 specification was followed by a separate 2028–2030 publication. Candidates should therefore select their learning resources according to the year they will sit the examination, not when lessons began.
The official Cambridge IGCSE Statistics page confirms the new course and administrative-zone availability. The 2027 syllabus PDF establishes the examination requirements explained here.
The school or centre should confirm actual availability, deadlines and permitted examination equipment. A student preparing in a country where the 2027 paper is not offered should not assume Cambridge will create a local sitting automatically.
What studying Statistics really means
Statistics is about learning from data while respecting uncertainty and the limitations of evidence. It combines collection, organisation, description, comparison, probability and interpretation.
Consider a claim that most students prefer a particular study method. Before trusting it, ask who was surveyed, how they were selected, what question they answered and whether the results represent the relevant student population.
Averages and graphs can help summarise results. But they do not automatically tell us whether the survey was fair or what caused an observed pattern.
Students should be comfortable explaining ordinary evidence rather than treating Statistics as only a set of calculations. Good examination answers may include a numerical result, a sentence interpreting it and a limitation or assumption.
The course provides preparation for research, business, health data, science and many other fields. It teaches disciplined reasoning rather than the ability to generate impressive-looking charts.
The twelve Cambridge syllabus topics
Cambridge lists twelve compulsory areas, and both examination papers can test any part of them.
- 1. Data and its collection: populations, samples, surveys, representativeness and types of data.
- 2. Representation of data: tables, charts, Venn diagrams, box plots and stem-and-leaf diagrams.
- 3. Frequency distributions: grouped data, histograms, frequency polygons and cumulative frequency.
- 4. Measures of central tendency: means, medians, modes and the reasons for choosing each.
- 5. Quartiles, percentiles and measures of dispersion: spread, interquartile range and standard deviation.
- 6. Transformations of data sets: effects of shifting or scaling values and comparison of distributions.
- 7. Probability: events, rules, tree diagrams, independence and simple conditional cases.
- 8. Probability distributions: outcome probabilities and expectation.
- 9. Crude and standardised rates: comparisons across populations with different structures.
- 10. Index numbers: price relatives, weighted aggregate indices and their limitations.
- 11. Bivariate distributions: scatter diagrams, correlation and lines of best fit.
- 12. Time series: trends, moving averages, seasonal effects and cautious forecasting.
Students should study relationships across the areas, not twelve disconnected definitions. For instance, interpreting a time series can require a graph, an average and a judgement about whether prediction is reasonable.
Two compulsory papers with the same structure of weighting
Paper 1 and Paper 2 are both written examinations lasting 135 minutes and carrying 100 marks each. Each contributes 50% towards the qualification. Cambridge does not label one the theory paper and the other the practical paper.
Both contain questions of varied length and style, and both can assess material from the entire syllabus. There are no separate Core and Extended question sets in the published 2027 assessment overview.
The student should be ready to recognise familiar techniques inside unfamiliar contexts. A question about rates might involve populations rather than classroom surveys; a time-series question could involve sales rather than rainfall.
A useful preparation rule is to mix topics once foundations are secure. If every question appears under a chapter title, the student may rely on the title to choose a method and struggle on an actual mixed paper.
Assessment objectives: technique and interpretation
Cambridge uses two main objectives. AO1 Knowledge and understanding of statistical techniques accounts for approximately 80–90% of the qualification. It includes calculation, procedures, probability and statistical representation.
AO2 Interpretation, communication and problem-solving contributes approximately 10–20%. It tests the student’s choice of methods, explanation of results, understanding of assumptions and ability to communicate meaningful conclusions.
The percentages do not mean interpretation is unimportant. A short interpretive step can determine whether a calculation answers the question correctly.
A student might correctly compute a mean but fail to say what it represents. Another might choose an unsuitable method even though their arithmetic is accurate.
Revision should therefore end each worked example with a brief explanation: What does this number say about the context, and what can it not tell us?
Calculator requirements: don’t bring the wrong device
A permitted calculator should be used on both Statistics papers. Cambridge explicitly states that algebraic and graphical calculators are not permitted in the 2027 assessment.
This differs from Cambridge IGCSE International Mathematics 0607, whose calculator rules and paper types are different. Do not confuse qualifications when buying equipment.
The 0479 syllabus also lists a pair of compasses, a protractor and a ruler for both papers, with relevant provisions about tracing paper. Centres should check the most recent examination materials list for the actual series.
