Alex B. Caldwell, Ph.D.
Hathaway (see Dahlstrom & Dahlstrom, 1980, pp. 73-75) primarily used gay vs. straight males to develop the scale. Secondarily he used the old Terman-Miles (1936) test, and thirdly (least weight) he looked at adult male vs. adult female item response differences. He sought to measure what he saw as important psychological differences between gay and straight males (in contrast to object choice prediction), and recent neuroscience research looks potentially supportive of his perception.
He did a separate Fm scale for lesbians vs. straight women, but the scale did not work very well, in my hindsight possibly because of a greater heterogeneity of gender identities in the sample of women. For example, Kinsey et al. (1953) showed that gays strongly tended to remain gay, and Hooker studied how difficult it is for gays to shift to (at least outwardly) straight gender roles. Women can shift from sexual relationships with men to women and back - or to being sexually inactive - much more easily than gay males. Hathaway et al. felt it would be less confusing to have a single scale, although four items that report being concerned about sex are scored in the opposite direction. Note that the norms go in the opposite direction (high is resemblance to the opposite gender, so a high score is feminine for men but a low score is feminine for women).
In understanding scale 5-Mf, I find it helpful to distinguish between gender identity, gender role, and object choice. Object choice is highly idiosyncratic and I think often susceptible to chance circumstances in terms of the person’s earliest and often acutely intense adolescent encounters. Self-consciousness and socially problematic behaviors often lead to object choices being concealed. This can also drive discrepancies between one’s role and identity; gender role can be a mask, and it often is. Regarding Mf and possible changes over the years, a followup study showed a high correspondence of the same descriptors showing up for high and low scoring male and female college students after a 30-year period (Todd & Gynther, 1988): there was a remarkable absence of change. Lists of descriptors reflect the expression of both role and identity, but I believe that if there were changes in the underlying gender identities, the behavioral expressions would show at least some changes. I like to think that there has been some increment in the general acceptance of cross-role behavior over time (e.g., unisex clothing, gay marriages, etc.) But to me, the Mf scale is primarily about gender identity, and I see little if any change in this basic underlying dimension.
Regarding interpretation, I propose that the underlying dimension is best summed up as being defined by an orientation toward actions on the masculine end and an orientation toward feelings on the feminine end, construing both "actions" and "feelings" broadly. This helps get us off the hook of such pejorative terms as fag, queer, bull dyke, etc. They hardly follow from the scale anyhow, especially since the use of such words is largely in response to gender role and object choice more than to identity. The lowest scoring males I have seen or known often have strong exploratory urges and individual needs for mastery over nature; they may not understand how women can apparently sit and talk all day. This does not, however, presume the abusive attitudes often attributed to machismo; other scales assess aggressive potentials. Some are well aware of the vulnerability of women, and they can be protective and gentle in a nevertheless very masculine way. High T-scoring males are much more interested in what is happening in the personal (emotional feeling) lives of the people around them as well as their own feelings; they may find hunting, boxing, etc., distasteful if not repugnant.
Adjectives more often descriptive of lower scoring women have included approachable, charitable, emotional, lazy, and tolerant; they have also been seen as worldly, sensitive, and self-dissatisfied. Many seem protective of their rights to the expression of their feminine identities (especially if they have felt exploited or oppressed). Higher scoring women often value physical strength and endurance and may be seen as adventurous, calculating, self-assured, exploitative, and self-confident. They often have tomboy elements in their histories and subtly masculine traits (largely independent of object choice). In one study, by far the strongest behavioral association with scale 5 in women was their answer to the question, "If you had to choose between your mate versus your work which would you choose?" The "action" end chose work and the "feeling" end chose mate.
I see the feminine end of the scale (both genders) as being more strongly associated with verbal expressions of anger and aggression (almost "verbally only" toward the extremes), and that physical expressions of aggression are somehow more a part of the nature of the world for those scoring at the masculine end (lower 5 men and higher 5 women). That would not be a specific prediction of behavior in an immediate sense but rather a shift of potential thresholds. I would add that I think that when low-5 women do strike out physically, it is apt to be more pain- and distress-inflicting (e.g., an old French movie with low SES women working in a laundry-a fight broke out and they were tearing out each others earrings and scratching each others’ faces), and that when high-5 women fight, it is more like male fighting over territoriality and prerogatives. Masculine aggression may often be seen as just enough for an efficient control of a child’s or another person’s behavior, and not necessarily as deliberately mean or cruel or tormenting; the latter, like machismo above, depends on other scale elevations such as 4-Pd, 6-Pa, 8-Sc, and 9-Ma.
Let me close with observations on the highest scale 5 score (female) I have ever seen. A woman came to the psychiatry unit at the University of Minnesota (1950's) wanting help in order to get married to another woman who also wanted the marriage. At a T just over 90, she reportedly had never worn a dress in her life, she was expert in using tractors, she went hunting alone, and she loved to go fishing alone. Psychologically she was "a man with a practical action problem." Given a satisfying complementarity of roles, their only problem was they could not have children, but they had accepted that. Unfortunately, the issue was a legal one and not psychiatric, so we were not really able to help her.
Dahlstrom, W. G., & Dahlstrom, L. E. (1980). Basic readings on the MMPI: A new selection on personality measurement. Minneapolis: University of Minnesota Press.
Kinsey, A. C., Pomeroy, W. B., Martin, C. E., & Gebhard, P. H. (1953). Sexual behavior in the human female. W. B. Saunders: Philadelphia & London.
Terman, L. M., & Miles, C. C. (1936). Attitude-interest analysis test. McGraw-Hill Book Company, Inc.
Todd, A. L., & Gynther, M. D. (1988). Have MMPI Mf correlates changed in the past 30 years? Journal of Clinical Psychology, 44, 505-510.