withingroup

Withingroup, often written within-group or within-group, is a term used across disciplines to describe analyses, comparisons, or processes that occur within the members of a single group, as opposed to between-group analyses that compare different groups. In statistics, within-group variation refers to variability among observations that share the same group label, contrasted with between-group variation. In an analysis of variance (ANOVA) framework, total variation is partitioned into within-group and between-group components. Within-group effects reflect differences that persist after accounting for group membership, and are central to understanding phenomena not attributable to group differences alone.

In repeated-measures or within-subject designs, measurements are taken from the same subjects under different conditions, and the analysis emphasizes within-subject or within-group changes, controlling for between-subject differences. In social sciences and psychology, within-group analyses study dynamics, norms, or outcomes that arise among members of the same demographic, organization, or social unit, often to identify processes less confounded by cross-group factors. The term is frequently encountered with hyphenation, though it can appear as a compound noun or adjective.

Withingroup contrasts are useful for isolating effects that operate within a common context and for improving statistical efficiency by exploiting the shared variance within groups. It is distinct from ingroup concepts in social identity theory, where the focus is on group membership and identity rather than on the scope of analysis. See also: within-subject design, within-group variance, between-group comparison, mixed-effects models, and hierarchical modeling.