10 pages in 60 hours no plagiarism
kettyg555
socialrelationship.pdf
Personal Relationships, 24 (2017), 350 – 369. Printed in the United States of America. Copyright © 2017 IARR; DOI: 10.1111/pere.12186
Intimate partner support: A comparison of gay, lesbian, and heterosexual relationships
LILLIAN ELLISa AND MARK DAVISb
aEastern Michigan University and bEckerd College
Abstract In this investigation, intimate partner support, relationship satisfaction, and separation proneness were assessed for four types of people: men in a relationship with a woman (MRW), men in a relationship with a man (MRM), women in a relationship with a man (WRM), and women in a relationship with a woman (WRW). Men and women in same-sex relationships received more support, were more satisfied, and reported fewer thoughts of separating than their counterparts in opposite-sex relationships. The effect of relationship type on satisfaction was not significant once the amount of received support was controlled. The implications of these findings for understanding the support process in same-sex relationships are discussed.
Are there gender differences in the provision or receipt of support within intimate relation- ships? One might expect, based on the nature of traditional gender roles in Western society, that there would be reliable differences in the social support offered by men and women. Given their cultural role as communal and other ori- ented (Bakan, 1966; Parsons & Bales, 1956), women on average might be especially likely to exhibit behaviors that reflect concern for their intimate partners, such as the offering of social support. Similarly, consistent with their cul- tural role as agentic and task oriented (Bakan, 1966; Parsons & Bales, 1956), men on average might be generally less likely to offer support to their partners. Indeed, in familial nonroman- tic relationships, women do provide more sup- port than men (Monin & Clark, 2011) and tend to have larger support networks (Sarason, Sara- son, Hacker, & Basham, 1985). Of course, this is not to say that all men are the same, nor that all women are. There is likely more variance within genders than between genders (Canary & Hause, 1993). However, the effect of gender
Lillian Ellis, Department of Psychology, Eastern Michigan University; Mark Davis, Department of Psychology, Eck- erd College.
Correspondence should be addressed to Lillian Ellis, 341 Science Complex, Eastern Michigan University, Ypsi- lanti, MI 48197, e-mail: [email protected].
expectations on social behavior is not trivial, and thus, predicting differences in such behav- ior for men and women is not unreasonable (Wharton & Erickson, 1993).
Gender differences in social support
Some evidence for such a gender-based pat- tern can be found in the early investigations of what has been termed the “support gap hypothesis” (Belle, 1982). A number of these investigations found that men report receiv- ing more support from their romantic partners than do women (e.g., Antonucci & Akiyama, 1987; Huston-Hoburg & Strange, 1986). Using a somewhat different methodology, Brock and Lawrence (2009) also found evidence consis- tent with such a gap. They had respondents indicate, for 48 different support behaviors, whether the amount of support received from their partner was more than they preferred, less than they preferred, or the right amount. Wives were significantly more likely to report being “underprovided” than were their husbands.
However, other research has suggested a pattern that is far less clear. Social support takes different forms, and some of these are more likely than others to produce gender differences. For example, emotional support (offering encouragement and reassurance) is
350
Intimate partner support 351
a form of support that seems to fall especially clearly within the female gender stereotype and, as a result, may be especially likely to show a gender difference. On the other hand, gender differences might be less reliable for another type of aid — instrumental support (providing advice and guidance) — because of its relevance to an agentic orientation toward task completion. Indeed, forms of support other than emotional support often fail to sup- port the “support gap” pattern. On measures of instrumental support (offering practical help, guidance, and material assistance), men have sometimes been found to provide more than women (e.g., Verhofstadt, Buysse, & Ickes, 2007); in other investigations, men and women do not differ at all (e.g., Cutrona & Suhr, 1994; Goldsmith & Dun, 1997; Pasch, Bradbury, & Davila, 1997).
In addition, Xu and Burleson (2001) iden- tified a separate issue that complicates the “support gap” question. They asked recently married individuals to report on the degree of social support received from their partners but also asked them to report on the degree of sup- port they desired. These judgments were made separately with regard to five different types of support: emotional, tangible, informational, esteem (providing respect and validation), and network (providing feelings of social connection). When participants were asked how much support they received, virtually no gender differences consistent with the support gap hypothesis were found for any form of support; however, women reported desiring higher levels of support for all five types. Thus, the Xu and Burleson findings suggest that the support “gap” might not reflect differences between men and women in the amount of support actually received but may instead be the result of differences in the amount of support that men and women prefer. The Brock and Lawrence (2009) findings are also consistent with this interpretation.
Finally, one additional complication in the issue of gender differences lies in the fact that in the research conducted to date, it is impos- sible to disentangle the possible effects of the support provider’s gender from the effects of the support recipient’s gender. That is, if women on average receive less support than
men, it could be something about men that make them less likely to offer support to their intimate partners, or it could be something about women that invites lower levels of sup- port. In heterosexual couples, such a con- founding is inescapable. The problem might be overcome, however, if same-sex as well as opposite-sex couples were studied.
Social support in same-sex relationships
To date, however, the vast majority of the research examining social support within inti- mate relationships has been carried out with opposite-sex relationships. When social sup- port has been studied in gay and lesbian pop- ulations, the focus has often been on other issues, such as the ways in which such sup- port helps buffer the stresses and strains of living with a sometimes stigmatized identity (e.g., Elizur & Mintzer, 2003; Kurdek, 1988), the relative importance of support from family versus friends for heterosexual and same-sex couples (e.g., Graham & Barnow, 2013), or the role of social support in coping with impor- tant health concerns such as HIV/AIDS (e.g., Darbes & Lewis, 2005). In such research, the focus is often on the aid provided by friendship networks. Relatively little research has specifi- cally focused on the role of social support pro- vided by the relationship partner.
Nevertheless, some investigations have addressed this issue, although the question of gender differences in partner support now takes a different form as both partners are of the same gender. Thus, the typical approach has been to directly compare gay men (or gay couples) with lesbians (or lesbian cou- ples). Investigations making such comparisons are relatively few in number and are further complicated by the fact that it is not entirely clear what kind of pattern to expect. If the support gap in heterosexual relationships is the result of men and women being socialized into distinct cultural roles, then men and women who tend to reject such traditional gender roles (e.g., Bell, Weinberg, & Hammersmith, 1981; Kurdek, 1987) might not differ much from one another. On the other hand, it might be hypoth- esized that in same-sex relationships, the couple has a “double dose” of the relevant cul- tural conditioning and that this may produce
352 L. Ellis and M. Davis
distinct patterns of social support. Lesbian relationships, for example, seem to be more characterized than gay relationships by recip- rocal expressiveness and a concern for equality of power and shared decision making (Kurdek, 1988; Kurdek & Schmitt, 1986), and it might be expected that at least some types of support may be higher as a result.
Evidence thus far favors the former pos- sibility, with little difference between gay and lesbian couples. Although Kurdek (1988) found those in lesbian relationships to report greater satisfaction, other studies have found no differences between gay and lesbian cou- ples in terms of either relationship satisfaction (Graham & Barnow, 2013; Julien, Chartrand, Simard, Bouthillier, & Bégin, 2003; Kurdek, 1988, 1991) or in partner support (Graham & Barnow, 2013; Julien et al., 2003). Graham and Barnow (2013) also examined the asso- ciation between support and satisfaction and found no difference between gay and lesbian respondents.
Direct comparisons of same-sex and opposite-sex relationships
If there is little research comparing gay and lesbian couples, there is even less that directly compares relationship support within same-sex and opposite-sex intimate relationships. However, the scant evidence that does exist suggests few differences (Kurdek, 2004). Kur- dek (2001) compared gay and lesbian couples to two kinds of heterosexual couples — those with children and those without. Using the heterosexual nonparents as the most relevant comparison, Kurdek (2001) found lesbians to report greater satisfaction; gay men did not differ from this heterosexual comparison group. Julien and colleagues (2003) found no differences between gay, lesbian, and hetero- sexual couples in the amount of social support offered in a lab setting. Graham and Barnow (2013) reported no differences between gay, lesbian, and heterosexual couples with regard to support from partners, satisfaction, or the support – satisfaction association.
