Compare distributions, inspect quartiles and outliers, and export a publication-ready box-and-whisker chart.
Compare delivery-time medians, quartiles, spread, means, and unusual orders across shipping methods.
| Group | Count | Minimum | Q1 | Median | Q3 | Maximum | Mean | Outliers |
|---|
The chart has not been calculated yet.
A box plot summarizes a full group of observations without hiding its shape behind a single average. The box marks the middle half of the data, the median divides that box, whiskers describe the typical range, and separate points reveal potential outliers. Use this maker when several samples need to be compared on the same measurement scale.
Place each sample group in its own numeric column; every row represents one observation position. For example, a delivery table might contain Standard, Express, and Priority shipping-time columns. Import CSV, TSV, TXT, JSON, XLSX, or XLS data, then select one or more numeric columns to compare.
Missing cells follow the selected data rule and are never converted to zero. Keep measurement units consistent across groups and include enough observations for quartiles to be meaningful. The editable table supports direct corrections, additional observations, imported columns, renamed groups, visibility toggles, and color selection.
Linear interpolation is a practical default and behaves smoothly for samples of many sizes. Tukey hinges split the ordered observations around the median, while the exclusive method estimates quartiles from positions beyond the endpoints. Different statistical packages can return slightly different quartiles, so select the method used by the analysis you need to reproduce.
The conventional 1.5× IQR rule places whiskers at the most extreme observations within the quartile fences and displays values beyond them as possible outliers. Minimum-to-maximum whiskers include every observation and therefore produce no outlier points. The percentile option uses the fifth and ninety-fifth percentiles for a robust overview of larger samples.
Enable the mean marker to compare the arithmetic average with the median. A large separation can indicate skew. Sample points provide additional context for small datasets, while hiding them keeps dense charts clean. Hover over a box to see its count, five-number summary, whiskers, mean, interquartile range, and outlier total.
Adjust box width, spacing, fill opacity, colors, grid lines, and orientation without changing the source values. Vertical boxes are familiar for a modest number of groups; horizontal boxes give longer group names more room. Zoom and pan along the category axis when the comparison contains many columns.
Compare median position, box height or width, whisker length, and outlier patterns together. A narrow box indicates a concentrated middle half, but it does not necessarily imply a small overall range. Box plots also do not show whether the distribution has multiple peaks unless individual sample points are displayed.
Download PNG for slides, documents, and web publishing, or SVG for scalable layouts. Preset and custom dimensions, aspect ratios, render scale, transparency, and a preview help ensure that category names and statistical marks remain legible in the final image.
Suppose three shipping methods have similar average delivery times. A box plot can still reveal an important operational difference: one method may have a narrow box and short whiskers, while another has a wider middle range and several late outliers. The first service is more predictable even when their means are close. Enter each method as a separate sample column, keep every value in the same unit, and compare the median, interquartile range, whiskers, and unusual observations together.
The box-and-whisker summary compresses a distribution. Different datasets can share the same five-number summary while having different clusters, gaps, or shapes. Small groups also produce unstable quartiles. When distribution shape matters, compare this view with a violin plot or histogram, and show sample points when the number of observations is manageable.
Confirm that groups use the same measurement process and units, document the quartile and whisker rules, investigate rather than silently delete outliers, and state the sample period or population. These details let another reader reproduce the summary and prevent a visual difference from being interpreted as evidence of a cause.