Oakes, P. J., Turner, J. C. & Haslam, S. A. (1991). British Journal of Social Psychology, 30(2), 125–144.
Not read. The summary is drawn from the abstract or from what the work is standardly cited for. A reading should check it first.
Why this is here
The primary experiment behind THEORY-1114’s fit clause, reported in Turner et al. 1994 (NOTE-1284) and Reimer et al. (NOTE-1251). That THEORY’s promote_when asks for a primary study read in full; this is one.
Read in its place
The work cannot be read here, so the record holds it at second hand.
Spears 2021 (NOTE-1268). It does not cite the 1991 paper. It defines comparative fit (groups differ on the dimension) and normative fit (they differ in the stereotyped direction) in the same terms (p. 373). It also reports a study of the same form from another lab. In Lea et al. 2007, cross-cutting nationality and sex subgroups each bipolarized only on topics that fitted their own category. Missing still: the 1991 design (six arts and science students in a staged discussion), its measure of salience by attributions to academic field, its data, and whether comparative and normative fit had separable effects.