The current study
Given the well-documented importance of social support from intimate partners (e.g.,
Acitelli & Antonucci, 1994; Jensen, Rauer, & Volling, 2013) and the striking dearth of evidence regarding this topic in same-sex relationships, it seemed to us important to investigate this issue and to do so in as com- prehensive a fashion as possible. Why so? In addition to the general, but very real, value of studying a topic that has been largely ignored to this point, we feel that there is a specific value in expanding our understanding in this particular case. Studies of relationship processes have tended to focus almost exclu- sively on heterosexual relationships and only rarely have directly compared same-sex and opposite-sex relationships in the same investi- gation. One result of this is that in the absence of any data, there is an implicit assumption that the social support processes operate in the same way in nonheterosexual relationships. This may be true, but it may not. To the degree that it is not accurate, our current level of understanding is insufficient, offering a view of social support that is too simple and that may not accurately reflect the more nuanced ways that relationship partners of all types may seek, provide, and receive support.
Thus, participants in this investigation were asked to provide information about several types of support and to do so in three ways. For each type of support, participants reported on the support they received from their partners, the support they desired from the partners, and the support they provided to the partner. As evidenced by the Xu and Burleson (2001) find- ings, it is important to distinguish among these different conceptualizations of support. To our knowledge, no research to date has used such a comprehensive approach when comparing those in same-sex and opposite-sex intimate relationships. We also included measures of two relationship outcomes — satisfaction and separation proneness — making it possible to determine whether such outcomes vary by relationship type.
To examine these questions, we gathered information from four types of individuals: (a) men who are in a relationship with another man (MRM), (b) women who are in a relationship with another woman (WRW), (c) men who are in a relationship with a woman (MRW), and (d) women who are in a relationship with a man
Intimate partner support 353
(WRM). We identified these four types of indi- viduals by asking participants to indicate their own gender (male or female) and that of their partners. (Thus, we asked explicitly about gen- der identity rather than biological sex or gender expression.) This approach therefore allows an examination of the separate and combined effects of participant gender and partner gen- der. The primary analytical strategy employed in this investigation was to examine social support and relationship outcomes through 2 (participant gender) × 2 (partner gender) mul- tivariate analyses of variance (MANOVAs). We can envision three broad types of findings, each of which might be expected on the basis of at least some previous research.
Gender of the participant
One possibility is that there will be main effects of participant gender on social sup- port, especially for emotional support. Prior research indicating a “support gap” suggests that women will, on average, report receiv- ing lower levels of support (especially emo- tional support) than men, and it might also be expected that women, on average, would report higher levels of desired support (Xu & Burleson, 2001). However, as noted previously, the evidence on this point is mixed.
Gender of the partner
Another possibility is that there will be main effects of partner gender on support, espe- cially for emotional support. Based on prior research with heterosexual relationships that has found a support gap (women receiving less support), it might be argued that people with male partners, on average, will report receiving lower levels of support, regardless of their own gender. Such an expectation would be consistent with the general view of women as “relationship experts” (Kurdek, 1991) who on average take a larger responsibility for the monitoring and maintenance of relationship functioning, with men on average less likely to do so (e.g., Ragsdale, 1996).
Nature of the relationship
Finally, it is possible that regardless of any main effects, there may be an interaction of participant gender and partner gender. Specifi- cally, it is hypothesized that those in same-sex
relationships will report different levels of support than those in opposite-sex relation- ships. On one hand, predicting such interaction effects may be viewed as somewhat speculative because prior evidence suggests that those in same-sex and opposite-sex romantic relation- ships typically do not differ in some important ways (e.g., Kurdek, 2004). On the other hand, there is some theoretical reason to suppose that the provision and receipt of social support might unfold at least somewhat differently in same-sex relationships than heterosexual ones. One argument for such a supposition is that those in same-sex relationships — by virtue of having partners of the same gender — will on average be more likely to desire the same type and amount of social support. That is, MRM and WRW are likely to have had his- tories of gender socialization that are more similar to those of their partners; MRW and WRM will have gender socialization histories that are more distinct. This difference may make it somewhat easier for those in same-sex relationships to give and receive the desired amounts and types of support, which “optimal matching” theory (e.g., Cutrona, Shaffer, Wes- ner, & Gardner, 2007) suggests will be the most effective. In addition, beyond the rather broad similarities in gender socialization to which men and women are subjected, there is some evidence that gendered expectations of tradi- tional roles also differ in light of sexual orien- tation (Lehavot & Lambert, 2007), especially for men (Blashill & Powlishta, 2009). Thus, both members of same-sex relationships may also share these more nuanced gender expecta- tions, which might act to increase the similarity of (and thus the effectiveness of) the support given and received in same-sex relationships.
Finally, one additional argument specif- ically identifies lesbian relationships as possibly possessing an advantage over other types. Kurdek (2001) describes the “double dose” hypothesis: that because lesbian relationships consist of two women, they enjoy “the salubrious effects of a double dose of the relationship- enhancing influences associated with feminine socialization” (p. 728). On the basis of this argument then, we might expect greater sup- port and satisfaction in lesbian relationships,
354 L. Ellis and M. Davis
even if gay relationships do not display this pattern. For example, Roisman, Clausell, Holland, Fortuna, and Elieff (2008) found les- bian couples to be more effective at working together harmoniously in a laboratory setting than gay or heterosexual couples.
Method
Participants and procedure
Participants in this study were individuals in same-sex and opposite-sex intimate rela- tionships and were recruited using a number of different methods. First, 21 lesbian, gay, bisexual (LGB)-related Facebook groups and 91 LGB-related Facebook pages were con- tacted and asked to share the survey with their members; many of these pages also included straight allies. An explanation and link to the survey was posted on the wall of all the Facebook groups contacted. Of the 91 pages, 80 allowed the survey to be posted on their walls, and 13 shared it with their followers. Second, over 450 leaders of organizations who meet in person were contacted and asked to share the survey with their members and to fill it out themselves. Many of these groups were made up of gay and lesbian individuals, but some included or targeted heterosexuals. For example, the 371 chapters of Parents and Fam- ilies of Lesbians and Gays (PFLAG) across the United States were contacted. In an effort to include organizations with more heterosexual men, we also contacted group leaders of Young Men’s Christian Associations (YMCAs), out- door clubs, and so on. In addition, students in Eckerd College’s adult education program were contacted via e-mail and invited to par- ticipate. Finally, three researchers attended the 2013 St. Petersburg Pride Parade and recruited people in gay and lesbian relationships who were willing to participate in the study. In the process of contacting groups and organiza- tions, some special effort was also put into contacting groups with diverse racial member- ship. Everyone contacted was also encouraged to share the survey with friends and family. All participants, no matter how they were recruited, were given a link and completed the questionnaire online using Google Docs Forms, a cloud-based survey software. No compensation was offered, and in order to
participate, the person had to currently be in an intimate relationship, with no requirement on length of relationship.
This recruitment process was successful, with 637 individuals eventually participating. When asked to indicate their gender (we did not ask about biological sex), 273 identified themselves as men, 351 as women, and 13 offered some other gender description, such as genderqueer, bigender, or transgender. The 13 people who identified as something other than a binary gender (i.e., male or female) were not included in analysis. Thus, unless otherwise stated, all analyses are based on the 624 remaining participants. Importantly, there were sizable numbers of participants in the four categories we were interested in comparing: (a) MRW (n = 182), (b) MRM (n = 91), (c) WRM (n = 231), and (d) WRW (n = 120).
The age range of participants was 16 – 84 years, with a mean of 38.03 (SD = 15.83). Despite our efforts to recruit a sample with racial diversity, 85% of the participants were White, 4% identified as Hispanic, 3% as Asian or Pacific Islander, 2% as Black, and 2% as Native American; 4% preferred not to answer. Most of the sample was from the United States (84%) and Canada (6%). The sample was highly educated, with 35% having a bachelor’s degree, and 30% having further postgraduate higher education. Thus, approximately two thirds of the sample had a college degree or higher. In addition, 13% reported having an associate’s degree or trade school, and 23% reported graduating from high school or sec- ondary school or getting a General Education Diploma (GED). An additional 1% preferred not to answer. Roughly one fourth of those in same-sex relationships were married (25.7%), as were almost half of those in opposite-sex relationships (48.3%). However, about the same percentage of same-sex and opposite-sex couples lived together (77.0% and 70.5%, respectively).
