Gender, gender roles and attitudes towards violence: Are viewpoints changing?

Social Behavior and Personality, 2001 by Smith, Stephen C, Ellis, Jon B, Benson, Trisha A

Androgyny is a gender role that has several positive characteristics such as adaptive coping strategies and less violent viewpoints. To examine the relationship between gender, gender roles, and attitudes towards violence, 161 individuals were administered the Bern Sex Role Inventory and a violence questionnaire. It was hypothesized that a significant number of nonandrogynous people would have a great predisposition towards violence. The mean violence score was higher for men than for women, however, there was no difference between androgynous individuals and others. There were no significant interaction effects. Results are surprising when compared to previous research. It may be that more and more people, regardless of their interpersonal style, are moving towards a view which does not condone violence.

When the term androgyny was introduced in the Journal of Consulting and Clinical Psychology by Bern in 1974, it was a technical concept. The concept itself was intended to denote individuals whose self-reports indicate that they exhibit traits that are stereotypically masculine and stereotypically feminine (Durkin, 1985). The advantages of androgyny result in an individual who is typically more flexible than individuals who are either masculine or feminine. The stereotypes used in the Bern Sex Role Inventory (BSRI, (Bem, 1993) are not defined by the selfreports of people: instead these stereotypes are defined by cultural ideas of what is masculine and what is feminine. Bern stated that society identifies assertiveness and independence with masculinity. Femininity is reflected by behaviors such as tenderness and understanding. The theory of androgynous adaptability and nonandrogynous behavioral restrictiveness is supported within several studies (e.g., Bem, 1974; Ho, 1981; Lee & Scheurer, 1983; Orlofsky & Windle, 1978; Pyke, 1985). Based on the postulation that individuals are more adaptive if they are androgynous, some studies have shown a correlation between violence, maladaptive behavior, and non-androgynous individuals. Worth, Matthews, and Coleman (1990) found that men who participated in abusive interactions had lower femininity scores than did men without abusive interactions. A correlation of non-androgynous sex roles and violence has also been frequently demonstrated (Bernard, Bernard & Bernard, 1985; Finn, 1986; McConaghy, Zamir, Manicavasagar, 1993; Rosenbaum, 1986; Thompson, 1991; Thoreson, Shaughnessy, Cook & Moore, 1993). This is not surprising, since people who are less flexible and adaptive are typically less likely to cope adaptively with issues in life, including frustration. This, in turn, can cause a violent response.

The purpose of this study was to investigate the relationship between gender, gender-roles and attitudes towards violence. Based on previous research, it was hypothesized that a significant number of non-androgynous people would have a greater predisposition towards violence when compared to androgynous people.

METHOD

SUBJECTS

All subjects were undergraduate students enrolled in a southern university with an enrolment of approximately 12,000. There were a total of 161 individuals (85 women, 76 men). The mean age was 21.7 years, with a range of 17 to 54 years. In terms of demographics, 83% were single, 88% were white, and 82% of the subjects did not have children.

MATERIALS

All subjects completed an informed consent form, a short demographic questionnaire, a violence survey, and a BSRI-short form (Bem, 1981). The BSRI-short form includes exactly half the items on the original BSRI (Bem, 1974), and correlates highly (r = .85 to .94) with the corresponding scales on the original BSRI. Product-moment correlation scores are highly reliable (r = .76 to .91) with the lowest test-retest reliability (.76) occurring for men describing themselves with 10 masculine adjectives, 10 feminine adjectives, and 10 neutral adjectives. All of the adjectives are rated on a Likert scale ranging from one (never to almost never true) to seven (always or almost always true). Using subject medians from this study, subjects were classified as masculine, feminine, both masculine and feminine (androgynous), or neither masculine nor feminine (undifferentiated). Those individuals who scored above the median on both femininity and masculinity were classified as androgynous. Those who scored above the median on femininity and below the median on masculinity were classified as feminine. Those who scored above the median on masculinity and below the median on femininity were classifled as masculine. Those who scored below the median on both femininity and masculinity were classified as undifferentiated.

The violence survey makes statements to be rated on a Likert scale ranging from one (strongly disagree) to five (strongly agree). Examples of the 59 statements in the violence survey are as follows: "It is OK to destroy another person's mailbox as a prank." It is OK for a teacher to hit a student who is disobedient." This violence survey has been previously used (MacIntyre & Cantrell, 1995a, 1995b) and has been shown to differentiate adults who have experienced violence from those who have not. The survey is a modified version of one which was conceived by Baron, Straus and Jaffee (1989).