Tokens in the tower: Perceptual processes and interaction dynamics in academic settings with `Skewed', `Tilted' and `Balanced' sex ratios

We tested Kanter's (1977a, 1977b) theory concerning the effects of group proportions (sex ratios) on visibility, polarization and assimilation, using natural groups of women and men in academia. Study 1 compared male-skewed and male-tilted settings and found evidence of greater polarization by minority women than majority men. The only effect of group proportions occurred for perceived dispersion as a measure of assimilation; replicating Brown and Smith (1989), men showed an out-group (OH), and women an in-group (IH), homogeneity effect, and both effects were accentuated in the skewed setting. Study 2 extended the research to include male-skewed, male-tilted, balanced and female-tilted sex ratios. Men's OH effect declined as relative out-group size increased, and women's IH effect declined as relative in-group size increased. There was also a linear decrease in relative perceived in-group impact and status as actual relative in-group size declined. We discuss our findings with respect to the validity of Kanter's theory, gender and group size as moderators of perceived variability, and methodological issues in studying diversity.