Learn how to use a permitted calculator for ordinary calculations and statistical work while still recording essential working in the booklet. A final decimal from the machine is not a substitute for the method where the question requires it.
Practise with the exact model approved by the centre, rather than relying on an unsuitable graphing device.
Showing working and choosing precision
The paper states that essential working must be shown for full marks. That means students should write enough steps to reveal their method, especially in multi-stage calculations.
If a question asks for a mean from frequency data, show the weighted total and the total frequency before division. If it asks for a standard deviation, identify the formula and how the numerical values were substituted.
Use a sensible number of significant figures or decimal places according to the instructions. Rounding intermediate values too early can distort a later percentage or comparison.
Check units before writing the final answer. A population rate expressed per thousand is not the same as a raw number of events.
A reliable habit is method → substitution → result → interpretation. The last step may be brief, but it prevents the learner from leaving a correct computation detached from the question.
Topic 1: population, sample and census
A population is the complete group or collection relevant to the statistical question. A sample is a subset used to obtain information. A census attempts to examine every relevant member.
Suppose a school wishes to understand the commute methods of its 1,200 students. The population for that question is the school’s students, not every teenager in the country. A survey of 120 selected students is a sample.
The sampling method matters. Surveying only students arriving before 7 am may give a distorted view because those students may use different transport from late arrivals.
A good answer identifies the intended population, explains the sample selection and considers whether the process could exclude important groups.
A larger sample can be useful, but sample size alone does not guarantee representativeness. The selection process is equally important.
Sampling methods: understand the mechanism
The syllabus includes simple random, systematic, stratified and quota sampling. These methods differ in how participants or items are selected.
A simple random sample gives an appropriate equal chance of selection using a defined sampling frame. A systematic sample selects at regular intervals after a suitable start. A stratified sample divides a population into relevant groups and selects within those groups. Quota sampling fills specified group targets using a different selection approach.
A student should not memorise only definitions. Consider what the method is designed to represent and what bias may remain.
For example, stratifying a school by year group can help preserve the proportion of students in each year, provided selection within the groups is handled appropriately.
Cambridge’s 2027 syllabus also includes selecting samples using supplied random-number tables. Practice the actual procedure rather than only explaining it in words.
Random numbers and representativeness
A random-number table can support participant selection when each population member has been identified appropriately. It does not create a valid sample if the list itself excludes important members.
Imagine a club with 80 members numbered 01 through 80. A random-number table can be read according to the problem’s instructions, rejecting numbers outside the defined range or duplicates when sampling without replacement.
The sample still represents the club, not every person in the region. Generalisation beyond the sampling frame would need additional justification.
A useful exam practice asks the student to select a small sample from a numbered list and explain why a different proposed selection method might be biased.
This trains both procedure and interpretation. Randomisation is a method, not a magic guarantee that every possible limitation disappears.
Open and closed survey questions
An open question allows a respondent to express an answer in their own words. A closed question offers defined alternatives or a specified response structure.
Open questions can reveal unexpected ideas but may be harder to classify consistently. Closed questions are often easier to compare but can restrict what respondents are able to express.
A poor survey may ask “You agree the new timetable is excellent, don’t you?” This pushes the respondent towards a particular answer. A more neutral question would make the options and wording less leading.
Students should evaluate whether a proposed question genuinely measures the topic of interest. If the answer categories overlap or fail to include plausible responses, the resulting data may mislead.
Designing a fair question is part of Statistics, not merely a preliminary administrative exercise.
Qualitative, quantitative, discrete and continuous data
Qualitative data describe categories, such as a preferred transport type. Quantitative data concern numerical quantities. Quantitative variables may be discrete, based on countable outcomes, or continuous, measured on a scale.
The number of people attending a class is discrete in the usual context. Time taken to travel can be continuous, even when recorded to the nearest minute.
A student should know which representation is suitable for the variable. A pie chart may be appropriate for categories, while a grouped histogram can suit continuous measurements.
Be careful about numbers used merely as labels. A postal code or student identification number is not meaningfully averaged simply because it consists of digits.
The correct classification depends on what the variable means, not only how it looks on the page.
Topic 2: choosing a representation
Cambridge includes frequency and two-way tables, bar charts, pie charts, Venn diagrams, box-and-whisker diagrams and stem-and-leaf displays.
A display should help answer a question. Bar charts can compare categories. A box plot summarises central position and spread. A stem-and-leaf display can preserve individual values in a small data set.
