Degrees of Freedom« Back to Glossary Index
Degrees of Freedom
A technical term in a statistical analysis that has to do with the power of the analysis. The more degrees of freedom, the more powerful the analysis. The degrees of freedom is typically the number of observations in a sample less the number of unknown parameters estimated for the model. It reflects a sort of adjusted sample size, with the adjustment based on the number of unknowns that need to be estimated in a model. For example, in a two-sample t-test the degrees of freedom is n1 + n2 – 1 – 1 since there are n1+n2 subjects altogether and 1 mean estimated in one group and one mean in another giving n1 + n2 – 2.