Using ratings provided by significant others, we examined characteristics of 819 norma! men whose scores or, the Masculinity-Femininity scale (Scale 5) of the Minnesota Multiphasic Personality Ir;ventory-2 (MMP1-2) ranged from low to high. Also examined were the possible effects of educational ievei as a moderator variable. Only one external characteristic was correlated with Scale 5 scores at a level of significance that could not be attributed tc chance, and that correlation was negative, whereas previous literature indicated a positive relationship. The findings were essentially the same whether or not the effects of education were controlled. Regardless of Scale 5 scores and the higher the men's educational levels, the more positive were ratings by their partners. In general, the results do not support the usefulness of Scale 5 in describing the personalities and behaviors of norma! men. © 1991, Taylor & Francis Group, LLC. All rights reserved.