Comparative box plots show several data distributions on the same number scale so you can compare groups quickly and fairly. Each box plot summarizes a data set using the minimum, first quartile, median, third quartile, and maximum, with possible outliers shown separately. This makes it easier to compare centers, spreads, skew, and unusual values without listing every data point.
They are useful in science, economics, sports, and any setting where groups must be compared using data.
When three box plots are stacked on one horizontal axis, differences in median show which group tends to have larger values. Differences in box length and whisker length show variability within each group. A longer right whisker or a median closer to the left side of the box suggests right skew, while the opposite suggests left skew.
Outliers help identify values that may be unusual, important, or worth checking for error.
Understanding Statistics: Comparative Box Plots
A fair comparison begins before any box is drawn. Every group must use the same measurement units and the same number scale. If one plot uses a narrow scale while another uses a wide scale, the visual comparison can mislead the reader.
The groups should represent similar kinds of observations too. For example, comparing test scores from two classes makes sense if both classes took the same test under similar conditions. Comparing heights measured in centimetres with heights measured in inches does not make sense until the units are converted.
To build each plot, place the data in order from smallest to largest. Find the middle value, then find the middle values of the lower half and upper half. These positions divide the ordered data into four sections.
The box contains the central half of the observations, so it shows where many values are clustered. The line inside the box marks the central value.
The whiskers extend toward values outside the box, but their exact endpoints depend on the class or software rule for outliers. This is why students should check the stated plotting rule rather than assume every whisker reaches the actual smallest or largest value.
When comparing groups, start with the typical values, then examine the amount of overlap. Two groups can have different median lines but still share many values. A small difference between medians does not prove that every member of one group is larger than every member of the other group.
Large overlap in the boxes suggests that the middle portions of the groups are similar. Little or no overlap suggests a clearer separation, though it still does not explain the cause. Sample size matters as well.
A box plot with data from ten people may change greatly when a few more people are measured. A plot from hundreds of people is usually more stable.
Outliers deserve careful attention because they can tell different stories. A very high electricity bill might come from an unusually cold month, a large household, a faulty meter, or a recording mistake. In a sports data set, an outlier performance may show an exceptional athlete rather than bad data.
Do not delete an outlier simply because it looks inconvenient. First check whether the value was measured and recorded correctly. If it is real, report it and consider how it affects the interpretation.
Box plots are summaries, so they hide details such as repeated values, gaps, and the exact number of observations. A dot plot or histogram can provide those missing details when a closer view is needed.
Key Facts
- The five-number summary is minimum, Q1, median, Q3, maximum.
- IQR = Q3 - Q1 measures the spread of the middle 50% of the data.
- Range = maximum - minimum measures the full spread of the data, excluding separated outliers if shown that way.
- Outlier rule: values below Q1 - 1.5(IQR) or above Q3 + 1.5(IQR) are often labeled outliers.
- A higher median line means that group has a higher typical value on the shared scale.
- A longer box or longer whiskers mean greater variability in that part of the distribution.
Vocabulary
- Median
- The median is the middle value of a data set when the values are ordered from least to greatest.
- Quartile
- A quartile is a value that divides ordered data into four parts with about 25% of the data in each part.
- Interquartile range
- The interquartile range is the distance from the first quartile to the third quartile and describes the spread of the middle half of the data.
- Skew
- Skew describes a distribution that stretches farther on one side than the other instead of being roughly balanced.
- Outlier
- An outlier is a data value that is unusually far from the rest of the values in a distribution.
Common Mistakes to Avoid
- Comparing box plots drawn on different scales is wrong because the visual lengths no longer represent the same numerical distances.
- Treating the whole box as the range is wrong because the box only runs from Q1 to Q3 and contains the middle 50% of the data.
- Assuming a larger median means every value in that group is larger is wrong because the distributions can overlap.
- Ignoring outliers is wrong because outliers can reveal unusual cases, measurement errors, or important features of the data.
Practice Questions
- 1 Group A has Q1 = 12, median = 18, and Q3 = 25. Find the IQR and interpret what it means.
- 2 Group B has Q1 = 40 and Q3 = 56. Use the 1.5(IQR) rule to find the lower and upper outlier fences.
- 3 Three groups have medians of 22, 27, and 25, but Group C has the longest box and whiskers. Explain which group has the greatest typical value and which group has the greatest variability.