Course: JustAcademy Tableau Training
Course Demo: Register for a Tableau Course Demo
A Box-and-Whisker Plot, also known as a Box Plot, is a statistical visualization used to understand the distribution, spread, median, quartiles, and potential outliers in numerical data. It provides a compact summary of how values are distributed within a dataset.
In Tableau, box-and-whisker plots are especially useful when comparing numerical values across categories, such as sales across regions, employee salaries across departments, product delivery times across shipping methods, or customer spending across customer segments.
Unlike a simple bar chart, which typically compares totals or averages, a box plot helps reveal the variation within each group. It can show whether values are closely grouped, widely spread, skewed, or contain unusually high or low observations.
Example: A business wants to compare delivery times across four shipping methods. A box plot can show the median delivery time, the middle 50% of delivery times, the overall spread represented by the whiskers, and potential outliers for each shipping method.
A box plot contains several important statistical elements. Understanding these components helps you interpret the chart correctly.
The minimum and maximum are the smallest and largest observed values in the dataset. However, in a conventional box plot using the 1.5 × IQR rule, the whiskers do not necessarily extend to the absolute minimum and maximum. They extend to the most extreme observations within the whisker limits, while values beyond those limits are displayed as potential outliers.
The first quartile, also called the lower quartile, is the 25th percentile. Approximately 25% of the observations are at or below this value, depending on the percentile calculation method.
The median is the middle value of an ordered dataset. It divides the observations into two halves. In a box plot, the median is represented by a line inside the box.
The third quartile, also called the upper quartile, is the 75th percentile. Approximately 75% of the observations are at or below this value.
The interquartile range measures the spread of the middle 50% of the data. It is calculated by subtracting the first quartile from the third quartile.
Formula: IQR = Q3 − Q1
A larger IQR indicates that the middle half of the observations is more spread out. A smaller IQR indicates that the middle half is more concentrated.
The box extends from Q1 to Q3 and represents the middle 50% of the observations. The median line inside the box shows the central value of the distribution.
Whiskers extend outward from the box. In a common box plot configuration, they reach the most extreme data points within the lower and upper limits calculated using 1.5 times the IQR. Tableau also allows whiskers to extend to the maximum extent of the data.
Potential outliers are observations that lie beyond the whisker limits. They are often displayed as individual marks outside the whiskers. An outlier is not automatically a data error; it may be a valid observation that deserves further investigation.
A commonly used box plot method calculates lower and upper limits to identify potential outliers.
Lower fence: Q1 − (1.5 × IQR)
Upper fence: Q3 + (1.5 × IQR)
Values below the lower fence or above the upper fence are considered potential outliers under this rule. The fences are not necessarily the actual endpoints of the whiskers. In the standard 1.5 × IQR configuration, the whiskers extend to the most extreme observed values that remain within the fences.
Suppose a dataset has the following values:
Step 1: Calculate the IQR.
IQR = Q3 − Q1 = 50 − 20 = 30
Step 2: Calculate the lower fence.
Lower fence = 20 − (1.5 × 30) = 20 − 45 = −25
Step 3: Calculate the upper fence.
Upper fence = 50 + (1.5 × 30) = 50 + 45 = 95
Interpretation: Under this rule, values below −25 or above 95 are potential outliers. The actual whiskers extend to the most extreme observed values within these limits. For a dataset that contains only non-negative values, the lower whisker might reach the smallest observed value if that value is within the lower fence.
| Component | Statistical Meaning | What It Helps You Understand |
|---|---|---|
| Q1 | 25th percentile | Lower boundary of the middle 50%. |
| Median | 50th percentile | Central position of the distribution. |
| Q3 | 75th percentile | Upper boundary of the middle 50%. |
| Box | Range from Q1 to Q3 | Spread of the middle half of the data. |
| IQR | Q3 − Q1 | Variation in the middle 50%. |
| Whiskers | Configured extent of the distribution | Spread beyond the box. |
| Outlier marks | Values beyond whisker limits | Observations that may warrant investigation. |
Before creating a box plot, make sure your data includes the following:
Important: A box plot based on only one aggregated value per category may appear as a flattened line. To display a meaningful distribution, Tableau needs enough underlying observations or marks within each group.
Show Me provides a convenient way to create a box-and-whisker plot in Tableau. It is suitable for beginners who want to create a distribution comparison quickly.
Tableau displays box plots for the selected groups. Each box summarizes the middle 50% of the data, the line inside the box indicates the median, and the whiskers show the configured extent of the distribution. Potential outliers may appear as individual marks beyond the whiskers.
