Which of the following is true regarding gender differences and similarities?

1.4 The Gender Similarities Hypothesis

The gender similarities hypothesis, proposed by Hyde (2005), states that males and females are similar on most, but not all, psychological variables. Based on a meta-analysis of 46 meta-analyses of psychological gender differences, 30% of effect sizes were trivial in magnitude (d between 0 and 0.10) and an additional 48% were small (0.10–0.35). Gender similarities were found for self-esteem, math performance, self-disclosure, and reading comprehension. Exceptions were found for some aspects of motor performance such as throwing distance, some measures of sexuality like the prevalence of masturbation, and physical aggression, all of which showed moderate or large gender differences.

Some gender meta-analyses have taken a developmental approach, computing effect sizes at different ages. When available, these findings are reviewed in later sections of this chapter. Hyde (2005) noted that the magnitude of gender differences varies considerably across age for some measures.

The gender similarities hypothesis has profound implications for research on gender development. Much gender development research aims at explaining why, in development, large gender differences are present by adolescence or adulthood. The gender similarities hypothesis holds that many gender similarities are found in adulthood, even in domains such as math performance. In short, in many areas there may be no gender difference to explain. This approach encourages researchers to balance emphasis on gender differences with emphasis on gender similarities.

In the context of the gender similarities hypothesis, the gender segregation of most adult occupations is striking. This observation encourages researchers to consider influences outside the individual, such as cultural sexism and stereotypes, as part of the explanation for the gender segregation of adult occupations.

I Introduction

Hyde's (2005) gender similarities hypothesis states that females and males are more similar than different on most psychological variables. She contrasts her hypothesis with the common view in the media and some areas of psychology emphasizing gender differences. In support of her argument, Hyde (2005) pointed to 46 meta-analyses that mostly indicated either nonsignificant or negligible average gender differences in various behaviors. In an analogous manner, the premise that I shall explore is that many contemporary theories of social development are more similar than different—yet most researchers tend to emphasize the differences. My corresponding plea is for more theoretical integration in the field.

Two related tendencies in the field of psychology impede theoretical synthesis (see Sternberg & Grigorenko, 2001). First, researchers sometimes present a new theory with constructs and explanations that are similar to those found in another theory. Thus, parallel theories may offer very similar ideas using different (but synonymous) terms. Second, theories that provide complementary explanations for phenomena are rarely merged to create a more complete picture of psychological processes and development. These two trends contribute to the fragmentation of the field and may slow scientific advancement. My goal is to encourage the dissolution of boundaries and the building of bridges regarding some of our theories. I shall use theory and research on gender development to illustrate some of the dilemmas and possible directions for the future; however, my arguments are meant to apply more generally to theory and research on social development.

The chapter is divided into four sections. First, I will introduce the issue of theoretical fragmentation in psychology. Second, I will summarize some of the reasons for this predicament. In the third section, I will consider the feasibility and advisability of theoretical synthesis. Finally, I will suggest some possible directions for theoretical integration among a few contemporary theories of social and gender development.

Essentialist “sex” model

Essentialists regard cognitive differences between women and men as natural, fixed, deep-seated, discrete, and related to biological sex (Haslam et al., 2000; Skewes et al., 2018).

Seen through an evolutionary lens, two possibilities arise if a cognitive trait is heritable; the trait can either be transmitted equally to both sexes or unequally. This gives rise to two hypotheses; the gender similarity hypothesis (Hyde, 2005, 2014) or a sexual difference hypothesis (e.g. Buss and Schmitt, 1993; Miller, 2000; Kimura, 2004; Baron-Cohen et al., 2005; Cahill, 2006).

The basis for an evolutionary pressure leading to sexually divergent traits could be a long-term division of labor between the sexes, hypothesized to be present in prehistoric groups of hunter/gatherer foragers (Marlowe, 2007). Another possibility is that divergent traits arise through within-sex competition or as sexual fitness display, usually in males (Miller, 2000; Locke and Bogin, 2006; Lange et al., 2014). These two theories predict opposite outcomes for sex differences in language skills. The hunter/gatherer theory posits that women have better developed social (and hence verbal) skills due to spending more time in groups and with offspring, while the sexual display hypothesis posits that language is a runaway trait selected for its aesthetic features like the tail-feather of the peacock. The consequence of the latter process, where language is used to signal a healthy cognitive system during sexual selection, would be that males compete to become better speakers while females would benefit from being better listeners (Rosenberg and Tunney, 2008).

