Social Exclusion and Resilience: Examining Social Network Stratification among People in Same-Sex and Different-Sex Relationships | Social Forces



Abstract

Social networks of minoritized societal groups may be exposed to a unique structural force, namely that of social exclusion. Using a national sample of people in same-sex and different-sex relationships in the Netherlands (N = 1,329), this study examines sexual orientation as stratifying factor in social networks. Specifically, it is a comparison of their size and composition. Overall, the networks are similar but a few differences stand out. People in same-sex relationships have larger networks than people in different-sex relationships, which are made up of fewer ties with the family-of-origin and more friends. This lends support to the families-of-choice hypothesis and suggests that people employ resilience strategies, such as alternative community building, to counteract social exclusion from families-of-origin. The results further show that men in same-sex relationships have the fewest same-gender ties in their networks out of both men and women in any relationship type. Overall, the results show that sexual orientation is a dimension worthwhile studying as a stratifying factor of social networks both standing alone and at the intersection with gender.

Introduction

People maintain personal social networks for survival and to realize a variety of different well-being goals (Bernstein 2016). This can range from accessing material, emotional or financial resources in times of need, over getting validation and recognition, to simple companionship. While there is a great deal of personal agency involved when it comes to choosing the people who make up personal networks, we know that there are structural forces, too. The physical spaces where we interact structure meeting opportunities (Feld 1981). Members of stigmatized societal groups may be exposed to a unique structural force, namely that of social exclusion. This could result in personal networks, which are systematically different from those of people who belong to the mainstream. For one, this raises questions pertaining to social cohesion, legitimacy, and citizenship within societies. Moreover, the opportunities and constraints associated with differently structured networks have been shown to reproduce existing social, economic, and health gradients in society (Chen and Volker 2016; Pena-López and Sánchez-Santos 2017; Pichler and Wallace 2009). Prolonged loneliness and ostracism have detrimental consequences for mental and physical health and mortality (Bernstein, 2016). This makes the study of network structures of sexual minority people relevant, not lastly because of what we know about the disadvantage of sexual minority people in various life domains, such as well-being and health, family life, education, and the labor market (e.g., Badgett 2003; Mize 2016; Waite and Denier 2015).

This study examines whether we can observe structural differences in the size and composition of social networks between people in same-sex and mixed-sex relationships as a way to better understand the social position of sexual minority people in society. The social networks include people with whom they discuss important matters and people with whom they like to spend time. In times where societal acceptance is higher in many countries compared to a few decades ago, this will provide a more nuanced view on the state of equity than relying on self-reported acceptance, which is prone to social desirability. The Netherlands is a great case to examine this since opinions are fairly positive in international comparison; yet the share of people who report having “reservations” about sexual diversity increases when it concerns close relationships (Kuyper 2016). This suggests that the story is not that simple. If the network structure is better understood, this could serve as important factor to understand the reproduction mechanisms of social inequality of sexual minority people compared to the heterosexual majority population.

The social networks of sexual minority people have not gone unexamined. One prominent strand of research has focused on sexual networks under the premise of disease control, predominantly among men who have sex with men (Amirkhanian 2014; Holloway et al. 2015). Another direction of research has examined social support among sexual minority people, highlighting its importance in buffering stress from living in heteronormative societies (McConnell, Birkett, and Mustanski 2015; Meyer 2003; Moran, Chen, and Tryon 2018; Snapp et al. 2015). Studies of sexual minority people’ social networks, which go beyond specific the subsets of (perceived) support ties or sexual contacts, are scarcer. Studies also tend to leave out heterosexual comparison groups, which prevents us from drawing conclusions about a possible stratification of social networks along sexual orientation lines.

There are a few welcome exceptions of studies, which have made creative efforts to circumvent the limitations in available data to study sexual minority and heterosexual people comparatively. Two studies of confidant networks found that people with minoritized sexual identities have fewer family-of-origin members and more friends as confidants compared to heterosexual identifying people. The first study is situated in Belgium and compared a convenience sample of 2,931 sexual minority people to heterosexual people from a different but comparable dataset (Dewaele et al. 2011); the second study is a report, which combined a non-probability web survey with German national survey data to arrive at a sample of 4,511 sexual minority respondents (Kasprowski et al. 2021). Similar findings about the presence of family-of-origin and friendship ties were made in a comparison of the joint social networks of eighty-three same-sex and fifty different-sex couples in a convenience sample in the United States (Julien, Chartrand, and Bégin 1999). The strategy of using a comparison group of people in different-sex relationships from a separate dataset was also used to compare friendship networks of 405 sexual minority people to those of heterosexual people in the United States; sexual minority people appeared to have more friendships with people of a different gender than their own compared to heterosexual people. Such comparative evidence is paramount in understanding structural differences in social networks and it pertains to issues of social cohesion in societies.

This small body of comparative evidence—while spread over two decades, involving different countries, different definitions of social networks and sexual orientation, and largely based on convenience samples that do not allow generalization—points toward potential differences between the social networks of sexual minority and heterosexual people. A joint rigorous study of the social networks of sexual minority and heterosexual people in one single dataset, with a focus beyond support or sexual ties, is needed to further examine a possible stratification of social networks along sexual orientation lines. To this end, the current study offers an analysis of the personal social networks of women and men in same-sex and different-sex relationships in the Netherlands. I use data from the Unions in Context (UNICON) project, which collected probability survey data among same-sex and different-sex couples and families by means of a population register sampling frame (Fischer, Kalmijn & Steinmetz, 2017). Note that all respondents in these data are partnered and generalizations to sexual minority people who do not live with or do not have a partner should be avoided. The overarching research question reads: How similar or different are the social networks of people in same-sex relationships compared to the social networks of people in different-sex relationships in the Netherlands?

Social Exclusion and Resilience

From a Sociological perspective, social exclusion involves multidimensional disadvantage. According to a working definition of Levitas et al. (2007), social exclusion may involve “the lack or denial of resources, rights, goods and services, and the inability to participate in the normal relationships and activities, available to the majority of people in a society (p. 25).” For sexual minority people, social exclusion may stem from the fact that sexual diversity remains marginalized in Western societies. Even when people state they are accepting in social surveys (which is a matter prone to social desirability), their actions may differ from their expressed attitudes. Indeed, acceptance of sexual diversity is often conditional. While 91% of the Dutch population agrees that “gay and lesbian people should be free to live their lives as they wish” (survey item), acceptance of a gay or lesbian teacher for their child drops to 83%, and only 71% would accept a same-sex partner for their child (Kuyper 2016). These numbers illustrate that the endorsement of sexual diversity in a general abstract sense is easier than in concrete instances in one’s personal relationships.

One possible manifestation of social exclusion in social networks could simply be in size. People in same-sex relationships could have smaller social networks than people in different-sex relationships, who do not experience exclusion on grounds of their relationship or sexual orientation. Yet, there is an abundant literature, which details the manifold strategies of resistance and resilience among sexual minority people (Freitas et al. 2017; McConnell et al. 2018, Mustanski et al. 2011). This may involve building strong and large communities of like-minded, supportive people to buffer against the negative experience of living in heterosexist societies (Oswald 2002). Being part of a minoritized group, which does not conform to the pervasive standard of cis-heterosexism, also requires an active effort on the part of the minoritized people to seek out role models and information related to their own lived experience (Frost et al. 2017). Such efforts could lead to larger networks compared to people who come by role models and examples how to live successful and fulfilled lives by default in their daily lives and popular culture. Heterosexual people do not require information on how to navigate life as a minoritized member of society and might therefore be less likely to actively seek out social contact with others that extends beyond the need for companionship that people need more generally. And indeed, the few comparative studies that exist so far suggest somewhat larger networks among sexual minority people and heterosexual people in the number of close friends and confidants (Dewaele et al. 2011; Galupo 2009). Against this background, I expect larger networks among people in same-sex relationships.

Hypothesis 1: The social networks of people in same-sex relationships are larger than the social networks of people in different-sex relationships.

Families-of-origins are both a site of important social contacts for some and a source of potential conflict and exclusion. For sexual minority people, problems with the family-of-origin may occur when they first disclose their sexual orientation to their parents (Goodrich, Trahan and Brammer, 2019). In the Netherlands, the first coming-out tends to occur during youth for people who are in same-sex relationships in adulthood; the relationships with their parents recover largely, with some degree of ambivalence and reduced contact persisting (Fischer and Kalmijn, 2020). It is possible then that family-of-origin members are still less likely to be included in the circle of people with whom people in same-sex relationships discuss important matters or like to spend their time. One often-theorized strategy of resilience is that of substituting complicated or absent ties to families-of-origin with selected supportive ties (Rostosky and Riggle 2017). The family-of-choice hypothesis holds that ties with others outside the family-of-origin are used to compensate for this potential exclusion to form chosen families.

Hypothesis 2 (families of choice hypothesis): The social networks of people in same-sex relationships consist of fewer family-of-origin ties and more friendship ties than the social networks of people in different-sex relationships.

Choice Homophily at the Intersection of Sexual Orientation and Gender

If social exclusion is understood as limited or absent access to resources, resilience involves the creation of strategies to gain access to these lacking resources in a different way. The question with whom sexual minority people build their personal networks merits closer examination. Not all ties guarantee access to the same (amount) of resources, particularly if they are similar to one self in terms of social position. The phenomenon of people sorting into relationships with others who are similar to themselves, known as homophily, is a well-studied phenomenon in the social network literature and one of the driving forces behind social segregation. We observe it in all kinds of relationships such as partner choice (Kalmijn 1998), friendships, support networks, or ties with colleagues (McPherson, Smith-Lovin, and Cook 2001). Social segregation is particularly relevant to social cohesion. If members of minoritized groups are socially excluded from society this is at odds with notions of societal cohesion, which entails a concern for the “belonging, inclusion, participation, recognition, and legitimacy” (Jenson 1998, p. 29) of all members of society. The legitimacy of being a full citizen and member of a society can be called into question under the social exclusion of social groups (e.g., when sexual minority people exclusively connect to each other with few ties to the majority society, and vice versa).