Measures
The survey consisted of four sections, each assessing a different set of variables: social support, trait empathy, relationship outcomes, and demographic information. (The trait
Intimate partner support 355
empathy findings are beyond the scope of this article and are not described further.)
Social support
The first and longest part of the survey was made up of items measuring social support within the participant’s intimate relationship. Within this section, there were questions mea- suring social support of three different forms: support received from the partner in the rela- tionship, support that the individual desires from the partner, and the support that the indi- vidual provides to the partner. To measure support, a modified version of the Support in Intimate Relationships Rating Scale was used (SIRRS; Dehle, Larsen, & Landers, 2001). The modification (SIRRS:EFA) was based on an exploratory factor analysis that was carried out by Barry, Bunde, Brock, and Lawrence (2009), which found that support fell into four related but unique domains: emotional/esteem sup- port, physical comfort, tangible support, and informational support.
Emotional/esteem support includes emo- tional validation and showing confidence in your partner (seven items, sample item: “Said he/she thought I handled a situation well”). Physical comfort is the physical demonstration of affection and support (three items, sample item: “Hugged me or cuddled with me”). None of the physical comfort questions asked about sexual activity. Informational support is the sharing of advice or insight into a situ- ation (eight items, sample item: “Helped me think about a situation in a new way”). Like informational support, tangible support is also solution-focused, but instead, it includes taking action, sharing resources, or at least offering to do so (four items, sample item: “Did something to help me directly”). These four types of support generally map onto the typical distinction between emotional support (em- otional/esteem) and instrumental support (information; tangible) that is typical of much social support research. The physical com- fort measure does not fit quite as neatly into this pattern, although it seems intuitively likely to reflect emotional support more than instrumental.
All participants completed three slightly different versions of the SIRRS:EFA. The first
version asked them about the amount of sup- port received from their partners, the second asked about the amount of support desired, and the third asked about the amount of sup- port they provided to their partners. All ver- sions of the SIRRS:EFA asked respondents to answer in terms of frequency of support: how often they or their partner had provided that type of support in the last month on a scale from 0 (never) to 5 (always). For each form of support (received, desired, and provided), four social support domain scores were computed by averaging all the individual items for each domain of support: emotional/esteem support, physical comfort support, tangible support, and informational support. Internal reliabilities of these indices (Cronbach’s α) ranged from .84 to .93.
Relationship outcome variables
Several items were included to measure two relationship outcomes: satisfaction and sep- aration proneness. Relationship satisfaction was measured using the four-question ver- sion of the Couples Satisfaction Index (CSI; Funk & Rogge, 2007). The four items were “Please indicate the degree of happiness, all things considered, of your relationship” (from extremely unhappy to perfect), “I have a warm and comfortable relationship with my part- ner” (from not at all true to completely true), “How rewarding is your relationship with your partner?” (from not at all to completely), and “In general, how satisfied are you with your relationship?” (from not at all to completely). Each item was answered on a 6-point scale; Cronbach’s α for the scale was .89. Separation proneness was measured using two items that have been used with gay and lesbian popula- tions (Kurdek, 2006). Participants rated how often in the last 6 months they had seriously considered ending the relationship and how often they had seriously discussed ending the relationship. For both items, the response scale was from one (never) to four (more than three times); Cronbach’s α for the index was .79. The two outcome measures were strongly correlated at −.54 (p <. 001).
356 L. Ellis and M. Davis
Demographic information
Finally, participants responded to a number of demographic items, including race, age, educa- tion, gender, gender of partner, self-identified sexual orientation, whether or not the partici- pant and his or her partner are cohabitating or long distance, and whether the participant and his or her partner are married. To measure sex- ual orientation, the Kinsey attraction subscale was used, which is a single question, “What gender are you attracted to?” and the rating is from 0 = entirely same gender to 6 = entirely opposite gender with 3 = evenly both same and opposite gender (Kinsey, Pomeroy, & Martin, 1948).
Results
Descriptive statistics
Table 1 displays, separately for each of the four relationship categories, the descriptive infor- mation for all of the major variables considered in this investigation. Table 2 displays the corre- lations among these variables, and three points are worth mentioning. First, there are small but numerous correlations between the four back- ground variables (age, education, marital sta- tus, and living arrangement) and the support and outcome indices; thus, we used those four variables as covariates in all of the remain- ing analyses. Second, the correlations among the four types of support within each domain (received, desired, provided) were substantial, averaging .59. Correlations across domains were roughly half that size, although the corre- lations between received and provided support were somewhat higher. Third, and unsurpris- ingly, received support displayed the strongest correlations with outcome measures, and pro- vided support displayed a similar but weaker pattern. Desired support had few relationships with either outcome variable.
Preliminary analysis
Prior to examining the differences in mean levels of support and relationship quality across relationship types, we first examined the correlations between support and relation- ship outcome for each type. Table 3 displays
the partial correlations between the outcome variables (satisfaction; separation proneness) and social support for each relationship type; in all cases, the four background variables are controlled. For all four relationship cate- gories, each domain of received support was significantly and positively correlated with relationship satisfaction, and inversely corre- lated with separation proneness. With regard to support provided, there were significant correlations between all domains and relation- ship satisfaction for all relationship categories except MRM’s provision of informational support. Separation proneness was more spo- radically associated with support provision. Finally, in contrast to the other two natures of support, desired support was almost com- pletely unrelated to either outcome variable.
Group differences in social support
To examine the effect of relationship type on social support, three separate MANCOVAs were conducted, one each for support received, support provided, and support desired; all four domains of support were included in each MANCOVA. When significant, univariate analyses of variance (ANOVAs) were then con- ducted, and follow-up t tests were used when necessary. In each analysis, we also controlled for four variables that might influence social support: age, education, marital status (married or not), and living arrangement (cohabiting, noncohabiting, long distance).
Support received
To examine the effect of relationship type on levels of received support, a 2 (participant gen- der) × 2 (partner gender) MANCOVA on the four domains of received social support was conducted, with age, education, marital status, and living arrangement as covariates. Age and living arrangement exerted significant multi- variate effects (both Fs > 4.70, both ps < .01). Neither participant gender nor partner gender displayed a main effect, but there was a sig- nificant interaction, F(4, 606) = 7.24, p < .001, Wilks’s Λ = 0.95, partial η2 = .05. Given the significance of the multivariate analysis, sep- arate 2 × 2 ANOVAs were carried out for each of the four support domains (see Figure 1).
Intimate partner support 357
Table 1. Descriptive statistics for major variables as a function of relationship type
MRW MRM WRM WRW
Married 51.1% 23.1% 33.3% 24.2% Living arrangement
Cohabiting 63.7% 63.7% 55.4% 69.2% Noncohabiting 22.0% 25.3% 26.8% 15.8% Long distance 14.3% 11.0% 17.8% 15.0%
M SD M SD M SD M SD
Age 43.97 15.11 48.52 16.97 31.82 13.83 33.01 11.76 Education 2.93 1.20 3.11 1.36 2.62 1.19 2.57 1.24 Received support
Information 3.01 0.98 3.33 0.96 3.20 1.08 3.71 0.85 Physical 3.62 1.49 4.08 1.13 4.02 1.39 4.46 1.01 Emotional 3.27 1.13 3.43 1.18 3.53 1.20 4.07 0.81 Tangible 3.08 1.27 3.42 1.25 3.36 1.32 3.89 1.13
Desired support Information 3.29 0.85 3.53 0.92 3.45 1.01 3.66 1.03 Physical 4.21 1.01 4.34 1.01 4.33 1.12 4.55 1.06 Emotional 3.73 0.91 3.90 0.82 4.06 0.98 4.19 0.99 Tangible 3.49 1.06 3.84 0.95 3.93 1.18 4.09 1.07
Provided support Information 3.57 0.89 3.75 0.70 3.72 0.80 3.97 0.78 Physical 3.84 1.36 4.13 1.16 4.12 1.27 4.57 0.90 Emotional 3.74 0.92 3.85 0.89 3.87 0.91 4.34 0.62 Tangible 3.64 0.99 3.96 0.86 3.94 0.94 4.21 0.77
Satisfaction 4.54 1.14 4.77 0.99 4.85 0.95 5.15 0.89 Separation-proneness 0.78 0.99 0.42 0.74 0.63 0.86 0.45 0.69
Note. Education is coded on a 5-point continuum running from high school/General Education Diploma (GED) (1) to doctorate/professional degree (5). Marital status is coded as not married (1) or married (2). Living arrangement is coded on a 3-point continuum: long-distance relationship (1), same city/not living together (2), living together (3). MRW = men in relationship with a woman; MRM = men in relationship with a man; WRM = women in a relationship with a man; WRW = women in a relationship with a woman.