Regardless of theory, however, an essentialist sexual difference hypothesis based on natural selection would, in the long run, lead to clearly separable distributions of the particular cognitive traits.

The gender similarity hypothesis, on the other hand, as formulated by Hyde (2005, 2014), states that males and females are similar on most, but not all, cognitive variables (Hyde, 2005). Hyde presented evidence for the hypothesis using meta-analyses on 124 cognitive and psychologic gender difference measures and found that 30% of the effect sizes were effectively 0 (Cohen's d: 0–0.10), and an additional 48% were in the range of a small difference, between 0.11 and 0.35. Thus, Hyde found that almost 80% of the gender difference measures evaluated were either small or zero, arguing against any categorical gender divide in psychological traits.

Conclusion

Women with cancer have been found to be more likely to appraise events as stressful, experience a perceived loss of control, and report more intrusive thoughts, body image concerns, anxiety, avoidance, and distress, whereas men report more emotional stability and potentially higher rates of depression. Women with cancer, particularly those with breast cancer, also report higher levels of PTG than men with cancer. As research on meaning making among patients with cancer is typically based on self-report data, it is difficult to say whether the gender differences identified in the chapter are based on differences in self-disclosure by men and women or differences in lived experiences with cancer.

According to the gender similarity hypothesis, males and females are actually far more similar than different regarding most psychological topics (Hyde, 2005). Empirical investigations of gender differences and similarities proliferated following Hyde’s hypothesis, and although Zell, Krizan, and Teeter’s (2015) meta-synthesis using data from over 20,000 individual studies found a relatively small overall difference between males and females across multiple domains in psychological science, the researchers “caution against the conclusion that gender differences are trivial or nonexistent” (p. 18). It is possible that the way in which males and females create meaning from their experience with cancer may vary more based on individual factors such as age, education level, and social support than based on gender. These individual differences, including cultural background, must been taken into consideration in the mental health care of patients with cancer.

Lastly, there is a dearth of knowledge about gender differences in meaning making and cognitive appraisal at the diagnosis and treatment stages of cancer. Further research is needed to assess gender differences in patients’ understanding of cancer; for example, whether they interpret cancer as a threat versus challenge, resulting in potential harm versus growth and how this impacts coping. This information would assist medical professionals and caregivers to better serve their patients with cancer.

3.1.2 Gender similarities and differences

The social structural perspective outlined here provides a framework to understand both gender similarities and differences in role navigation. The intersection of diffuse and specific roles is especially relevant, because the processes and position of each type of role might influence behavior at a particular time. Consistent with the gender similarities hypothesis (Hyde, 2005), we anticipate that the behaviors of men and women will frequently be similar. Similarities in behavior across gender are expected when the diffuse role of gender is less salient, or when specific roles override expectations of gender roles. For example, both male and female elementary school teachers enact communal behaviors such as caring for children; other aspects of behavior certainly might differ by gender, but the core requirements of the specific role of teacher are to attend to students' development.

Gender similarities in attitudes and behavior might also emerge because men and women share fundamental motives. Motives related to connection to others, to achievement and mastery, and purpose appear throughout theories that seek to understand human motivation (Dweck, 2017; Fiske, 2004; Maslow, 1943; Ryan & Deci, 2000). The data collected in our laboratory noted above clearly show fairly high endorsements of both communal and agentic goals for both women and men. Thus, both men and women will prioritize roles that afford opportunities to pursue foundational motivational dimensions, relative to roles that do not afford such opportunities.

