Course:PSYC305/2013ST2/ClassProject/2.2.1 Introduction - GD Studies

Studies using Gender Diagnosticity

Lippa and Connelly’s (1990) gender diagnosticity approach offers a unique method for assessing gender-related individual differences and has stimulated fellow researchers to adopt this measure in addressing their own research questions. The following section will briefly summarize various studies that have utilized a gender diagnosticity approach.

Gender-Atypical Behaviour and Adolescent Bullying

Young and Sweeting (2004) conducted a study to determine whether gender-atypical behaviour was related to gender roles (masculinity and femininity), bullying (victimization or bully perpetration), peer relationships and psychological well-being. A cohort of 15 year-olds from a longitudinal investigation formulated the sample for this study. Participants completed questionnaires assessing the five variables listed above. Gender diagnosticity was divided into three groups: “gender atypical,” “extremely gender-typed," and “gender typical.”

Results revealed that “gender atypical” boys reported more victimization, more loneliness, fewer male friends, and fewer friends overall (Young & Sweeting, 2004). Conversely, loneliness and peer friendships did not vary amongst the different gender diagnosticity groups in girls. “Gender atypical” girls did report more victimization; however, this relationship was not statistically significant. In terms of psychological well-being, both “gender atypical” boys and girls reported more psychological distress, but only the rates for boys reached significance. Interestingly, “gender atypical” behaviour was unrelated to bully perpetration. That is, “gender atypical” boys and girls were not less likely than their counterparts to report bully perpetration. Rather, being a bully was related to gender role: masculinity was positively and femininity was negatively correlated with bully perpetration.

The findings provide support for the contention that masculinity, femininity, and gender diagnosticity are distinct constructs, as demonstrated by their different relationships with bully perpetration and victimization (Young & Sweeting, 2004). Moreover, the authors stress how professionals working with children need to be vigilant of adolescents who display “gender atypical” behaviours, as they are at an increased risk of experiencing victimization, loneliness and psychological distress.

Masculinity and Aggression

Researchers Weisbuch, Beal and O'Neal (1999) looked at the correlation between participants Bem Sex Role Inventory (BSRI) scores in three different situations and their completion of a computer version of the Taylor reaction-time aggression paradigm. The participants consisted of 86 male undergraduate students participating for course credit in a beginners psychology course. They each completed the BSRI as they are (their actual score), as they thought others wanted them to be, and as they thought they should ideally be. Each participant then competed against a fake competitor in a rigged computer game with varying levels of sounds that is meant to measure levels of overt and covert aggression.

The results present two interesting results: men who think others want them to be highly masculine showed greater covert aggression, and men who scored high in masculinity showed greater overt aggression (Weisbuch et al., 1999). The latter outcome corresponds with the masculine stereotype that males are aggressive, as well as the idea that aggression is an integral part of the male gender role. The first outcome is interesting as it suggests these males feel insecure about how masculine they should be and thus adopt covert aggression which is more common amongst females (Weisbuch et al., 1999).

These findings find support for the prominence of gender roles and stereotypes in our society. It is often assumed that males are more aggressive than females and are often described as having more aggressive personalities as well. Males scoring high in masculinity as well as overt aggression also provides support for studies on bullying, antisocial personality disorder, conduct disorder, and other areas looking into the links between masculinity and aggressive behaviours.

Masculinity and Mortality

Researchers have explored the relationship between gender-related traits and health outcomes: for example, one study demonstrated that negative masculinity (arrogance, hostility, egotism) was associated with coronary heart disease (Helgeson, 1995). While Helgeson’s work has promoted research in the domain of personality and illness, Lippa, Martin, and Friedman (2000) note that this study did not address masculinity and mortality – nor did it include other measures of gender-related traits, such as gender diagnosticity measures.

This deficit prompted the authors to adopt a gender diagnosticity approach and investigate links between masculinity and mortality (Lippa et al., 2000). The sample was comprised of a group of school-aged children from a longitudinal study that launched in 1920 (Terman, 1947). The participants completed questionnaires in mid-adulthood that assessed masculinity and femininity, vocational interests, and risk factors (e.g., smoking, alcohol consumption, obesity, physical activity, and riskiness of hobbies) (Terman, 1947). The authors used the participants’ vocational preference ratings to obtain gender diagnostic probabilities (Lippa et al., 2000). Masculine men and masculine women were defined as expressing greater preference for male-typical occupations. Results revealed that gender diagnosticity scores were significantly related to mortality scores. That is, masculine men and masculine women were more likely to die at any given age than individuals who were less male-typical. These results remained significant when researchers controlled for health risk factors. In addition, scores from the masculinity and femininity scale correlated similarly with mortality rates. Participants classified as feminine women showed the greatest longevity and were thus the most advantaged group.

