A number of tests have been designed to measure the degree of masculinity and femininity. The first to be developed, and the most comprehensive, is the Attitude-Interest Analysis Test of Terman and Miles, usually called the Masculinity-Femininity Test (1938). More recent tests, such as the scales included in the Strong Vocational Interest Blank, the Minnesota Multiphasic Personality Inventory, the Guilford-Zim- merman Temperament Survey as Well as the Gough Femininity Scale, are based on fewer and more limited items. These tests have some value in their specific context but do not correlate highly with each other.All these tests are based on the common principle of comparing men and women on responses that have proved most characteristic of each sex in recent American culture. The Guilford- Zimmerman Survey presents an inventory based on questions of the yes-no variety; one of the ten personality traits tested is masculinity-femininity. Among the characteristics indicative of masculinity are: masculine activities, not easily disgusted, hard-boiled, inhibits emotional expression, and little interest in clothes and style. Femininity is judged in terms of such characteristics as interest in feminine activities and vocations, easily disgusted, fearful, romantic, and emotionally expressive. The Strong Blank includes a masculinity- femininity scale based on interest patterns fond to be highly characteristic of men and women.The Masculinity-Femininity scale of Terman and Miles was based on particularly extensive research. In seeking test items that would successfully discriminate between the sexes, they first searched the psychological literature for types of test material that yielded the most marked sex differences, then prepared a preliminary set of items which were administered to elementary, high school, and college students, as well as unselected adults, members of different occupations, and a number of special groups such as athletes, juvenile delinquents, and adult homosexuals. On the basis of this research they retained items that yielded significant differences, and constructed a seven-part test consisting of word association, inkblot association, information, emotional and ethical attitudes, interests, opinions, and introvertive response. The scale proved successful in differentiating between male and female groups at all age levels from teen-agers to octogenarians, and threw considerable light on the more prominent sex differences in our culture. Here is a portion of Terman and Miles’ summary of these differences: “From whatever angle we have examined them the males included in the standardization groups evidenced a distinctive interest in exploit and adventure, in outdoor and physically strenuous occupations, in machinery and tools, in science, physical phenomena, and inventions; and from rather occasional evidence, in business and commerce. On the other hand, the females of our groups have evidenced a distinctive interest in domestic affairs and in esthetic objects and occupations; they have distinctly preferred more sedentary and indoor occupations, and occupations more directly ministrative, particularly to the young, the helpless, the distressed. Supporting and supplementing these are the more subjective differences—those in emotional disposition and direction. The males directly or indirectly manifest the greater self- assertion and aggressiveness; they express more hardihood and fearlessness, and more roughness of manners, language, and sentiment. The females express themselves as more compassionate and sympathetic, more timid, more fastidious, and esthetically sensitive, more emotional in general (or at least more expressive of the four emotions considered), severer moralists; they admit in themselves weaknesses in emotional control, but less noticeably in physique.”In interpreting these results, it must be recognized that the scores on the Terman-Miles Masculinity-Feminin- ity Test (and on similar scales) are based largely on the differences rather than the similarities between the sexes. Moreover, they tend to reflect the particular culture in which they were developed, and therefore emphasize social differences of the more extreme sort. They do not necessarily apply to other cultures—for instance, only a few of the items were found to differentiate between the sexes in Holland.more, there is some question of their value even in our own society. In applying their test, Terman and Miles found that dressmakers, domestic employees, and women over sixty rate highest in femininity, whereas twenty- year-old women (who may be at the height of their sexual attractiveness) are only moderately feminine since they are likely to have many “masculine” interests such as sports. Similarly, a young man who would be considered highly masculine in appearance and physical strength, and would be at the peak of his interest in women and his attractiveness to women, would nevertheless rate only moderately masculine on this test if he showed a special interest in music and religion. In other words, the scale gives too much consideration to custom and culture, and too little to sexual and physical differences.A number of other investigators have attempted to construct more accurate and penetrating tests. The California Psychological Inventory (Gough, 1957) is based on the same approach as the Masculinity-Femininity Test, but eliminates some of the more extreme social items such as attendance at pool halls as an indication of masculinity and at beauty parlors as an indication of femininity. A completely different approach is represented by the projective test devised by Franck and Rosen (1949), which is based on the assumption that men and women differ in fantasy and imaginative productions. On this test the subject is shown a series of simple lines and geometrical forms and is asked to draw a picture incorporating them. The results are then compared to productions which have been found to be typical of representative members of the subject’s sex. It is interesting that the scores do not correlate with the scores on the question and answer tests, and there is some evidence that they reveal latent aspects of masculinity and femininity that are not revealed on the social type of test. (Miller and Swanson, 1960)