Evidence brief
Does social connection differ across genders?
March 29, 2023
Kiffer Card, Cindy Yu, Jocelle Refol, Adam Frost, Pete Bombaci
Background
“Men are from Mars, Women are from Venus.” That’s the thesis of John Gray’s 1992best seller that describes how men and women differ with respect to their social and emotional needs and behavior. But how much truth is there to this popular idea? Certainly, the lived experiences of men and women differ in manifold meaningful ways.However, to understand better how the biology, culture, and individual experiences of gender shape our social health and wellbeing it is important to critically examine the scientific evidence related to gender differences in social connections. Such an understanding can improve our relationships with our same- and opposite-sex relationships and help us better understand our relationship with gender.
Purpose
The purpose of this evidence brief is to explore how genders differ with respect to their social behavior, including an assessment of differences across preferences and needs for certain types of connection, interpersonal communication styles, and other salient features of social relationships and personalities. In undertaking this review we acknowledge the separate constructs of sex (i.e., aspects related to biological attributes) and gender (i.e., aspects related to socially constructed identities, expressions and behaviours), while noting the scientific literature has not consistently distinguished between these factors and treat sex and gender as a binary category, when in reality people experience sex and gender in very different ways. In discussing differences according to sex and gender we also emphasize that most research has not sufficiently distinguished the extent to which biological vs. social factors give rise to the observed differences we discuss. In reality it is likely a complex interaction between these factors that gives rise to the behaviours associated with sex and gender (Leaper, 2011). Furthermore, it is important to recognize from the start that most psychological and social differences between men and women arise from differences in statistical averages, but in reality the distributions of these characteristics in men and women are largely overlapping and that there is little meaning and utility in describing any of these traits as “gender-typical” (Carothers & Reis, 2013; Staley & Cohen, 1988; Maes et al., 2019; Weisberg et al., 2011; Eagly & Wood, 1991). Despite these limitations, gender is a common frame of reference for understanding our social lives – motivating this review.
Evidence from Existing Studies
Perhaps the most obvious and important way that gender impacts our social life is the high degree of sex-segregation in our social networks (Bleske & Buss, 2016; Ghosh et al., 2019; Mehta et al., 2021; Machin & Dunbar, 2013) – a phenomenon that appears to deepen as we age (Kalmijn, 2002). While, sexual orientation appears to disrupt this, with greater cross-sexted friendships among and with sexual minorities (Gillespie et al., 2015; Galupo et al., 2014; Navvab et al., 2013; Muraco et al., 2012), it is clear that there is a strong tendency for individuals to spend more time with people who are similar to themselves (Mewa, 2020). While this may reflect an explicit preference for same-sex friendships and social interactions (Eisenbruch & Roney, 2020; Dunbar, 2016), Friebel et al. (2021; 2017) argues that the higher emotional closeness between women results in greater stability and selectivity in the friendship dyads of women – which alone is sufficient to generate the observed homophily in same-gender interactions. These patterns are supported in analyses of social media use, which indicate women use social media sites more for relationship maintenance while men use it for interacting with new people (Muscanell & Guadagno et al., 2012; Haferkamp et al., 2012) – reflecting differences in the social orientations of men and women.
Gendered patterns of social behavior begin early in life. Even toddlers begin exhibiting gender patterns in their play styles and social behavior. Boys engage in more rough and tumble play and girls in more collaborative, nurturing, pro-social play (Kung et al., 2021; Dunn & Munn et al., 2016; Hanish et al., 2013). These playstyles evolve throughout life, giving rise to differences across the lifecourse (Eagly , 1993). For example, Baumeister & Sommer (1997) argue that women areoriented toward dyadic close relationships while men are oriented towards a larger collective groups – reflecting important differences in the basic forms of relational intimacy exhibited across the genders (Seeley et al., 2003; David-Barrett et al., 2015).