Not every representation suits every variable. A pie chart about mutually exclusive categories needs a meaningful whole, and the scale must match the interpretation.
Students should read the title, axes, units and key before making a statement. A truncated scale or unclear category description can exaggerate differences visually.
A good exam answer can explain why this display is useful and recognise one limitation, not just draw a tidy chart.
Two-way tables and conditional descriptions
A two-way table relates two categorical variables. For instance, a school table might show travel methods by year group.
The total in a row is different from the total in a column. A percentage calculated using the wrong denominator can reverse the meaning of a comparison.
Suppose 24 of 60 students in Year A travel by bus, while 30 of 100 students in Year B do so. The shares are 40% and 30%, respectively. Year B has more bus users numerically, but Year A has the higher proportion.
The correct interpretation depends on whether the question asks for count, percentage of a group or percentage of the whole sample.
Practise writing the denominator in words before calculating. That small habit prevents many avoidable errors.
Box-and-whisker and stem-and-leaf plots
Box plots display features such as median, quartiles and range, helping readers compare distributions. Stem-and-leaf diagrams can preserve actual values while showing a distribution’s shape.
A student should not automatically interpret a long whisker as proof of an incorrect result. It may reflect greater spread in that part of the observations.
When comparing box plots, consider both central tendency and variability. Two groups may have the same median but different interquartile ranges.
For stem-and-leaf diagrams, use the key and place value accurately. A stem of 2 with a leaf of 7 might represent 27 under one key and 2.7 under another.
The 2027 syllabus notes that formal identification of outliers is not required in this representation section. Follow the exact content rather than borrowing every rule from a more advanced Statistics course.
Topic 3: grouped frequency distributions
Grouped distributions summarise many observations into intervals. This can make patterns easier to see but loses some information about the exact individual values.
If travel times are recorded in groups such as 10 to under 20 minutes and 20 to under 30 minutes, a student needs to understand the interval boundaries and avoid double counting.
The class midpoint can support estimates of a grouped mean. The class width matters especially when constructing histograms.
An ungrouped frequency table can preserve exact observed values, while grouped data are often better for displaying a large continuous range.
A good answer recognises the trade-off between simplicity and detail. A grouped estimate is not the same as a calculation from every original value.
Practise class limits, boundaries, widths and midpoints with several different tables.
Histograms: frequency density matters
For unequal class widths, a histogram’s bar area, rather than height alone, represents frequency. The vertical axis is frequency density.
Consider two classes: an interval of width 5 with frequency 20, and an interval of width 10 with frequency 20. Their frequency densities are 4 and 2. Equal frequencies do not produce equal heights when the widths differ.
A student who draws both bars with the same height has confused histogram rules with an ordinary bar chart.
Always label the horizontal intervals accurately and use the class boundaries supplied. Adjacent continuous intervals usually touch in the histogram.
A useful practice routine computes frequency density = frequency ÷ class width before drawing the bars. Then check that area recovers the intended frequency.
This is a highly transferable skill for grouped data questions.
Frequency polygons and cumulative frequency
A frequency polygon connects suitably plotted frequency information, often using class midpoints. Cambridge’s specified comparison tasks use equal-width classes for the frequency-polygon situations described in the syllabus.
Cumulative frequency accumulates counts up to specified boundaries and can help estimate medians, quartiles and percentiles for a grouped distribution.
Students should distinguish ordinary frequency from accumulated frequency. If the first four intervals contain 3, 5, 7 and 4 observations, the cumulative totals are 3, 8, 15 and 19.
A cumulative curve is not a picture of how quickly each individual event occurred. It displays how many observations fall up to a value under the distribution’s definition.
Practice converting a grouped table to cumulative counts before attempting graphical interpolation. A small addition error can propagate through an entire curve.
Topic 4: mean, median and mode
The mean is calculated by dividing the total of values by their number. The median is the central value after ordering. The mode is a most frequent value.
Consider 2, 3, 3, 4, 18. The mean is 6, the median is 3, and the mode is 3. The high final value pulls the mean upwards.
No single measure is always best. A mean uses every value but can be influenced by extremes. The median can be useful when a distribution is skewed. The mode is relevant to the most common category or value.
When comparing groups, explain why the chosen measure answers the specific question. A statement that one school has a higher average needs to identify which average and its meaning.
Accurate calculation is necessary, but interpretation completes the statistical answer.
