Introduction

The purpose of the current study was to investigate relationship quality within same-sex and heterosexual romantic relationships with a particular focus on men in relationships. Given that quality romantic relationships contribute to psychological well-being (Cohen and Willis 1985; Diener et al. 1999; Rook and Pietromonaco 1987) and physical health and longevity (Schuster et al. 1990; Taylor 1995; Willis 1985), it would be important to understand those factors that contribute to quality romantic relationships for men. Therefore, we used descriptive research to explore the relationships between masculinity ideology, male identity, and men’s perceptions of the quality of their romantic relationship in a sample of gay and heterosexual men.

The construct of “relationship quality” for romantic relationships stems from the “marital quality” construct of Lewis and Spanier (1979), which they defined as, “the subjective evaluation of a married couple’s relationship on a number of dimensions and evaluations” (Spanier and Lewis 1980, p. 826). High marital quality is “associated with good adjustment, adequate communication, a high level of marital happiness, integration, and a high degree of satisfaction with the relationship” (Lewis and Spanier 1979, p. 269). The more general research that has examined predictors of relationship quality has tended to focus on interpersonal factors, such as interactions between partners, or has emphasized intrapersonal factors, such as personality traits and attachment styles (Watson et al. 2000). However, this informative line of research inquiry typically has not had gender-related issues, or men in particular, as its focus of investigation. A smaller body of research has focused on gender-related issues and relationship quality, examining how men and women function in intimate relationships. Self-disclosure, or sharing of personal information, thoughts and feelings is associated with relationship satisfaction for both men and women (Siavelis and Lamke 1992; Sprecher and Hendrick 2004). Research has also shown a positive relationship between trait femininity and relationship satisfaction and quality in both men and women (Kurdek and Schmitt 1986b; Steiner-Pappalardo and Gurung 2002) (femininity and masculinity being the socially defined traits associated with the female and male gender, respectively). Yet, some research indicates men are less likely to include feminine characteristics as part of their gender role self-concept (Di-Dio et al. 1996; Feingold 1994).

Relationship quality has also been examined in hetero-sex versus same-sex partnerships. Generally, research indicates that the quality of lesbians’ and gay men’s relationships is similar to that of heterosexual couples (Duffy and Rusbult 1986; Kurdek 1994; Kurdek and Schmitt 1986a; Means-Christensen 2003). Kurdek and Schmitt (1986a) found no differences in relationship quality, in terms of love for one’s partner, liking of one’s partner, and general relationship satisfaction, between heterosexual married couples and cohabiting gay and lesbian couples. In 2004, Kurdek examined psychological distress, neuroticism, equality in the relationship, conflict resolution style, and satisfaction with social support as they relate to the quality of romantic relationships in gay and lesbian cohabiting couples and heterosexual married couples. No significant differences were found between the couples on these factors, suggesting that relationship factors operate the same despite the sexual orientation of the couple. Duffy and Rusbult (1986) found greater relationship satisfaction was associated with higher levels of rewards and lower levels of costs in both same-sex and hetero-sex couples. Generally, research has demonstrated that factors predicting relationship quality (e.g., personality traits, level of trust, communication and conflict-resolution styles) tend to be as strong in both same-sex and heterosexual couples (Kurdek 2005). Particular to same-sex couples however, identity confusion, homonegativity, and stigma sensitivity can negatively impact relationship quality (Mohr and Fassinger 2006), whereas a positive gay identity is associated with better relationship quality (Elizur and Mintzer 2003).

Research in the new psychology of men (Levant 1996) has explored how traditional masculinity ideology and male identity relate to men’s psychosocial functioning. There has been some research that has examined the relationship between traditional masculinity ideology and men’s romantic relationships, whereas no previous research has examined relationship quality and how it relates to male identity or nontraditional masculinity ideology (i.e., a masculinity that in some way diverts from traditional masculinity).

