Histogram

A histogram shows the distribution of a single continuous variable by dividing its range into consecutive intervals (bins) and drawing a bar for each, where the bar's height is the count of values falling in that bin. The bars touch because the bins are contiguous. It reveals the shape of the data — where values cluster, how spread out they are, whether the distribution is skewed, and where outliers sit.

A bar chart compares values across distinct categories and its bars have gaps because the categories are separate. A histogram shows the distribution of one numeric variable by binning it into ranges, and its bars touch because the bins are continuous. Rule of thumb: categorical x-axis → bar chart; numeric x-axis split into ranges → histogram.

Bin count controls the story. Too few bins hide the shape (everything looks like one lump); too many make it noisy and spiky. Common starting points are the square-root rule (bins ≈ √n) or Sturges' / Freedman–Diaconis rules, but the best move is to try a few widths and pick the one that shows the real structure without exaggerating noise. Keep bin widths equal so heights are comparable.

A lot: a single central peak suggests one dominant group; two peaks (bimodal) hint at two underlying populations; a long right tail means right-skew (a few large values), common in revenue or session-length data; a flat shape means values are spread evenly. The center tells you the typical value, the width tells you variability, and isolated bars far out flag outliers.

A histogram shows the full shape of one distribution in detail — peaks, skew, gaps — which is ideal when the shape matters. A box plot summarizes a distribution into median, quartiles, and outliers, which is more compact and far better for comparing several groups side by side. Use a histogram to understand one distribution deeply; use box plots to compare many distributions at a glance.