Yet, one likely path of resilience in the face of mainstream exclusion is to connect to other people with similar experiences for validation and support. The presence of similar others is an important source of well-being for individuals with stigmatized identities, as they can provide identity affirmation, role model function, and the understanding of shared experiences (Ghavami et al. 2011). It has been shown that knowing other same-sex couples can serve as a group-based resource for same-sex couples against negative mental health outcomes (Frost and Meyer 2012). And indeed, there is evidence that same-sex couples are at least superficially connected to similar others, with same-sex couples in a study in Atlanta and San Francisco knowing a median of twelve other same-sex couples (LeBlanc et al. 2015). While knowing other same-sex couples does of course not say much about whether they are part of the personal social networks, the authors rightly conclude that this is suggestive of homophily based on sexual orientation.

Yet, there are also constraints to homophily among sexual minority people. For one, the pool of available sexual minority ties is relatively small compared to that of heterosexual others, particularly outside the urban hubs. In the Netherlands, about 5–8% of the population is estimated to be identified as lesbian, gay, or bisexual (Kuyper 2016). Access to sexual minority community spaces is presumably much more difficult outside urban areas, which may limit meeting opportunities. Further, the desire to connect to other sexual minority people may vary depending on the salience of the sexual identity to people’s sense of self. Despite explanations that may impose limitations to the extent of homophily on the basis of sexual orientation, it remains a conceivable reality that people in same-sex relationships know more sexual minority people on average than the presumably mostly heterosexual people in mixed-sex relationships.

Hypothesis 3: People in same-sex relationships have more ties with other sexual minority people than people in different-sex relationships do.

In the network literature that exists on presumed heterosexual people, gender homophily emerges as one dominant force of network segregation, particularly in friendship networks (Galupo 2009; Kalmijn 2002). Taking sexual orientation into account, however, may complicate what we know about gender homophily. It is precisely this presumed heterosexuality in network literature why sexual orientation is rarely studied as a standalone dimension of homophily and as a factor that intersects with known phenomena like gender homophily (exceptions see Galupo 2007; Ueno et al. 2012). There are a number of points where sexual orientation may alter known processes of gender homophily.

One structural limitation to gender homophily may be the strongly gendered attitudes toward sexual diversity (Worthen 2013). Individual attitudes and preferences matter for the formation of friendships (Kalmijn 2002), both on the supply and the demand side. On the supply side, heterosexual people may not be available for friendships with sexual minority people if they hold negative attitudes toward sexual diversity. On the demand side, sexual minority people may not want to interact with people with negative attitudes toward sexual diversity. Attitudes toward sexual diversity depend both on the gender of the target person and on the gender of the person holding the attitude (Worthen 2013). People’s attitudes tend to be more negative toward sexual minority men compared to sexual minority women. Attitudes in heterosexual men tend to be more negative toward sexual diversity in everyone and particularly negative toward sexual diversity in men. If these attitudes translate into (the absence of) social ties, then heterosexual men are less available to all sexual minority people as potential ties. This would mean lower gender segregation among sexual minority men and higher gender segregation among sexual minority women. The particularly negative attitudes of heterosexual men toward sexual minority men would lead to an even lower degree of gender segregation among sexual minority men.

In addition to this pathway of social exclusion, structural meeting opportunities may play a role, such as at the workplace and in voluntary organizations. These environments can be fairly gender segregated, meaning more opportunities to form in-group ties, which increases gender segregation in friendships among heterosexual people (McPherson, Smith-Lovin, and Cook 2001). Yet, sexual minority people select into gender atypical industries at a higher rate than heterosexuals do (Waite and Denier 2015), suggesting that they have more opportunities to form cross-gender friendships, again resulting in lower gender segregation. One factor unique to sexual minority people, which may increase gender segregation, is that LGBTQI* community1 spaces are often highly gendered. If sexual minority people build up resilience in the form of ties with other sexual minority people and by frequenting LGBTQI* community spaces, this would expose them to more same-gender meeting opportunities.

Sexual orientation, moreover, can complicate what we know about non-romantic same-gender ties; an intersection that is an understudied area in friendship literature (Galupo 2009). One factor that can affect the gender composition of (friendship) networks is that of possible attraction. Among heterosexual people, this would lead to more same-gender friendships to minimize the complication of possible attraction, whereas it would suggest more cross-gender friendships among sexual minority people (Weger 2015). It is conceivable, however, that this logic based on heterosexist notions of monogamy and sexual attraction may not apply in the same way to sexual minority people.

Overall, the literature more strongly suggests lower gender segregation among people in same-sex relationships, particularly so among men.

Hypothesis 4: People in same-sex relationships have fewer same-gender ties compared to people in different-sex relationships. This difference is larger among men than it is among women.

Table 1 summarizes all hypotheses regarding potential network stratification along sexual orientation lines.

Dependent variable Expected relationship with being in a same-sex vs. different-sex relationship
Hypothesis 1: Network size  (+) 
Hypothesis 2: Number of family-of-origin ties,  (−) 
number of friendship ties  (+) 
Hypothesis 3: Number of sexual minority ties  (+) 
Hypothesis 4: Number of same-gender ties  (−), stronger among men 
Dependent variable Expected relationship with being in a same-sex vs. different-sex relationship
Hypothesis 1: Network size  (+) 
Hypothesis 2: Number of family-of-origin ties,  (−) 
number of friendship ties  (+) 
Hypothesis 3: Number of sexual minority ties  (+) 
Hypothesis 4: Number of same-gender ties  (−), stronger among men 
Dependent variable Expected relationship with being in a same-sex vs. different-sex relationship
Hypothesis 1: Network size  (+) 
Hypothesis 2: Number of family-of-origin ties,  (−) 
number of friendship ties  (+) 
Hypothesis 3: Number of sexual minority ties  (+) 
Hypothesis 4: Number of same-gender ties  (−), stronger among men 
Dependent variable Expected relationship with being in a same-sex vs. different-sex relationship
Hypothesis 1: Network size  (+) 
Hypothesis 2: Number of family-of-origin ties,  (−) 
number of friendship ties  (+) 
Hypothesis 3: Number of sexual minority ties  (+) 
Hypothesis 4: Number of same-gender ties  (−), stronger among men 

Data and Method

Data

This study is based on survey data from the Unions in Context (UNICON; Fischer, Kalmijn and Steinmetz, 2017) project, a web-based probability survey among same-sex and different-sex couples and families in the Netherlands (1,353 valid responses). The data were collected between July and December 2016. Respondents were sampled with a two-stage stratified sampling strategy. In the first stage, Dutch municipalities were selected across three geographical regions (North/East, West, and South) and three degrees of urbanization (marginal, moderate, and strong). In a second step, the local authorities drew random samples from three groups of addresses from their population registers: different-sex couples and same-sex couples with and without children. The same-sex households were oversampled in order to achieve group sizes large enough for comparison. An age limit of 30 to 65 years was imposed on the sample in order to maximize the number of participants given limited funds. The lower limit aimed at avoiding contacting student households, and the upper limit meant to exclude respondents who might be less likely to participate in web mode surveys. The final dataset includes weights, which allow correction for the geographic stratification of the municipality sample and for the oversampling of same-sex couples with and without children.

Respondents were contacted by post and invited to participate in the web mode survey. Response was higher among same-sex couples with children (34%) and same-sex couples without children (27%) and lower among different-sex couples (20%). After excluding respondents with missing information on relevant demographic variables, the analytical sample of partnered individuals in this study includes 521 women in same-sex relationships (in 320 households), 309 men in same-sex relationships (in 192 households), and 261 women and 238 men in different-sex relationships (in 346 households). The missing values showed no nested patterns and were therefore excluded under the assumption of missing at random (N = 24, 0.2%). The number of observations varies between analyses when respondents did not respond to the respective dependent variable. The survey does not contain measures of sexual or gender identity. This leaves some ambiguity as to who is in mixed-sex relationships. Sexual minority people who are in mixed-sex relationships remain invisible here. The binary gender measure conceals gender diverse identities and women or men with a history of transitioning gender.

Table 2.