The pattern was largely the same for each variable. There was never a main effect for partner gender, but participant gender displayed such an effect for emotional and tan- gible support (both Fs > 5.25, both ps < .05), with women receiving more support. In addi- tion, all domains showed a strong interaction effect (all Fs > 13.45, all ps < .001). The level of support that both men and women received from a same-sex partner was greater than that received from an opposite-sex partner; this was true for almost every domain (all ts > 2.04, all ps < .05). The sole exception was
the comparison between MRM and MRW on emotional support, which was not significant.
Support desired
An identical 2 (participant gender) × 2 (partner gender) MANCOVA was conducted on the four domains of desired support. Age had a multivariate effect, F(4, 606) = 2.77, p < .05. The main effect for participant gender was also significant, F(4, 606) = 5.04, p < .001, Wilks’s Λ = 0.97, partial η2 = .032. Separate 2 × 2 ANOVAs were then carried out for each of the four support domains (see Figure 2). The main effect of participant gender was
358 L. Ellis and M. Davis
T ab
le 2.
C o rr
el a ti
o n s
a m
o n g
b a ck
g ro
u n d
va ri
a b le
s, su
p p o rt
m ea
su re
s, a n d
o u tc
o m
e m
ea su
re s
2 3
4 5
6 7
8 9
10 11
12 13
14 15
16 17
18
1. A
ge .4
6* **
.3 8*
** .2
7* **
− .1
4* **
− .1
7* **
− .2
3* **
− .0
9* − .1
0* **
− .1
5* **
− .1
4* **
− .0
6 − .1
9* **
− .1
9* **
− .1
9* **
− .0
8* − .1
3* **
− .0
8 2.
E du
ca ti
on .2
2* **
.1 7*
** − .0
2 − .0
8* − .1
1* **
− .0
5 − .0
3 − .0
7 − .0
5 − .0
2 − .1
1* **
− .1
2* **
− .1
1* **
− .1
0* − .0
4 − .1
0* **
3. M
ar it
al st
at us
.4 7*
** − . 1
0* **
− .1
6* **
− .1
3* **
− .0
4 − .1
2* **
− .1
6* **
− .1
0* − .0
7 − .1
1* **
− .1
8* **
− .1
4* **
− .1
0* − .0
1 − .1
7* **
4. L
iv in
g ar
ra ng
em en
t − .0
4 − .0
1 − .0
9* .0
6 − .0
7 − .1
0* − .0
2 .0
4 − .0
7 − .0
1 − .1
4* **
.0 3
.0 5
− .1
7* **
R ec
ei ve
d su
p p o rt
5. In
fo rm
at io
na l
.4 4*
** .6
9* **
.6 7*
** .4
8* **
.1 5*
** .2
5* **
.2 2*
** .4
9* **
.3 6*
** .4
8* **
. 4 5*
** .4
7* **
− .2
9* **
6. P
hy si
ca l
.5 9*
** .6
0* **
.2 5*
** .3
6* **
.1 8*
** .2
4* **
.3 3*
** .7
6* **
.4 4*
** .4
5* **
.5 3*
** − .3
2* **
7. E
m ot
io na
l .7
5* **
.3 6*
** .1
8* **
.2 5*
** .1
9* **
.4 6*
** .4
5* **
.6 3*
** .4
3* **
.5 6*
** − .3
7* **
8. T
an gi
bl e
.3 8*
** .2
1* **
.2 4*
** .3
3* **
.4 8*
** .4
4* **
.4 9*
** .5
0* **
.5 4*
** − .3
9* **
D es
ir ed
su p p o rt
9. In
fo rm
at io
na l
.4 8*
** .6
5* **
.6 2*
** .4
4* **
.2 1*
** .3
4* **
.2 7*
** .1
6* **
− .0
8* 10
.P hy
si ca
l .5
8* **
.5 5*
** .2
4* **
.4 8*
** .3
1* **
.2 2*
** .1
7* **
− .0
5 11
.E m
ot io
na l
.7 0*
** .2
9* **
.2 2*
** .3
9* **
.2 2*
** .1
0* .0
1 12
.T an
gi bl
e .3
0* **
.2 1*
** .2
5* **
.3 4*
** .0
8 − .0
4 P
ro vi
d ed
su p p o rt
13 .I
nf or
m at
io na
l .3
7* **
.5 9*
** .6
0* **
.2 2*
** − .0
3 14
.P hy
si ca
l .5
5* **
.5 2*
** .4
2* **
− .2
0* **
15 .E
m ot
io na
l .5
8* **
.4 1*
** − .2
3* **
16 .T
an gi
bl e
.3 3*
** − .1
6* **
O u tc
o m
es 17
.S at
is fa
ct io
n − .5
5* **
18 .S
ep ar
at io
n pr
on en
es s
*p <
.0 5.
** *p
< .0
01 .
Intimate partner support 359
T ab
le 3.
P a rt
ia l
co rr
el a ti
o n s
b et
w ee
n so
ci a l
su p p o rt
a n d
re la
ti o n sh
ip o u tc
o m
es a s
a fu
n ct
io n
o f
re la
ti o n sh
ip ty
p e
M R
W M
R M
W R
M W
R W
R el
at io
ns hi
p sa
ti sf
ac ti
on S
ep ar
at io
n pr
on en
es s
R el
at io
ns hi
p sa
ti sf
ac ti
on S
ep ar
at io
n pr
on en
es s
R el
at io
ns hi
p sa
ti sf
ac ti
on S
ep ar
at io
n pr
on en
es s
R el
at io
ns hi
p sa
ti sf
ac ti
on S
ep ar
at io
n pr
on en
es s
S u p p o rt
re ce
iv ed
In fo
rm at
io na
l .4
4* **
− .2
6* **
.4 1*
** − .2
4* .4
0* **
− .1
4* .6
0* **
− .2
9* **
P hy
si ca
l .5
8* **
− .3
4* **
.5 5*
** − .3
5* **
.4 4*
** − .1
2* **
.5 0*
** − .1
5 E
m ot
io na
l .5
6* **
− .3
6* **
.5 6*
** − .3
9* **
.5 2*
** − .2
5* **
.5 8*
** − .2
4* T
an gi
bl e
.5 0*
** − .3
3* **
.5 1*
** − .3
3* **
.5 0*
** − .2
5* **
.6 5*
** − .3
6* **
S u p p o rt
p ro
vi d ed
In fo
rm at
io na
l .1
5* .0
6 .0
8 − .0
5 .1
8* .0
1 .3
8* **
− .2
3* P
hy si
ca l
.4 3*
** − .1
8* .4
8* **
− .3
9* **
.3 3*
** − .0
3 .4
6* **
− .1
1 E
m ot
io na
l .3
3* **
− .1
3 .4
9* **
− .3
0* .4
2* **
− .2
4* **
.4 1*
** − .3
1* **
T an
gi bl
e .2
9* **
− .0
8 .2
8* − .1
0 .2
5* **
− .0
7 .4
5* **
− .3
2* **
S u p p o rt
d es
ir ed
In fo
rm at
io na
l .1
6* − .0
6 .0
5 − .0
7 .1
2 − .0
2 .2
2* .0
0 P
hy si
ca l
.2 1*
− .1
0 .1
8 − .1
2 .0
9 .0
2 .1
9* .0
1 E
m ot
io na
l .0
7 .0
0 .1
3 .0
9 .0
3 .0
6 .0
7 .0
5 T
an gi
bl e
.0 6
.0 5
− .0
1 − .0
2 .0
0 .0
2 .1
3 − .0
4
N o
te . M
R W
= m
en in
re la
ti on
sh ip
w it
h a
w om
an ;
M R
M =
m en
in re
la ti
on sh
ip w
it h
a m
an ;
W R
M =
w om
en in
a re
la ti
on sh
ip w
it h
a m
an ;
W R
W =
w om
en in
a re
la ti
on sh
ip w
it h
a w
om an
. *p
< .0
5. **
*p <
.0 01
.
