Creating and interpreting box plots

A box plot turns a set of observations into a compact picture of a distribution. Its central box shows where the middle half of the data lies, the line inside marks the median, and the whiskers show how far the values extend. The diagram makes it easier to compare spread, centre and unusual results without drawing every point.

Before calculating, decide what each observation represents and which units apply. In Australian maths classrooms, students may hear “box-and-whisker plot”, “lower quartile” and “upper quartile”, while software often uses the shorter “box plot”. Clear labels prevent a neat diagram from hiding an unclear question.

Starting with the five-number summary

Sort the observations and find the minimum, lower quartile (Q_1), median, upper quartile (Q_3), and maximum. The median is the middle value, or the mean of the two middle values when there is an even number of observations. These five values form the five-number summary from which the main features of the plot are built.

Quartile calculations need a stated method. For example, with the ordered marks (5,7,8,9,10,11,12,13,15,17,20,24), the median is (11.5). Taking the median of each half gives (Q_1=8.5) and (Q_3=16), so the interquartile range is (16-8.5=7.5). Some calculators use percentile rules that give slightly different quartiles, so the method belongs in the working.

Drawing the box and whiskers

Draw a horizontal number line with a regular scale, then place a box from (Q_1) to (Q_3). Add a vertical line at the median and extend whiskers towards the lowest and highest relevant values. Label the axis, include units, and mark the numerical scale clearly; a box with no scale communicates very little.

The common modified box plot uses whiskers reaching the furthest values within (1.5) times the interquartile range from the box. More distant observations are plotted individually as outliers. Some classroom plots instead use whiskers for the minimum and maximum, so state which convention is being used and distinguish the observed minimum from the lower whisker when outliers occur.

Choosing a scale that informs

The scale should fit the data and make comparisons honest. Equal intervals help readers judge distances, and every value on the axis must carry the same unit. A truncated axis does not invalidate a box plot as it would a bar chart, because the plot uses positions rather than bar lengths, yet a narrow scale can still make a modest difference look dramatic.

Local examples show why units matter. Temperature might be measured in degrees Celsius, rainfall in millimetres, and household costs in Australian dollars. A plot of daily maximum temperatures in Melbourne and Sydney must use the same Celsius scale on both axes. Adding context such as the month, year and location turns a statistical shape into a meaningful comparison.

Reading the centre and spread

The median describes the middle observation: half the values are at or below it and half are at or above it. The interquartile range, (IQR=Q_3-Q_1), measures the spread of the central half of the data. It is more resistant to extreme values than the full range, which extends from the minimum to the maximum.

For the sample marks above, a median of (11.5) and an IQR of (7.5) show the location and spread of the middle results. The full range is (24-5=19) marks. Useful written interpretations compare these measures and refer to the sample, rather than claiming that every student or every future result must fall within the whiskers.

Interpreting shape and unusual values

A box plot suggests symmetry when the median sits near the centre of the box and the whiskers are broadly balanced. A median closer to one edge of the box often indicates skewness: a tail extends towards the larger values on the right or smaller values on the left. These are visual clues, not exact tests of distribution shape.

Outliers need investigation, not automatic deletion. They may represent genuine extreme observations, recording errors, unusual events or results from a different process. Record each outlier and check the source. Also remember that a five-number summary cannot reveal every cluster: two groups of values can produce the same quartiles and still have a different internal structure.

Comparing distributions

Place box plots for different groups on the same scale when making direct comparisons. Compare medians, IQRs, whisker lengths and outliers in the units of the data. Similar boxes do not prove that the groups are alike, and visible overlap does not establish a statistically significant difference.

Australian weather offers useful practice. Melbourne’s famously changeable days can create a broad temperature spread, while Sydney’s milder winter nights and coastal humidity produce a different seasonal profile. Darwin’s strong wet–dry season, Perth’s dry summers and Adelaide’s hot summer afternoons provide further contrasts, but climate summaries should come from clearly defined places and periods rather than casual impressions.

A dependable interpretation routine

Begin with context, sample size and units, then state the five-number summary. Describe the centre, the middle-half spread, the overall range and any unusual observations. Use cautious language such as “in this sample” and avoid inferring causes from the diagram alone. A dot plot or frequency table should accompany the box plot when individual patterns matter.

Box plots are powerful summaries, but they conceal details such as the number of observations, repeated values and gaps. Check for those by consulting the raw data, and confirm how software calculated quartiles. For Australian students using online homework assistance, CurryPlum’s step-by-step maths resources, calculators and math tools can help verify both the arithmetic and the final interpretation.

The practical takeaway is simple: write the five-number summary beside the diagram, name the units, and explain what the centre, spread and unusual points show while acknowledging what the box plot cannot reveal.