Every statistical graph and measure at GCSE explained — including the histogram and cumulative frequency skills that separate grade 6 from grade 8.
Statistics questions appear on every GCSE Maths paper and cover a wider range of skills than most students realise. It's not just averages — it's interpreting graphs, comparing distributions, calculating from grouped data, and drawing conclusions with appropriate statistical language. This guide covers all of it.
These four measures are the foundation of GCSE statistics. Most students know the definitions but consistently make errors when applying them to grouped data or frequency tables.
This is where most errors happen. You don't add up the frequencies — you multiply each value by its frequency, sum the products, then divide by the total frequency.
Example: scores 1, 2, 3, 4 with frequencies 3, 5, 7, 5. Mean = (1×3 + 2×5 + 3×7 + 4×5) / (3+5+7+5) = (3+10+21+20)/20 = 54/20 = 2.7
When data is grouped into class intervals, you can't find the exact mean — you estimate it using the midpoint of each class. This gives an estimated mean, not the true mean, and it's correct to call it an estimate.
❌ Most common error: using the class boundaries instead of midpoints. For the class 10 ≤ x < 20, the midpoint is 15, not 10 or 20.
Histograms look like bar charts but they are fundamentally different. In a histogram, the area of each bar represents frequency — not the height. The y-axis shows frequency density, not frequency.
This distinction matters enormously. If class widths are unequal (which they often are in exam questions), you cannot compare bar heights directly. A taller bar with a narrower width might represent fewer people than a shorter bar with a wider width.
To find the frequency for a class from a histogram: read the frequency density from the y-axis, then multiply by the class width. Always do this step — never just read the height as the frequency.
Example: a bar has frequency density 4 and spans from 20 to 25. Class width = 5. Frequency = 4 × 5 = 20.
Convert the frequency table to frequency densities (divide each frequency by class width). Plot frequency density on the y-axis. Draw bars with no gaps between them. Label the y-axis "Frequency Density" — not "Frequency".
In exam questions, histograms with unequal class widths are almost always designed specifically to catch students who confuse height with frequency. Always calculate frequency density — never plot raw frequencies in a histogram.
A cumulative frequency curve (or ogive) shows the running total of frequencies up to each value. It's always S-shaped (or part of an S-shape) and is used to estimate medians, quartiles and interquartile ranges.
From a cumulative frequency curve with total frequency n:
The IQR is a better measure of spread than the range because it ignores outliers — and exam questions often ask you to compare two distributions using the IQR.
A box plot summarises a distribution using five values: minimum, lower quartile, median, upper quartile, and maximum. These five values are exactly what you read from a cumulative frequency curve.
Drawing a box plot: draw a number line. Mark the five values. Draw a box from Q1 to Q3. Mark the median inside the box with a vertical line. Draw whiskers from the box to the minimum and maximum.
Interpreting box plots: a longer box means greater spread in the middle 50% of data. A median closer to Q1 means the distribution is skewed to the right. Comparing two box plots — comment on median (average), IQR (spread), and any overlap between distributions.
When a question asks you to compare two distributions (using box plots, frequency polygons, or any graph), always make two separate points: one about average (which has a higher/lower median or mean) and one about spread (which has a greater/smaller range or IQR). A one-sentence answer comparing only the averages will not get full marks.
Scatter graphs show the relationship between two variables. The key skills are drawing a line of best fit, describing the correlation, and using the line of best fit to make predictions.
The line of best fit should pass through the mean point (mean of x, mean of y) and have roughly equal numbers of points on each side. Don't force it through the origin unless the data clearly demands it.
Interpolation (predicting within the data range) is reliable. Extrapolation (predicting outside the data range) is unreliable — this distinction is explicitly tested in exam questions asking whether a prediction is reliable.
Correlation does not imply causation. Two things can be correlated without one causing the other — this is a common question in GCSE Statistics. The full statistics content for each exam board is specified in the AQA specification and the OCR GCSE Maths specification.
PaperPlus generates GCSE statistics questions including histograms, cumulative frequency and scatter graphs — with full mark schemes for AQA, Edexcel and OCR.
Start Practising Free →