360 L. Ellis and M. Davis
2.5
3
3.5
4
4.5
Informational Support
S u
p p
o rt
S co
re
Support Tangible Support Emotional Support Physical Comfort
Support Received
MRW MRM WRM WRW
Figure 1. Support received from partner as a function of participant gender and partner gen- der. MRW = men in relationship with a woman; MRM = men in relationship with a man; WRM = women in a relationship with a man; WRW = women in a relationship with a woman.
significant only for emotional and tangible support (both Fs > 7.50, both ps < .01). In each case, women desired more support than men. Also, significant interaction effects were found for informational support and tangible support (both Fs > 3.97, both ps < .05). The interactions resulted from a pattern similar to that found for received support. For men, the level of informational and tangible support desired from same-sex partners was greater than that desired from opposite-sex partners (both ts > 2.10, both ps < .05); for women, the same pattern was present but was not significant.
Support provided
An identical 2 (participant gender) × 2 (part- ner gender) MANCOVA was conducted to examine the effect of relationship type on support provided. Age, marital status, and living arrangement had significant multivariate effects (all Fs > 3.23, all ps < .05). There was a significant main effect of participant gender, F(4, 606) = 2.90, p < .05, Wilks’s Λ = 0.98, partial η2 = .019, and one for partner gender, F(4, 606) = 2.54, p < .05, Wilks’s Λ = 0.98, partial η2 = .016. The interaction effect was also significant, F(4, 606) = 5.40, p < .001, Wilks’s Λ = 0.97, partial η2 = .034. Separate 2 × 2 ANOVAs were conducted for each of
the four support domains (see Figure 3). Emotional and tangible support had signifi- cant main effects of participant gender (both Fs > 5.80, both ps < .05), with women pro- viding more support. The only significant main effect of partner gender was found for emotional support, F(1, 609) = 5.08, p < .05. In addition, the Participant Gender × Partner Gender interaction was significant for all four domains of support provided (all Fs > 7.90, all ps < .001). Similar to support received, for men, the level of support provided to a same-sex partner was at least marginally significantly greater than that provided to an opposite-sex partner, except emotional sup- port (all ts > 1. 74, all ps < .08). For women, support provided to a same-sex partner was significantly greater than that provided to an opposite-sex partner for all four domains (all ts > 2.68, all ps < .01).
Group differences in outcome variables
An identical 2 (Participant Gender) × 2 (Part- ner Gender) MANCOVA was carried out on the two outcome variables. Age and mari- tal status had significant multivariate effects (both Fs > 5.60, both ps < .01). There was a significant main effect of participant gender, F(2, 594) = 3.41, p < .05, Wilks’s Λ = 0.99, partial η2 = .011, and a significant interaction
Intimate partner support 361
3
3.2
3.4
3.6
3.8
4
4.2
4.4
4.6
4.8
Informational Physical
Support Desired
Tangible Emotional
MRW MRM WRM WRW
Figure 2. Support desired from partner as a function of participant gender and partner gen- der. MRW = men in relationship with a woman; MRM = men in relationship with a man; WRM = women in a relationship with a man; WRW = women in a relationship with a woman.
3
3.2
3.4
3.6
3.8
4
4.2
4.4
4.6
4.8
Informational PhysicalTangible Emotional
Support Provided
Support Domain
MRW WRMMRM WRW
Figure 3. Support provided to partner as a function of participant gender and partner gen- der. MRW = men in relationship with a woman; MRM = men in relationship with a man; WRM = women in a relationship with a man; WRW = women in a relationship with a woman.
effect, F(2, 594) = 8.57, p < .001, Wilks’s Λ = 0.97, partial η2 = .029. Separate 2 × 2 ANOVAs were then carried out. Relationship satisfaction displayed a pattern similar to the one for received support; the main effect of participant gender and the interaction effect were significant (both Fs > 6.43, both ps < .05; see Figure 4). Women, and participants in same-sex relationships, reported higher levels of satisfaction. For separation proneness, there was a marginal main effect for partner gender, F(1, 595) = 3.73, p = .054, and for
the interaction, F(1, 595) = 14.94, p < .001. Separation proneness was lower for partici- pants with same-sex partners than for those with opposite-sex partners (both ts > 2.03, both ps < .01; see Figure 5).
Does relationship type explain variance in outcome beyond that explained by social support?
Analyses thus far have revealed a significant interaction between participant gender and partner gender when predicting relationship
362 L. Ellis and M. Davis
4
4.2
4.4
4.6
4.8
5
5.2
5.4
MRW MRM WRM WRW
Relationship Satisfaction
Figure 4. Relationship satisfaction as a function of participant gender and partner gen- der. MRW = men in relationship with a woman; MRM = men in relationship with a man; WRM = women in a relationship with a man; WRW = women in a relationship with a woman.
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
MRW MRM WRM WRW
Separation Proneness
Figure 5. Separation proneness as a function of participant gender and partner gender. MRW = men in relationship with a woman; MRM = men in relationship with a man; WRM = women in a relationship with a man; WRW = women in a relationship with a woman.
satisfaction and a very similar interaction when received support was the dependent variable. A logical question then becomes: Does the effect of relationship type on sat- isfaction depend entirely on its covariance with received support, or does relationship type influence relationship outcomes in a unique way? To answer this question, we carried out two series of hierarchical multiple regression analyses in which relationship satisfaction and separation proneness served as dependent variables. In the first regression, the four covariates were entered on the first step. On the second step, three variables were
entered — participant gender, partner gender, and a Participant Gender × Partner Gender interaction term. This regression essentially reproduces the MANOVAs reported earlier. In the second regression, the four measures of received support were also included on the first step, allowing us to determine whether the gender interaction term remains significant when the effect of social support on relation- ship outcome is controlled. Table 4 contains the result of these analyses.
For both satisfaction and separation proneness, there were significant effects of participant gender, partner gender, and their
Intimate partner support 363
Table 4. Multiple regression analyses predicting relationship satisfaction and separation prone- ness (standardized regression coefficients)
Model 1
Relationship satisfaction Separation proneness
Step 1 Age −.17*** .03 Education .03 −.07 Marital status .01 −.11* Living arrangement .08 −.11*
R2 .02** .04*** Step 2
Participant gender −.56*** .60*** Partner gender −.45*** .49* Gender × Gender .53*** −.64***
R2 .05*** .08*** R2 change .03*** .04***
Model 2
Relationship satisfaction Separation proneness
Step 1 Age −.07 −.05 Education .05 −.08 Marital status .07 −.15*** Living arrangement .04 −.09* Informational support .12** −.02 Physical support .29*** −.14** Emotional support .23*** −.23*** Tangible support .14** −.14*
R2 .42** .24*** Step 2
Participant gender −.08 .29* Partner gender −.05 .21 Gender × Gender .07 −.33*
R2 .42*** .25*** R2 change .00 .01*
*p < .05. **p < .01. ***p < .001.
interaction when the four covariates were entered on the first step. When the four mea- sures of received support were also included on the first step, satisfaction and separation proneness displayed differing patterns. For satisfaction, the effects of gender disappeared entirely, suggesting that relationship type has no unique effect on satisfaction above and beyond that of social support. In contrast, for separation proneness, small but significant
gender effects remained, most importantly the Participant Gender × Partner Gender inter- action. Thus, relationship type has a unique effect on separation proneness that is distinct from the effect on social support.