Their findings offer preliminary support for the case that masculinity may be harmful to one’s health (Lippa et al., 2000); however, they recognize that the observed relationship between masculinity and mortality is constrained by the use of archival data and the possibility of third variables. For instance, masculine individuals and feminine individuals may differ in the use of medical care: feminine individuals may engage in more preventive versus remedial health care. Moreover, masculine individuals may have weaker social support networks than feminine individuals.

Gender-Related Traits in Transsexuals and Nontranssexuals

Lippa (2001) strived to extend the work already done in regards to the difference between transsexual persons and nontranssexual persons by focusing on whether these individuals differ on gender-related interests. Through using GD measures, Lippa (2001) sought to determine whether individuals from either group differed in terms of instrumentality, expressiveness, and self-ascribed masculinity and femininity, believing that this research would help yield more information about about transsexualism and the nature of gender-related traits as a whole.

The participants, which consisted of volunteer transsexual persons from support groups and volunteer nontranssexual, heterosexual students from a post-secondary institution, were given self-report styled questionnaires which they were to fill out and anonymously mail back in provided stamped envelopes. The result analysis found that--for gender-related traits--the differences for M-to-F transsexuals and nontranssexual men, F-to-M transsexuals and nontranssexual women, and nontranssexual men and nontranssexual women were all statistically significant. In addition, Lippa found that the differences between M-to-F transsexuals and nontranssexual men were significant for all gender-related traits, in which there was an especially large effect size for GD based on occupational preference, hobby preference, and self-ascribed femininity. For F-to-M transsexuals and nontranssexual women, there were four gender-related traits that showed significant difference, these being occupational preference, hobby preference, self-ascribed masculinity, and self-ascribed femininity. Similarly, nontranssexual men and nontranssexual women also differed significantly on all six gender-related traits, and the effect sizes were particularly large for GD measures and self-ascribed masculinity and femininity (Lippa, 2001).

Lippa (2001) concluded that GD measures and self-ascribed masculinity and femininity scales could assess masculinity and femininity more validly than measures of instrumentality or expressiveness. Furthermore, he determined from his study that GD measures and self-ascribed masculinity and femininity scales could strongly distinguish transsexual persons from nontranssexual persons as well as nontranssexual men from nontranssexual women, and that these findings could be potentially relevant to counselling and screening transsexual individuals. He acknowledges that despite the strength of the current results, more research can be done on this particular topic.

The Structure of Sexual Orientation and Its Relation to Masculinity, Femininity, and Gender Diagnosticity: Different for Men and Women

Lippa and Arad (1997) conducted a study to determine the relationship between gender-related individual differences and sexual behaviour. The participants, comprised of 394 psychology students from California State University, completed three questionnaires anonymously. The first questionnaire evaluated the participant's occupational preferences, every day activities, and hobbies. The second questionnaire, which was a measure of their personality traits, contained items derived from the scales of the Big Five Personality Factors. The last questionnaire completed by the participants surveyed their sexual behaviours and attitudes that typically reveal gender differences. These domains include sexual orientation, emotional commitment, level of sex drive, and finally, interest in visual stimuli and fantasy. In order to compute gender diagnostic probabilities, multiple discriminant analyses were applied to non-overlapping subsets of the participant's preference ratings for hobbies, occupations, and every day activities. This was done to obtain the probability that an individual is predicted to be male or female on the basis of the patterns in their preference ratings.

The results of the study indicate that the structure of sexual orientation differs for men and women. For example, while men's sexual orientation tends to be bipolar, women's sexual orientation proved to be two-dimensional, suggesting that women's attraction to men is independent of their attraction to women. In addition, the data obtained from this study also suggests that among men, gender diagnosticity and masculinity, but not femininity, are correlated with sexual orientation. In fact, the data at hand suggests that the link between gender diagnosticity, as based on occupational preferences, and sexual orientation are very significant. For example, men who reported preferences for female-typical occupations were more likely to report same-sex attraction, whereas males who reported more interest in male-typical occupations were less likely to do so. However, for women, there is little to no relationship between GD, M, F, and the two dimensions related to women's sexual orientation. In other words, the personality or occupational preferences of a women are not predictive of their sexual behaviours and attractions.