Descriptive statistics (weighted; with standard errors) of demographic characteristics by relationship type and gender

Women in same-sex relationships Women in different-sex relationships Men in same-sex relationships Men in different-sex relationships
M/% SE M/% SE M/% SE M/% SE
Age (Mean)  46  .52  47  .81  49  .62  50  .88 
Education                 
low  6%  .02  20%  .04  9%  .02  12%  .03 
average  22%  .03  28%  .04  29%  .03  31%  .04 
high  71%  .03  52%  .04  63%  .03  58%  .05 
Children, 1 = yes  45%  .03  64%  .04  8%  .02  61%  .04 
Marital status                 
Married  61%  .03  77%  .03  59%  .03  77%  .04 
Civil union  12%  .02  6%  .02  14%  .02  11%  .03 
Single  28%  .03  18%  .03  27%  .03  12%  .02 
Living standard                 
low  7%  .02  8%  .02  6%  .01  5%  .02 
average  30%  .03  26%  .04  23%  .03  24%  .04 
high  62%  .03  66%  .04  72%  .03  71%  .04 
Urbanization                 
marginal  18%  .03  22%  .05  13%  .03  18%  .05 
average  13%  .02  29%  .04  13%  .03  32%  .05 
high  70%  .03  49%  .04  75%  .04  51%  .05 
N  521    261    309    238   
Women in same-sex relationships Women in different-sex relationships Men in same-sex relationships Men in different-sex relationships
M/% SE M/% SE M/% SE M/% SE
Age (Mean)  46  .52  47  .81  49  .62  50  .88 
Education                 
low  6%  .02  20%  .04  9%  .02  12%  .03 
average  22%  .03  28%  .04  29%  .03  31%  .04 
high  71%  .03  52%  .04  63%  .03  58%  .05 
Children, 1 = yes  45%  .03  64%  .04  8%  .02  61%  .04 
Marital status                 
Married  61%  .03  77%  .03  59%  .03  77%  .04 
Civil union  12%  .02  6%  .02  14%  .02  11%  .03 
Single  28%  .03  18%  .03  27%  .03  12%  .02 
Living standard                 
low  7%  .02  8%  .02  6%  .01  5%  .02 
average  30%  .03  26%  .04  23%  .03  24%  .04 
high  62%  .03  66%  .04  72%  .03  71%  .04 
Urbanization                 
marginal  18%  .03  22%  .05  13%  .03  18%  .05 
average  13%  .02  29%  .04  13%  .03  32%  .05 
high  70%  .03  49%  .04  75%  .04  51%  .05 
N  521    261    309    238   
Table 2.

Descriptive statistics (weighted; with standard errors) of demographic characteristics by relationship type and gender

Women in same-sex relationships Women in different-sex relationships Men in same-sex relationships Men in different-sex relationships
M/% SE M/% SE M/% SE M/% SE
Age (Mean)  46  .52  47  .81  49  .62  50  .88 
Education                 
low  6%  .02  20%  .04  9%  .02  12%  .03 
average  22%  .03  28%  .04  29%  .03  31%  .04 
high  71%  .03  52%  .04  63%  .03  58%  .05 
Children, 1 = yes  45%  .03  64%  .04  8%  .02  61%  .04 
Marital status                 
Married  61%  .03  77%  .03  59%  .03  77%  .04 
Civil union  12%  .02  6%  .02  14%  .02  11%  .03 
Single  28%  .03  18%  .03  27%  .03  12%  .02 
Living standard                 
low  7%  .02  8%  .02  6%  .01  5%  .02 
average  30%  .03  26%  .04  23%  .03  24%  .04 
high  62%  .03  66%  .04  72%  .03  71%  .04 
Urbanization                 
marginal  18%  .03  22%  .05  13%  .03  18%  .05 
average  13%  .02  29%  .04  13%  .03  32%  .05 
high  70%  .03  49%  .04  75%  .04  51%  .05 
N  521    261    309    238   
Women in same-sex relationships Women in different-sex relationships Men in same-sex relationships Men in different-sex relationships
M/% SE M/% SE M/% SE M/% SE
Age (Mean)  46  .52  47  .81  49  .62  50  .88 
Education                 
low  6%  .02  20%  .04  9%  .02  12%  .03 
average  22%  .03  28%  .04  29%  .03  31%  .04 
high  71%  .03  52%  .04  63%  .03  58%  .05 
Children, 1 = yes  45%  .03  64%  .04  8%  .02  61%  .04 
Marital status                 
Married  61%  .03  77%  .03  59%  .03  77%  .04 
Civil union  12%  .02  6%  .02  14%  .02  11%  .03 
Single  28%  .03  18%  .03  27%  .03  12%  .02 
Living standard                 
low  7%  .02  8%  .02  6%  .01  5%  .02 
average  30%  .03  26%  .04  23%  .03  24%  .04 
high  62%  .03  66%  .04  72%  .03  71%  .04 
Urbanization                 
marginal  18%  .03  22%  .05  13%  .03  18%  .05 
average  13%  .02  29%  .04  13%  .03  32%  .05 
high  70%  .03  49%  .04  75%  .04  51%  .05 
N  521    261    309    238   

Table 2 contains weighted descriptive statistics for people in same-sex and different-sex relationships. The point estimates of means and percentages can be generalized to the population of people in the Netherlands, who are between 30 and 65 years old and live together with a partner. The standard errors are reported as opposed to standard deviations to express the uncertainty in inference. Table 2 shows that the individuals are on average between 46 and 50 years old. Seventy-one percent of women in same-sex relationships and 63% of men in same-sex relationships obtained a higher education degree, compared to 52% of women in different-sex relationships and 58% of men in different-sex relationships. We see a low prevalence of parenthood among men in same-sex relationships (8%), while 45% of women in same-sex relationships have one or more children. Among people in different-sex relationships, about 60% have children. In the Netherlands, both marriage and civil unions are open to same-sex and different-sex couples. People in same-sex relationships are married in about 60% of the cases, compared to 77% of people in different-sex relationships. Civil unions are most prevalent among men in same-sex relationships (14%), followed by women in same-sex relationships (12%), men in different-sex relationships (11%), and women in different-sex relationships (6%). About 70% of all men, regardless of relationship type, can live very comfortably on their current household income. This is the case for 66% of women in different-sex relationships and 62% of women in same-sex relationships. People in same-sex relationships frequently live in urban areas (70%–75%), compared to half of all people in different-sex relationships.

Measures

Network variables

The social networks are measured by a name generator question, which is an extension of the classic discussion network instrument by Marsden (1987). Respondents were asked “Who are the people with whom you like to do things or discuss important matters with?” The original discussion partner measure has been extended by the addition of people with whom one likes to spend time. This is aimed at correcting some of the bias brought on by the fact that some people either do not have important things to talk about or do not talk about them with other people (Bearman and Parigi 2004). This alteration to the measure has the intent of better grasping something we might call “daily social networks”—a mixture of people that provide companionship and people who are there if the respondent is in need of a discussion partner (other than the cohabiting partner or spouse). Respondents were asked to report a maximum of eight people who live outside of their household, which excludes the partner they live with.

Network size is measured by the total number of people reported, with a minimum of zero and a maximum of eight. For each network member, a number of name interpreter questions were asked. For the role composition of the network, I consider the total number of family-of-origin members mentioned and the total number of friends. Similar others in the network are measured by the number of sexual minority ties and the number of same-gender ties.

Independent variable

The main independent variable is relationship type (same-sex or mixed-sex). The classification is based on the self-reported gender of the respondent and the gender of the partner with whom the respondent lives (as reported by the respondent). The gender of the respondent is included as moderator variable where appropriate, otherwise as control variable. The binary gender variable conceals gender diversity that may exist within either relationship type.

Control variables

At the individual level, age is a predictor of network size and composition (Kalmijn 2012; Wrzus et al. 2013). A mean-centered version is included in the analyses. Education is measured in three categories, namely low (up to mavo; ISCED 0-2), average (mbo, havo-vwo; ISCED 3-4), and high education (HBO and higher; ISCED 5-6). It has been observed that social networks of highly educated individuals are larger, less dense, and more diverse (McPherson, Smith-Lovin, and Brashears 2006; Mollenhorst, Volker, and Flap 2008). Socioeconomic position is also relevant as more financial resources are linked to larger networks. In this study, socioeconomic position is approximated by subjective evaluation of the current household income. The final variable has three categories indicating whether the living standard is perceived to be low (very difficult to cope; difficult to cope), average (possible to cope), or high (living comfortably; living very comfortably). There is a significant positive association between individual net income and subjective living standard (gamma = .53). Subjective living standard has fewer missing values than household income and is therefore preferred. The variable likely underestimates variation in income since people’s subjective evaluation of their situation is usually relative to their environment. Prior research shows that sexual minority men tend to have lower earnings than heterosexual men; sexual minority women have similar or slightly higher earnings than heterosexual women (but still lower than heterosexual men) (Mize 2016). In light of this, the size of sexual minority men’s networks may be underestimated in the analyses. There is also ample evidence that the family situation is relevant to social networks (Rözer, Mollenhorst, and Volker 2015). Marital status is distinguished in three categories (married, civil union, single), and the presence of at least one child under the age of 18 in the household is indicated by a binary variable (no, yes). The degree of urbanization is measured in three categories (marginal, moderate, strong). The measure is based on the official five-category measure by Statistics Netherlands, whereby the two highest and two lowest degrees of urbanization are collapsed into two respective categories. The demographic variables are used in the regression models to account for compositional differences between the two groups and to control for omitted variable bias.

Analyses

For each of the six network variables, I run multiple ordinary-least squares (OLS) regression models. The first model does not include any control variables to show whether there is a mean difference between people in same-sex and different-sex relationships. The following models include relevant variables, which may explain part of the gap. Regression models which have number of family-of-origin ties, friends, same-gender ties, and sexual minority people as outcome are controlled for network size. Non-adjusted coefficients of these variables are not meaningful. Poisson models for count variables have also been calculated as robustness check since the variables for number of family-of-origin members and number of sexual minority ties are heavily skewed. The Poisson models produce the same substantive conclusions as the OLS models. Therefore, the main text presents OLS models. By sampling design, individuals are nested in municipalities and households in these data. This needs to be accounted for in the analyses. Therefore, I use clustered standard errors at the household level in an effort to statistically correct for the dependence of two observations per household. In an effort to account for municipal-level variance, I add a control variable at the municipality-level for urbanization. Twenty municipalities are not enough contextual units to reliably employ multilevel analyses.

