Background and Aim: Gay men’s identity harbors a secondary sexual role or self-label that can affect many aspects of their lives. Studies have shown that many gay men express a secondary self-label (i.e., top, bottom, and versatile) based on their role during anal intercourse. Considering the unwelcoming social atmosphere and religious and legal restrictions in Iran regarding the issues related to LGBT people, a few studies have been conducted on attachment styles and the quality of relationships with primary caregivers in this sexual minority in Iran. This study hence aimed to compare gay men playing different sexual roles with their heterosexual peers in attachment styles.
Materials and methods: In a causal-comparative research, 197 gay men (30 top, 36 bottom, and 131 versatile) and 49 heterosexual men were selected by snowball and purposive sampling methods to fill out the Revised Adult Attachment Scale (Collins & Read, 1996).
Results: The findings showed that there was a significant difference between versatile gay men and heterosexual men in the avoidance attachment style, as heterosexual men gained a higher mean score in this attachment style. There was also a significant difference between the gay men with different sex roles and heterosexual men in the anxious-ambivalent attachment style, as the bottom gay men obtained the highest mean scores.
Conclusion: The imitation of the generally accepted masculinity criteria by Iranian gay men in an attempt to avoid rejection from parent and peers increases their anxiety levels and may leads to the emergence of the anxious attachment style in them.
 

Despite the remarkable scientific advances of Iranian researchers in various psychological fields, few studies have been conducted on homosexuality and issues related to this sexual minority mainly due to cultural, social, and political issues in today’s Iranian society; for example, same-sex relationships are considered a crime and subject to strict punishments. This research gap has deprived the Iranian society of psychologists of sufficient and accurate information about this sexual minority and made it difficult for psychiatrists to effectively deal with gay clients and identify their problems. The personal prejudices and biases of therapists, such as homophobia, resulting from living in a patriarchal society sometimes add to the complexity of such situations and problems of gay clients (Pourmohseni Shakib, 2021).

To get to know this sexual minority as much as possible in the context of Iranian society, it is necessary to first address the issues that introduce us to the most fundamental worldview of members of this minority, i.e., how they look at themselves, others, and the surrounding world. Accordingly, this study compares gay men with different sex roles with their heterosexual (straight) peers in terms of attachment styles. Homosexuality is a pattern of romantic, emotional, and sexual attraction between people of the same sex (American Psychological Association, 2019). The identity of gay men includes a secondary sexual self-label that can affect many aspects of their lives, from physical characteristics to preferences for choosing an emotional-sexual partner (Moskowitz, 2008). Sociological, psychological, and general health studies have shown that many gay men express secondary sexual self-label (i.e., top, bottom, and versatile) based on the role they play during anal intercourse, i.e., their sex role (Moskowitz, 2011). It is noteworthy that sex roles deal with how one describes themselves and affect their preferences for sexual positions, while sexual positions in anal relationships refer to sexual practices and behaviors (Kippax & Smith, 2001; Johns & Pingel, 2012). During a same-sex relationship, sexual partners may take different sex roles. This simple rule allows us to distinguish two distinct patterns of sex roles in homosexuals as follows:

1. Those who establish themselves in only one of the two sex roles over time:

• Insertive partner (top): those who penetrate their penis into their partner’s anus or mouth or prefer Insertive roles.
• Receptive partner (bottom): those who receive penetration during anal or oral sex and prefer receptive roles.