Masculinity Ideology

Masculinity ideology refers to men’s acceptance or internalization of a culture’s definition of masculinity, and beliefs about adherence to culturally defined standards of male behavior (Pleck et al. 1993). Although there may be many masculinity ideologies, the masculinity ideology that has been examined most within the literature has been referred to as “traditional,” which has been described by several researchers (e.g., David and Brannon 1976; Franklin 1984; Harris 1995; Levant et al. 1992; O’Neil 1981). Conceptual formulations of traditional masculinity ideology in contemporary American culture have focused on those standards and expectations that have various negative consequences (Pleck 1995). Such standards and expectations include anti-femininity, homophobia, emotional restrictiveness, competitiveness, toughness, and aggressiveness.

Research on men that has examined traditional masculinity and heterosexual romantic relationships indicates traditional masculinity negatively affects relationship quality (e.g., Pleck et al. 1993; Sinn 1997; Truman et al. 1996). Burn and Ward (2005) assessed how traditional masculinity affects relationship quality for both men and women. Conformity to traditional masculine norms related negatively to relationship satisfaction for both women and men. Specifically, women who perceived that their male partners conformed more to traditional masculinity norms were less satisfied with their relationship, and men who were more conforming to traditional male norms were also less satisfied with their relationship. However, Fitzpatrick et al. (2004) studied men’s gender-role ideology, specifically their attitudes toward the equality of men and women, and found it had no influence on relationship quality.

A few studies have focused on how gender role conflict (O’Neil et al. 1986) relates to relationship quality. Men’s gender role conflict is a consequence of conforming to or deviating from traditional masculinity (O’Neil et al. 1995). When relationship and marital satisfaction have been examined, greater gender role conflict in men was associated with lower levels of relationship satisfaction in heterosexual men (Campbell and Snow 1992; Sharpe et al. 1995) and gay men (Wester et al. 2005). Additionally, women who perceived their partners as having less gender role conflict reported greater relationship satisfaction (Rochlen and Mahalik 2004).

Nontraditional attitudes toward masculinity have rarely been examined in the literature. However, the literature suggests that gay men are more nontraditional in their masculinity (Bailey et al. 1997; Lippa and Tan 2001) and more nontraditional in their intimate relationships (Huston and Schwartz 2002; Peplau 1993) as compared to heterosexual men. If traditional masculinity ideology negatively effects the quality of men’s romantic relationships, then it may be that nontraditional attitudes toward masculinity has the opposite effect in both heterosexual and gay men’s relationships.

Male Identity

Wade’s (1998) theory of male identity suggests another possible explanation for differences in the quality of men’s romantic relationships. Wade conceptualized male identity in terms of “male reference group identity dependence,” defined as the extent to which males are dependent on a reference group for their gender role self-concept. The gender role self-concept is one’s self-concept with regard to gender roles and includes one’s gender-related attributes, attitudes, and behaviors (McCreary 1990). There are three male reference group identity dependence statuses characterized by a man’s feelings of psychological relatedness to other men. The No Reference Group status is characterized by a lack of psychological relatedness to other males. There is no particular group or image of males that the individual feels he is similar to, connected to, or he identifies with, and the gender role self-concept is therefore relatively undefined or fragmented. The individual feels disconnected from, and confusion about, other males. He feels there are no males like oneself or with whom he identifies or feels connected. The Reference Group Dependent status is characterized by psychological relatedness to some males and not others. There is a particular group or image of males the individual feels he is similar to, connected to, has a sense of commonality or identification with while this is not so with males perceived to be unlike or dissimilar to oneself. Here, the gender role self-concept is dependent on a male reference group and therefore externally defined, stereotyped, conformist, and rigid. The Reference Group Nondependent status is characterized by psychological relatedness to all males. Although the individual recognizes there are differences among males, there is a sense of commonality, similarity, connectedness, identification, and association with all or various types of males. The gender role self-concept is not dependent on a male reference group and therefore is internally defined, pluralistic, flexible, and autonomous.

