Bem Sex-Role Inventory
KeywordsPersonality Characteristic Masculinity Score Individual High Masculinity Scale Femininity Score
The Bem Sex-Role Inventory (BSRI) is a self-rating inventory of masculine, feminine, and neutral personality characteristics; the BSRI score is calculated in a way that reflects the relative weight of the scores on the independently measured Masculinity scale and Femininity scale.
Prior to the 1970s, masculinity and femininity were conceptualized and measured as opposite ends of a singular continuum, such that an individual high on masculinity was low on femininity and an individual high on femininity was low on masculinity. In the early 1970s, Sandra Bem (1974) began developing the Bem Sex-Role Inventory (BSRI), an instrument designed so that respondents could rate themselves on personality characteristics in a way that allowed the separate measures of masculinity and femininity to vary independently of one another. On the BSRI, the score on the Masculinity scale and the score on the Femininity scale could then be used together in a calculation that classifies respondents into one of four groups: androgynous (high masculine-high feminine), masculine (high masculine-low feminine), feminine (low masculine-high feminine), and undifferentiated (low masculine-low feminine) (Bem 1981).
Development of the BSRI
In order to develop the BSRI, Bem (1974) gave a list of approximately 400 personality traits to 100 Stanford University undergraduate students in the winter (N = 40) and summer (N = 60) of 1972. Respondents were instructed: “We would like you to indicate how desirable it is in American society for a man/woman to possess each of these characteristics. Note: We are not interested in your personal opinion of how desirable each of these characteristics is. Rather, we want your judgment of how our society evaluates each of these characteristics in a man/woman” (1981, p. 17). The 50 male and 50 female respondents acting as judges were asked to rate the desirability of each characteristic using a seven-point scale ranging from 1 (“Not at all desirable”) to 7 (“Extremely desirable”) (Bem 1974, p. 157). Each respondent was asked about the desirability of traits for a man or for a woman but not for both.
In order to develop the BSRI, characteristics were labeled as masculine, feminine, or neutral based on the ratings of these 100 undergraduate students about the desirability of the characteristics. “A personality characteristic qualified as masculine if it was independently judged by both males and females in both samples to be significantly more desirable for a man than for a woman (p < 0.05). Similarly, a personality characteristic qualified as feminine if it was independently judged by both males and females in both samples to be significantly more desirable for a woman than for a man (p < 0.05). Of those personality characteristics that satisfied these criteria, 20 were selected for the Masculinity scale and 20 were selected for the Femininity scale” (1974, p. 157).
The Masculinity scale is composed of the following personality characteristics: acts as a leader, aggressive, ambitious, analytical, assertive, athletic, competitive, defends own beliefs, dominant, forceful, has leadership abilities, independent, individualistic, makes decisions easily, masculine, self-reliant, self-sufficient, strong personality, willing to take a stand, and willing to take risks. The Femininity scale is composed of the following personality characteristics: affectionate, cheerful, childlike, compassionate, does not use harsh language, eager to soothe hurt feelings, feminine, flatterable, gentle, gullible, loves children, loyal, sensitive to the needs of others, shy, soft spoken, sympathetic, tender, understanding, warm, and yielding. The BSRI also includes 20 neutral personality characteristics, that is, those that could not be classified as either masculine or feminine based on the criteria used by Bem. The 20 neutral items are as follows: adaptable, conceited, conscientious, conventional, friendly, happy, helpful, inefficient, jealous, likable, moody, reliable, secretive, sincere, solemn, tactful, theatrical, truthful, unpredictable, and unsystematic. In addition to the original 60-item BSRI, a short form BSRI was developed using statistical techniques, including factor analysis and item-total correlations, and consists of a 30-item subset of the 60 personality characters that appear on the long form BSRI (Bem 1981). The 30-item short form BSRI is composed of a 10-item Masculinity scale, a 10-item Femininity scale, and a 10-item scale of neutral personality characteristics.
Administration and Scoring of the BSRI
Respondents are given either the 60-item original form or the 30-item short form and instructed to consider the personality characteristics on the list and indicate the degree they believe each personality characteristic describes them by using a 7-point scale (1, Never or almost never true, to 7, Almost or almost always true) for their self-ratings. For both the long and short form BSRI, the list begins with a personality characteristic from the Masculine scale (e.g., assertive) followed by a personality characteristic from the Feminine scale (e.g., gentle) and then by a neutral personality characteristic (e.g., helpful). This pattern is repeated for the listing of the remaining personality characteristics on the instrument, though at the time of administration, the respondent is not made aware of this pattern nor to which scale (Masculinity scale, Femininity scale, Neutral scale) each personality characteristic belongs.
Although a respondent is asked to provide self-ratings of all 60 of the personality characteristics, the 20 neutral personality characteristics are not used in the calculation of the BSRI score. Measuring the extent to which a respondent adheres to sex-typical expectations or not based on their BSRI score, also referred to as the androgyny score, involves several calculations. The first two calculations determine a respondent’s (1) masculinity score, based on the mean of the self-ratings of the 20 personality characteristics on the Masculinity scale, and (2) femininity score, based on the mean of the self-ratings of the 20 personality characteristics on the Femininity scale. These two scores are then used to determine the final BSRI score and classification, which takes into account the relative weight of the masculinity score and femininity score (Bem 1974, 1981). The simplest version of the BSRI score is calculated by subtracting the Masculinity score from the Femininity score (Bem 1977, p. 322). Another more complex version of scoring involves separately converting the Masculinity score and the Femininity score to standardized T-scores and calculating the difference between the two standardized T-scores (Bem 1981, p. 5).
