Definition
A histogram shows the distribution of a continuous variable: values are divided into intervals (bins) and the height of each bar shows how many observations fall into it. It reveals the shape of the distribution — symmetry, skewness, multiple peaks, truncation or gaps — and with a fitted normal curve and tolerance limits it is the basis of capability analysis.
Data & visualisation
The appearance depends on the number of bins: too few smooth the shape away, too many fragment it. Try several widths; Sturges' rule or the square root of n gives a sensible start. Two peaks often mean two mixed populations (two machines, two shifts).
Complement the histogram with a cumulative curve or a probability plot when judging normality. For discrete data and categories use a bar chart of frequencies instead.
In Statistica
Histograms are in Graphs → Histograms (simple, categorized by group, with a fitted normal or other distribution and a goodness-of-fit test) and are part of Descriptive statistics and Process Analysis, where they are drawn with tolerance limits. Bin count, boundaries and colours are adjusted interactively in the graph.
Related terms
- Normal distributionThe normal (Gaussian) distribution is a symmetric bell-shaped distribution described by its mean and standard deviation.
- Box plotA box plot shows a distribution with five numbers: the box spans from the lower to the upper quartile (the middle 50 %…
- Process capability (Cp, Cpk)Capability indices compare process variability with the customer's tolerance limits.
- Normality testA normality test checks whether data come from a normal distribution — an assumption of the t-test, ANOVA, regression and capability indices.
Knowledgebase guides
FAQ
- How many bins should I choose?
- Roughly the square root of the number of observations (10 bins for 100 values); Statistica sets a default automatically and you can change it with a slider.
- How do I recognize a non-normal distribution?
- By skewness (a long tail on one side), multiple peaks or a shape too flat or too peaked compared with the fitted curve; confirm with a normality test.
Try it on your own data
Statistica free for 30 days
Full version, no credit card. Or get a pricing estimate in a minute.