Grouped estimates: understand what is approximate
For grouped data, the exact individual values may not be available. A grouped mean often uses class midpoints, and a grouped median can be estimated from cumulative frequency or suitable interpolation.
Suppose a class interval contains several observations. Using its midpoint assumes a representative position within the class for estimation; it does not reveal every true value.
Students should label a grouped result as an estimate when appropriate. A calculator may produce many decimal places, but that does not restore information lost through grouping.
A strong answer states the method and recognises the assumption behind it. If two groups have different interval widths, the method and display must respect that structure.
Practice one question using raw values and one using grouped values, then explain why their results have different levels of certainty.
Topic 5: quartiles and percentiles
Quartiles divide ordered data into four sections by position under the relevant convention. Percentiles indicate position relative to a distribution.
The interquartile range, IQR, is upper quartile minus lower quartile. It summarises the spread of the middle portion of data and is less sensitive to extremes than the full range.
Students should use the quartile convention and interpolation instructions appropriate to the actual question and syllabus. Different educational resources may use slightly different positional methods, so an unexplained mismatch between guides can be confusing.
If a student’s score lies near a specified percentile, that is a statement about relative position in the comparison group. It does not automatically reveal the proportion of marks achieved.
Interpret the distribution and its population before using a percentile as a judgement about performance.
Standard deviation: what variation means
Standard deviation measures spread around a mean. Under the population-style formula used in the 2027 syllabus examples, it is the square root of the mean squared deviation from the mean.
For the values 2, 4, 6, the mean is 4. The squared deviations are 4, 0 and 4, giving a variance of 8/3 and a standard deviation of approximately 1.63.
The number has the same units as the original observations. It is a measure of dispersion, not a measure of whether every observation is correct.
Students must know the formulae specified by Cambridge and show essential working. Calculator use should support the method rather than conceal it.
A good comparison explains whether one distribution tends to have more variable values, while recognising that groups may also differ in their means.
Topic 6: transformations of data sets
Adding a constant to every value shifts measures of location such as the mean and median by that constant. It does not change the standard deviation or interquartile range.
Multiplying every observation by a positive factor scales the mean and median by that factor, and scales the standard deviation and IQR by the factor’s magnitude.
For example, the values 3, 5, 7 have mean 5. Adding 10 creates 13, 15, 17, with mean 15 while the spread stays the same.
Multiplying the original values by 2 creates 6, 10, 14, with mean 10 and twice the original standard deviation.
The syllabus also covers combining distributions and scaling to a specified mean and standard deviation. Students should understand the mechanism instead of memorising a separate result for every measure.
Check whether the transformation is a shift, scale or change in membership of the data set.
Combining data sets: weight by the number of values
Suppose one class has 20 students with a mean score of 60, and another has 30 students with a mean of 70. The combined mean is not simply (60 + 70)/2.
The total scores represented by the means are 20 × 60 = 1200 and 30 × 70 = 2100. The combined mean is 3300 ÷ 50 = 66.
The group sizes matter because the means summarise different numbers of observations.
Students should also learn the Cambridge requirements for combining standard deviations and means, with careful interpretation of the information given.
A common error is assuming that summary statistics can be combined through unweighted averages regardless of sample size. The repair is to reconstruct totals and denominators before calculating.
The result 66 describes the combined group under the supplied assumptions, not either group separately.
Topic 7: probability and event notation
Probability describes uncertainty under a stated model. Students use notation for an event, its complement, union and intersection.
If P(A) = 0.3, then P(not A) = 0.7. The complement rule follows because the event and its complement exhaust the possible outcomes.
For two events, P(A or B) = P(A) + P(B) − P(A and B). The overlap is subtracted because it would otherwise be counted twice.
This addition rule is useful for overlapping categories in a survey or for events described by a Venn diagram.
A strong answer defines A and B before calculating. A probability without a clear event can be ambiguous even if the arithmetic is correct.
Use fractions, decimals or percentages consistently and check that the final probability lies between 0 and 1.
Mutually exclusive and independent are different
Mutually exclusive events cannot occur together on a single trial. Independent events are events where learning one outcome does not change the probability of the other under the relevant model.
A single fair die roll cannot be both a 2 and a 5, so those events are mutually exclusive. Two successive independent fair die rolls can both produce a 2.
Do not confuse the two relationships. Except in special zero-probability cases, mutually exclusive events with positive probabilities cannot also be independent.