Regardless of the method used, larger absolute values, either positive or negative, indicate that the respondent is more sex typed. As a result of the way in which the masculinity and femininity scores are used in the calculation, a large positive value reflects a much higher mean score on the Femininity scale than on the Masculinity scale, whereas a large negative value reflects a much higher mean score on the Masculinity scale than on the Femininity scale. When a respondent’s BSRI score is closer to zero, this reveals more equal self-ratings of personality characteristics on the two scales with moderate to high scores on both scales yielding a classification of the respondent as androgynous. For Bem, this particular classification category was part of the purpose of developing an instrument that allows the measures of masculinity and femininity to vary independently of one another. That is, the instrument allows individuals to provide responses that reflect a mix of both masculine and feminine personality characteristics.
Bem suggested that androgynous individuals may “come to define a more human standard of psychological health” (1974, p. 162). Using the BSRI score as an independent variable, Bem and her colleagues (Bem, 1975; Bem et al. 1976) later conducted experiments that revealed that those whose scores classified them as androgynous had a wider repertoire of behavior that allowed them to “engage in situationally effective behavior without regard for its stereotype as masculine or feminine” (1975, p. 643).
Assessment and Use of the BSRI
More than 25 years after the development of the BSRI, researchers asked respondents to rate the 60 personality characteristics on the BSRI as desirable for a man and for woman using the same wording and criteria as Bem (Auster and Ohm 2000). The results showed that all but 2 (i.e., childlike, yielding) of the 20 personality characteristics on the Femininity scale still qualified as feminine, but only 8 (i.e., acts as a leader, aggressive, ambitious, dominant, forceful, has leadership abilities, independent, and masculine) of the 20 personality characteristics on the Masculinity scale still qualified as masculine. The belief by both male and female respondents that most of personality characteristics on the Femininity scale were significantly more desirable “for a woman” than “for a man” but that this was no longer the case for a majority of the personality characteristics on the Masculinity scale revealed societal changes over that time period regarding expectations for women and men, most notably that many characteristics previously labeled as masculine are now seen as desirable for both women and men. Konrad and Harris (2002) examined the impact of gender, race (European American, African American), region (northeast, south), and size of urban area (large, small) and found variations in the ratings of the desirability of the 40 personality characteristics that compose the Masculinity and Femininity scales. These two studies show that the desirability ratings have varied over time as well as by the demographics of the respondents.
In addition to the assessing the BSRI, researchers have used the BSRI in a variety of ways. Some interested in how BSRI self-ratings might vary by the context asked respondents to think of themselves as a parent, student, or romantic partner (Uleman and Weston 1986) or to consider the social context, such as interacting at home, at work, and at school, with same-sex friends, with other-sex friends, and where one does not know many people (Smith et al. 1999). The findings of such studies revealed that the social role and context have some impact on self-ratings on the BSRI. Other researchers have looked at changes in BSRI self-rating scores over time. A meta-analysis of 63 studies that used the BSRI revealed that women’s self-ratings on the Masculinity scale have been increasing steadily over time; this has also resulted in male-female differences in self-ratings on the Masculinity scale decreasing over time (Twenge 1997). In addition, more than 25 years after the development of the BSRI, when respondents were given a list of the 60 personality characteristics on the BSRI and asked to indicate the importance of these personality characteristics for themselves rather than actually rating themselves, 13 of the highest rated 15 personality characteristics were the same for women and men and were a mix of the personality characteristics that make up the Masculinity scale and the Femininity scale on the BSRI (Auster and Ohm 2000). Most recently, researchers have used the BSRI instrument to compare respondents’ ratings of personality characteristics for ideal political candidates and actual women and men running for office (Powell and Butterfield 2016).
The Bem Sex-Role Inventory was groundbreaking in that it allowed masculinity and femininity to be measured independently rather than as part of a singular continuum, and as a result, the BSRI became a commonly used measure. Despite changes in gender expectations, variations in desirability ratings by respondents’ demographic characteristics, and variations in respondents’ BSRI score depending on social context or social role, the BSRI persists as a popular instrument that is still used to measure the extent of adherence to traditional, perhaps stereotypical, expectations for women and men.
- Bem, S. L. (1977). Psychological androgyny. In A. G. Sargent (Ed.), Beyond sex roles (pp. 319–324). St. Paul: West Publishing Company.Google Scholar
- Bem, S. L. (1981). Bem Sex-Role Inventory: Professional manual. Palo Alto: Consulting Psychologists Press.Google Scholar
- Powell, G. N., & Butterfield, D. A. (2016). Sex, gender, and leadership in the public sector: Is the role of femininity greater at lower levels? In M. F. Karsten (Ed.), Gender, race, and ethnicity in the workplace: Emerging issues and enduring challenges (pp. 181–200). Santa Barbara: Praeger.Google Scholar