«This article describes the development and psychometric evaluation of a twenty-four-item scale to measure attitudes toward gender norms among young ...»
Men and Masculinities
Volume 10 Number 3
April 2008 322-338
© 2008 Sage Publications
Measuring Attitudes toward http://jmm.sagepub.com
Gender Norms among Young http://online.sagepub.com
Men in Brazil
Development and Psychometric
Evaluation of the GEM Scale
This article describes the development and psychometric evaluation of a twenty-four-item scale to measure attitudes toward gender norms among young men: the Gender-Equitable Men (GEM) Scale. Scale items on gender norms related to sexual and reproductive health, sexual relations, violence, domestic work, and homophobia are designed. Items are based on previous qualitative work in the community and a literature review and administered to a household sample of 742 men, including 223 young men ages fifteen to twenty-four, in Rio de Janeiro, Brazil. The current analysis focuses on the young men, as they were the main audience for a planned intervention to promote gender equitable and HIV risk reduction behaviors. Factor analyses support two subscales, and the scale is internally consistent (alpha =.81). As hypothesized, more support for equitable norms (i.e., higher GEM Scale scores) is significantly associated with less self-reported partner violence, more contraceptive use, and a higher education level.
Keywords: gender; norms; masculinity; HIV; violence; measurement A discussion of gender norms—defined here as social expectations for appropriate behaviors of men as compared to women—has been at the forefront in recent years of international efforts to achieve gender equity, including, for example, the 1994 Authors’ Note: This research was funded through the Horizons Program/Population Council (via USAID Cooperative Agreement HRN-A-000-97-00012-00), the MacArthur Foundation, and SSL International.
Special thanks to Marcio Segundo and Marcos Nascimento, Instituto Promundo, for their substantive technical input, and Bebhinn NiDhonnail, formerly of Instituto Promundo. Thanks to Fernando Acosta, Instituto Noos, Rio de Janeiro, for his collaboration on this baseline testing of the GEM Scale and to Marina Teixeira for coordinating the data collection. Thanks to Louis Apicella, Horizons/Population Council, and Roberto Valverde, formerly of Horizons/Population Council, for their assistance with data analysis and Ellen Weiss, Horizons/ICRW, for her comments. The authors would also like to acknowledge the founding members of the Program H Alliance, including Salud y Genero, Instituto Papai, ECOS, JohnSnowBrasil, PAHO, and IPPF/WHR.
Pulerwitz, Barker / Gender Norms Among Young Men in Brazil 323 International Conference on Population and Development. It is increasingly accepted that support for inequitable gender norms negatively influences sexual and reproductive health-related behaviors and disease prevention as well as men’s use of violence against women (e.g., Amaro 1995; Campbell 1995; Cohen and Burger 2000; Worth 1989). Of note, reviews and studies with men across the globe have offered tremendous insights into how inequitable gender norms influence the way men interact with their female partners on a wide range of issues, including HIV/sexually transmitted infection (STI) prevention, contraceptive use, physical violence, domestic chores, and parenting (Barker 2000b; Kaufman 1993; Kimmel 2000; Marsiglio 1988; Rivers and Aggleton 1998). In sum, increasing evidence suggests that men’s collective and individual attitudes about gender norms as well as the social reproduction of these norms in institutions and cultural practices are directly related to many of men’s behaviors, with health implications for themselves and their partners.
Many programs have described gender equity as a program goal but have rarely assessed how the program interventions contributed to achieving gender equity and gender-equitable attitudes or behaviors among men (White, Greene, and Murphy 2005).
Recently, a number of pilot programs that specifically promote gender-equitable norms and related behaviors have begun to be implemented in different cultural settings (see White et al. 2005 for review of programs). It is important to measure the impact of these programs on gender-related attitudes as well as on related risk and prevention behaviors. This article describes the development and psychometric evaluation of a scale, particularly intended for program evaluation use, to measure attitudes toward gender norms among young men in Brazil: the Gender-Equitable Men (GEM) Scale.
Scale development in the field, especially related to a construct called “masculine ideologies,” is not new. Since the 1970s, various researchers have sought to measure masculine ideologies, defined as “beliefs about the importance of men adhering to culturally defined standards for male behavior” (Pleck, Sonenstein, and Ku 1993, 11).
A number of scales have been developed and affirmed to be valid and reliable (see Thompson and Pleck 1995 for a review of these scales). These scales on the whole assess the extent to which individuals agree with a specific belief system about masculinity. Similarly, other researchers have developed scales to measure sex role egalitarianism, which measure the propensity to hold views about others independent of whether they are male or female (King, King, Carter, Surface, and Stepanski 1994).
This scale addresses a number of domains, including educational roles, employment roles, parental roles, marital roles, and social roles.
Although this previous scale development informed the development of the GEM Scale, the existing scales did not entirely meet the demands for a meaningful evaluation of interventions to engage young men in questioning gender norms related to sexual and reproductive health and intimate relationships, and developing a new measure was deemed important. First, few of these scales have been tested and grounded in developing country realities and were developed with and for 324 Men and Masculinities U.S.-based audiences. Second, few, if any, of these scales were developed and used with the explicit purpose of program evaluation—that is, a measure that examines potential changes in attitudes toward gender norms as a result of a programmatic intervention. Third, the GEM Scale authors were especially interested in certain domains within the construct of gender norms—those related to intimate relationships, sexual and reproductive health, and disease and violence prevention, which addressed the main goals of the interventions in question. Although the sex role egalitarian scale includes a number of important domains, such as marriage, education, and work, it does not focus on some of the key domains in question for the authors, particularly those related to sexual relationships, reproductive health, disease prevention, and violence. Finally, although there is some overlap between the constructs, the authors would argue that there is a distinction between the concepts of masculine ideology and support for equitable gender norms in intimate relationships (the focus of the proposed scale). Masculine ideology scales for the most part measure how men define themselves as men. Although some of the issues addressed are relational, in that they involve comparisons between how men should be or are, as compared to women, these scales mainly do not assess how men view relationships with women and the degree of equality or inequality in those relationships.