For independent events, the probability of both occurring is the product of their probabilities. For mutually exclusive events, the overlap is zero.
A useful practice task asks the student to explain the physical situation before choosing an addition or multiplication rule.
Understanding the event relationship is more important than remembering a formula alone.
Tree diagrams and conditional situations
Tree diagrams show stages of an experiment and help organise probability calculations. A question may specify selection with replacement or without replacement, which changes whether later probabilities stay the same.
Suppose a bag contains three blue and two red counters. One is selected without replacement. If the first is blue, the second-stage bag contains two blue and two red. The probability of a blue second counter on that branch is 2/4.
If the first is returned before the second draw, the probability remains 3/5 under the original conditions.
The diagram should reflect the actual experiment. A student who writes identical second-stage probabilities in a without-replacement case may have misunderstood the setup.
Cambridge includes simple conditional probability but does not require memorising every advanced conditional formula. Follow the published scope.
Topic 8: probability distributions and expectation
A probability distribution assigns probabilities to the possible values of a random variable. The probabilities must sum to 1.
Imagine a simple original example where X takes values 0, 1 and 2 with probabilities 0.2, 0.5 and 0.3. The expected value is 0 × 0.2 + 1 × 0.5 + 2 × 0.3 = 1.1.
The expectation is an average across the model’s probability distribution. It is not necessarily a value that can occur on one individual trial.
Students should calculate and then interpret expected value in context. A result of 1.1 items per trial does not imply anyone receives one-tenth of an item in a single trial.
Before calculating, check that all outcomes are represented and that the probabilities form a valid distribution.
This idea supports decisions involving games, quality control and other uncertain outcomes.
Topic 9: crude and standardised rates
A crude rate expresses events relative to a population, often using a scale such as events per thousand people. For example, 50 events in 10,000 people correspond to 5 per thousand, if that is the requested rate.
But comparing crude rates can be misleading when populations have different age or other subgroup structures. A population with more older people may have a different crude rate even if subgroup-specific risks are similar.
A standardised rate uses a common reference population to make a more meaningful comparison across groups with differing compositions.
Students should understand that standardisation is a weighted comparison using defined subgroup rates and the chosen standard population. It is not an adjustment that makes every population identical.
A good answer explains what was controlled for and what the comparison still cannot establish.
An original standardisation example
Imagine Population A has 80% younger people and 20% older people. Population B has 40% younger and 60% older people. Even if each age group experiences the same event rate in both populations, the overall crude rates could differ because the age mix is different.
Standardisation applies a common set of weights to subgroup-specific rates. If the chosen standard population is 50% younger and 50% older, the standardised rate is the corresponding weighted average.
The example illustrates why a fair comparison sometimes needs more than one overall count.
It does not establish anything about the actual health or safety of a real country. The data are hypothetical and the conclusion depends on the specified event and groups.
Practise identifying which rate is being reported and which population forms the denominator.
Topic 10: price relatives and index numbers
An index number can show a value relative to a selected base, often set to 100. This helps compare changes across time.
Suppose a product cost $40 in a base year and $50 later. The price relative is 50/40 × 100 = 125. That indicates a 25% increase from the base-year price.
Index numbers help compare changing costs, but they depend on what items are included and how they are weighted.
An index of 125 does not mean the price increased by 125%. It means the indexed level is 125% of the base level.
Use the correct denominator and distinguish the index value from the percentage change.
The 2027 syllabus also includes weighted aggregate index numbers and interpretation of their limitations.
Weighted indices: why a basket matters
A weighted index reflects that different goods or services contribute differently to the expenditure or other total being represented.
Imagine a hypothetical basket where food has a greater share of spending than stationery. A change in food prices should generally influence an expenditure-weighted price index more than the same percentage change in stationery prices.
The exact weights come from the relevant base-year information under the syllabus method. Students should calculate the weights correctly and then apply them to the price relatives.
A weighted index remains an approximation of the selected basket’s change. It may not reflect the spending pattern of every household or business.
A strong evaluation asks whose consumption pattern the weights represent, how old they are and whether spending habits have shifted.
Numbers become meaningful only when the basis of the index is understood.
Topic 11: correlation and two-variable data
Bivariate distributions describe pairs of observations. A scatter diagram helps reveal whether two variables tend to change together.
A positive association means larger values of one variable often accompany larger values of the other. A negative association means they tend to move in opposite directions. A weak association shows less consistent pattern.