Supporting this view, studies have shown that men and women take different approaches to organizing their social networks (Szell & Thurner, 2013; Rose & Rudolph, 2006) and building intimacy (Schoenfeld et al., 2012). Men are seen to be more competitive, status-seeking, aggressive and risk-taking, while women more cooperative, prosocial, empathetic, and relational (Szell & Thurner, 2013; Thoni et al., 2021; Espinosa & Kovarik, 2015; Carpenter et al., 2018; Eagly & Steffen, 1986; Suh et al., 2004). Men engage in more shared activities (e.g., recreation, externally-focused discussions) to build intimacy and rely on each other for instrumental support. Meanwhile, women provide more direct social and emotional support through inwardly focused conversation (Radmachet & Azmitia, 2016; Aukett et al., 1988; David-Barrett, 2022). Similarly, men and women also exhibit minor differences in social communication (Salavera et al., 2018; Bamman et al., 2012; Goldsmith et al., 2016) – with, for example, men being less emotionally expressive and responsive (Vigil et al., 2009; Neff & Karney, 2005; Liebler & Sandefur, 2002; Hall, 1990), but these differences are often small.
Adding important context for understanding differences in relational intimacy and social support across genders, Blyth & Foster Clark (1987) show that while young boys and girls report an array of differences in perceived intimacy with different types of relationships, the direction of effects vary with boys and girls being more intimate with different parts of their networks. Authors have hypothesized that these patterns of intimacy arise from social pressures and social roles that dictate how people form attachments and what kind of attachments they should form with different parts of their networks (Eagly, 1987). Thus these norms give rise to the overall structure of individuals social networks, but in different ways for men and women (Moore, 1990). For men the pressure appears to be to form less emotionally intimate and close relationships (Pauriyal et al., 2017; Bank & Hansfor, 2000) – possibly due to the social pressures driven by homophobia that have discouraged closeness between men over the past century (Eagly, 1987).
Differences in social styles appear to impact men and women differently (Zhao et al., 2022) – likely a reflection of the gendered values and expectations that individuals hold for themselves and others (Rudolph & Dobson, 2022; Hibbard & Buhrmester, 1998). Nevertheless, the more emotive social style displayed by women does appear to result in “closer” relationships – at least as “closeness” is typically defined (Mosley et al., 1987). Notably these differences are largely based on studies of same-sex friendships and there is some evidence suggesting that men and women adapt to the social styles of their cross-sex friends (Hacker, 1981; Leaper et al., 1995).
The different social styles are also reflected in the social networks of men and women. For example, Sterling (2018) showed that women have more friends and men have more coworker ties and larger social networks overall. While these differences have decreased since women began entering the workforce, it has not been eliminated. Women continue to play an outsized role in organizing the social lives of families and men engage more actively with weak ties (Fang et al, 2021). Evidence suggest that this may have considerable consequences even for men, who upon divorce or death of their spouse have less capacity to find the support they need (Mclaughlin et al., 2010). Meanwhile, the lesser engagement with weak ties among women can have important career consequences.
In reflecting on the impact of social style on wellbeing, reviews have found that healthy supportive social relationships are critical for both men and women and that low-quality strained relationships contribute to distress among both genders (Umberson et al., 1996). Similarly, the particular ways that social networks are organized have been identified as an important determinant of wellbeing. For example, some evidence suggesting that cross-sex relationships (which may violate certain social norms) might be particularly beneficial and rewarding for individuals (Xiao et al., 2022; Baumgarte & Nelson, 2009). Supporting this, Baiocco et al. (2014) found that gay men with female best friends had elevated wellbeing. Similarly, students from single-sex schools have been shown to be less socially skilled and more socially anxious in mixed-sex settings – highlighting the importance of cross-sex socialization for healthy development (Wong et al., 2018). Studies have also reported that men have higher levels of perceived social support (Soman et al., 2016), but not necessarily objective social support (Stokes & Wilson, 1984). Gender differences in support may be due to differences in the expected level and kinds of support that contributes to a higher satisfaction among men (who have lower expectations for traditionally defined emotional supports). For example, Hall (2010) suggests that women have greater friendship expectations for loyalty, intimacy, companionship, but men have more expectations for agency and physical activity. Clark & Ayers (1993) report similarly high expectations for reciprocity among women. In turn, these expectations are reflected in the reciprocal behaviours of men and women (Parker & de Vries, 2016).
Perhaps driven by unmet expectations for support, some studies have shown that women experience greater loneliness than men, but that this relationship may vary with age such that younger women and older men experience higher levels of loneliness (Wang et al., 2023; Jung et al., 2022). Differences in men and women’s evaluation of loneliness may arise from the social styles discussed above, with women valuing dyadic connections and men valuing group-oriented connections (Stokes & Levin, 1986). Of course, other studies show that men are more lonely – driven primarily by higher individualism (Barreto et al., 2021). In other words, greater social disconnection among men likely arises from cultural pressures related to independence that disproportionately shape the boundaries of men’s sociotrophy (Yang & Girgus, 2019). Evidence has also suggested that women avoid isolation and exclusion more strongly than men and react more negatively when they are excluded (Benenson et al., 2013) – perhaps buffering them against risk for loneliness.