Results

Bivariate Results

I begin by discussing social network differences and commonalities between people in same-sex and mixed-sex relationships. Table 3 shows that the average network size consists of five members, except for men in mixed-sex relationships, who have only four ties. Several respondents indicated that they have nobody with whom they discuss important matters or like to spend time (not shown in table). This share was highest among men in different-sex relationships (18%), followed by men in same-sex relationships (8%), women in different-sex relationships (5%), and women in same-sex relationships (3%). Prior research suggests (McPherson et al. 2006) that women have more family-of-origin ties in their social networks than men. This seems to be true for women in mixed-sex relationships who named two family-of-origin ties, on average; all other groups have fewer family-of-origin ties at about 1.6 ties. The number of friends in the personal network differs mostly between relationship types, but not between men and women: people in same-sex relationships have more friends in their social networks (about 3.2) than people in different-sex relationships (2.5 for women and 2.4 for men).

Table 3.

Descriptives (weighted; with standard errors) of social networks by relationship type and gender

Women in same-sex relationships Women in different-sex relationships Men in same-sex relationships Men in different-sex relationships
N Mean (SD) N Mean (SD) N Mean (SD) N Mean (SD)
Network size  521  5.44  261  5.14  309  5.12  238  3.98 
    (.13)    (.19)    (.18)    (.26) 
Family  507  1.64  247  2.06  286  1.68  198  1.63 
    (.09)    (.14)    (.13)    (.16) 
Friends  507  3.15  247  2.52  286  3.18  198  2.36 
    (.12)    (.17)    (.13)    (.20) 
Sexual minority ties  504  1.19  237  0.12  280  1.32  189  0.17 
    (.07)    (.25)    (.10)    (.09) 
Same gender ties  484  4.42  231  4.37  274  2.81  189  3.55 
    (.11)    (.18)    (.14)    (.18) 
N  521    261    309    238   
Women in same-sex relationships Women in different-sex relationships Men in same-sex relationships Men in different-sex relationships
N Mean (SD) N Mean (SD) N Mean (SD) N Mean (SD)
Network size  521  5.44  261  5.14  309  5.12  238  3.98 
    (.13)    (.19)    (.18)    (.26) 
Family  507  1.64  247  2.06  286  1.68  198  1.63 
    (.09)    (.14)    (.13)    (.16) 
Friends  507  3.15  247  2.52  286  3.18  198  2.36 
    (.12)    (.17)    (.13)    (.20) 
Sexual minority ties  504  1.19  237  0.12  280  1.32  189  0.17 
    (.07)    (.25)    (.10)    (.09) 
Same gender ties  484  4.42  231  4.37  274  2.81  189  3.55 
    (.11)    (.18)    (.14)    (.18) 
N  521    261    309    238   
Table 3.

Descriptives (weighted; with standard errors) of social networks by relationship type and gender

Women in same-sex relationships Women in different-sex relationships Men in same-sex relationships Men in different-sex relationships
N Mean (SD) N Mean (SD) N Mean (SD) N Mean (SD)
Network size  521  5.44  261  5.14  309  5.12  238  3.98 
    (.13)    (.19)    (.18)    (.26) 
Family  507  1.64  247  2.06  286  1.68  198  1.63 
    (.09)    (.14)    (.13)    (.16) 
Friends  507  3.15  247  2.52  286  3.18  198  2.36 
    (.12)    (.17)    (.13)    (.20) 
Sexual minority ties  504  1.19  237  0.12  280  1.32  189  0.17 
    (.07)    (.25)    (.10)    (.09) 
Same gender ties  484  4.42  231  4.37  274  2.81  189  3.55 
    (.11)    (.18)    (.14)    (.18) 
N  521    261    309    238   
Women in same-sex relationships Women in different-sex relationships Men in same-sex relationships Men in different-sex relationships
N Mean (SD) N Mean (SD) N Mean (SD) N Mean (SD)
Network size  521  5.44  261  5.14  309  5.12  238  3.98 
    (.13)    (.19)    (.18)    (.26) 
Family  507  1.64  247  2.06  286  1.68  198  1.63 
    (.09)    (.14)    (.13)    (.16) 
Friends  507  3.15  247  2.52  286  3.18  198  2.36 
    (.12)    (.17)    (.13)    (.20) 
Sexual minority ties  504  1.19  237  0.12  280  1.32  189  0.17 
    (.07)    (.25)    (.10)    (.09) 
Same gender ties  484  4.42  231  4.37  274  2.81  189  3.55 
    (.11)    (.18)    (.14)    (.18) 
N  521    261    309    238   

Moving on to the presence of similar others, women and men in same-sex relationships have an average of 1.2 and 1.3 sexual minority people in their social networks, respectively. People in different-sex relationships have very few sexual minority people in their social networks, namely 0.12 for women and 0.17 for men, on average. Moreover, social networks consist for a large part of same-gender ties. Women in both different- and same-sex relationships have an average of 4.4 ties with other women. Men in different-sex relationships report 3.6 ties with other men, whereas men in same-sex relationships have significantly fewer ties with other men (2.8). In the next section, I explore whether these differences in social networks between people in same-sex and different-sex relationships can be explained by demographic characteristics or by features of their social networks.

Multivariate Results

I begin by discussing the regression results for network size (see table 4). The regression results further support what we have already seen in the bivariate analyses: people in same-sex relationships have significantly larger networks than their counterparts in different-sex relationships (b = .855). In the second model, I include demographic control variables. Being a woman, highly educated and having a high living standard is associated with larger networks and having children with smaller networks. The coefficient reduces somewhat in size but continues to show larger networks among people in same-sex relationships. Next, I include a control variable for urbanization, which does not impact the coefficient. This is evidence in support of hypothesis 1: people in same-sex relationships have larger networks than people in different-sex relationships, with “half a person” more on average.

Table 4.

OLS regression models of network size, role composition (family, friends) and similar others (sexual minority people, same-gender ties) on relationship type