The meaning of a relationship depends on the variables and the context. If hours spent studying and a test score are positively associated in a sample, that does not by itself prove additional hours caused the score difference.
Other factors, selection and measurement could matter.
The student should be able to describe the direction and relative strength of an observed pattern without inventing precise results that the scatter diagram does not show.
Lines of best fit and semi-averages
The 2027 syllabus includes drawing a line of best fit by eye and using the method of semi-averages, as well as finding equations in the form y = mx + c.
A line of best fit summarises a relationship for the data. It is not a claim that every point lies on the line.
For semi-averages practice, students should follow the syllabus method and the specified condition of an even number of data pairs. Identify the two relevant summary points and construct the line accurately.
The slope and intercept should be interpreted in terms of the variables and units. A slope of 2 could mean dollars per item, kilometres per hour or a different rate depending on the question.
A good answer checks whether the fitted line makes sense across the observed range.
Prediction and the danger of extrapolation
Using a fitted relationship to estimate a value within the observed data range is different from extending it far beyond that range.
Suppose a short data set shows sales rising with advertising expenditure. A line of best fit could provide an estimate within the recorded range, but a prediction at ten times the original expenditure assumes that the relationship continues unchanged.
That assumption may be unreasonable. Businesses face limits, competing influences and changing customer behaviour.
Cambridge explicitly includes the assumptions and limitations of line-of-best-fit estimates, including extrapolation.
A strong response states the predicted value when requested, then explains whether the new input is inside or outside the observed range and why that matters.
Statistical confidence should follow evidence, not a straight line drawn beyond the page.
Topic 12: time series and moving averages
A time series records values in chronological order. It can reveal longer-term direction, regular seasonal changes and irregular fluctuations.
A moving average can smooth short-term variation to make an underlying trend easier to see. Students should know how to calculate and plot moving averages, including centring when appropriate.
Imagine monthly sales that rise and fall because of school holidays. Comparing only one month with the previous one may exaggerate or obscure longer-term change.
A moving average provides a different perspective, but it also loses some fine detail and may not capture sudden structural changes.
The 2027 syllabus includes trend lines, seasonal components and prediction using trend plus seasonal variation.
Students should label the time unit and distinguish observed history from forecast values.
Seasonal components and cautious forecasting
Seasonal variation describes patterns associated with repeated positions in a year or other cycle under the appropriate model.
A shop might sell more cold drinks during a hotter season. But the pattern could vary with location, promotions or other events, so a seasonal component is an estimate rather than a permanent rule.
The syllabus asks students to calculate seasonal components from suitable data and use them with a trend to make predictions.
A forecast should be interpreted with assumptions. If a new transport line opens or a major event changes demand, earlier seasonal patterns might no longer predict accurately.
Avoid saying an estimate is guaranteed simply because it came from a formula.
The useful skill is calculation plus explanation of the conditions under which the model is reasonable.
Common errors that cost Statistics marks
- Wrong denominator: comparing proportions with different group sizes incorrectly. Repair by labelling the population for each percentage.
- Sampling bias: treating convenient volunteers as representative without justification. Repair by examining selection.
- Frequency-density confusion: using frequency as histogram height despite unequal class widths. Repair with class-width calculations.
- Mean versus median: interpreting an average without checking skew or extreme values. Repair by comparing measures.
- Premature rounding: producing an inaccurate later calculation. Repair by retaining suitable precision.
- Mutually exclusive versus independent: selecting the wrong probability rule. Repair through event descriptions.
- Correlation as causation: presenting a relationship as proof of mechanism. Repair with an alternative explanation.
- Out-of-range forecast: using a line of best fit far beyond data without caution. Repair by identifying the observed range.
- Wrong syllabus year: assuming all 2028 rules equal the 2027 ones. Repair by checking the official document.
Each error should lead to a small targeted task and a later independent retry.
Repair, stabilise and extend
Repair is for students whose fundamentals are unreliable: fractions, percentages, graph scales, frequency and basic probability. Start with small correct examples and show how the calculations connect to a question.
Stabilisation suits students who can solve familiar chapter exercises but struggle to recognise which technique a mixed paper requires. Interleave topics and explain method choice before calculating.
Extension challenges students who have secure procedures to evaluate evidence, compare distributions, examine limitations and justify decisions without overclaiming.
