Histogram

A histogram divides the value range of a metric characteristic into equally wide bins and draws one bar per bin whose height gives the count of values inside it. It answers three questions at once: where does the distribution sit, how wide is it, and what shape does it have?

• Location — where is the centre of mass? Compare with mean and median
• Spread — how wide is the distribution? Relates to the standard deviation
• Shape — symmetric, skewed, bimodal or truncated
• Outliers — isolated bars far from the main body

The number of bins is the only real tuning knob and changes the picture considerably: too few smooth away real structure, too many create random spikes. Common rules of thumb are $k\approx \sqrt{n}$ or Sturges' rule; Scott's bin width additionally accounts for the spread.