Research examining correlates of male reference group identity has found the statuses to differentially relate to masculinity ideology and other variables (see Wade 2001; Wade and Brittan-Powell 2000, 2001; Wade and Gelso 1998). For example, the no reference group status has related positively to nontraditional masculinity ideology and negatively to trait masculinity. The reference group dependent status has related positively to gender role conflict, traditional masculinity ideology, attitudes unsupportive of race and gender equity, and attitudes conducive to sexual harassment, and negatively to nontraditional masculinity ideology. The reference group nondependent status has related positively to trait femininity, trait masculinity, and a universal-diverse orientation (i.e., awareness and acceptance of the similarities and differences that exist among people), and negatively to gender role conflict.

Although no research has previously examined male reference group identity as it relates to the quality of men’s romantic relationships, given the above findings it was expected that the three male identity statuses would also differentially relate to relationship quality. In that the reference group dependent status has been associated with traditional masculinity ideology, the expectation was this status would be related to low quality romantic relationships. Conversely, the reference group nondependent status has been associated with nontraditional masculinity and would be related to high quality romantic relationships. The salient characteristic of the no reference group status is the lack of a male identity, which presupposes that one would have difficulty with romantic relationships.

On reviewing the literature on same-sex couples and heterosexual couples, the notions of masculinity ideology and reference group identity dependence emerge as possible explanatory variables in considering the quality of men’s romantic relationships. Although previous research on same-sex and heterosexual couples indicates no differences in relationship quality, we first assessed whether relationship quality differed for same-sex versus heterosexual couples for the men in our sample. Second, we examined how masculinity ideology and male reference group identity related to relationship quality separately for gay men and heterosexual men. Consistent with the research literature, we hypothesized that traditional masculinity ideology would negatively relate to relationship quality. Conversely, we hypothesized nontraditional masculinity would be related to higher relationship quality. In that male identity is related to masculinity ideology, we explored the extent to which male reference group identity is a factor associated with relationship quality. Specifically, we hypothesized the reference group dependent status and no reference group status would relate negatively to relationship quality whereas the reference group nondependent status would relate positively.

Method

Participants

Participants were 100 men recruited from an urban area in the Northeast United States who volunteered without compensation for their participation. Fifty of the participants identified as heterosexual/straight and 50 of the participants identified as homosexual/gay. For the sample of gay men, participants’ ages ranged from 19 to 32 years, with an average age of 22 (SD = 3.41). Most (80%) were undergraduate students: 10% freshman, 10% sophomores, 36% juniors, and 24% seniors. Twenty percent reported not attending college and working a full-time job. They reported the following ethnicities: White/European American (54%), Hispanic/Latino (36%), Black/African American (4%), and Asian American (6%). Forty percent of the men were currently involved in a relationship, 24% were currently dating, 4% were currently involved in a relationship and dating, and 32% were neither dating or in a relationship.

For the sample of heterosexual men, participants’ ages ranged from 18 to 29 years, with an average age of 20.6 (SD = 2.15). Most (88%) were undergraduate students: 20% freshman, 30% sophomores, 18% juniors, and 20% seniors. Twelve percent reported not attending college and working a full-time job. They reported the following ethnicities: White/European American (62%), Hispanic/Latino (18%), Black/African American (16%), and Asian American (4%). Forty-eight percent of the men were currently involved in a relationship, 12% were currently dating, 2% were currently involved in a relationship and dating, and 38% were neither dating or in a relationship.

We examined for possible demographic differences between the undergraduate students and men who were working full-time. The working participants were significantly older, t(98) = −8.790, p < .001, and had significantly longer romantic relationships, t(98) = −3.120, p < .005.