1. Those who play both insertive and receptive roles, called versatile (Goodreau & Peinado, 2007; Mcgill, 2014).

One of the differences between homosexuals and heterosexuals is the greater importance of romantic and social relationships for homosexuals (Mohr, 2003; Grossman, D’augelli, & Hershberger, 2008). Despite all the general stereotypes about same-sex relationships, many gay men and lesbian women usually live together for a long time. The data obtained from different resources indicate that 30-60% of gays and 45-80% of lesbians are always engaged in a monogamous romantic relationship (Elizur, 2003; Allan, 2018). A study conducted by Colgan (1987) showed that many homosexuals face intimacy dysfunctions originating from problems with interpersonal communication, unresolved intrapsychic issues, interpersonal stress, and behavioral patterns formed to deal with unresolved stress. According to these concepts, identity and functional disorders in intimate relationships are regarded as efforts made by a person to recover their state of well-being.The attachment theory deals with the need to make close relationships with others. John Bowlby (1969) states that this pervasive need is an evolutionary advantage to ensure closeness between the child and the caregiver in times of danger, anxiety, and ambiguity. Accordingly, a child’s experiences in the relationship with a caregiver form internal working models that indicate how one interacts with themselves and others and how interprets the surrounding world in the face of stressful and threatening situations (Allan, 2018). It is noteworthy that researchers classify attachment in different ways. For example, based on the child's expectations of the caregiver's presence and responsiveness, a sense of comfort to build an intimate relationship with independence and an attitude about being lovable, Ainsworth (1978) introduced three attachment styles: secure, avoidance, and anxious-ambivalent. Although there is a long history of studies on the relationship between attachment styles and the quality of one’s relationships in adulthood, most of these studies have focused on heterosexuals and a few of them have dealt with homosexuals. This is the result of the sovereignty of heteronormative systems in society (Mohr, 2008). Based on a scoping review conducted by Allan (2018), most studies conducted on this subject can be divided into 4 categories. The first category is called “universal attachment dynamics” because they emphasize the universal and dynamic nature of attachment regardless of sexual orientation. Such studies, in fact, argue that homosexuals and heterosexuals share the same structure of attachment dynamics. For example, the findings of Ridge (1998) showed that the frequency of attachment styles in heterosexual and gay samples was similar. The second category, which is known as “particular attachment dynamics”, includes studies emphasizing that gay men face attachment dynamics different from those of heterosexual men due to their special experiences such as homophobia and rejection. For example, Landolt (2004) investigated the independent effect of being rejected by father and peers on predicting anxious attachment in gay men, and Shenkman (2019) studied the relationship between minority stress and higher levels of avoidance attachment style. The third category, known as “the impact of attachment narratives”, includes studies that investigate the effects of problems such as homophobia, homonegative expressions, and heterosexism on patterns of romantic relationships among gay men.For example, Sherry (2007) studied the relationship between internalized homophobia and adulthood attachment and reported that insecure attachment style exhibited the strongest relationship with internalized homophobia, shame, and sense of guilt. The fourth category, titled “Monogamish”, includes studies on attachment dynamics and non-monogamous relationships of gay men. For example, Ramirez (2010) found no sign of avoidance attachment style in gay men with open relationships and Mohr et al. (2013) showed that there was a negative relationship between open relationships and the level of satisfaction with a relationship when a person or their sexual-emotional partner are suffering from mild to severe anxiety. This is consistent with the findings of similar studies conducted on heterosexual couples.Given all challenges toward gay men in Iran, this study hypothesized attachment style of self-identified gay men with different sex roles (top, bottom, and versatile) would differ from attachment style of self-identified heterosexual.

Participants and Procedure

The statistical population consisted of Iranian gay and heterosexual men, especially those living in Rasht and Tehran, in 2019-2020. In a causal-comparative research, gay men and heterosexual men were selected by snowball and purposive sampling methods and compared in term of different adult attachment styles. The reason for selecting these two cities was the easier in-person sampling and data collection. Before the official beginning of the study, the first author met and befriended several gay men by attending their private parties and tried to convince them to participate in a study about the issues and problems related to gay men. The author received a warm welcome from the gay men he met. In addition, the author signed up on a gay dating app named Hornet and followed the gay groups and channels on Telegram, WhatsApp and Instagram. However, the socio-cultural-judicial constraints on homosexuality as well as the abuse of members of this minority by some sexual and mental abusers led to their distrust and unwillingness to participate in the research either in-person or online. There fore, an online questionnaire was developed on Google Forms to be filled out without the need for personal information. The link of this online questionnaire was sent to the participants in two different ways. In the first method, based on snowball sampling, the link was sent to Telegram or WhatsApp accounts of gay men who were identified at parties and private circles and they were asked to send the link to other gay men they knew. In the second method, based on purposive sampling, the author found the personal accounts of gay men on social networks and then sent the questionnaire link after making an introductory interview and obtaining their informed consent. Considering the atmosphere of fear about homosexuality in Iranian society, only 332 (24.11%) questionnaires of the total 1377 questionnaires sent to participants were filled out. The inclusion criteria were being physiologically male based on self-report, self-expression about sexual orientation (gay and heterosexual), and informed consent. In addition, the exclusion criteria were being bisexual or transsexual based on self-report, being physiologically female based on self-report, and being aged under 21 years. Based on the inclusion and exclusion criteria, 49 heterosexual men and 197 gay men (including 30 top, 36 bottom, and 131 versatile) were selected to enter the study. It is noteworthy that the convenience sampling method was used to select participants from among the matched heterosexual men.