The Impact of Sex and Gendered Organizational Culture Preferences on MBA’s Income Attainment

Differences between male and female incomes have been explained through two main theories:

1. Gender Role Theory, in which socially-constructed beliefs about women incongruent to men affect managerial decisions (eg. hiring, promotion, compensation), resulting in women earning less than men.
2. Tournament Theory, in which employees compete with one another for career advancement. This theory emphasises effort and ability, not sex.

In previous studies, masculinity, a preference for an aggressive organizational culture, has been be correlated with career success (Abele, 2003), higher salaries for men and women (Kirchmeyer and Bullin, 1997), and leader emergence (Kent and Moss, 1994).

In order to test the plausibility of this, O’Neill and O’Reilly (2009) conducted a longitudinal study of newly-enrolled MBA students at three different time periods. At Time 1, the participant's initial personality was assessed, and their degree of masculinity and femininity was also indexed based on their preferences for organizational cultures. Those who preferred a “masculine” organizational culture preferred work environments that were high on aggressiveness and low on supportiveness, while those who preferred a “feminine” organizational culture took a preference for work environments that were high on supportiveness and low on aggressiveness. At Time 2 (four years after graduation) and Time 3 (eight years after graduation), a follow up was conducted to determine the participant's occupational attainment and major life events, which were examined against their initial personality and preference assessments at Time 1.

The results revealed that masculinity has a short-run positive impact on income attainment (at Time 2), and gender roles are a driving force only for early occupational success. Thus, employees (men or women) who prefer masculine organizational cultures are likely to achieve occupational success during the initial stages of their career. However, in the long run, sex differences still come into play and men eventually have significantly higher incomes than women. This is because men gradually work more hours per week for pay, while women have been more likely than men to prefer fewer working hours (Eastman, 1998), focusing their efforts on a work-life balance instead.

Ultimately, the amount of effort an employee (man or woman) puts into their work is more predictive of a potential higher income than compliance with an idealised, socially constructed masculine role or a preference for a masculine organizational culture.

Gender-Related Traits of Heterosexual and Homosexual Men and Women

Lippa (2002) examined the ways in which sexual orientation interacted with gender-related traits. Previous studies found that gay men are more feminine than heterosexual men, while lesbian women are more masculine than heterosexual women (Dunne, Bailey, Kirk, & Martin, 2000; Haslam, 1997; Lippa, 2000; Pillard, 1991).

Biological and psychosocial theories have been used to explain why sex-typed behaviours are related to sexual orientation:

1. Biological theories, which suggest that genetics, biological-environmental factors, (eg. prenatal stress), or fraternal birth order may affect the masculinisation of a developing fetus.
2. Psychosocial theories, which consider the degree of parental reinforcement of sex-typed behaviours, and the adoption and labelling of sex-role identities. D.J. Bem’s “Exotic Becomes Erotic” Theory also falls under the psychosocial theory, which focuses on children’s friend groups and the perception of opposite sex similarity or dissimilarity, resulting in an attraction (erotic) to the dissimilar sex (exotic).

However, both theories only suggested weak links between sex-typed behaviours and sexual orientation, prompting further investigation.

Lippa conducted two studies: The first study analyzed personal attributes and occupational preferences. It was conducted online, allowing for a large geographically diverse sample. The second study was conducted using a survey that asked questions regarding demographics, expressiveness vs. instrumentality traits, and preference ratings for various occupations.

The findings were consistent with previous studies. Study 1 showed that gay men are different from heterosexual men, as well as lesbian women and heterosexuals women, when measuring gender diagnosticity (GD) on career interests which is instrumentality versus expressiveness (900). Study 2 found when measuring instrumentality versus expressiveness, there are large differences between gay and heterosexual men, while there are much smaller differences between lesbian and heterosexual women. Homosexual and heterosexual individuals differed most on gender diagnosticity and on self-ascribed masculinity-femininity. That is, gay men reported a more feminine profile and lesbian women a more masculine profile as compared to their heterosexual counterparts. Additionally, the gay men preferred female-typical occupations, while the lesbian women preferred more male-typical occupations.

The study also allowed for the development of gay- and lesbian-heterosexual diagnosticity measures, which were successful at distinguishing homosexual individuals from heterosexual individuals. (75% men and 81% women’s sexual orientations were correctly classified). The occupational preferences that distinguished homosexual from heterosexual individuals were the same as those that differentiated heterosexual men and women. Thus, one’s gender, as well as sexual orientation is strongly linked to occupational preferences.