Network size (1) Network size (2) Network size (3)
Constant  4.579*** (.134)  3.800*** (.259)  3.431*** (.353) 
Main independent variables       
Same-sex relationship (ref. is different-sex)  .855*** (.161)  .492** (.168)  .464** (.169) 
Woman (ref. is man)    .791*** (.143)  .796*** (.142) 
Interaction effect       
Same-sex relationship x woman       
Control variables       
Age    −.017* (.008)  −.017* (.008) 
Education       
low    −.506 (.291)  −.518 (.289) 
high    .650*** (.177)  .618*** (.178) 
Marital status       
Married    .115 (.173)  .145 (.174) 
Civil union    −.357 (.250)  −.337 (.250) 
Children, 1 = yes    −.450** (.160)  −.457** (.159) 
Living standard       
low    −.727** (.278)  −.748** (.277) 
high    .550*** (.164)  .550*** (.164) 
Urbanization       
marginal      .383 (.424) 
strong      .425 (.272) 
Network variables       
Network size       
Observations  1,265  1,265  1,265 
Adjusted R2  .027  .108  .109 
Constant  1.805*** (.080)  .242 (.169)  .435 (.228)  2.495*** (.100)  −.456** (.173)  −.584* (.241) 
Main independent variables             
Same-sex relationship  −.240* (.099)  −.315*** (.095)  −.302** (.095)  .853*** (.129)  .447*** (.109)  .439*** (.110) 
Woman (ref. is man)    .181* (.090)  .172 (.089)    −.177 (.100)  −.168 (.099) 
Interaction effect             
Same-sex relationship x woman             
Control variables             
Age    −.009 (.005)  −.009 (.005)    −.005 (.006)  −.004 (.005) 
Education             
low    .188 (.170)  .184 (.172)    .103 (.173)  .109 (.175) 
high    −.419*** (.104)  −.390*** (.106)    .383*** (.113)  .355** (.116) 
Marital status             
Married    .131 (.101)  .098 (.102)    −.120 (.117)  −.088 (.118) 
Civil union    −.006 (.150)  −.024 (.151)    .257 (.156)  .273 (.155) 
Children, 1 = yes    .007 (.101)  .011 (.100)    −.132 (.113)  −.135 (.112) 
Living standard             
low    −.085 (.148)  −.074 (.148)    .126 (.169)  .119 (.172) 
high    .046 (.102)  .046 (.102)    .039 (.108)  .039 (.108) 
Urbanization             
marginal      .148 (.241)      −.305 (.303) 
strong      −.241 (.166)      .171(.181) 
Network variables             
Network size    .303*** (.020)  .304*** (.020)    .567*** (.022)  .565*** (.022) 
Observations  1,178  1,178  1,178  1,178  1,178  1,178 
Adjusted R2  .005  .187  .189  .040  .413  .415 
Constant  .186*** (.035)  −.617*** (.150)  −.832*** (.202)  3.901*** (.103)  −.495** (.153)  −.310 (.189)  .072 (.192) 
Main independent variables               
Same-sex relationship  1.192*** (.073)  1.051*** (.072)  1.036*** (.071)  −.020 (.132)  −.372*** (.086)  −.357*** (.087)  −1.003*** (.139) 
Woman (ref. is man)    −.076 (.088)  −.068 (.089)    1.183*** (.087)  1.183*** (.087)  .544*** (.110) 
Interaction effect               
Same-sex relationship x woman              1.048*** (.160) 
Control variables               
Age    .009 (.005)  .009 (.005)    .005 (.004)  .005 (.004)  .004 (.004) 
Education               
low    −.205 (.135)  −.203 (.135)    −.059 (.128)  −.055 (.127)  −.023 (.131) 
high    −.068 (.099)  −.101 (.099)    .130 (.099)  .140 (.100)  .125 (.099) 
Marital status               
Married    −.236* (.102)  −.202 (.103)    −.032 (.089)  −.042 (.090)  −.011 (.088) 
Civil union    −.179 (.149)  −.160 (.149)    .056 (.135)  .049 (.135)  .074 (.133) 
Children, 1 = yes    −.061 (.091)  −.066 (.091)    .282** (.089)  .285** (.089)  .159 (.090) 
Living standard               
low    .342* (.134)  .330* (.135)    .150 (.152)  .163 (.152)  .205 (.151) 
high    .170* (.086)  .172* (.087)    .091 (.090)  .091 (.090)  .097 (.089) 
Urbanization               
marginal      −.120 (.193)      −.277 (.194)  −.281 (.193) 
strong      .268 (.140)      −.213 (.137)  −.212 (.135) 
Network variables               
Network size    .192*** (.019)  .190*** (.019)    .670*** (.017)  .671*** (.017)  .676*** (.017) 
Observations  1,161  1,161  1,161  1,121  1,121  1,121  1,121 
Adjusted R2  .163  .254  .258  .000  .643  .643  .657 
Network size (1) Network size (2) Network size (3)
Constant  4.579*** (.134)  3.800*** (.259)  3.431*** (.353) 
Main independent variables       
Same-sex relationship (ref. is different-sex)  .855*** (.161)  .492** (.168)  .464** (.169) 
Woman (ref. is man)    .791*** (.143)  .796*** (.142) 
Interaction effect       
Same-sex relationship x woman       
Control variables       
Age    −.017* (.008)  −.017* (.008) 
Education       
low    −.506 (.291)  −.518 (.289) 
high    .650*** (.177)  .618*** (.178) 
Marital status       
Married    .115 (.173)  .145 (.174) 
Civil union    −.357 (.250)  −.337 (.250) 
Children, 1 = yes    −.450** (.160)  −.457** (.159) 
Living standard       
low    −.727** (.278)  −.748** (.277) 
high    .550*** (.164)  .550*** (.164) 
Urbanization       
marginal      .383 (.424) 
strong      .425 (.272) 
Network variables       
Network size       
Observations  1,265  1,265  1,265 
Adjusted R2  .027  .108  .109 
Constant  1.805*** (.080)  .242 (.169)  .435 (.228)  2.495*** (.100)  −.456** (.173)  −.584* (.241) 
Main independent variables             
Same-sex relationship  −.240* (.099)  −.315*** (.095)  −.302** (.095)  .853*** (.129)  .447*** (.109)  .439*** (.110) 
Woman (ref. is man)    .181* (.090)  .172 (.089)    −.177 (.100)  −.168 (.099) 
Interaction effect             
Same-sex relationship x woman             
Control variables             
Age    −.009 (.005)  −.009 (.005)    −.005 (.006)  −.004 (.005) 
Education             
low    .188 (.170)  .184 (.172)    .103 (.173)  .109 (.175) 
high    −.419*** (.104)  −.390*** (.106)    .383*** (.113)  .355** (.116) 
Marital status             
Married    .131 (.101)  .098 (.102)    −.120 (.117)  −.088 (.118) 
Civil union    −.006 (.150)  −.024 (.151)    .257 (.156)  .273 (.155) 
Children, 1 = yes    .007 (.101)  .011 (.100)    −.132 (.113)  −.135 (.112) 
Living standard             
low    −.085 (.148)  −.074 (.148)    .126 (.169)  .119 (.172) 
high    .046 (.102)  .046 (.102)    .039 (.108)  .039 (.108) 
Urbanization             
marginal      .148 (.241)      −.305 (.303) 
strong      −.241 (.166)      .171(.181) 
Network variables             
Network size    .303*** (.020)  .304*** (.020)    .567*** (.022)  .565*** (.022) 
Observations  1,178  1,178  1,178  1,178  1,178  1,178 
Adjusted R2  .005  .187  .189  .040  .413  .415 
Constant  .186*** (.035)  −.617*** (.150)  −.832*** (.202)  3.901*** (.103)  −.495** (.153)  −.310 (.189)  .072 (.192) 
Main independent variables               
Same-sex relationship  1.192*** (.073)  1.051*** (.072)  1.036*** (.071)  −.020 (.132)  −.372*** (.086)  −.357*** (.087)  −1.003*** (.139) 
Woman (ref. is man)    −.076 (.088)  −.068 (.089)    1.183*** (.087)  1.183*** (.087)  .544*** (.110) 
Interaction effect               
Same-sex relationship x woman              1.048*** (.160) 
Control variables               
Age    .009 (.005)  .009 (.005)    .005 (.004)  .005 (.004)  .004 (.004) 
Education               
low    −.205 (.135)  −.203 (.135)    −.059 (.128)  −.055 (.127)  −.023 (.131) 
high    −.068 (.099)  −.101 (.099)    .130 (.099)  .140 (.100)  .125 (.099) 
Marital status               
Married    −.236* (.102)  −.202 (.103)    −.032 (.089)  −.042 (.090)  −.011 (.088) 
Civil union    −.179 (.149)  −.160 (.149)    .056 (.135)  .049 (.135)  .074 (.133) 
Children, 1 = yes    −.061 (.091)  −.066 (.091)    .282** (.089)  .285** (.089)  .159 (.090) 
Living standard               
low    .342* (.134)  .330* (.135)    .150 (.152)  .163 (.152)  .205 (.151) 
high    .170* (.086)  .172* (.087)    .091 (.090)  .091 (.090)  .097 (.089) 
Urbanization               
marginal      −.120 (.193)      −.277 (.194)  −.281 (.193) 
strong      .268 (.140)      −.213 (.137)  −.212 (.135) 
Network variables               
Network size    .192*** (.019)  .190*** (.019)    .670*** (.017)  .671*** (.017)  .676*** (.017) 
Observations  1,161  1,161  1,161  1,121  1,121  1,121  1,121 
Adjusted R2  .163  .254  .258  .000  .643  .643  .657 
Table 4.

OLS regression models of network size, role composition (family, friends) and similar others (sexual minority people, same-gender ties) on relationship type