Yet, even in a framework that recognizes similarities, gender differences occur in thought, motives, and action. When the diffuse gender role is especially salient or when specific roles have sufficient latitude to be enacted in a range of ways, men and women might diverge in their responses. Precise predictions of divergence follow from elements of gender roles. In contemporary data, women and men report similar levels of agentic traits and expectations; however, a gender gap persists for communal traits and expectations, as noted earlier. Consequently, when differences emerge, they tend to take the form of women particularly valuing communal opportunities.

An important consequence of identifying communal orientation as a key psychological process underlying gender differences in STEM attitudes is that it allows for understanding both gender similarities and gender differences. One important implication of this recognition is that it emphasizes that perceptions of STEM culture as impeding communal pursuits can deter communally-oriented men as well as women (Boucher, Fuesting, Diekman, & Murphy, 2017). Hence, gender is not expected to always moderate the effect of communal affordances on interest in the role. Given the logic of the model, this moderation would depend on (a) the strength of the affordance cue and (b) whether gender differences in communal orientation emerge. To return to the model (Fig. 2), the important question is whether the opportunities presented in a role are motivationally relevant to the individual. Continued research is needed to understand how features of the individual and the context interact to produce more or less gender-differentiated behaviors, attitudes, and cognitions (Deaux & Major, 1987).

Limitations of Women’s Health Studies

The analysis of the distribution and causes of disease and mortality by sex has led to impressive improvements in allocation of resources and in attention to previously neglected health issues affecting men and women. However, inadvertently they have reinforced binary constructions of sex (male vs. female) and gender (masculine vs. feminine), as well as treating sex and gender as easily separable. Indeed, the existence of male/female differences in health outcomes is often considered self-evident, and confirming these differences is often privileged over exploring similarities (Springer, Stellman, et al., 2012). The fact that researchers, systematically or without any theoretical guide, analyze data from men and women separately, and also the insistence by some editors and reviewers that researchers routinely analyze their data for gender differences implies belief in at least the possibility that men’s and women’s behaviors may be categorically distinct (Carothers & Reis, 2013).

The emphasis on sex differences has led to categorical thinking and essentialism risks. The conceptual weakness of categorical thinking has been apparent in gender analysis since the 1980s. Categorical thinking persistently underplays diversity within the gender categories, when “within group” would be more relevant than “between group” differences. Although biological sex is clearly a categorical variable, the variables commonly of interest to researchers tend to be dimensional varying along a continuum. When a significant sex difference is obtained, implicitly it is assumed that all people of a group (men) are different from all people of another group (women). This assumption treats an observed mean difference between men and women as a special kind of category called a taxon (Carothers & Reis, 2013). Although most scholars are skeptical about the general idea that in terms of social behavior men and women represent natural kinds, categorical thinking persists with the higher risk of essentialism bias, that is, the belief that both categories possess distinctive and unchanging properties grounded in biological explanations. In spite of these gender biases, research indicates that similarity between sexes is common (Hyde, 2005) and that even those variables for which males and females are not alike may be evidence of variations along a continuous dimension rather than categorical difference (Carothers & Reis, 2013).

One of the most relevant meta-analyses about psychological sex differences shows that the main pattern found in psychological research is sex similarities (Hyde, 2005). For most differences, the magnitude of sex-related differences is usually quite small and they seem not necessarily relevant for daily life. The gender-similarities hypothesis holds that men and women are more alike than they are different (Hyde, 2005, p. 581). In terms of effect sizes, the gender-similarities hypothesis states that most psychological gender differences are in the range of close to zero (d < 0.10) or small (0.11 < d <0.35), a few are in the moderate range (0.36 < d < 0.65), and very few are large (d < 0.66–1.00) or very large (d > 1.00). Hyde’s meta-analyses on psychological gender differences included categories of variables such as cognitive abilities, verbal or nonverbal communication, social presence or personality, psychological wellbeing, and motor behaviors. The striking result was that 30% (37 of 124 variables) of the effect sizes were in the close-to-zero range, 48% (59 variables) were in the small range and 23% (28 variables) were in the large range. The largest gender differences were in variables such as motor performance (throwing velocity and distance), sexuality (uncommitted relationship), and physical aggression. Apart from age trends in the magnitude of gender differences, gender researchers have emphasized the importance of context, which could mediate or moderate psychological sex/gender differences. Hyde (2005) highlights several examples about the influence of context in sex/gender differences in a deindividuation situation, stereotype threat, helping behavior, conversational interruption, and smiling (see Hyde, 2005 for more details). Here, the first three examples were emphasized. First, in a study about conformity to gender roles where deindividuation (state in which the person became anonymous) was emphasized, people should feel no obligation to conform to gender roles; sex differences on aggressive behavior disappeared (number of bombs in an interactive video game). The second example was about how a gendered stereotype threat can exert a decisive role in academic performance. Sex differences in math performance were moderated by previous information about gender differences (when participants expected gender differences, women underperformed compared with men). And the third example was about how sex difference in helping behavior is moderated by context. In general, men help more than women (d = 0.34). However, heroic helping involves danger to the self and is facilitated when onlookers are present, whereas women’s nurturant helping more often occurs in private with no onlookers. When studies are classified into onlookers vs. nononlookers, size effect was d = −0.02.