Measure

The Male Role Norms Inventory (MRNI)

The MRNI (Levant et al. 1992) is a 57-item measure that assesses masculinity ideology. There are seven subscales consisting of 45 items used to assess traditional masculinity ideology: Avoidance of Femininity, Homophobia, Self-Reliance, Aggression, Restrictive Emotionality, Achievement/Status, and Attitudes Toward Sex. Only the full scale was used for the analyses. One subscale of 12 items assesses nontraditional attitudes toward masculinity (e.g., “A man should love his sex partner” and “Men should be allowed to kiss their fathers”). Items are responded to on a 7-point Likert-type scale where 1 = strongly disagree and 7 = strongly agree, with higher scores representing greater endorsement of traditional or nontraditional masculinity ideology. Internal consistency reliability was .91 for Traditional Masculinity and .59 for Nontraditional Masculinity.

Reference Group Identity Dependence Scale (RGIDS)

The RGIDS (Wade and Gelso 1998) was developed to assess male reference group identity dependence. The RGIDS has 30 items that are responded to on a 6-point Likert-type scale where 1 = strongly disagree and 6 = strongly agree. The measure consists of four subscales. The No Reference Group scale (ten items) assesses one’s lack of psychological relatedness and feelings of disconnectedness with other men. The Reference Group Dependent scale (eight items) assesses men’s psychological relatedness and feelings of connectedness with some males but not others. The Reference Group Nondependent status is characterized by psychological relatedness and feelings of connectedness with all males and is represented by two subscales: Similarity and Diversity. The Similarity scale (six items) assesses feelings of similarity with all males. The Diversity scale (six items) assesses one’s appreciation of differences among men. Scores are continuous with higher scores on the subscales indicating higher levels of the relevant feelings and beliefs associated with each male identity status. Internal consistency reliability for the subscales in the current study was as follows: No Reference Group, .74; Reference Group Nondependent Similarity, .75; Reference Group Nondependent Diversity, .80; Reference Group Dependent, .64.

Dyadic Adjustment Scale (DAS)

The DAS is a 32-item questionnaire developed by Spanier (1976) to measure the quality of a romantic relationship. Spanier defined dyadic adjustment as, “...a process, the outcome of which is determined by the degree of (1) troublesome dyadic differences; (2) interpersonal tensions and personal anxiety; (3) dyadic satisfaction; (4) dyadic cohesion; (5) consensus on matters important to dyadic functioning” (p. 17). The scale assesses four empirically verified components of dyadic adjustment, which comprise the four subscales. The Dyadic Satisfaction subscale (ten items) assesses feelings of positive and global contentment with the relationship. Dyadic Cohesion (five items) assesses the degree to which the couple engages in pleasant activities together (e.g., work together on a project, laugh together, etc.). Dyadic Consensus (nine items) assesses the degree to which the couple can agree on matters of importance to the relationship. Affectional Expression (eight items) assesses agreement around issues of sexual expression and intimacy. Each subscale differs with regard to possible responses to the items and their scoring. For the purposes of this study, the scale was worded to gather participants’ responses based on one’s current romantic relationship or the last relationship the participant was involved in. Only the full scale was used for the analyses. Internal consistency reliability for the full scale was .78.

Modified Interpersonal Relationship Scale (MIRS)

The MIRS (Garthoeffner et al. 1993) is a revision of the Interpersonal Relationship Scale developed by Schlein et al. (1990). The 49-item measure assesses the quality of one’s romantic relationship and has six subscales: Trust, Self-Disclosure, Genuineness, Empathy, Comfort, and Communication. Participants rate each of the items using a 5-point scale in Likert-type format anchored by 1 = strongly disagree and 5 = strongly agree. For the purposes of this study, the scale was worded to gather participants’ responses based on one’s current romantic relationship or the last relationship the participant was involved in. Only the full scale was used for the analyses. Internal consistency reliability of the full scale was .93.

Personal Data Sheet

The personal data sheet consisted of the following items: participant’s age; year in college or employment if not in college; race/ethnicity; sexual orientation; whether or not the participant is currently in a relationship; whether or not the participant is currently dating; how long the participant has been in their current relationship (in months), or if not currently in a relationship how long was their last/past relationship.