Network size (1) Network size (2) Network size (3)
Constant  4.579*** (.134)  3.800*** (.259)  3.431*** (.353) 
Main independent variables       
Same-sex relationship (ref. is different-sex)  .855*** (.161)  .492** (.168)  .464** (.169) 
Woman (ref. is man)    .791*** (.143)  .796*** (.142) 
Interaction effect       
Same-sex relationship x woman       
Control variables       
Age    −.017* (.008)  −.017* (.008) 
Education       
low    −.506 (.291)  −.518 (.289) 
high    .650*** (.177)  .618*** (.178) 
Marital status       
Married    .115 (.173)  .145 (.174) 
Civil union    −.357 (.250)  −.337 (.250) 
Children, 1 = yes    −.450** (.160)  −.457** (.159) 
Living standard       
low    −.727** (.278)  −.748** (.277) 
high    .550*** (.164)  .550*** (.164) 
Urbanization       
marginal      .383 (.424) 
strong      .425 (.272) 
Network variables       
Network size       
Observations  1,265  1,265  1,265 
Adjusted R2  .027  .108  .109 
Constant  1.805*** (.080)  .242 (.169)  .435 (.228)  2.495*** (.100)  −.456** (.173)  −.584* (.241) 
Main independent variables             
Same-sex relationship  −.240* (.099)  −.315*** (.095)  −.302** (.095)  .853*** (.129)  .447*** (.109)  .439*** (.110) 
Woman (ref. is man)    .181* (.090)  .172 (.089)    −.177 (.100)  −.168 (.099) 
Interaction effect             
Same-sex relationship x woman             
Control variables             
Age    −.009 (.005)  −.009 (.005)    −.005 (.006)  −.004 (.005) 
Education             
low    .188 (.170)  .184 (.172)    .103 (.173)  .109 (.175) 
high    −.419*** (.104)  −.390*** (.106)    .383*** (.113)  .355** (.116) 
Marital status             
Married    .131 (.101)  .098 (.102)    −.120 (.117)  −.088 (.118) 
Civil union    −.006 (.150)  −.024 (.151)    .257 (.156)  .273 (.155) 
Children, 1 = yes    .007 (.101)  .011 (.100)    −.132 (.113)  −.135 (.112) 
Living standard             
low    −.085 (.148)  −.074 (.148)    .126 (.169)  .119 (.172) 
high    .046 (.102)  .046 (.102)    .039 (.108)  .039 (.108) 
Urbanization             
marginal      .148 (.241)      −.305 (.303) 
strong      −.241 (.166)      .171(.181) 
Network variables             
Network size    .303*** (.020)  .304*** (.020)    .567*** (.022)  .565*** (.022) 
Observations  1,178  1,178  1,178  1,178  1,178  1,178 
Adjusted R2  .005  .187  .189  .040  .413  .415 
Constant  .186*** (.035)  −.617*** (.150)  −.832*** (.202)  3.901*** (.103)  −.495** (.153)  −.310 (.189)  .072 (.192) 
Main independent variables               
Same-sex relationship  1.192*** (.073)  1.051*** (.072)  1.036*** (.071)  −.020 (.132)  −.372*** (.086)  −.357*** (.087)  −1.003*** (.139) 
Woman (ref. is man)    −.076 (.088)  −.068 (.089)    1.183*** (.087)  1.183*** (.087)  .544*** (.110) 
Interaction effect               
Same-sex relationship x woman              1.048*** (.160) 
Control variables               
Age    .009 (.005)  .009 (.005)    .005 (.004)  .005 (.004)  .004 (.004) 
Education               
low    −.205 (.135)  −.203 (.135)    −.059 (.128)  −.055 (.127)  −.023 (.131) 
high    −.068 (.099)  −.101 (.099)    .130 (.099)  .140 (.100)  .125 (.099) 
Marital status               
Married    −.236* (.102)  −.202 (.103)    −.032 (.089)  −.042 (.090)  −.011 (.088) 
Civil union    −.179 (.149)  −.160 (.149)    .056 (.135)  .049 (.135)  .074 (.133) 
Children, 1 = yes    −.061 (.091)  −.066 (.091)    .282** (.089)  .285** (.089)  .159 (.090) 
Living standard               
low    .342* (.134)  .330* (.135)    .150 (.152)  .163 (.152)  .205 (.151) 
high    .170* (.086)  .172* (.087)    .091 (.090)  .091 (.090)  .097 (.089) 
Urbanization               
marginal      −.120 (.193)      −.277 (.194)  −.281 (.193) 
strong      .268 (.140)      −.213 (.137)  −.212 (.135) 
Network variables               
Network size    .192*** (.019)  .190*** (.019)    .670*** (.017)  .671*** (.017)  .676*** (.017) 
Observations  1,161  1,161  1,161  1,121  1,121  1,121  1,121 
Adjusted R2  .163  .254  .258  .000  .643  .643  .657 
Network size (1) Network size (2) Network size (3)
Constant  4.579*** (.134)  3.800*** (.259)  3.431*** (.353) 
Main independent variables       
Same-sex relationship (ref. is different-sex)  .855*** (.161)  .492** (.168)  .464** (.169) 
Woman (ref. is man)    .791*** (.143)  .796*** (.142) 
Interaction effect       
Same-sex relationship x woman       
Control variables       
Age    −.017* (.008)  −.017* (.008) 
Education       
low    −.506 (.291)  −.518 (.289) 
high    .650*** (.177)  .618*** (.178) 
Marital status       
Married    .115 (.173)  .145 (.174) 
Civil union    −.357 (.250)  −.337 (.250) 
Children, 1 = yes    −.450** (.160)  −.457** (.159) 
Living standard       
low    −.727** (.278)  −.748** (.277) 
high    .550*** (.164)  .550*** (.164) 
Urbanization       
marginal      .383 (.424) 
strong      .425 (.272) 
Network variables       
Network size       
Observations  1,265  1,265  1,265 
Adjusted R2  .027  .108  .109 
Constant  1.805*** (.080)  .242 (.169)  .435 (.228)  2.495*** (.100)  −.456** (.173)  −.584* (.241) 
Main independent variables             
Same-sex relationship  −.240* (.099)  −.315*** (.095)  −.302** (.095)  .853*** (.129)  .447*** (.109)  .439*** (.110) 
Woman (ref. is man)    .181* (.090)  .172 (.089)    −.177 (.100)  −.168 (.099) 
Interaction effect             
Same-sex relationship x woman             
Control variables             
Age    −.009 (.005)  −.009 (.005)    −.005 (.006)  −.004 (.005) 
Education             
low    .188 (.170)  .184 (.172)    .103 (.173)  .109 (.175) 
high    −.419*** (.104)  −.390*** (.106)    .383*** (.113)  .355** (.116) 
Marital status             
Married    .131 (.101)  .098 (.102)    −.120 (.117)  −.088 (.118) 
Civil union    −.006 (.150)  −.024 (.151)    .257 (.156)  .273 (.155) 
Children, 1 = yes    .007 (.101)  .011 (.100)    −.132 (.113)  −.135 (.112) 
Living standard             
low    −.085 (.148)  −.074 (.148)    .126 (.169)  .119 (.172) 
high    .046 (.102)  .046 (.102)    .039 (.108)  .039 (.108) 
Urbanization             
marginal      .148 (.241)      −.305 (.303) 
strong      −.241 (.166)      .171(.181) 
Network variables             
Network size    .303*** (.020)  .304*** (.020)    .567*** (.022)  .565*** (.022) 
Observations  1,178  1,178  1,178  1,178  1,178  1,178 
Adjusted R2  .005  .187  .189  .040  .413  .415 
Constant  .186*** (.035)  −.617*** (.150)  −.832*** (.202)  3.901*** (.103)  −.495** (.153)  −.310 (.189)  .072 (.192) 
Main independent variables               
Same-sex relationship  1.192*** (.073)  1.051*** (.072)  1.036*** (.071)  −.020 (.132)  −.372*** (.086)  −.357*** (.087)  −1.003*** (.139) 
Woman (ref. is man)    −.076 (.088)  −.068 (.089)    1.183*** (.087)  1.183*** (.087)  .544*** (.110) 
Interaction effect               
Same-sex relationship x woman              1.048*** (.160) 
Control variables               
Age    .009 (.005)  .009 (.005)    .005 (.004)  .005 (.004)  .004 (.004) 
Education               
low    −.205 (.135)  −.203 (.135)    −.059 (.128)  −.055 (.127)  −.023 (.131) 
high    −.068 (.099)  −.101 (.099)    .130 (.099)  .140 (.100)  .125 (.099) 
Marital status               
Married    −.236* (.102)  −.202 (.103)    −.032 (.089)  −.042 (.090)  −.011 (.088) 
Civil union    −.179 (.149)  −.160 (.149)    .056 (.135)  .049 (.135)  .074 (.133) 
Children, 1 = yes    −.061 (.091)  −.066 (.091)    .282** (.089)  .285** (.089)  .159 (.090) 
Living standard               
low    .342* (.134)  .330* (.135)    .150 (.152)  .163 (.152)  .205 (.151) 
high    .170* (.086)  .172* (.087)    .091 (.090)  .091 (.090)  .097 (.089) 
Urbanization               
marginal      −.120 (.193)      −.277 (.194)  −.281 (.193) 
strong      .268 (.140)      −.213 (.137)  −.212 (.135) 
Network variables               
Network size    .192*** (.019)  .190*** (.019)    .670*** (.017)  .671*** (.017)  .676*** (.017) 
Observations  1,161  1,161  1,161  1,121  1,121  1,121  1,121 
Adjusted R2  .163  .254  .258  .000  .643  .643  .657 

Is there evidence for the families-of-choice hypothesis? It appears that people in same-sex relationships have somewhat fewer family-of-origin ties in their social networks than people in different-sex relationships (b = −.240). The coefficient is modest in size and does not indicate a large difference. When including control variables for individual demographic characteristics and urbanization, the difference increases somewhat, suggesting a compositional explanation behind the initially smaller coefficient. The bivariate analyses have already suggested that people in same-sex relationships have more friends in their networks than people in different-sex relationships. This result is affirmed in the regression analyses, where I adjust for network size (b = .853). The coefficient reduces in size when controlling for demographic explanations and urbanization (b = .439) but the substantive conclusion remains unchanged: people in same-sex relationships have somewhat fewer family-of-origin ties and more friendship ties than people in different-sex relationships. The evidence supports the families-of-choice hypothesis. Additional analyses did not show a gender difference in this dynamic.

Next, I examine homophily on the basis of sexual orientation and gender. Unsurprisingly, people in same-sex relationships have more sexual minority ties in their networks compared to people in different-sex relationships; specifically, about one more sexual minority tie (b = 1.192). The coefficients remain stable in size and significance when taking demographic characteristics and urbanization into account. This supports hypothesis 3 which suggests more sexual minority ties among people in same-sex relationships. Interestingly, being in a same-sex relationship compared to a mixed-sex relationship explains only 16% of the variance in the number of sexual minority ties in the personal network. Turning to the gender composition of the networks, no difference between relationship types is found initially. Controlling for demographic characteristics and urbanization reveal a negative coefficient for being in a same-sex relationship, suggesting less gender segregation among people in same-sex relationships. Notably, the gender of the respondent emerges as the strongest correlate, which is in line with the literature on gender segregation in social networks (Kalmijn 2002). In light of this, the interaction between relationship type and gender is certainly relevant. As hypothesized, gender segregation seems higher among men in same-sex relationships as indicated by the interaction effect between gender and relationship type (b = 1.048). The predicted number of same-gender people in the personal network is lowest among men in same-sex relationships (2.8), followed by men in mixed-sex relationships (3.8) and women in both relationship types (4.4). This means that hypothesis 4, which expects fewer same-gender ties, receives supporting evidence for men only. Among women there is no observed difference.

Discussion and Conclusions

What is the state of social exclusion and resilience in the face of stigmatization in a place like the Netherlands? This study uses a national sample of women and men in same-sex and different-sex relationships in the Netherlands to provide one answer to this question. Specifically, I examine network stratification along sexual orientation lines by comparing the social networks of people in same-sex relationships to those of people in different-sex relationships in terms of size and composition. The results illustrate that there are mostly commonalities. Yet, a few meaningful differences between people in same-sex and different-sex relationships emerge.

One hypothesis that is often anecdotally referred to is that sexual minority people tend to have more problematic family-of-origin ties, which are then substituted by a chosen family of supportive friendship ties (families-of-choice hypothesis). The current study empirically tests the families-of-choice hypothesis on probability-based survey data among people in same-sex relationships. I find evidence in support of this hypothesis. While potentially difficult ties with the family-of-origin, in particular with the parents, seem to recover in adulthood (Fischer and Kalmijn, 2020), family-of-origin ties are still less integrated into the circle with whom people discuss matters important to them or with whom they spend their time with. For women in same-sex relationships, this finding of fewer family-of-origin ties is quite concerning. Frost et al. (2016) have shown that—like people in different-sex relationships—women in same-sex relationships predominantly rely on family-of-origin ties for major support (like borrowing money). The fact that women in same-sex relationships appear to have fewer family-of-origin ties, therefore, can be detrimental in times of need. For sexual minority men friendship ties are the primary source of support (Frost, Meyer, and Schwartz 2016). The somewhat larger networks among people in same-sex relationships are suggestive of successful resilience strategies that involve building alternative, accepting communities.