Measurement tools:

A personal information form was used to collect data on demographics and sexual self-label of participants.  The demographics section contained questions about gender (male, female, or transgender), educational attainment (junior high school, high school diploma, associate’s, bachelor’s, master’s, doctorate, and post-doctorate), age group, order of birth (first, middle, last, or single child), emotional relationship status (engaged, not engaged, or open relationship), marital status (single, married to a woman, or divorced), and email address for receiving the general recommendations of  the study. In the sexual self-label section, the participants were asked three simple questions to determine their sexual orientation and sex role. The first question, which aimed to determine the sexual orientation of participants, consisted of three items as follows: heterosexual (attracted to people of the opposite sex), gay (attracted to people of the same sex), and bisexual (attracted to people of both sexes). To add to the accuracy of the information, the second questions was about their romantic (“Exclusively heterosexual”, “Predominantly heterosexual, only incidentally homosexual”, “Predominantly heterosexual, but more than incidentally homosexual”, “Equally heterosexual and homosexual”, “Predominantly homosexual, but more than incidentally heterosexual”, “Predominantly homosexual, only incidentally heterosexual”, and “Exclusively homosexual”). If the participants select one of the items from 0 to 6, they are regarded as heterosexual [0 or 1], bisexual [2 to 4], and gay [5 or 6], respectively (Besharat et al., 2016). To determine the sex role of participants, they were asked which of the sex roles they could better describe. This third question consisted of three items for homosexuals (top, bottom, or versatile) and one item for heterosexuals (none).

Revised Adult Attachment Scale (RAAS): This scale was developed by Collins and Read in 1990 to measure attachment styles. The RAAS initially consisted of 21 items that were later (1996) reduced to 18 items. The items are scored based on a 5-point Likert scale (from Not at all characteristic of me to Very characteristic of me). The items of this scale, such as “In relationships, I often worry that my partner does not really love me”, measure three attachment styles of secure, avoidance, and anxious-ambivalent. Secure attachment style: The caregiver gives predictable and sincere answers to the child and leaves him/her while the child is confident about the caregiver’s availability. Avoidance attachment style: This attachment style refers to the situation in which the child realizes that the caregiver is emotionally distant and physically inaccessible. Avoidant children have little inclination to rely on their caregiver when needed, always keeping a distance between the caregiver and themselves. People with this attachment style believe that no one is available to help them when they are in stressful and threatening situations. Anxious-ambivalent attachment style: This attachment style refers to the situation in which children feel that the caregiver is not aware of their needs and are faced with unpredictable responses from them. Such children are more attached to their caregiver, ask them greater demands, and show less desire to explore the world around them. This attachment style is usually observed in individuals who are not comfortable with emotional closeness (Allan, 2018). This scale also measures three other factors using 18 items. The first factor is “anxiety” that deals with stresses such as fear of being abandoned and not being loved in a relationship, the second factor is “dependence” that measures one’s degree of trust in others and their availability, and the last factor is “closeness” that examines how one feels uncomfortable with intimate relationships (Teixeira, 2019). Collins and Read assessed the reliability of this scale by the repeatability test on a sample of 101 members who filled out the scale at an interval of two months. The correlation between scores on secure, avoidant, and anxiety-ambivalent attachment styles was obtained 0.68, 0.72, and 0.52, respectively (Collins and Read, 1990). In Iran, Pakdaman et al. assessed the reliability of this scale by the repeatability test on a random sample of 100 male and female junior high school students who filled out the scale twice at an interval of one month. The results showed that this scale was reliable at a 95% level of confidence. The construct validity of this scale was also evaluated by divergent or diagnostic validity. At the 0.001 level of significance, the correlation coefficient between subscales was -0.313 and -0.336, respectively. It is noteworthy that the correlation coefficient between “closeness” and “dependence” was equal to 0.264 at the 0.014 level of significance (Pakdaman, 2001; Tardast, 2015).

Statistical analysis:

All statistical analyses were performed in SPSS-24. The multivariate analysis of variance was employed to compare different levels of attachment styles in three groups of gay men and a group of normal heterosexual men. Since matching groups no confounding demographic or background variable was found, there was no need to control their values by analysis of covariance. Considering the sample size of subgroups and the homogeneity of variances, the Games-Howell post-hoc test was used to compare the mean difference between the four groups.

