Bin

A bin is a value range in a histogram into which data points are grouped, so the chart can show how many values fall in each range.

Continuous data has too many distinct values to plot one bar each. Binning solves this by slicing the value axis into adjacent ranges — the bins — and counting how many points land in each. Every bin becomes one bar, and its height is the count for that range. Bins are the building blocks that turn a list of measurements into a picture of their distribution.

How bins work in a chart

Bins are usually equal in width and sit side by side with no gaps, because together they cover the whole range of the data with no overlaps. The crucial design choice is bin width. Wide bins lump many values together, giving a smooth, blocky shape that can hide detail. Narrow bins reveal fine structure but can look spiky and noisy. The same data can look very different depending on how the bins are sized — which is why bin width should always be considered, not accepted blindly.

A concrete example

Suppose you have test scores from 0 to 100 and choose a bin width of 20. That gives five bins: 0–20, 20–40, 40–60, 60–80 and 80–100. If the counts in those bins are 2, 8, 25, 12 and 3, they sum to 50 scores, and the histogram shows a clear peak in the 40–60 bin where most scores cluster. Halve the bin width to 10 and you would get ten thinner bars, exposing more shape but with smaller counts in each.

Related terms

Bins divide the overall range of the data, and they are the defining feature of the histogram. The values being binned are usually continuous data. Each bin's height is a count, much like a data point per range.