Procedure

Participants were a combination of university students and workingmen. University participants were recruited through announcements made in psychology classes and completed the survey packet on campus in small groups. The workingmen were primarily a convenience sample of men with whom the researcher (i.e., second author) was acquainted or who were referred to the researcher by acquaintances. Non-university participants were provided the survey packet and returned it to the researcher once completed. Recruitment ended once we had received completed surveys for 50 gay men and 50 heterosexual men. All participants were asked to read and sign an informed consent form that explained the nature of the study, confidentiality of responses, and their rights as participants. The survey packet included the informed consent form, and the four measures (MIRS, RGIDS, MRNI, DAS) followed by the personal data sheet. When survey packets were completed and returned to the researcher, participants were provided a debriefing statement explaining the study and contact information if the participant was interested in receiving the results of the study once it was completed.

Results

Descriptive Statistics

Scale score means and standard deviations for the measures in the study are provided in Table 1. On average the participants in this sample reported having high quality intimate relationships. They tended to endorse characteristics of a reference group nondependent male identity with an appreciation of diversity among males. Participants were overall more nontraditional than traditional in their masculinity ideology being somewhat neutral with regard to traditional masculinity ideology.

Table 1 Scale means and standard deviations.
Full size table

Preliminary Analyses

Preliminary analyses involved examining the extent to which participants that differ in sexual orientation differed in male identity and masculinity ideology, and the relationships between male identity and masculinity ideology in both samples. Comparison of group means indicated that gay participants’ scores were higher for No Reference Group, Reference Group Dependent, and Nontraditional Masculinity, and lower for Traditional Masculinity than heterosexual participants’ scores (see Table 1).

Table 2 provides the correlations between male identity and masculinity ideology for the gay and heterosexual participants separately. For the gay participants, Reference Group Nondependent Similarity significantly positively correlated with traditional masculinity and negatively with nontraditional masculinity. Thus, the stronger one’s feelings of similarity and connectedness with all men the more traditional one’s masculinity ideology and the less one endorsed nontraditional masculinity ideology. For the heterosexual participants, Reference Group Nondependent Diversity significantly negatively correlated with traditional masculinity and positively with nontraditional masculinity. These results indicate that having an appreciation of differences among men was associated with endorsement of nontraditional masculinity while rejecting traditional masculinity ideology. Additionally, Reference Group Dependent significantly positively correlated with traditional masculinity ideology, and No Reference Group significantly positively correlated with nontraditional masculinity ideology. Thus, being dependent on a reference group for one’s gender role self-concept was associated with traditional masculinity, whereas having no reference group for one’s gender role self-concept was associated with nontraditional masculinity ideology.

Table 2 Correlation matrix for gay men and heterosexual men
Full size table

Primary Analyses

We first wanted to assess whether relationship quality differed for same-sex versus heterosexual couples for the men in our sample. This question was examined by using t tests to compare scale means for the two groups on the Dyadic Adjustment Scale (DAS) and Modified Interpersonal Relationship Scale (MIRS). No significant differences were found between the groups on the DAS, t(98) = .749, p = .456, or MIRS, t(98) = .273, p = .785. We further examined relationship quality by separating the samples by whether they were undergraduate students or men who were working full-time. A significant difference was found only between gay working men (M = 164.90) and heterosexual working men (M = 197.33) on the MIRS, t(14) = −2.736, p < .05, with heterosexual men having a significantly higher mean.