We also carried out an additional regression predicting separation proneness in which rela- tionship satisfaction was included as an addi- tional variable on the first step of the analysis. Although satisfaction was strongly associated
364 L. Ellis and M. Davis
with separation proneness, the interaction of participant gender and partner gender remained significant (β = −.33, p < .05). Thus, the effect of relationship type on separation proneness still remained, suggesting that same-sex rela- tionships are associated with lower separation proneness for reasons other than their higher levels of support, and even satisfaction.
Discussion
The evidence from this investigation reveals three broad patterns. First, as expected, there were differences in social support and relation- ship outcomes as a function of gender. Women consistently reported providing and receiv- ing more support than men, with the pattern especially strong for emotional and tangible support. In contrast, the gender of one’s partner had very little effect on any support measure. Most importantly, the interaction between participant gender and partner gender was significant for all forms of support — received, provided, and desired. Respondents in same-sex intimate relationships reported receiving more support, providing more support, and to some degree desiring more support. The same pattern was apparent for the relationship outcome variables; participants in same-sex relationships reported greater satisfaction and lower separation proneness than those in opposite-sex relationships.
Second, although the mean levels of sup- port, satisfaction, and separation proneness varied as a function of relationship type, the associations between support and outcomes did not. Consistent and very similar correla- tions were found for all four groups: MRM, MRW, WRM, and WRW. Generally speaking, for all four groups, better outcomes were associated with providing more support and receiving more support; desired support was essentially unrelated to outcomes.
Third, analyses revealed different patterns for the two relationship outcome variables. Dif- ferences in satisfaction between various types of relationships appear to be accounted for by differences in social support. However, differ- ences in separation proneness do not show this pattern — even when relationship satisfaction was included as a covariate — suggesting that
other characteristics of same-sex relationships are at least partially responsible for the differences in separation proneness.
Gender differences: Is there a support gap?
Given our analytical strategy, we had envi- sioned three patterns of results that might shed light on the question of a gender gap in rela- tionship support. First, there might be gender effects in which women report receiving, pro- viding, and/or desiring more (or less) sup- port than men. Second, there might be partner effects such that those with male partners might report receiving, providing, and/or desir- ing more (or less) support than those with female partners. Third, there might be interac- tion effects such that those in same-sex rela- tionships might report receiving, providing, and/or desiring more (or less) support than those in opposite-sex relationships.
Regarding received support, we found consistent gender effects that did not support the gender gap hypothesis; women reported receiving more support of every kind than did men, especially emotional and tangible. However, women also consistently reported providing more support than men, which is broadly consistent with the gender gap hypoth- esis. This seemingly inconsistent pattern of women both receiving and providing more sup- port is accounted for by the support patterns of one group: women in a relationship with another woman. The consistently significant Participant Gender × Partner Gender interac- tions reveal what the gender main effects do not — that those in same-sex relationships, and especially WRW, both provide and receive the highest levels of support. Thus, the heightened tendency to provide/receive support was not found for women in general but primarily for a particular subset of women — those who are in relationships with another woman.
What might account for this pattern? One possibility, and the one underlying our initial predictions, is that it may be easier to provide effective support when one’s romantic partner is of the same gender. Relationships are fraught with opportunities for partners to misunder- stand one another, to misread intentions, and to incorrectly perceive the other person’s needs.
Intimate partner support 365
To the degree that such misunderstandings result from gender-related differences (and certainly not all misunderstandings do), having a partner of the same gender may on average reduce such errors and confer an advantage on same-sex relationships. In particular, to the degree that men and women desire different types or amounts of social support, same-sex relationships may have the edge. Certainly, the results from Xu and Burleson (2001) and the analyses of desired support in this investiga- tion are consistent with the idea that there are at least some gender differences in the desired amount of support.
Pearlin and McCall’s (1990) model of support provision provides a useful way to understand how this same-sex advantage might operate. This model describes three sequential stages in the support process. First, potential support providers must perceive that their part- ners face some problem and that offering sup- port is a possible response. Second, they must evaluate the situation to determine whether or not to offer support, and if so, what form the support should take. Finally, based on the anal- ysis occurring at the second stage, actual sup- port is provided (or not) to the support seeker.
Successfully passing through the first stage can be difficult because potential recipients do not always actively seek support and instead may use hints or nonverbal cues, assum- ing that their relationship partners will be able to interpret such subtle signals without being explicitly informed (Cutrona, Suhr, & MacFarlane, 1990). At this stage, cue recog- nition might be easier for same-sex couples because partners may more easily recognize and respond to gender-specific cues that they themselves are prone to use. In support of this, there is at least some evidence that men and women differ in the way that they solicit social support. For example, women are generally more likely than men to seek a particular type of support — emotional (e.g., Ashton & Fuehrer, 1993; Burda, Vaux, & Schill, 1984). In addition, Verhofstadt and colleagues (2007) found that both husbands and wives perceive wives as soliciting marital support more frequently than husbands, in ways both positive (calmly, appreciatively) and negative (demanding, complaining).
Same-sex couples may also benefit at the second stage, in which support providers evaluate the nature of the challenge facing the potential recipient. How well equipped is the partner to deal with the stressor he or she is fac- ing? What type of support would be the most effective at this moment? What type of support does the partner desire? At this stage, those in same-sex relationships may benefit from a simple and common strategy — projection. Considerable evidence suggests that we are prone to use projection in a wide variety of cir- cumstances (see Nickerson, 1999). Moreover, we may be especially likely to use projection when we perceive greater similarity between ourselves and some target (Ames, 2004), which would seem to be the case in same-sex relationships. To the degree that same-sex part- ners actually are more similar in their support preferences, simply assuming that intimate partners want or need the same things that we do is advantageous; projection will lead to more accuracy. Thus, it is entirely plausible that same-sex couples may be especially likely to reap the benefits of this particular social inference strategy. Again, and to be clear, this argument in no way relies on an assump- tion that all women (or men) are the same. Many factors undoubtedly influence intimate partners’ success at discerning each other’s support needs. However, in the aggregate, having a partner of the same gender reduces the likelihood of some kinds of errors.
Regardless of the particular mechanism — cue recognition, projection, or both — the end result will be to make same-sex partners at least somewhat more likely to offer support in the form and amount that is desired by the other. Thus, in Cutrona’s (Cutrona & Russell, 1990) terms, same-sex partners may be more successful at achieving “optimal matching” of support in which the recipient’s desires and the provider’s behavior are correctly aligned. Such a state of affairs is likely to then pro- duce a more favorable view of the provider and greater satisfaction in the recipient. Con- sistent with this interpretation, regression anal- yses in this investigation revealed exactly this pattern; once the levels of received support were controlled, the effect of relationship type on satisfaction disappeared. Thus, the greater
366 L. Ellis and M. Davis
satisfaction experienced by those in same-sex relationships may be largely due to the superior partner support that they receive.
Interestingly, separation proneness did not show this pattern. When social support was statistically controlled — and relationship satisfaction as well — those in same-sex rela- tionships remained significantly less likely to report thinking or talking about ending the relationship. What might account for this? Although it is speculative, the decreased sep- aration proneness in same-sex relationships might be explained by a difference in the available alternatives. Considerable evidence suggests that individuals are more likely to leave a relationship if there are good alter- natives available (e.g., Edwards, Gidycz, & Murphy, 2011; Miller, Lund, & Weatherly, 2012), and it is possible that those in same-sex relationships might generally perceive them- selves as having fewer alternatives than those in heterosexual relationships. Such a percep- tion might result from simple numbers (there are fewer people in same-sex relationships than opposite-sex relationships). In support of this, Lehmiller and Agnew (2006) conducted an investigation of three types of “marginal- ized couples,” defined as relationships that are somehow socially devalued; the three they studied were interracial couples, same-sex couples, and couples in which younger women were dating older men. Compared to those in nonmarginalized relationships, those in marginalized relationships perceived their relationship alternatives to be poorer.
