PlotSet LogoPlotSet
全部图表类型

Histogram Maker

See the shape of your numbers, not just their average. Upload a spreadsheet, pick one numeric column, and publish an interactive histogram you can embed anywhere — free.

免费创建 Histogram查看图库

什么是 Histogram？

A histogram takes one column of numbers, chops the range it covers into equal intervals called bins, and draws a bar for each bin whose height is the count of values that fall inside it. Unlike a bar chart, the bars touch: the axis underneath is a continuous number line, not a list of separate categories. What you get back is the shape of the data — where the bulk of the values sit, how wide the spread is, whether it leans one way, and whether there is a second cluster or a straggler far out on the tail that the mean was quietly hiding.

什么时候使用 Histogram

  • You want to know how one variable is distributed

    Salaries, response times, exam scores, order values. One numeric column, one picture of where the mass of the data actually sits.

  • The average is not telling the whole story

    Two datasets with the same mean can look nothing alike. A histogram is the fastest way to show a room full of people that their headline number is hiding a split, a skew or a long tail.

  • You are checking data before you model it

    Skew, bimodality and impossible values all show up immediately in a histogram. It is the standard first look at a column, and it costs one chart.

  • You need to justify a cut-off

    Where does 'slow' start? A histogram of response times shows the reader the gap you are drawing the line in, which is far more persuasive than the threshold on its own.

什么时候不该用 Histogram

  • Your data is categorical

    Countries, departments and product names have no numeric order, so there is nothing to bin. Use a bar chart, which keeps the gaps between bars precisely because the categories are separate.

  • You are comparing several groups at once

    Overlaid histograms turn to mud after two series. Box plots or a ridgeline chart compare many distributions side by side without occlusion.

  • You only have a handful of values

    Below about thirty observations the bin heights are noise: change the bin width and the shape changes with it. Plot the individual points instead.

  • Every exact value matters

    Binning is deliberate information loss. If the reader needs to look up individual figures, publish a table alongside — or instead.

如何制作 Histogram

从电子表格到已发布的图表，一步一步来。

01

Start from the histogram template

Open the PlotSet histogram template. Sample data comes bound, so you can see exactly which shape of spreadsheet it expects before you upload your own.

02

Bring one column of numbers

A histogram needs raw observations — one row per thing measured, one numeric column. Do not pre-summarise: if you have already counted the rows per bracket, you have a bar chart, not a histogram.

03

Upload or connect your data

Import a CSV or Excel file, paste straight from a spreadsheet, or connect a Google Sheet so the chart can be refreshed when the data moves.

04

Set the bin width deliberately

Bin width is the one setting that changes the story. Too wide and you flatten a second peak into a hump; too narrow and every wobble looks like a feature. Try a few, and pick round numbers a reader can hold in their head — 5, 10, 250 — rather than whatever the automatic rule produces.

05

Publish, embed or export

Publish for a live interactive embed with hover counts, or export a PNG, SVG or PDF for a report or a deck.

从已加载示例数据的 Histogram 模板开始，再替换成你自己的数据。

打开模板

Histogram 的组成部分

Bin
One interval on the number line. Every observation falls in exactly one bin, and every bin is the same width — unequal bins make the bar heights incomparable.
Frequency axis
The vertical scale, counting how many observations landed in each bin. It starts at zero, always: bar height is the encoding.
Value axis
The horizontal scale — a continuous number line, which is why the bars sit flush against each other with no gaps.
Bin width
The size of each interval. It is a judgement call, not a fact about the data, and it deserves to be stated in the caption.
Tail
The thin stretch of bins at one or both ends. Long tails are usually the most interesting part of the chart and the first thing an average destroys.
Mode
The tallest bin — the most common range of values. A distribution with two distinct peaks is bimodal and almost always means two populations have been mixed together.

Histogram 的常见错误

Confusing a histogram with a bar chart

Gaps between the bars say 'separate categories'. Touching bars say 'one continuous scale'. Getting this wrong tells the reader the wrong thing about the data before they read a single label.

Accepting the default bin width

The default is a starting point, not an answer. Always look at the chart with two or three different bin widths before publishing, because a real feature survives the change and an artefact does not.

Using unequal bins without adjusting height

If one bin is twice as wide as its neighbours it collects roughly twice as many values for free. Either keep every bin the same width, or plot density rather than raw counts.

Binning data that is already summarised

Feeding in counts per bracket produces a chart of your own bins, not the data's distribution. Histograms need one row per observation.

Cutting off the tail to tidy the chart

The outliers you truncate to make the picture neater are frequently the finding. Keep them, and annotate them.

Histogram：常见问题

Showing how one numeric variable is distributed — where values cluster, how spread out they are, and whether the distribution is skewed or has more than one peak. It is the standard first look at a column of numbers.

立即制作你的 Histogram

免费开始，300 多个模板，无需信用卡。

免费创建 Histogram全部图表类型