Second, we hypothesized that traditional masculinity ideology would negatively effect relationship quality. Conversely, we expected nontraditional masculinity would be related to high relationship quality. We tested these two hypotheses by examining correlations between the Male Role Norms Inventory Traditional and Nontraditional subscales and the DAS and MIRS scales. However, we first examined the relationships between the demographic (i.e., age, race/ethnicity, employed versus student status, currently in relationship, currently dating, and how long in current relationship) and measurement variables (masculinity ideology, male identity, and relationship quality). Ethnicity was examined using ANOVA. It was found that the gay Hispanic and African American participants were significantly more traditional in their masculinity ideology. We therefore combined the racial ethnic minority sample when doing our analyses, coding the groups 0 = racial ethnic minority, 1 = European American. For analyses we also coded groups according to employed (1) versus student (0), currently in relationship (1) versus not in relationship (0), dating (1) versus not currently dating (0).

Results for the gay men and heterosexual men are presented separately in Table 2. For the heterosexual men, of the demographic variables age and employment/student status significantly correlated with traditional masculinity ideology. The older the participant and/or being a student related negatively to traditional masculinity. After controlling for these two variables, there was a significant negative partial correlation between traditional masculinity ideology and the MIRS. For the gay men, of the demographic variables age and ethnicity significantly correlated with traditional masculinity ideology. The older the participant the more likely he would endorse traditional masculinity, and European American participants were more likely to not endorse traditional masculinity. Ethnicity and employment/student status significantly correlated with the MIRS. European American correlated positively with the MIRS, and being a student correlated negatively with the MIRS. After controlling for these three variables, a significant negative partial correlation was also found between traditional masculinity ideology and the MIRS. Thus, higher traditional masculinity ideology related to lower relationship quality for both gay and heterosexual men. No significant correlation was found between nontraditional masculinity and relationship quality.

Lastly, we explored the extent to which male reference group identity is a factor associated with relationship quality. We hypothesized the reference group dependent status and no reference group status would relate negatively to relationship quality whereas the reference group nondependent status would relate positively. Correlations between the Reference Group Identity Dependence subscales and the DAS and MIRS scales were examined. Results for the gay men and heterosexual men are presented separately in Table 2. For the heterosexual men, Reference Group Dependent significantly positively correlated with dyadic adjustment (DAS), and Reference Group Nondependent Diversity significantly negatively correlated with dyadic adjustment. Thus, contrary to our expectation the more reference group dependent the higher the relationship quality, and the more appreciation of diversity among males the lower the relationship quality. Additionally, No Reference Group significantly negatively correlated with the MIRS. For gay men, of the demographic variables age significantly negatively correlated with No Reference Group. After controlling for age, there was a significant negative partial correlation between No Reference Group and the DAS. Thus, for both heterosexual and gay men, having no reference group for one’s gender role self-concept was associated with lower relationship quality.

Regression Analyses

To evaluate the contribution and significance of the independent variables as a group (i.e., the four reference group identity dependence statuses, traditional masculinity, and nontraditional masculinity as predictors) to relationship quality, four hierarchical multiple regressions were performed, two each for the gay sample and the heterosexual sample. Correlations between the demographic variables and relationship quality scales were examined for significant relationships. Those variables that correlated significantly with MIRS or DAS were entered first in the regression equation.

Table 3 provides results of the regression analyses on MIRS scores for the gay and heterosexual samples separately. For the sample of gay men, race/ethnicity, employed versus student status, and currently dating significantly correlated with the MIRS and were therefore entered in the first step of the equation. These variables accounted for 30% of the variance and the model was significant. At the second step the independent variables accounted for an additional 14% of the variance and the model was significant, with no significant R 2 change. Being employed and currently dating were significant negative predictors of relationship quality. For the sample of heterosexual men, currently in a relationship (versus not currently in a relationship) significantly correlated with the MIRS and was therefore entered in the first step of the equation. Currently in a relationship accounted for 20% of the variance and the model was significant. At the second step the independent variables accounted for an additional 27% of the variance and the model was significant. The R 2 change was also significant. No Reference Group was a significant negative predictor, and currently being in a relationship was a significant positive predictor of relationship quality.

