Histogram: How Bin Width Shapes the Story

What is a histogram?

A histogram groups continuous data into intervals (bins) and shows the frequency (count) of values in each bin as a bar. Unlike a bar chart, the bars touch — the x-axis is a continuous number line.

How bin width changes the story

The same data can tell very different visual stories depending on the bin width:

• Too few bins (wide): The histogram looks like 2-3 big blocks. It hides the shape of the distribution — you can't see if it's bimodal, skewed, or normal.
• Too many bins (narrow): Each bar is very short and jagged. Noise overwhelms the pattern — you see random spikes, not the underlying distribution.
• Just right: The shape of the distribution is clear — you can see peaks, symmetry, skewness, and gaps.

Rules of thumb for choosing bin count

• Square root rule: $\text{bins} \approx \sqrt{n}$ (for $n = 200$: about 14 bins).
• Sturges' formula: $\text{bins} = 1 + 3.322 \log_{10} n$ (for $n = 200$: about 9 bins).
• Freedman-Diaconis: bin width $= 2 \times \text{IQR} \times n^{-1/3}$ (accounts for data spread).

Histogram vs bar chart

• Histogram: continuous data, bars touch, order matters (x-axis is numerical).
• Bar chart: categorical data, bars separated, order is flexible.

Try it in the visualization

Drag the bin width slider. Watch the same data reshape dramatically. Too wide hides bimodality; too narrow shows noise. The "just right" zone reveals the true distribution shape.