Carothers and Reis (2013) analyzed the underlying structure of gender differences, evaluating the likelihood that existing gender differences (large size effect) are better represented as dimensional or taxometric models. They reanalyzed data from 122 indicators from 13 studies assembled in sex-stereotyped activities and physical measures (sex-stereotyped activities, physical strength, anthropometric measures), sexuality and mating (sexual attitudes and behaviors, mate selectivity, sociosexual orientation), interpersonal orientation (empathy, relational interdependence), gender-related dispositions (masculinity, femininity, unmitigated communion, care orientation, fear to success, science inclination, and Big-5 personality traits), and intimacy (intimacy prototype, intimacy stage, intimacy with best friend, social provisions in intimate relationships). They used taxometric methods to determine whether gender differences are taxonic (ie, representing the existence of distinct categories) or dimensional (ie, reflecting differences of degree). The results indicated that sex-stereotyped activities were better represented by distinct categories (taxon), whereas sexuality and mating, interpersonal orientation, gender-related dispositions, and intimacy areas showed good fit to the dimensional model. In Carothers and Reis’s words, “data provide clear empirical evidence to support the belief of researchers who see psychological gender differences in dimensional terms” (p. 17).

The most relevant limitations of this approach can be summarized in (1) the consideration of sex differences as a static or fixed category (more than dimensional) without the consideration of interaction with other social and biological variables (for instance, age trends), (2) the lack of analysis of context where sex differences are studied (exemplified by Hyde, 2005), and (3) the absence of integration between different levels of analysis (Hammarström et al., 2014).

Connell (2012) and Springer, Hankivsky, et al. (2012) noted these limitations (categorical thinking and essentialism) and emphasized the multidimensional aspects of gender differences (gender-based analysis). Gender differences are commonly confused with sex differences (categorical thinking). For example, sex assignment based on biological characteristics is commonly confused by gender identity, which based on self-categorization as male, female, or other labels. Gender is considered as a static variable. Categorical thinking underlies a very widespread problem in the health literature, where gender is named, but actually women are spoken about. In such studies men and boys are not seen in active relations with women and girls, nor as actors in gender change (Connell, 2012). Emphasis in differences between sex and gender has drifted to a lack of understanding gender differences within the gender categories, such as between hegemonic and subordinated masculinities, or between lesbian and heterosexual women’s sexuality.

The presence of a biological health-related difference indicates exclusively a biological cause (essentialism). Health outcomes, particularly physical health outcomes, by definition have a physical manifestation or presentation. These physical presentations can often be linked with a biological process or predictor. In the case of gender and health, this is often articulated in a way that attributes male/female differences in outcomes to male/female physiology. One key problem with this logic is the assumption (or even assertion, in some cases) that this physiological difference is innate, or at least not socially determined.

Sex differences do not account for the dynamics of gender, that is, the ways gender orders are created and gender inequalities are created and challenged. Therefore mainstream policy documents and quantitative researchers typically invoke external forces to explain changes in gender phenomena: technology, economic development, globalization, or some vague notion of cultural change (Connell, 2012). Considering the dynamics of gender categories, gender can shape health in multiplicative ways.