Analyses from the Canadian Social Connection Survey
Using data from the pooled 2021 and 2022 Canadian Social Connection Surveys, we examined gender differences, across ages, between men (n = 2496), women (n= 3789) and non-binary people (n = 165) with respect to several indicators of social connectedness. Findings showed that emotional loneliness decreased with older age (p < 0.0001) and was statistically higher for both women (p = 0.0048) and non-binary people (p = 0.0007). An interaction effect between age and gender revealed that declines in emotional loneliness with age were greater for men compared to women (p = 0.001) and non-binary people (p = 0.006).Social loneliness also decreased with age (p = 0.037), but there were no statistical differences across genders.
Social support from family declined with age (p < 0.0001), but increased in older age. Non-binary people had lower support from family (p = 0.03). Similarly, social support from friends declined with age (p = 0.005), before increasing among older adults. No statistically significant differences in social support from friends were observed across gender. Finally, social support from significant others did not differ by age statistically (p = 0.373), but does appear to be higher among the elderly. No significant gender differences were observed.
Number of close friends declined with age (p < 0.0001), as did time with friends (p < 0.0001), time with family (p = 0.001), time with coworkers and classmates (p = 0.0002), and time with acquaintances (p < 0.0001). Gender differences were observed for time with family, wherein women spent more time with family (p = 0.02) and less time with acquaintances (p < 0.0001). Non-binary people also spent less time with acquaintances (p = 0.01).
As a second aim, we explored whether these factors had a differential effect on the self-rated mental health of men and women. Emotional loneliness had a more negative affect on women (p = 0.14) and non-binary people (p = 0.012), compared to men. Social loneliness had a more negative effect for women (p = 0.01) than for men. Effects of social support on self-rated mental health did not differ across gender groups. Neither did the effect of time spent with family. However, the effect of spending time with friends (p = 0.007), coworkers and classmates (p = 0.25), and acquaintances (p = 0.006) was a stronger predictor of self-rated mental health for women compared to men. Taken together these findings suggest that social time with others and adverting loneliness may be marginally more important for the mental health of women – though it is possible that these differences arise from a stronger cultural expectation for relational closeness among women as reviewed above.
Discussion
Research studies, including analyses from the Canadian Social Connection Survey, highlight a variety of gender differences in the ways men and women relate to others. Of course, the differences that are attributable to gender or sex alone are generally small – and there is more individual variation within genders than between genders. As such, individuals must assess for themselves the extent to which their relationships are meeting their social health needs.
Accounting for these differences are an array of factors. The bulk of evidence reviewed suggests that social roles and expectancies are the primary reason for gender differences in social behavior. However, some studies do show that biological processes are also relevant (Altman & Roth, 2018). For example, gendered differences in neurobiological rewards processing have been identified as a possible mechanism that shape the intrinsic rewards that people experience when socializing (Soutschek et al., 2017). Nevertheless, the findings reviewed should not be read as proscriptive. Rather, we acknowledge that our social world is constructed and that the choices we make about its construction have significant impact on us and our relationships. As such, we call for more research to help us understand the evolving and nuanced relationships between sex, gender, role expectations, sexual orientation, and social connectedness.
Conclusion
Based on the available evidence and our analyses of the Canadian Social Connection Survey, we recommend that all levels of society engage in efforts to support social connectedness and wellbeing among all genders – removing systemic barriers to inclusion and supporting healthy relationships across sex and gender groups. We acknowledge that existing social norms and role expectations have a powerful influence on the quality and enjoyment of relationships and activities, but acknowledge the wealth of theoretical and empirical work that demonstrates a need to combat cultural drivers of social inequities. These include gender power imbalances and roles that disadvantage women in the workplace and ridding our society of homophobia and toxic masculinity which harm the relationships of all of us, but most especially those who live with stigma and discrimination due to their sexuality or gender. Efforts to ensure that members belonging to these communities are fully integrated into warm, caring communities are critically needed. As such, we endorse structural and systemic interventions that support community development and social cohesion within these communities.