To further evaluate this possibility, we car- ried out an additional analysis of our data using the respondents’ self-reported sexual orienta- tion: gay/lesbian, heterosexual, or bisexual. In a very straightforward way, it might be argued that bisexuals will generally perceive them- selves to have a larger pool of potential partners available to them and thus should be some- what less likely to remain in a relationship. When the three groups are compared on lev- els of received social support and relationship satisfaction, bisexuals and gays/lesbians did not differ, and both were significantly higher than heterosexuals. For separation proneness, however, the pattern was different. Bisexuals’ separation proneness was significantly higher
than those identifying as gay/lesbian and did not differ from that of heterosexuals. This pat- tern is therefore generally consistent with the argument that commitment to a relationship will be lower when there are other alternatives. However, given the speculative nature of this argument, it is far too early to conclude that bisexuals are more prone to separating primar- ily because they have the largest pool of poten- tial partners.
What is the role of gender roles?
Finally, there is another variable likely to play a role in intimate partner support but that was not included in this investigation: the degree to which individual participants adhere to cultural gender roles. Much of the theory and research addressing the gender gap is predicated on the assumptions that men and women are social- ized into distinct gender roles and that those roles have implications for the provision of social support. In general, men do score higher on measures of psychological masculinity and women on measures of psychological feminin- ity; however, there is also considerable varia- tion in the degree to which individual men and women adhere to such roles (e.g., Bem, 1977; Spence & Helmreich, 1978).
Thus, it is possible that variability in gen- der role identity is responsible for the major finding reported here — that is, the relative superiority of same-sex relationships in offer- ing effective support. One way in which this might occur is if MRM and WRW are more psychologically androgynous (i.e., relatively high in both masculine and feminine gender identities) than MRW and WRM. After all, if a more masculine, instrumental identity pro- motes the offering of instrumental support, and a more feminine, expressive identity promotes emotional support, then possessing both may make one better equipped to provide the full range of support behaviors to one’s partner. In support of this possibility, it does appear that in general, gay men are equivalent to straight men in their levels of instrumentality — but are higher on expressiveness — and that lesbians are equivalent to straight women in their levels of expressiveness but are higher on instru- mentality (Lippa, 2005). Thus, in general,
Intimate partner support 367
people who are LGB are more psycholog- ically androgynous than their heterosexual counterparts.
There is also some evidence that each gen- der identity is associated in some way with social support processes. Higher scores on measures of psychological femininity (com- munal, expressive) tend to be associated with a greater willingness to both seek and receive social support (e.g., Ashton & Fuehrer, 1993; Burda et al., 1984). With regard to offering support, there is evidence from heterosexual intimate relationships that a masculine gender identity is associated with offering more instru- mental support, and a feminine identity is asso- ciated with offering more emotional support (Verhofstadt & Weytens, 2013).
Thus, based on this pattern, it is certainly plausible that at least some — or perhaps all — of the differences in intimate partner support found in this investigation between same-sex and opposite-sex relationships are the result of differential adherence to masculine and femi- nine gender roles. However, because the cur- rent investigation did not include any measures of gender role identification, it is impossible to examine its effects here. However, future research would do well to include such mea- sures so as to directly evaluate this possibility.
Limitations
We believe that this investigation has con- siderable value because it represents the first attempt to directly compare social support in same-sex and opposite-sex relationships in a fine-grained fashion. Nevertheless, it is also limited in some important ways that must con- dition our interpretation of the findings. Most importantly, the sample cannot be considered a random or representative sample of same-sex and opposite-sex relationships. Participants were recruited in a number of ways, but there was a heavy emphasis on contacting people via social media, especially Facebook pages and groups with an lesbian, gay, bisexual, trans, and queer (LGBTQ) focus. Thus, the sample is subject to clear self-selection issues. For example, the MRM and WRW in this investi- gation may be more interested in gender issues generally, and gender orientation might also
be a more important part of their identities. It is also possible that such self-selection at least partially contributed to the greater satisfaction reported by these groups in this study, which differs from the usual pattern of no differences between same-sex and opposite-sex couples (see Kurdek, 2004).
In addition to the effects of self-selection for the respondents in same-sex relationships, there may be similar concerns for the hetero- sexual participants as well. Because the sample was recruited for the most part in LGBTQ safe spaces, these heterosexual participants might differ from heterosexuals in general. This may be a special concern for the MRW in our sample. Because they were most likely allies of the LGBTQ movement, it is possible that they may differ in some ways from a more representative sample of MRW. On the other hand, to the degree that our sample of male heterosexuals differs meaningfully from males in general, this also means that the differences found between MRW and other groups in this investigation may actually underestimate the differences that would be found with a more representative sample.
We included several background variables that might account for differences among the four couple types and found that relationship type had an effect even when the influence of these variables were controlled. However, there are other factors that might differ across relationship types that we did not assess. For example, we did not ask about length of the relationship, and it is possible that this would also influence the provision and receipt of support.
Although these limitations are real, it is also true that some of the findings from this investigation are similar enough to prior work to assuage our concerns. For example, several investigations have found evidence of better relationship functioning in lesbian couples, particularly greater emotional expressivity and concern for egalitarian values (Kurdek, 1988, 1991, 1993, 1998). The fact that WRW in this investigation were more likely to give and receive support is consistent with such a pattern. In addition, the association between social support and satisfaction did not vary across relationship types, consistent
368 L. Ellis and M. Davis
with prior work finding that relationships processes appear to operate similarly for both same-sex and opposite-sex couples (Kurdek, 2004).
Conclusion
Support from romantic partners is a uniquely important element in social life. It is therefore important to identify the factors influencing the offering and receipt of effective support. This investigation provides important initial support for the argument that same-sex relationships may have an advantage in this regard. Because of shared socialization experiences, those in same-sex relationships may possess advan- tages in recognizing partner need and offering more effective support, which then con- tributes to partner satisfaction. More research is required to determine precisely how this same-sex advantage is created, but the current study suggests the real value of doing so.
References
Acitelli, L. K., & Antonucci, T. C. (1994). Gender differ- ences in the link between marital support and satisfac- tion in older couples. Journal of Personality and Social Psychology, 67, 688 – 698.
Ames, D. R. (2004). Inside the mind reader’s tool kit: Projection and stereotyping in mental state inference. Journal of Personality and Social Psychology, 87, 340 – 353.
Antonucci, T. C., & Akiyama, H. (1987). An examination of sex differences in social support among older men and women. Sex Roles, 17, 737 – 749.
Ashton, W. A., & Fuehrer, A. (1993). Effects of gender and gender role identification of participant and type of social support resource on support seeking. Sex Roles, 28, 461 – 476.
Bakan, D. (1966). The duality of human existence: Isola- tion and communion in Western man. Oxford, England: Rand-McNally.
Barry, R. A., Bunde, M., Brock, R. L., & Lawrence, E. (2009). Validity and utility of a multidimensional model of received support in intimate relationships. Journal of Family Psychology, 23, 48 – 57.
Bell, A. P., Weinberg, M. S., & Hammersmith, S. K. (1981). Sexual preference: Its development in men and women (Vol. 2). Bloomington: Indiana University Press.
Belle, D. (1982). The stress of caring: Women as providers of social support. In L. Goldberger & S. Breznitz (Eds.), Handbook of stress: Theoretical and clinical aspects (pp. 496 – 505). New York, NY: Free Press.
Bem, S. L. (1977). On the utility of alternative proce- dures for assessing psychological androgyny. Journal of Consulting and Clinical Psychology, 45, 196 – 205.
Blashill, A. J., & Powlishta, K. (2009). Gay stereo- types: The use of sexual orientation as a cue for gender-related attributes. Sex Roles, 61, 783 – 793.
Brock, R. L., & Lawrence, E. (2009). Too much of a good thing: Underprovision versus overprovision of partner support. Journal of Family Psychology, 23, 181 – 192.
Burda, P. C., Jr., Vaux, A., & Schill, T. (1984). Social sup- port resources: Variation across sex and sex role. Per- sonality and Social Psychology Bulletin, 10, 119 – 126.
Canary, D. J., & Hause, K. S. (1993). Is there any reason to research sex differences in communication? Communi- cation Quarterly, 41, 129 – 144.
Cutrona, C. E., Shaffer, P. A., Wesner, K. A., & Gardner, K. A. (2007). Optimally matching support and perceived spousal sensitivity. Journal of Family Psychology, 21, 754 – 758.