Table 3 Summary of regression analyses for variables predicting relationship quality: Modified Interpersonal Relationship Scale.
Full size table

Table 4 provides results of the regression analyses on DAS scores for the gay and heterosexual samples separately. For the sample of gay men, currently dating and currently in a relationship significantly correlated with the DAS and were therefore entered in the first step of the equation. These variables accounted for 15% of the variance and the model was significant. At the second step the independent variables accounted for an additional 24% of the variance and the model was significant. The R 2 change was also significant. Traditional Masculinity and No Reference Group were significant negative predictors, and being currently in a relationship was a significant positive predictor of relationship quality. For the sample of heterosexual men, being currently in a relationship and length of relationship significantly correlated with the DAS and were therefore entered in the first step of the equation. These variables accounted for 26% of the variance and the model was significant. At the second step the independent variables accounted for an additional 26% of the variance and the model was significant. The R 2 change was also significant. Nontraditional Masculinity, Reference Group Dependent, currently being in a relationship, and longevity of relationship (the longer the relationship) were significant positive predictors of relationship quality.

Table 4 Summary of regression analyses for variables predicting relationship quality: Dyadic Adjustment Scale.
Full size table

Discussion

The purpose of this study was to examine men’s perceptions of the quality of their romantic relationships, and the extent to which masculinity ideology and male identity were related to relationship quality. Both heterosexual and gay men were included in the study to examine whether there were differences in the quality of their romantic relationships, and what gender-related factors contributed to relationship quality in gay and heterosexual romantic relationships. Both gay and heterosexual men in this study perceived their intimate relationships to be of high quality. As measured in this study, relationship quality concerned whether the relationship had such characteristics as trust, genuineness, empathy, comfort with one another, expressions of affection, communication, agreement on important matters, doing things together, and general satisfaction with the relationship. Initially, no difference was found in the quality of relationships based on sexual orientation. However, when examining the groups based on whether the participant was a student versus working full-time, there was a significant difference between gay and heterosexual workingmen. Being a working gay man in this sample, as opposed to being an undergraduate student, related to lower relationship quality while this was not found in the heterosexual men. This finding may be particular to this sample, as there were only ten gay men who were employed and six heterosexual men who were employed.

In this study, traditional masculinity ideology was related to lower relationship quality for both gay and heterosexual men. Thus, men who are more traditional in their masculinity are likely to have more difficulty in their romantic relationships. This finding is consistent with previous research examining traditional masculinity and heterosexual relationships (e.g., Pleck et al. 1993; Sinn 1997; Truman et al. 1996). However, based on the regression analyses there was the new finding that indicates nontraditional masculinity ideology may contribute to having higher quality romantic relationships for heterosexual men.

With regard to male identity, it was found that a no reference group identity, characterized by a lack of psychological relatedness to and connection with other males, was related to lower quality romantic relationships for gay and heterosexual men. This result is consistent with male reference group identity dependence theory (Wade 1998), and with Erikson’s (1968) stages of psychosexual development. According to Erikson’s developmental model, identity must be achieved before intimacy can be successfully achieved. Indeed, research has found support for the link between identity development and intimacy (see Markstrom and Kalmanir 2001). In previous research an undifferentiated gender role identity related to low relationship quality (Kurdek and Schmitt 1986b), and the no reference group status related to an undifferentiated ego identity (Wade and Gelso 1998). With regard to male identity then, perhaps if one doesn’t have a clear sense of one’s identity as a man (i.e., feeling characteristically similar to other males) it may be difficult to achieve high quality romantic relationships.