Note: The exact field placement and appearance may vary depending on the Tableau version, the fields selected, and the level of detail in the view.
The Analytics pane allows you to add a box plot to an existing view. This is useful when you want to preserve the structure of a view and add statistical distribution information to it.
In Tableau Desktop, you can also right-click a continuous axis and select Add Reference Line. In the dialog box, select Box Plot to add a box plot to the view. The precise options available can vary by Tableau product and authoring environment.
This method is useful when you already have a chart and want to add a box plot without rebuilding the entire view.
Difference between Show Me and Analytics: Show Me creates the box plot as the main visualization. The Analytics pane or axis-based method adds a box plot to an existing view.
Data aggregation is an important concept when creating box plots in Tableau. Tableau commonly aggregates measures by default. If each category is represented by only one aggregated value, there may not be enough marks to display a meaningful distribution.
Suppose you want to compare sales across three regions: East, West, and Central.
Warning: Disaggregating data changes the level of detail in the view and may affect performance or chart interpretation. Use it only when it matches the analytical objective.
Tableau supports different whisker configurations for box plots. The selected method affects how the spread of the data is represented and how potential outliers appear.
| Whisker Configuration | Description | When It Is Useful |
|---|---|---|
| 1.5 × IQR | Whiskers extend to the most extreme observations within the lower and upper limits calculated from 1.5 times the IQR. | When you want to identify potential outliers using a common statistical convention. |
| Maximum extent of data | Whiskers extend to the minimum and maximum extent of the data. | When you want the whiskers to show the full observed range. |
When using the 1.5 × IQR method, points outside the whiskers are potential outliers under that rule. When using the maximum extent of the data, the whiskers reach the observed extremes, so those same points will not appear outside the whiskers under this configuration.
Use a descriptive title that identifies the measure and grouping, such as Sales Distribution by Region, Employee Salary Distribution by Department, or Delivery Time by Shipping Method.
Use meaningful labels. For example, label the numerical axis Sales Amount and the category axis Region.
Use consistent colors when comparing similar groups. You may use Color on the Marks card to distinguish categories, but avoid excessive color variations that distract from the statistical summary.
Use the box plot's formatting or edit options to adjust fill, border, and whisker appearance when those settings are available in your Tableau environment.
Depending on the box plot configuration, you can hide underlying marks while retaining outliers, or display more of the individual observations. Showing marks can help explain the distribution but may make a chart crowded when the dataset is large.
Include relevant details such as region, product category, measure value, and other useful dimensions. Avoid adding unrelated fields that make the tooltip difficult to read.
Where appropriate, sort categories by median, average, or another relevant statistic to make comparisons easier. Ensure the sorting method matches the purpose of the analysis.
Business scenario: A retail company wants to understand how sales values vary across East, West, Central, and South regions. The goal is to compare typical values, variability, and potential unusually large or small sales records.
Important: These are analysis questions, not pre-established findings. Use the actual Tableau view to determine the answers.
Business scenario: An HR team wants to compare salary distributions across departments to understand typical pay levels, variation, and unusual salary values.
| Field | Role |
|---|---|
| Department | Dimension used to group employees. |
| Salary | Numerical measure used to build the box plot. |
| Job Level | Optional dimension for additional analysis. |
| Employee ID | Optional detail field to help preserve individual-level observations. |
Business benefit: The chart can help HR analysts identify differences in salary distributions. However, a box plot alone cannot determine whether a salary difference is fair or justified; further context and analysis are needed.
An e-commerce company wants to compare delivery performance across shipping methods.
A shipping method with a low median and a narrow IQR may offer both faster and more consistent delivery, although the conclusion should also consider cost, service coverage, and sample size.
The median line inside each box indicates the central value. Comparing medians helps identify which group generally has higher or lower values. A higher median does not necessarily mean that every observation in that group is higher than observations in another group.
A taller box in a vertical plot or a longer box in a horizontal plot indicates a larger IQR and therefore more spread in the middle 50% of the data.
Longer whiskers indicate a greater spread beyond the box under the selected whisker configuration. Differences in whisker length can suggest that the groups have different ranges or asymmetry.
Marks beyond the whiskers may indicate unusual observations. Investigate their context before classifying them as errors or removing them.
If the median is not centered in the box, or if one whisker is considerably longer than the other, the distribution may be asymmetric. These patterns are clues rather than definitive proof of skewness.
