Definition
Sample size is the count of observations included in a statistical study — typically denoted n. It governs how closely a sample-based estimate (a mean, a proportion, a regression coefficient) is expected to track the true value in the underlying population. Doubling the data does not double the precision, but precision does improve in a predictable, formula-driven way as n grows.
Choosing the right sample size is one of the few decisions in a study that can be made in advance with arithmetic rather than judgement alone. Statisticians call this calculation a power analysis, and it is the standard first step in designing any experiment or survey.
Why it matters
How it works
The relationship between sample size and precision runs through the standard error. For a simple mean, the standard error equals the population standard deviation divided by the square root of n. That square-root term is the key intuition: to cut your uncertainty in half you need four times the data, and to cut it in tenths you need a hundred times the data. Diminishing returns are baked into the mathematics, which is why most well-designed studies converge on sample sizes in a particular range rather than chasing maximum possible n.
Sample size also drives statistical power — the probability that a test will detect a true effect of a given size. A small sample can miss a real difference simply because random noise swamps the signal; this produces a false negative even when the underlying effect is genuine. Conversely, very large samples make tests so sensitive that practically meaningless differences register as statistically significant. The discipline is to choose n deliberately for the effect size that matters in context — large enough to catch real effects, not so large that trivial ones dominate the headline.