The GEM Scale described in this article is intended to have a few key characteristics. It is intended to (a) be multifaceted and measure multiple domains within the construct of gender norms, with a focus on support for equitable or inequitable gender norms; (b) address program goals related to sexual and intimate relationships and sexual and reproductive health and disease prevention; (c) be broadly applicable yet culturally sensitive, so indicators can be applied in and compared across varied settings and be sufficiently relevant for specific cultural contexts; and (d) be easily administered, so that a number of actors—including the organizations that are implementing interventions—can take on this type of evaluation.
The GEM Scale emerges out of a social constructionist perspective of gender identity (e.g., Connell 1987, 1995; Kimmel 2000). According to this overarching conceptual framework, any given cultural setting provides a version, or multiple versions, of appropriate behaviors for men and women. These gender norms, which are passed on to boys and young men by their families, peer groups, and social institutions among others, are interpreted and internalized by individual men. Individuals also “reconstruct” these norms, by in essence putting their own “subjective spin” on the gender norms around them (Barker 2001), and as members of society, these individuals also influence the broader norms. This conceptual framework highlights that certain models of manhood or masculinity are promoted in specific cultural settings but that individual men will vary according to how much they adhere to these norms Pulerwitz, Barker / Gender Norms Among Young Men in Brazil 325 and that norms can evolve or change over time as individuals and groups reconstruct them. Furthermore, this conceptual framework also recognizes gender as based in power relations and as relational or created and reinforced through ongoing interactions between men and women.
A literature review was conducted to explore existing evidence related to associations between gender norms and key issues to be included in the planned program, such as HIV/STIs, sexual behavior and relationships, domestic life and child care, and partner violence. As more exhaustive reviews have been described elsewhere in the literature (e.g., Rivers and Aggleton 1998 on gender and HIV), only a few of the most salient points for this discussion are highlighted here. Various publications describe how attitudes and behaviors stemming from inequitable gender norms play an important role in sexual relationships and sexual and reproductive health and risk, such as risk of HIV/STIs and violence. As one example of a specific norm or expectation, young men can view sexual initiation and having regular sexual relations as a way to affirm their identity as men (Marsiglio 1988), and therefore early sexual initiation and maintaining multiple sexual partnerships are potential risk behaviors that are normatively encouraged. As another example, women often feel that they “cannot” buy or carry condoms, as if they were to do so, this would suggest that they intended to have sex and they could be labeled, very negatively, as “promiscuous” women (Childhope 1997). Turning to risk of violence, more than thirty studies from different cultural contexts have shown that between one-fifth and one-half of women interviewed have been subject to physical violence by a male partner (Heise 1994).
The causes and factors associated with men’s use of physical and sexual violence against women are complex but among them, various authors posit, are aspects of the social construction of masculinity (e.g., Kaufman 1993). The literature also describes how boys are socialized into an environment with norms about household roles and childrearing. For example, studies across the globe find that fathers tend to contribute about one-third to one-fourth of the time that mothers do in direct child care (Population Council 2001). In sum, existing literature supports the notion that boys and young men are socialized around a constellation of gender norms related to sexual and reproductive health and risk, sexuality, fatherhood, use or acceptability of violence against women, and participation in domestic chores.
Formative Research: Operationalizing Gender-Equitable Norms The development of the GEM Scale is grounded in formative, qualitative research carried out by one of the authors with young men in low income settings in Rio de 326 Men and Masculinities Janeiro (Barker 2000a, 2001) and with colleagues in a second study with both younger and older men (Instituto Promundo and Instituto Noos 2003). These studies explored (a) the norms that men perceived about male-female relationships and interactions (almost all of the men self-identified as heterosexual); (b) phrases and expressions they used to describe those norms, and in some cases used to justify or describe their own behavior; and (c) the dimensions or domains of male-female interactions in this setting.
Research methods for the 1999-2000 study included observation and life history interviews with twenty-five young men ages fifteen to twenty-one for one year, interviews with family members of some of the young men, focus group discussions, and key informant interviews. Based on results from this qualitative research with the study population (and the literature review), the term gender-equitable young man
has been operationalized here as a man who:
Seeks relationships with women based on equality, respect, and intimacy rather than sexual conquest. This includes believing that men and women have equal rights and that women have as much “right” to sexual agency as do men.
Seeks to be involved in household chores and child care, meaning that they support taking both financial and care-giving responsibility for their children and household.
Assumes some responsibility for sexually transmitted infection prevention and reproductive health in their relationships. This includes taking the initiative to discuss reproductive health concerns with their partner, using condoms, or assisting their partner in acquiring or using a contraceptive method.
Is opposed to violence against women under all circumstances, even those that are commonly used to justify violence (e.g., sexual infidelity).
Is opposed to homophobia and violence against homosexuals. (Although not directly related to male-female interactions, in the formative research, men often included “nonhomosexual” in their definition of what it was to be a “real” man, and homophobic comments were reportedly frequently used as a way to pressure or ridicule any man seen as being too “soft” on women (e.g., nonviolent). Thus, this domain was considered part of the locally defined notion of gender-equitable.)