Introduction

Recent reviews suggest that human psychological sex differences are typically small [1,2]. For example, Hyde's [2] ‘gender similarities hypothesis’ states that ‘men and women, as well as boys and girls, are more alike than they are different’. Of course, males and females belong to the same species, so we should naturally expect similarity on many dimensions. But the hypothesis that males and females are more alike than different lacks predictive power about where sex differences are likely to lie, and about the direction and magnitude of these differences. Sexual selection theory offers this resolution.

Sexual selection is the type of natural selection that favors traits that aid in obtaining mates. It tends to be strongest where potential mates differ greatly in quality, and especially where members of one sex can monopolize multiple mates, leaving many unmated same-sex competitors [3]. When reproductive success hinges on winning mates, sexual selection may favor even those traits that harm survival if they compensate by boosting mating success. Here, I review evidence, focusing on recent findings, regarding the strength and forms of sexual selection operating over human evolution and consider how sexual selection has shaped human psychology, including psychological sex differences.

2.4.3 Why most sex differences are small

Following her review of 46 meta-analyses of sex differences in cognition and behavior, Hyde (2005) concluded that the vast majority of sex differences are quite minor. More specifically, Hyde found that 78% of the differences exhibited effect sizes of .35 or less. This lead her to use the term gender similarities hypothesis to describe most sex differences, which was then contrasted with the popularized “men are from Mars and women are from Venus” hypothesis.

While Hyde’s concept can be applied to most sex differences, it should not obscure the fact that several effect sizes have been shown to far exceed 0.35. These include 0.66 for mental rotation (Voyer, Voyer, & Bryden, 1995), 0.96 for throwing accuracy (Ashmore, 1990), 0.81 for toleration of casual sex (Oliver & Hyde, 1993), 0.87 for desired number of sex partners (Buss & Schmitt, 1993), 0.98 for mechanical reasoning (Findgold, 1992, p. 76), and 0.93 for interest in objects versus people (Su, Rounds, & Armstrong, 2009). Also, in proportional terms, males throughout the world are three- to four-times more likely to commit violent crimes, especially those that are most serious such as murder, robbery, and rape (Brookman, 2003; Ellis, Beaver, & Wright, 2009). Nonetheless, Hyde convincingly argues that most sex differences in cognition and behavior are in fact of a minor nature.

ENA theory is compatible with Hyde’s gender similarities hypothesis. The reasoning behind this compatibility is as follows: Even though males and females have been subjected to quite different natural selection pressure for eons, the fact that they share nearly 99% of the same genes has prevented the sexes from separating very far from one another for most traits. To illustrate, adult males are universally taller than adult females (Ellis et al., 2008, p. 14) even though females throughout the world have a distinct preference for mates who are taller than themselves (Ellis et al., 2008, p. 441). Theoretically, average sex differences in height continue to be small because nearly all the genes controlling human height are located on chromosomes other than the Y chromosome (Weedon et al., 2008). This makes it difficult for height-influencing genes to give males more than a slight edge. Put another way, because males and females share approximately 99% of the same genes, major restraints are imposed on the extent to which males can be genetically programmed to surpass females in height. It should be added, however, that genetic programming for substantial sex differences in height is not impossible, given that in some species, one sex (nearly always males) are more than twice the size of the other sex in adulthood (Leutenegger, 1982).

Paralleling the argument just made for height, nearly all of the same genes are involved in brain development for males and for females. Therefore, most of whatever sex differences in neurology exist are likely to be subtle no matter how strong the natural selection pressure may be. This leads one to expect that most (but not necessarily all) of the cognitive and behavioral differences between the sexes will also be modest. Stated another way, the human genome only has a narrow range within which to respond to natural selection pressure to program males and females differently. The natural (including sexual) selection pressure may be great, but the ways that the brains of males and females can be genetically configured to respond to this pressure are primarily restricted to one relatively small chromosome, the Y chromosome.