Effect of Culture on Relationship Between Sexual Orientation and Personality Traits

Previous psychological research has shown that sexual orientation is sometimes linked to gender-related personality traits (Pillard, 1991). Social constructionist theories argue that sexual orientation "categories", such as homosexual-heterosexual and masculine-feminine, are cultural inventions; therefore, these theories suggest that culture plays a role in determining the relationship between sexual orientation and cultural definitions of masculinity and femininity.

The goal of this cross-cultural research (Lippa & Tan, 2001) was to investigate whether cultural factors moderate the relationship between sexual orientation and gender-related personality traits. They assessed large groups of gay men, lesbian women, heterosexual men, and heterosexual women coming from three broad cultural groups: Asians, Hispanics, and Whites. Participants were asked to complete an anonymous questionnaire booklet that started with questions about demographic information. All participants also completed the Personal Attributes Questionnaire (PAQ), a six-item scale developed by Storms that assessed self-ascribed masculinity and femininity, and a 32-item Sexual Behaviour and Attitudes Questionnaire. Some were asked to complete questionnaires targeted for their specific culture and sexual orientation groups. For example, Hispanics and Asians completed the Suinn-Lew scale of acculturation (Suinn, Rickard-Figueroa, Lew, & Vigil, 1987) and gay and lesbian participants answered a 10-item scale that assessed their degree of discomfort with being "out" about their homosexuality.

Their findings were consistent their working hypothesis that the relationship between sexual orientation and gender-related traits is moderated by culture (Lippa & Tan, 2001). Hispanics and Asians, from traditional, gender-polarized cultures, showed larger homosexual-heterosexual differences in assessed gender-related traits than did Whites, who came from non-traditional and gender-non-polarized culture. The moderating effects of culture were present for measures of gender-related occupation and hobby preferences as well as for measures of self-ascribed masculinity and femininity. Lippa and Tan noted the significant main effects for culture that emerged from their results were due to differences among gay and lesbian participants from the three cultural groups.

Sex Role Orientation and Self-Esteem

Current Research on sex-role and self-esteem have focused on one of three models (Whitley, 1983). There is the traditional congruence model which argues that self-esteem and psychological well-being are high only when ones sex role orientation matches their gender; the androgyny model which suggests self-esteem and well-being are highest when ones sex role orientation is high in both masculinity and femininity regardless of their gender; the masculinity model which states that self-esteem and well-being is positively correlated with masculinity regardless of ones gender (Whitley, 1983). The study “Sex Role Orientation and Self Esteem” by Bernard Whitley (1983) is a meta-analytic review aimed at answering the question of which model is best supported by the literature.

The measures used in each study to assess sex roles were either the BSRI (Bem Sex Role Inventory) or the PAQ (Personal Attributes Questionnaire). Both scales measure the same construct but they are usually only moderately correlated (Kelly, Furman, & Young, 1978). The self-esteem measures used in the various studies measured global self-esteem, a self-evaluation across a number of areas, and social self-esteem, a self-evaluation of adequacy in social interactions (Whitley, 1983).

The results from the meta-analysis find no support for the congruence model and the greatest support for the masculinity model. There was a positive correlation between masculinity, femininity, and the interaction between them with self-esteem, however masculinity had the greatest correlation (Whitley, 1983). These results are important as low-self esteem is linked to neuroticism (Bagley & Evan-Wong, 1975), anxiety (Percell, Berwick, & Beigel, 1974) and poor adjustment (Ellis & Greiger, 1977). Gender roles and gender schemas are implicated in low and high self-esteem.

Stereotype Threat and The Gender Gap in Political Knowledge

According to McGlone et al., it has been found that when surveys of political knowledge are conducted, men are usually more accurate in their responses compared to women. They further inspect such gender gap by exploring the reasons for why women may perform less accurately on questions concerning politics, compared to men. McGlone and his colleagues argued that the reason for poor performance in such a survey is that the survey is communicated in a way that is threatening to women, therefore causing women to fall under the stereotype threat. In order to support such argument, McGlone et al. conducted a telephone survey, which was used to determine the political knowledge of college students. They used gender diagnostic set of questions and the gender of the interviewer to shape and manipulate the survey.

The study consisted of 141 subjects (71 women, 70 men). Those participants who were randomly selected into the diagnostic set of questions were told that the survey has had results creating gender differences in past research, while those who were in the non-diagnostic set of questions were told the opposite.

Results showed that men still scored higher on political knowledge testings, but despite this, a change was seen when the gender diagnostic set of questions were eliminated and portrayed as non-diagnostic and female interviewers conducted the telephone survey. In this case women were more likely to score higher on surveys of political knowledge. This study is beneficial for future use in creating political knowledge surveys because it will create a more accurate response from all respondents.