It should be noted that the above-mentioned findings are consistent with previous research that shows few differences in those factors predictive of relationship quality in gay and heterosexual romantic relationships. However, there was a finding that was particular to the heterosexual men in this sample. We found that the reference group nondependent status, specifically the diversity factor, was associated with lower relationship quality. This result runs contrary to Wade’s (1998) theory. The diversity aspect of this male identity status is characterized by an appreciation of diversity among men. Items on this scale reflect men’s comfort with differences among men (e.g., race, age, sexual orientation, etc.) and association with different types of men, which is a characteristic in opposition to the male identity status of reference group dependent. In a previous research study by Wade and Brittan-Powell (2001), as well as in this study, the diversity factor was associated with nontraditional attitudes about masculinity. So, it is surprising that this status would be associated with lower relationship quality when nontraditional masculinity predicted better relationship quality. The converse finding to this one is the positive relationship found between the reference group dependent status and relationship quality, and this status was associated with traditional masculinity. The reference group dependent male identity status is associated with a conformist ego identity (Wade and Gelso 1998). The individual conforms to the standards and norms of the male reference group (Wade 1998). The results suggest that the male reference group norms are likely to be traditional masculinity, which is hegemonic. Perhaps in this sample of heterosexual men, conforming to the dominant standards of masculinity is to be just like other men. Having the sense that one’s masculinity is consistent with other men may contribute to the perception that the quality of one’s romantic relationship is also consistent with other men, i.e., that of high quality. However, it would be important to get the perspective of the women involved in the relationship with these men. It is possible that either the woman’s perspective may be contrary to the man’s, or her female identity and masculinity ideology may be complementary to that of the man’s. Future research could shed some light on this complex relationship dynamic.

Implications

The results of this study have implications for clinicians working with men. Traditional masculinity ideology was related to lower relationship quality for both gay and heterosexual men. For a man having marital or romantic relationship difficulties, it would be important to assess the extent to which attitudes about masculinity are affecting the quality of his relationship. If so, interventions would need to address such attitudes, possibly through examining gender role socialization experiences and the cost and benefits of traditional masculinity to his relationship. Case examples of psychotherapy with men whose masculinity ideology has affected their romantic relationships are provided in the book, In the Room With Men: A Casebook of Therapeutic Change (Englar-Carlson and Stevens 2006).

The no reference group male identity also was related to lower quality romantic relationships for gay and heterosexual men. This male identity status is associated with an undifferentiated ego identity (Wade and Gelso 1998), which suggests there may be other areas of identity in which there is confusion (e.g., racial, sexual, religious, etc.). However, with respect to male identity it would be important to assess the extent to which the client feels connected to, and identifies with other men. Feeling essentially different from other men could potentially effect how secure and confident he feels if and when he is involved in a romantic relationship, thereby affecting relationship quality. Interventions would involve helping the client identify other men with whom he has similar qualities, and encouraging developing peer relationships with similar men. For some young men, it may even be necessary to have male friends for support when initiating a romantic relationship.

The results also have relevance for research on the psychology of men and masculinity. The majority of research has focused on traditional masculinity ideology and gender role conflict as they relate to relationship quality. The results of this study indicate nontraditional masculinity and male reference group identity are other possible explanatory constructs that have utility for research on gay men and heterosexual men in romantic relationships.

Limitations and Future Research Suggestions

This exploratory descriptive study found significant relationships between the variables examined. However, it is important to note that the findings of this study do not mean that there is any causality between any of the factors for which correlations were found. Further, many of the correlations were small leaving a lot of the variance unaccounted for. Additionally, the internal consistency reliability for the Nontraditional Masculinity Scale and Reference Group Dependent Scale may have attenuated the correlations these measures had with other scales. Although generally consistent with what has been reported in other research studies that have used these scales (Levant and Richmond 2004; Liu 2002; Wade and Brittan-Powell 2000, 2001; Wade and Gelso 1998), the internal consistency reliability was less than the reliability coefficient of .70 that is considered “acceptable” in most social science research situations. Future studies should attempt to discover what other factors contribute to romantic relationships in gay and heterosexual persons and couples, using more sophisticated statistical models. In order for this study to have greater significance and generalizability, a larger sample size with a diverse sample is needed, perhaps by using a web-based study. Additionally, it would be important to examine couples for both partner’s perspectives on the quality of their relationship. Future research in this area would provide insights into the gender-related characteristics that contribute to having quality romantic relationships.