A learner may be excellent in probability but weak in sampling, or confident with histograms but uncertain about crude rates. Diagnose the precise gap instead of assigning an overall label.
The aim is independent statistical judgement supported by accurate calculation.
A four-session weekly study routine
Session one — data and representation: collect or analyse a small safe dataset, choose a chart and explain why it suits the variable.
Session two — calculations: practise means, spread, probability or rates, showing essential working and checking units.
Session three — interpretation: read an unfamiliar graph or claim and separate observation from assumption or causal explanation.
Session four — mixed exam practice: complete questions from different topics, identify the first wrong step and reattempt a related task after correction.
Use a consistent manageable schedule. Full papers are useful closer to the exam but should not replace targeted learning.
A simple error log can be more valuable than many unreviewed worksheets.
Twelve-week revision roadmap
Weeks 1–2: confirm Cambridge 0479, the examination year and permitted calculator. Diagnose percentages, graphs, sampling and basic probability.
Weeks 3–5: strengthen topics 1–6, including frequency, central tendency, dispersion and transformations. Use real-world-style data without collecting private information.
Weeks 6–8: integrate probability, distributions, rates, index numbers, bivariate data and time series. Practise interpretation alongside calculation.
Weeks 9–10: use 2027 specimen materials for both 135-minute, 100-mark papers. Record timing and recurring errors by topic.
Weeks 11–12: revisit difficult methods, practise formula recall and show working clearly. Confirm centre instructions and calculator restrictions.
This is an illustrative schedule, not a Cambridge-required teaching order or a grade guarantee. The first 2027 examination means students should rely on specimen materials rather than expect a long archive of 0479 past papers.
Frequently asked questions
Is Cambridge IGCSE Statistics 0479 a new qualification?
Yes. The published 2027 syllabus is for first examination in June 2027. Cambridge also publishes a separate 2028–2030 syllabus.
How many exam papers are compulsory?
Two. Each is 2 hours 15 minutes, 100 marks and 50% of the qualification. Both may draw on all twelve topics.
Can I use a graphical calculator?
No. The 2027 rules permit a suitable ordinary calculator but prohibit algebraic and graphical calculators. Check the latest Cambridge exam materials list.
Must I show working even with a calculator?
Yes. Cambridge explicitly requires essential working to be shown for full marks.
Is this the same as studying Statistics within IGCSE Mathematics?
No. Some skills overlap, but Statistics 0479 has its own twelve-topic syllabus and two 100-mark examinations.
Is coursework compulsory?
The published 2027 assessment consists of two externally assessed written papers, without a coursework component.
What makes the exam difficult?
Students must choose appropriate methods and understand assumptions, as well as calculate correctly. Grouped data, rates, indices and time-series reasoning require careful interpretation.
Can students in any zone take the 2027 exam?
No. Cambridge lists 2027 availability in administrative zones 3 and 4, in the June series. A wider set of zones is listed from 2028. Confirm with the centre.
Official sources and next reading
The primary references are Cambridge IGCSE Statistics (0479) and the 2027 syllabus PDF. The official page also links the separate 2028–2030 syllabus, specimen materials and updates.
For related learning, read How IGCSE Mathematics Works, How IGCSE Economics Works and How IGCSE Geography Works. They develop related quantitative reasoning but do not share the 0479 paper structure.
Final student checklist
- I have confirmed syllabus 0479 and the correct exam year.
- I know both papers last 135 minutes, carry 100 marks and count 50% each.
- I have a permitted calculator, not a prohibited graphical or algebraic model.
- I can distinguish populations, samples, selection methods and data types.
- I can read and construct suitable charts and frequency distributions.
- I understand average, quartiles, spread and transformations.
- I can apply probability rules and expectation correctly.
- I can interpret crude and standardised rates and weighted indices.
- I can describe correlations and time-series trends without claiming unsupported causes.
- I show working, explain assumptions and revisit recurring mistakes with new data.
IGCSE Statistics works when students can move from question → data → appropriate method → calculation → honest interpretation. A strong candidate does not merely produce an answer; they explain what it means, why the method suits the situation and what the data do not establish.
Official source: Cambridge IGCSE Statistics (0479). This eduKateSG guide is independent educational commentary, not an official Cambridge publication.
More IGCSE Student Guides in This Batch
Explore the other eduKateSG Cambridge IGCSE guides below. Their syllabus codes, examination years and paper structures differ; use the exact requirements for the subject entered.