Is there such a thing as choice homophily on the basis of sexual orientation? Studied as standalone dimension, network diversity in terms of sexual orientation unsurprisingly emerges as a matter of relationship type. People in same-sex relationships have about one more tie with other sexual minority people compared to people in different-sex relationships. Since I do not measure sexual identity here, this difference may be underestimated, as some sexual minority people are in mixed-sex relationships. Relationship type only accounts for 16% of the variance in sexual minority ties; sexual identity may be able to account for more.

People in different-sex relationships have very few sexual minority ties, in fact close to none. It is possible that, here too, underestimation plays a role if sexual minority people conceal their sexual orientation from heterosexual friends. The sexual orientation of network members is reported by the respondent. Nonetheless, this is evidence in support of homophily based on sexual orientation. Considering that there are persistent gender differences in how accepting heterosexual people are of sexual diversity (Worthen 2013), it is interesting to see that this does not seem to translate into differences in actual sexual minority ties among heterosexuals (no significant gender coefficient). A larger sample would be needed to tease out gender differences within a group that is already so small (i.e., the group of people in different-sex relationships who have least one sexual minority person in their social networks).

What can we learn if we take into account sexual diversity when studying gender segregation of social networks? Among men, a clear sexual orientation difference emerged. This finding is in line with literature that suggests traditional standards of masculinity strongly sanction same-sex attraction in men (Worthen 2013). Without implying that same-sex attraction in women is not socially stigmatized—the processes differ from those applying to masculinities—men in same-sex relationships emerge as the ones showing structural gender differences in their social networks. The fact that negative attitudes toward sexual diversity among men show up in the form of the gender composition of men in same-sex relationships but not among men in different-sex relationships in the number of sexual minority ties likely has to do with group size. Such effects would be more noticeable in the numerically small group of sexual minority men compared to the numerically large group heterosexual men. The fact that this dynamic is observable in the Netherlands, where acceptance of sexual diversity is relatively high in international comparison, shows that social exclusion still plays a substantial role. Among women, gender segregation is relatively high, regardless of the relationship type. The current data do not allow for a breakdown of the social networks according to sexual orientation and gender of the network members simultaneously, as the number of observations would become too low. It would be insightful to conduct further research into whether sexual minority men’s ties are with other sexual minority men or with heterosexual men to gain a more nuanced understanding of the role of heteronormative masculinities. Similarly, it would enhance our understanding of women’s networks; perhaps the same number of same-gender ties conceals heterogeneity in terms of sexual orientation of the network members.

As any study, the present one comes with certain limitations. It is important to highlight again that these findings apply to people in same-sex and different-sex relationships between 30 and 65 years and who are currently living together with a partner in the Netherlands. This makes this group selective with limited applicability to singles and other age groups. Based on existing research of (presumed) heterosexual people, we know that the networks of partnered individuals tend to differ from those of singles in meaningful ways. Couples have fewer friends and more family-focused networks (Rözer, Mollenhorst, and Volker 2015). It is reasonable to assume that these processes apply to women and men in same-sex relationships in a similar way. Yet, given that the presence of a same-sex partner does not comply with the prevailing heteronormative standard, it is advisable to examine in what way these processes possibly differ for sexual minority people. In addition, a life course perspective on social networks of people in same-sex relationships could give more detailed insights into changes over time in relation to significant events, such as coming out to the family-of-origin and partnership formation. Networks are not time-invariant and the current analysis must be considered as a snapshot of differences, which may shift over time and across the life course.

The original measure of discussion partners, which the current measure expanded on, has received a number of critiques relating to gender differences in recall bias and the interpretation what important topics are to different people (Bearman and Parigi 2004). Since there is no reason to believe that these dynamics differ for people in different relationship types, this does not threaten the validity of the central comparison between relationship types. Another critique relates to the fact that not all people have things to discuss with others or, if they do, they chose not to discuss them (Bearman and Parigi 2004). In order to counteract this, the extended measure in this study allows people to list others with whom they like to spend time with. This study does not use a measure that purely captures discussion partners but a broader range of people.

What are the implications of this study? Two contributions stand out. First, I have shown that sexual orientation is a meaningful axis of difference in social networks. Sexual orientation is a dimension worthwhile studying as a stratifying factor of social networks both standing alone and at the intersection with gender. Second, I have shown that differences in social networks between people in same-sex and different-sex relationships can be detected in a place as tolerant as the Netherlands. This highlights once more that tolerance cannot be equated to true acceptance. Differences in interpersonal relationships persist and indicate that social exclusion continues to plays a role. The study of social exclusion and resilience strategies is a more nuanced view into the state of equity than attitudes of the majority population. This insight can complement current explanation models for inequalities faced by sexual minority people across various life domains.

ABOUT THE AUTHORS

Mirjam Fischer is a Sociologist working as postdoctoral researcher at the University of Cologne. She received her PhD at the University of Amsterdam. Her research interests include structural inequalities related to (mental) health, well-being, social networks, and family ties between people with minoritized sexual and gender identities and the cis-heterosexual majority, and sampling and measurement issues related to LGBTQI* populations. Her most recent publication dealt with intergenerational ties of people in same-sex relationships.

References

Amirkhanian

,

Yuri A.

 

2014

. “

Social Networks, Sexual Networks and HIV Risk in Men Who Have Sex with Men

.”

Current HIV/AIDS Reports

 

11

(

1

):

81

92

. doi: .

Badgett

,

M.V. Lee

 

2003

.

Money, Myths, and Change: The Economic Lives of Women and men in same-sex relationships

.

Chicago and London: University of Chicago Press

.

Bearman

,

P.

, &

Parigi

,

P.

(

2004

). “

Cloning Headless Frogs and Other Important Matters: Conversation Topics and Network Structure

.”

Social Forces

,

83

(

2

),

535

57

. doi:

Bernstein

,

M. J.

(

2016

). Research in Social Psychology: Consequences of Short- and Long-Term Social Exclusion. In

P.

 

Riva

&

J.

 

Eck

(Eds.),

Social Exclusion (pp. 51–72)

.

Cham: Springer

. doi:

Chen

,

Yunsong

, and

Beate

 

Volker

.

2016

. “

Social Capital and Homophily both Matter for Labor Market Outcomes–Evidence from Replication and Extension

.”

Social Networks

 

45

(

March

):

18

31

. doi:

David Kasprowski, Mirjam M. Fischer, Xiao Chen, Lisa de Vries, Martin Kroh, Simon Kühne, David Richter, and Zaza Zindel.

2021

.

LGBTQI* People in Germany Face Staggering Health Disparities

.

DIW Weekly Report 5/2021: 42–50

.

Berlin: German Institute for Economic Research

. https://doi.org/10.18723/diw_dwr:2021-5-1.

Dewaele

,

Alexis

,

Nele

 

Cox

,

Wim

 

Van den Berghe

, and

John

 

Vincke

.

2011

. “

Families of Choice? Exploring the Supportive Networks of Lesbians, Men in Same-Sex Relationships, and Bisexuals

.”

Journal of Applied Social Psychology

 

41

(

2

):

312

31

.

Feld

,

Scott L.

 

1981

. “

The Focused Organization of Social Ties

.”

American Journal of Sociology

,

86

(

5

):

1015

35

. doi: .

Fischer, Mirjam M. and Kalmijn, Matthijs (

2020

).

Do Adult Men and Women in Same-Sex Relationships Have Weaker Ties to Their Parents?

 

Journal of Family Psychology

. https://doi.org/10.1037/fam0000696.

Fischer, Mirjam M., Kalmijn, Matthijs, and Steinmetz, Stephanie (

2017

).

Unions in Context (UNICON) study: Lesbian, gay and heterosexual couples and families in the Netherlands (Dataset)

. Amsterdam, the Netherlands: University of Amsterdam/DANS. https://doi.org/10.17026/dans-z5y-xa6w.

Freitas

,

D. F.

,

Coimbra

,

S.

, and

Fontaine

,

A. M.

(

2017

).

Resilience in LGB Youths: A Systematic Review of Protection Mechanisms

.

Paideia

,

27

(

66

):

69

79

. doi:

Frost

,

David M.

, and

Ilan H.

 

Meyer

.

2012

. “

Measuring Community Connectedness among Diverse Sexual Minority Populations

.”

Journal of Sex Research

 

49

(

1

):

36

49

. doi: .

Frost

,

David M.

,

Ilan H.

 

Meyer

, and

Sharon

 

Schwartz

.

2016

. “

Social Support Networks among Diverse Sexual Minority Populations

.”

American Journal of Orthopsychiatry

 

86

(

1

):

91

102

. doi: .

Frost

,

David M.

,

LeBlanc

,

Allen J.

,

de

 

Vries

,

Brian

,

Alston-Stepnitz

,

Eli

,

Stephenson

,

Rob

, and

Woodyatt

,

Cory

.

2017

. “

Couple-Level Minority Stress: An Examination of Same-Sex Couples’ Unique Experiences

.”

Journal of Health and Social Behavior

 

58

(

4

):

455

472

. doi:

Galupo

,

M. Paz

.

2007

. “

Women’s Close Friendships across Sexual Orientation: A Comparative Analysis of Lesbian-Heterosexual and Bisexual-Women in Different-Sex relationships’s Friendships

.”

Sex Roles

 

56

(

7–8

):

473

82

. doi:

Galupo

,

M. Paz

.

2009

. “

Cross-Category Friendship Patterns: Comparison of Heterosexual and Sexual Minority Adults

.”

Journal of Social and Personal Relationships

 

26

(

6–7

):

811

31

.

Ghavami

,

Negin

,

Adam

 

Fingerhut

,

Letitia A.

 

Peplau

,

Sheila K.

 

Grant

, and

Michele A.

