Degrees of Freedom Calculator

Degrees of freedom describe how many independent pieces of information remain after accounting for constraints or estimated parameters. They determine the shape of many reference distributions used in hypothesis tests and confidence intervals. Although the phrase can sound abstract, the basic idea is practical: once some values are fixed by a condition, the remaining values are free to vary. For a one-sample t-test, estimating the sample mean imposes one constraint, so n observations provide n - 1 degrees of freedom. A paired t-test applies the same rule to the number of paired differences. The pooled two-sample t-test estimates two group means and combines variance information, giving n₁ + n₂ - 2. Welch's unequal-variance t-test uses a more complicated approximate formula and is not represented by the pooled option in this calculator. For a chi-square contingency table, the row and column totals constrain the cell counts. Once counts in (r - 1)(c - 1) cells are known, the remaining cells follow from those margins, so df = (r - 1)(c - 1). A one-way ANOVA separates degrees of freedom into between-groups and within-groups components. With k groups and N total observations, the between-groups value is k - 1 and the within-groups value is N - k. Their sum is the total N - 1. Multiple linear regression with n observations and p predictor variables estimates p slopes plus one intercept. The residual degrees of freedom are therefore n - p - 1. This value controls the denominator used to estimate residual variance and must be positive. Models with too many predictors relative to observations have little or no remaining information for estimating error, even if software can produce fitted values. Choose the option that exactly matches the method used in your analysis. Degrees of freedom are not interchangeable across tests, and special designs can require different calculations. Repeated-measures models, Welch tests, mixed models, corrected chi-square procedures, and regressions with penalties or complex survey weights may use adjusted or approximate values. The calculator covers common textbook formulas for independent observations. Report the test statistic together with its degrees of freedom, p-value, effect size, and assumptions so readers can understand how the inference was formed.