Overlapping boxes do not, by themselves, prove that two distributions are the same or statistically indistinguishable. A box plot is a descriptive visualization. Formal statistical conclusions may require additional analysis.
| Feature | Box Plot | Histogram | Bar Chart |
|---|---|---|---|
| Primary purpose | Summarizes distribution and spread. | Shows frequency across numerical bins. | Compares categories. |
| Median | Shown directly. | Not shown directly. | Not shown directly unless separately encoded. |
| Quartiles | Shown directly. | Not shown directly. | Not shown directly. |
| Distribution shape | Summarized through quartiles, whiskers, and outliers. | Shown through the shape of the bars. | Usually not designed for continuous distribution analysis. |
| Potential outliers | Often shown as marks outside the whiskers. | May appear in low-frequency extreme bins. | Usually not shown as a distribution feature. |
| Best use | Comparing distributions across groups. | Exploring the frequency pattern of numerical values. | Comparing totals or values by category. |
| Problem | Possible Cause | Solution |
|---|---|---|
| The box plot looks like a flat line. | The view may contain only one aggregated mark per group or very little variation. | Check data granularity and ensure there are enough underlying observations. |
| The box plot option is unavailable. | The selected fields or view may not meet the requirements for Show Me. | Start with a numerical measure and a relevant grouping dimension, then review the Show Me options. |
| No outliers are visible. | The data may have no observations beyond the selected whisker limits, or the whiskers may extend to the full data range. | Review the whisker configuration and inspect the underlying data. |
| The chart contains too many marks. | The view may show individual observations for a very large dataset. | Consider hiding underlying marks, using suitable filters, or simplifying the view. |
| Groups are difficult to compare. | There may be too many categories or inconsistent axis scales. | Reduce the number of groups, use a consistent scale, and sort categories meaningfully. |
| Quartile interpretation seems unexpected. | The data distribution, sample size, or percentile calculation method may affect the result. | Inspect the underlying values and Tableau's statistical summary. |
| The measure shows totals instead of individual values. | The measure is aggregated in the current view. | Review the aggregation and level of detail before interpreting the box plot. |
| Outliers are automatically treated as errors. | Potential outliers are being interpreted without business context. | Investigate the records before deciding whether to correct or exclude them. |
Create a box plot of Profit by Region. Compare medians, IQRs, whiskers, and potential outliers.
Create a box plot showing Discount by Segment. Ensure the view contains enough underlying observations and explain the differences between the groups.
Build a box plot of Sales by Category. Identify which categories have the largest IQR and investigate potential outliers.
Using an employee dataset, compare Salary by Department. Discuss differences in median and variation.
Create a box plot to compare delivery-time distributions. Identify the method with the lowest median and investigate unusual delivery times.
Project Objective: Create an interactive Tableau dashboard that compares the distribution of a numerical business measure across multiple groups.
Answer: A box-and-whisker plot is a statistical chart that summarizes the distribution of numerical data using quartiles, a median, whiskers, and sometimes individual marks for potential outliers.
Answer: The box extends from the first quartile (Q1) to the third quartile (Q3), representing the middle 50% of the observations.
Answer: The median is the middle value of an ordered dataset and is represented by a line inside the box.
Answer: The IQR is the difference between the third quartile and the first quartile: IQR = Q3 − Q1.
Answer: Under the common 1.5 × IQR rule, values below Q1 − 1.5 × IQR or above Q3 + 1.5 × IQR are potential outliers.
Answer: Place a suitable numerical measure and grouping dimension in the view, then use Show Me to select Box-and-Whisker Plot. Alternatively, add a box plot from the Analytics pane or the appropriate axis options.
Answer: A box plot summarizes the distribution using quartiles, median, whiskers, and potential outliers. A histogram shows the frequency of values within numerical intervals.
Answer: It may be based on only one aggregated mark per category or on values with very little variation. Check the granularity and aggregation of the view.
Answer: Longer whiskers indicate greater spread beyond the box under the selected whisker configuration.
Answer: No. An outlier is an observation that falls beyond the selected statistical limits. It may be valid and should be investigated before being corrected or removed.
Answer: The 1.5 × IQR method extends whiskers to the most extreme observations within the calculated limits. The maximum-extent method extends them to the minimum and maximum observed values.
Answer: Use a box plot when you want to compare the median, spread, and potential outliers of numerical data across categories.
Box-and-whisker plots are powerful Tableau visualizations for comparing numerical distributions. They summarize the median, quartiles, interquartile range, whiskers, and potential outliers in a compact format. By choosing the right level of detail, configuring the whiskers correctly, and interpreting the chart in context, analysts can compare business performance and investigate unusual values more effectively.
To continue learning Tableau, visit JustAcademy Tableau Training or Register for a Tableau Course Demo.