 

Wittig

.

2011

. “

Testing a Model of Minority Identity Achievement, Identity Affirmation, and Psychological Well-Being among Ethnic Minority and Sexual Minority Individuals

.”

Cultural Diversity and Ethnic Minority Psychology

 

17

(

1

):

79

88

. doi: .

Goodrich, Kristopher M., Trahan, Don P., and Brammer, M. Kathryn. (

2019

).

Family Dynamics Following Disclosure as LGB: A Narratology

.

Family Journal

 

27

(

2

):

122

32

. https://doi.org/10.1177/1066480719832516.

Holloway

,

Ian W.

,

Craig A.

 

Pulsipher

,

Jeremy

 

Gibbs

,

Anamika

 

Barman-Adhikari

, and

Eric

 

Rice

.

2015

. “

Network Influences on the Sexual Risk Behaviors of Gay, Bisexual and Other Men Who Have Sex with Men Using Geosocial Networking Applications

.”

AIDS and Behavior

 

19

(

S2

):

112

22

. doi: .

Jenson

,

Jane

 

1998

.

Mapping Social Cohesion: The State of Canadian Research

.

Ottawa, CA

:

Canadian Policy Research Networks Inc. Renouf Publishing Co. Ltd.

Julien

,

Danielle

,

Elise

 

Chartrand

, and

Jean

 

Bégin

.

1999

. “

Social Networks, Structural Interdependence, and Conjugal Adjustment in Heterosexual, Gay, and Lesbian Couples

.”

Journal of Marriage and Family

 

61

(

2

):

516

30

. doi:

Kalmijn

,

Matthijs

.

1998

. “

Intermarriage and Homogamy: Causes, Patterns and Trends

.”

The Annual Review of Sociology

 

24

:

395

421

doi:  

1

.

Kalmijn

,

Matthijs

.

2002

. “

Sex Segregation of Friendship Networks. Individual and Structural Determinants of Having Cross-Sex Friends

.”

European Sociological Review

 

18

(

1

):

101

117

.

Kalmijn

,

Matthijs

.

2012

. “

Longitudinal Analyses of the Effects of Age, Marriage, and Parenthood on Social Contacts and Support

.”

Advances in Life Course Research

 

17

(

4

):

177

90

. doi: .

Kuyper

,

Lisette

 

2016

.

LGBT Monitor 2016: Opinions Towards and Experiences of Lesbian, Gay, Bisexual and Transgender People

, 2nd edn.

Den Haag

:

Sociaal en Cultureel Planbureau

.

LeBlanc

,

Allen J.

,

David M.

 

Frost

,

Eli

 

Alston-Stepnitz

,

Jose

 

Bauermeister

,

Rob

 

Stephenson

,

Cory R.

 

Woodyatt

, and

Brian

 

de

 

Vries

.

2015

. “

Similar Others in Same-Sex Couples’ Social Networks

.”

Journal of Homosexuality

 

62

(

11

):

1599

1610

. doi: .

Levitas

,

R.

,

Pantazis

,

C.

,

Fahmy

,

E.

,

Gordon

,

D.

,

Patsios

,

E. L.

, &

Patsios

,

D

. (

2007

).

The Multi-Dimensional Analysis of Social Exclusion

. In

London, Department for Communities and Local Government (DCLG), London

. http://www.cabinetoffice.gov.uk/social_exclusion_task_force/publications/multidimensional.aspx.

Marsden

,

Peter V.

 

1987

. “

Core Discussion Networks of Americans

.”

American Sociological Review

 

52

(

1

):

122

. doi: .

McConnell

,

Elizabeth A.

,

Janulis

,

Patrick

,

Phillips

,

Gregory

 II,

Truong

,

Roky

, and

Michelle A.

 

Birkett

.

2018

.

Multiple Minority Stress and LGBT Community Resilience among Sexual Minority Men

.

Psychology of Sexual Orientation and Gender Diversity

,

5

(

1

):

1

12

. doi: .

McConnell

,

Elizabeth A.

,

Michelle A.

 

Birkett

, and

Brian

 

Mustanski

.

2015

. “

Typologies of Social Support and Associations with Mental Health Outcomes among LGBT Youth

.”

LGBT Health

 

2

(

1

):

55

61

. doi: .

McPherson

,

Miller

,

Lynn

 

Smith-Lovin

and

Matthew E.

 

Brashears

 

2006

. “

Social Isolation in America: Changes in Core Discussion Networks over Two Decades

.”

American Sociological Review

 

71

(3): 353–75. doi: .

McPherson

,

Miller

,

Lynn

 

Smith-Lovin

, and

James M.

 

Cook

.

2001

. “

Birds of a Feather: Homophily in Social Networks

.”

Annual Review of Sociology

 

27

(

1

):

415

44

.

Meyer

,

Ilan H.

 

2003

. “

Prejudice, Social Stress, and Mental Health in Lesbian, Gay, and Bisexual Populations: Conceptual Issues and Research Evidence

.”

Psychological Bulletin

 

129

(

5

):

674

97

. doi: .

Mize

,

Trenton D

.

2016

. “

Sexual Orientation in the Labor Market

.”

American Sociological Review

 

81

(

6

):

1132

1160

.

Mollenhorst

,

Gerald

,

Beate

 

Volker

, and

Henk

 

Flap

.

2008

. “

Social Contexts and Core Discussion Networks: Using a Choice Approach to Study Similarity in Intimate Relationships

.”

Social Forces

 

86

(

3

):

937

65

.

Moran

,

Theresa E.

,

Cliff Yung-Chi

 

Chen

, and

Georgiana Shick

 

Tryon

.

2018

. “

Bully Victimization, Depression, and the Role of Protective Factors among College LGBTQ Students

.”

Journal of Community Psychology

. doi: .

Mustanski

,

B.

,

Newcomb

,

M. E.

, and

Garofalo

,

R.

(

2011

). “

Mental Health of Lesbian, Gay, and Bisexual Youths: A Developmental Resiliency Perspective

.”

Journal of Gay and Lesbian Social Services

,

23

(

2

),

204

25

. doi: .

Oswald

,

R. F.

(

2002

). “

Resilience within the Family Networks of Lesbians and Gay Men: Intentionality and Redefinition

.”

Journal of Marriage and Family

,

64

(

May

),

374

83

. doi: .

Pena-López

,

José Atilano

, and

José Manuel

 

Sánchez-Santos

.

2017

. “

Individual Social Capital: Accessibility and Mobilization of Resources Embedded in Social Networks

.”

Social Networks

 

49

(

May

):

1

11

. doi: .

Pichler

,

Florian

, and

Claire

 

Wallace

.

2009

. “

Social Capital and Social Class in Europe: The Role of Social Networks in Social Stratification

.”

European Sociological Review

 

25

(

3

):

319

32

. doi: .

Rostosky

,

Sharon Scales

, and

Ellen DB

 

Riggle

.

2017

. “

Same-Sex Relationships and Minority Stress

.”

Current Opinion in Psychology

 

13

(

February

):

29

38

. doi: .

Rözer

,

Jesper Jelle

,

Gerald

 

Mollenhorst

, and

Beate

 

Volker

.

2015

. “

Romantic Relationship Formation, Maintenance and Changes in Personal Networks

.”

Advances in Life Course Research

 

23

(

March

):

86

97

. doi:

Snapp

,

Shannon D.

,

Ryan J.

 

Watson

,

Stephen T.

 

Russell

,

Rafael M.

 

Diaz

, and

Caitlin

 

Ryan

.

2015

. “

Social Support Networks for LGBT Young Adults: Low Cost Strategies for Positive Adjustment

.”

Family Relations

 

64

(

3

):

420

30

. doi: .

Ueno

,

K.

,

E. R.

 

Wright

,

M. D.

 

Gayman

, and

J. M.

 

McCabe

.

2012

. “

Segregation in Gay, Lesbian and Bisexual Youth’s Personal Networks: Testing Structural Constraint, Choice Homophily and Compartmentalization Hypotheses

.”

Social Forces

 

90

(

3

):

971

91

. doi: .

Waite

,

Sean

, and

Nicole

 

Denier

.

2015

. “

Gay Pay for Straight Work: Mechanisms Generating Disadvantage

.”

Gender & Society

 

29

(

4

):

561

88

. doi: .

Weger

,

Harry Jr

.

2015

. Cross-Sex Friendships.

The International Encyclopedia of Interpersonal Communication

(Editors C.R. Berger, M.E. Roloff, S.R. Wilson, J.P. Dillard, J. Caughlin and D. Solomon). New York: John Wiley & Sons Inc. https://doi.org/10.1002/9781118540190.wbeic131.

Worthen

,

M.G.F.

 

2013

. “

An Argument for Separate Analyses of Attitudes toward Lesbian, Gay, Bisexual Men, Bisexual Women, MtF and FtM Transgender Individuals

.”

Sex Roles

 

68

:

703–23

. doi: .

Wrzus

,

Cornelia

,

Martha

 

Hänel

,

Jenny

 

Wagner

, and

Franz J.

 

Neyer

.

2013

. “

Social Network Changes and Life Events across the Life Span: A Meta-Analysis

.”

Psychological Bulletin

 

139

(

1

):

53

80

. doi: .

Author notes

© The Author(s) 2021. Published by Oxford University Press on behalf of the University of North Carolina at Chapel Hill. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.



Source link

xandoblogs

An open minded personality.. fun to be with, because of my positive vibes. God fearing, for without God I am nothing.. Moved with compassion when dealing with you, not selfish or self-centered...

Leave a Reply

Your email address will not be published. Required fields are marked *

error

Enjoy this blog? Please spread the word :)