Box Plot & IQR Calculator

Box Plot Statistics Guide

What's the key thing to understand about Five Number Summary?

A box plot (box-and-whisker plot) displays 5 key statistics: Minimum — smallest non-outlier value (whisker end). Q1 — lower quartile (25th percentile). Median (Q2) — middle value (50th percentile). Q3 — upper quartile (75th percentile). Maximum — largest non-outlier value (whisker end). IQR = Q3 - Q1 (the box width). The box contains the middle 50% of data. The whiskers extend to the furthest values within 1.5×IQR of the box. Points beyond 1.5×IQR are plotted individually as outliers.

What should I know about Calculating Quartiles?

For a dataset of n values sorted in order: Median: if n is odd, median = middle value. If n is even, median = average of two middle values. Q1: lower half of data (below median) — median of lower half. Q3: upper half of data (above median) — median of upper half. Whether to include the overall median in the halves (inclusive/exclusive) varies by convention — GCSE in the UK typically uses the inclusive method. Interpolation method, used by many calculators and statistical software: Q1 is the value at a specific fractional position in the ordered data, giving a more precise result than the simple inclusive method when the dataset doesn't divide evenly.

What should I know about Outliers?

Definition: a value is an outlier if it falls more than 1.5×IQR below Q1 or above Q3. Lower fence = Q1 − 1.5×IQR. Upper fence = Q3 + 1.5×IQR. Example: IQR = 10, Q1 = 20, Q3 = 30. Lower fence = 20 − 15 = 5. Upper fence = 30 + 15 = 45. A value of 50 would be an outlier. Extreme outliers (beyond 3×IQR): sometimes plotted differently from mild outliers. Outliers are NOT necessarily errors — they may be genuine extreme values. Always investigate outliers before removing them. In GCSE and A-level statistics, outliers are typically identified using the 1.5 × IQR rule, marking any point beyond this range from the quartiles as worth investigating separately.

What's the key thing to understand about Comparing Box Plots?

Box plots are particularly useful for comparing distributions of two or more groups. Comparing: median positions (central tendency). Box widths/IQR (spread of middle 50%). Overall range (total spread). Symmetry: symmetric box (median in centre of box) vs skewed (median nearer one end). Skewness: positive skew — longer upper whisker and median nearer Q1. Negative skew — longer lower whisker and median nearer Q3. Multiple box plots on the same scale allow direct comparison of distributions. GCSE students are commonly asked to compare the spread, median, and skew of two or more box plots side by side, describing the differences in plain, comparative language.

What's the key thing to understand about Skewness and Box Plot Shape?

A box plot's shape tells you about skewness at a glance. If the median line sits roughly in the middle of the box, and the whiskers are similar lengths, the data is roughly symmetric. If the box is compressed toward the bottom with a long whisker (and most outliers) stretching upward, the data is right-skewed (positively skewed) — common in income or house price data, where most values cluster low but a few very high values pull the tail out. If the opposite is true — a long lower whisker and a box pushed toward the top — the data is left-skewed. Comparing the two halves of the box (Q1-to-median vs median-to-Q3) is a quick, visual way to spot skew without calculating a skewness coefficient, and it's one reason box plots are preferred over just reporting a mean and standard deviation when a dataset might not be symmetric.

What's the key thing to understand about Box Plots vs Histograms: When to Use Each?

Box plots and histograms both summarize a distribution but answer different questions. A histogram shows the full shape of the data — how many values fall in each bin — so it reveals multiple peaks (bimodal data), gaps, or unusual clustering that a box plot would hide. A box plot compresses that detail into five numbers (minimum, Q1, median, Q3, maximum) plus outliers, which makes it excellent for comparing several groups side by side on one axis — for example, exam scores across five different classes. Use a histogram when you need to understand one dataset's shape in detail; use a box plot when you need to compare the spread and center of multiple datasets at a glance, or when outliers specifically need to be flagged rather than just seen.

What should I know about Notched Box Plots?

A standard box plot shows the five-number summary, but a notched box plot adds a visual cue for comparing medians between groups: a narrowed 'waist' cut into each side of the box around the median line. The notch width approximates a confidence interval for the median, commonly calculated as median ± 1.57 × IQR/√n, where IQR is the interquartile range and n is the sample size — a formula derived from the standard error of the median under an approximately normal distribution assumption. The practical rule of thumb is simple: if the notches of two box plots do not overlap, there is reasonably strong visual evidence that the two groups' medians genuinely differ, while overlapping notches suggest the difference could plausibly be due to sampling variation alone. This makes notched box plots a quick, informal visual alternative to running a formal hypothesis test when comparing several groups at once, though the notch is only an approximation and becomes less reliable for small sample sizes or strongly skewed distributions, where the usual normal-based assumptions behind the confidence interval break down. Analysts typically still follow up with a proper statistical test before drawing firm conclusions, treating the notch as a useful first visual signal rather than a final answer.

Worked Example: Building a Box Plot from Raw Data

Consider 11 test scores, already sorted: 55, 62, 68, 70, 72, 75, 78, 80, 85, 90, 95. The median is the middle value (position 6 of 11): 75. Splitting the data at the median gives a lower half of 55, 62, 68, 70, 72 and an upper half of 78, 80, 85, 90, 95. Q1 is the median of the lower half: 68. Q3 is the median of the upper half: 85. The interquartile range is IQR = Q3 − Q1 = 85 − 68 = 17. The outlier fences are Q1 − 1.5×IQR = 68 − 25.5 = 42.5 and Q3 + 1.5×IQR = 85 + 25.5 = 110.5. Since every value in the data set (55 to 95) falls inside this 42.5-to-110.5 range, there are no outliers. The five-number summary used to draw the box plot is: Minimum = 55, Q1 = 68, Median = 75, Q3 = 85, Maximum = 95.

Reference Table: Reading a Box Plot's Features

Each visual feature of a box plot maps to a specific statistical statement about the data:

Standard Outliers vs. Extreme Outliers

Some box plot conventions distinguish between "mild" outliers (beyond 1.5×IQR) and "extreme" outliers (beyond 3×IQR), plotting them with different markers. Using the earlier example data set (five-number summary: Min 55, Q1 68, Median 75, Q3 85, Max 95, IQR 17), the standard 1.5×IQR fences were 42.5 and 110.5. The extreme 3×IQR fences are Q1 − 3×IQR = 68 − 51 = 17 and Q3 + 3×IQR = 85 + 51 = 136. Since every value in the data set already fell inside the milder 1.5×IQR fences, it's automatically true that none fall outside the wider 3×IQR fences either — extreme-outlier fences are always wider than mild-outlier fences, so a point can never be an "extreme" outlier without first also being a "mild" one.

A Caution About Box Plots with Very Small Samples

With only a handful of data points, quartiles computed by different methods (inclusive vs. exclusive median splitting) can disagree noticeably, and a box plot can visually suggest more structure in the data than a small sample actually supports. As a practical guideline, box plots are most informative with at least 10-15 data points per group; below that, consider showing the individual data points alongside or instead of the box, so viewers aren't misled by summary statistics built from very few numbers.