Cutrona, C. E., & Suhr, J. A. (1994). Social support com- munication in the context of marriage: An analysis of couples’ supportive interactions. In B. R. Burleson, T. L. Albrecht, & I. G. Sarason (Eds.), Communication of social support: Messages, interactions, relationships, and community (pp. 113 – 135). Thousand Oaks, CA: Sage.
Cutrona, C. E., Suhr, J. A., & MacFarlane, R. (1990). Interpersonal transactions and the psychological sense of support. In S. Duck (Ed.), Personal relationships and social support (pp. 30 – 45). London, England: Sage.
Cutrona, C. E., & Russell, D. W. (1990). Type of social support and specific stress: Toward a theory of optimal matching. In B. R. Sarason, I. G. Sarason, & G. R. Pierce (Eds.), Social support: An interactional view (pp. 319 – 366). Oxford, England: Wiley.
Darbes, L. A., & Lewis, M. A. (2005). HIV-specific social support predicts less sexual risk behavior in gay male couples. Health Psychology, 24, 617 – 622.
Dehle, C., Larsen, D., & Landers, J. E. (2001). Social sup- port in marriage. American Journal of Family Therapy, 29, 307 – 324.
Edwards, K. M., Gidycz, C. A., & Murphy, M. J. (2011). College women’s stay/leave decisions in abusive dating relationships: A prospective analysis of an expanded investment model. Journal of Interpersonal Violence, 26, 1446 – 1462.
Elizur, Y., & Mintzer, A. (2003). Gay males’ intimate rela- tionship quality: The roles of attachment security, gay identity, social support, and income. Personal Rela- tionships, 10, 411 – 435.
Funk, J. L., & Rogge, R. D. (2007). Testing the ruler with item response theory: Increasing precision of mea- surement for relationship satisfaction with the Couples Satisfaction Index. Journal of Family Psychology, 21, 572 – 583.
Goldsmith, D. J., & Dun, S. A. (1997). Sex differences and similarities in the communication of social sup- port. Journal of Social and Personal Relationships, 14, 317 – 337.
Intimate partner support 369
Graham, J. M., & Barnow, Z. B. (2013). Stress and social support in gay, lesbian, and heterosexual cou- ples: Direct effects and buffering models. Journal of Family Psychology, 27, 569 – 578.
Huston-Hoburg, L., & Strange, C. (1986). Spouse support among male and female returning adult students. Jour- nal of College Student Personnel, 27, 388 – 394.
Jensen, J., Rauer, A., & Volling, B. (2013). A dyadic view of support in marriage: The critical role of men’s support provision. Sex Roles, 68, 427 – 438.
Julien, D., Chartrand, E., Simard, M. C., Bouthillier, D., & Bégin, J. (2003). Conflict, social support and relation- ship quality: An observational study of heterosexual, gay male and lesbian couples’ communication. Journal of Family Psychology, 17, 419 – 428.
Kinsey, A. C., Pomeroy, W. B., & Martin, C. E. (1948). Sexual behavior in the human male. Philadelphia, PA: Saunders.
Kurdek, L. A. (1987). Sex role self schema and psycholog- ical adjustment in coupled homosexual and heterosex- ual men and women. Sex Roles, 17, 549 – 562.
Kurdek, L. A. (1988). Perceived social support in gays and lesbians in cohabitating relationships. Journal of Personality and Social Psychology, 54, 504 – 509.
Kurdek, L. A. (1991). Correlates of relationship satisfac- tion in cohabiting gay and lesbian couples: Integration of contextual, investment, and problem-solving mod- els. Journal of Personality and Social Psychology, 61, 910 – 922.
Kurdek, L. A. (1993). Predicting marital dissolution: A 5-year prospective longitudinal study of newlywed couples. Journal of Personality and Social Psychology, 64, 221 – 242.
Kurdek, L. A. (1998). Relationship outcomes and their pre- dictors: Longitudinal evidence from heterosexual mar- ried, gay cohabiting, and lesbian cohabiting couples. Journal of Marriage and the Family, 60, 553 – 568.
Kurdek, L. A. (2001). Differences between heterosexual-nonparent couples and gay, lesbian, and heterosexual-parent couples. Journal of Family Issues, 22, 727 – 754.
Kurdek, L. A. (2004). Are gay and lesbian cohabiting cou- ples really different from heterosexual married cou- ples? Journal of Marriage and Family, 66, 880 – 900.
Kurdek, L. A. (2006). The nature and correlates of deter- rents to leaving a relationship. Personal Relationships, 13, 521 – 535.
Kurdek, L. A., & Schmitt, J. P. (1986). Relationship qual- ity of partners in heterosexual married, heterosexual cohabiting, and gay and lesbian relationships. Journal of Personality and Social Psychology, 51, 711 – 720.
Lehavot, K., & Lambert, A. J. (2007). Toward a greater understanding of antigay prejudice: On the role of sexual orientation and gender role violation. Basic and Applied Social Psychology, 29, 279 – 292.
Lehmiller, J. J., & Agnew, C. R. (2006). Marginalized rela- tionships: The impact of social disapproval on romantic relationship commitment. Personality and Social Psy- chology Bulletin, 32, 40 – 51.
Lippa, R. A. (2005). Sexual orientation and personality. Annual Review of Sex Research, 16, 119 – 153.
Miller, K. B., Lund, E., & Weatherly, J. (2012). Apply- ing operant learning to the stay-leave decision in domestic violence. Behavior and Social Issues, 21, 135 – 151.
Monin, J., & Clark, M. (2011). Why do men benefit more from marriage than do women? Thinking more broadly about interpersonal processes that occur within and outside of marriage. Sex Roles, 65, 320 – 326.
Nickerson, R. S. (1999). How we know — and some- times misjudge — what others know: Imputing one’s own knowledge to others. Psychological Bulletin, 125, 737 – 759.
Parsons, T., & Bales, R. F. (1956). Family socialization and interaction process (Vol. 7). New York, NY: Psychol- ogy Press.
Pasch, L. A., Bradbury, T. N., & Davila, J. (1997). Gen- der, negative affectivity, and observed social support behavior in marital interaction. Personal Relationships, 4, 361 – 378.
Pearlin, L. I., & McCall, M. E. (1990). Occupational stress and marital support. In J. Eckenrode & S. Gore (Eds.), Stress between work and family (pp. 39 – 60). New York, NY: Springer.
Ragsdale, J. D. (1996). Gender, satisfaction level, and the use of relational maintenance strategies in marriage. Communications Monographs, 63, 354 – 369.
Roisman, G. I., Clausell, E., Holland, A., Fortuna, K., & Elieff, C. (2008). Adult romantic relationships as contexts of human development: A multimethod com- parison of same-sex couples with opposite-sex dating, engaged, and married dyads. Developmental Psychol- ogy, 44, 91 – 101.
Sarason, B. R., Sarason, I. G., Hacker, T. A., & Basham, R. B. (1985). Concomitants of social support: Social skills, physical attractiveness, and gender. Journal of Personality and Social Psychology, 49, 469 – 480.
Spence, J. T., & Helmreich, R. L. (1978). Masculinity and femininity: Their psychological dimensions, cor- relates, and antecedents. Austin: University of Texas Press.
Verhofstadt, L. L., Buysse, A., & Ickes, W. (2007). Social support in couples: An examination of gender differ- ences using self-report and observational methods. Sex Roles, 57, 267 – 282.
Verhofstadt, L. L., & Weytens, F. (2013). Biological sex and gender role identity as predictors of spousal sup- port provision: A scenario-based study. Journal of Gender Studies, 22, 166 – 177.
Wharton, A. S., & Erickson, R. I. (1993). Managing emo- tions on the job and at home: Understanding the conse- quences of multiple emotional roles. Academy of Man- agement Review, 18, 457 – 486.
Xu, Y., & Burleson, B. R. (2001). Effects of sex, culture, and support type on perceptions of spousal social sup- port: An assessment of the “support gap” hypothesis in early marriage. Human Communication Research, 27, 535 – 566.
Copyright of Personal Relationships is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