The data obtained from 332 participants were statistically analyzed. Table 1 shows the frequency distribution and percentage of sexual orientation by the gender of participants separately

Table 1: Frequency distribution and percentage of sexual orientation by the gender of participants (n=332)

Based on the exclusion criteria, the participants who stated that they were female or transsexual and the males who reported a bisexual orientation were excluded from the study. Of the remaining 208 gay men, 3.3% of them due to being under 21 years and 1.9% of them because of being categorized as bisexual (based on their answers to the question about romantic affairs) were excluded from the study. As a result, a total of 49 heterosexual men and 197 gay men advanced to the next stage. The results indicated that 30 (15.2%), 36 (18.3%), and 131 (66.5%) of the gay men were playing the role of top, bottom, and versatile, respectively. Table 2 presents the frequency distribution and percentage of age groups for each sexual orientation and role.

Table 2: Frequency distribution and percentage of age groups for each sexual orientation and role

Table 3 shows the frequency distribution and percentage of marital status for each sexual orientation and role. The results showed that the highest 

rates of marriage and separation were observed among versatile gay men (6.9%) and top gay men (3.3%), respectively.

Table 3: Frequency distribution and percentage of marital status for each sexual orientation and role

Table 4: The mean and standard deviation of attachment style in studied groups

Levene’s test was employed to compare the research variables in terms of the homogeneity of variance. The results indicated that the F-statistic of Levene’s test to evaluate the homogeneity of variance of variables in study groups was not statistically significant for the avoidance attachment style (F=1.28, P=0.279) and the anxious-ambivalent attachment style (F=1.85, P=0.138). This means that the variance of these variables was homogeneous in the studied groups. By contrast, the F-statistic of Levene’s test was statistically significant for the secure attachment style (F=4.54, P=0.004). Allen and Bennett (2008) suggest that if the homogeneity of variances is not established for one or more dependent variables, it is better to use a stricter alpha or significance level, such as 0.001 than 0.05. Therefore, the significance levels of the -statistic of Levene’s test were processed based on the suggestion of Allen and Bennett (2008).

The Box's M test was used to investigate the homogeneity of the variance-covariance matrix of dependent variables in the studied groups. The results demonstrated that the F-statistic of Box's M test was statistically significant for all attachment styles (F=1.72, Box’s M=32.054, P=0.029). Allen and Bennett (2008) suggest that the Box's M test is resistant to the heterogeneity of variance-covariance matrices when the sample size of each group is greater than 30. In this study, the sample size of all gay and heterosexual groups was greater than 30.

The Games-Howell post-hoc test for unequal variance was also employed for pairwise comparisons of means. The results of Wilks' lambda multivariate analysis of variance showed that there was a significant difference between the groups in attachment styles at the 0.0001 level of significance (Wilks' lambda=0.87, F(9.584632)=3.75, P<0>

Table 5: One-way analysis of variance on attachment styles for difference between the groups

The data contained in Table 5 show that the F-statistic of avoidance style (4.05) and anxious-ambivalent style (7.81) were statistically significant at the 0.01 level. This means that there was a significant difference between the study groups in this attachment style. However, the F-statistic of secure attachment style (F= 0.95, P>0.05) was not statistically significant. Considering the effect size of the avoidance attunement style (η2= 0.04) and the anxious-ambivalent attachment style (η2= 0.08), it can be stated that the difference between the population members in these two attachment styles was at a moderate level. Figures 1 and 2 presents the results of pairwise comparisons between the studied groups in the mean scores of avoidant and anxious-ambivalent attachment styles.

ell post-hoc test were statistically significant for avoidant and anxious-ambivalent attachment styles. The results of this post-hoc test indicated that there was a significant difference between versatile gay men and heterosexual men in the avoidant attachment style. In addition, there was a significant difference between top, bottom, versatile gay men with the heterosexual men in terms of the anxious-ambivalent attachment style (P=0.02, P<0>

The study findings revealed that there was a significant difference between versatile gay men and heterosexual men in the avoidance attachment style, as heterosexual men obtained higher mean scores. A high score on the avoidance attachment style means that one evaluates others as controller, hostile, and inattentive. Accordingly, such individuals do not find physical relationships and emotional intimacy soothing and have any curiosity about other people's psycho-inner world. Such people are self-sufficient, self-confident, and independent and do not make any requests for help and deal with anxiety by ignoring emotions in stressful situations. In fact, the more internally angry they are, the more socially isolated they will be. Based on the study results, the level of avoidant attachment style was lower in versatile gay men compared to heterosexual men. This is not consistent with the findings of Shenkman and Boss (2019) and Shenkman and Stein (2021) who reported a higher level of the avoidant attachment style in gay men compared to their heterosexual peers.In a qualitative study, Gil (2007) showed that the level of internalized homophobia was lower in versatile gay men than their peers who play the top role. Moreover, versatile gay men assumed that they were psychologically, emotionally, and sexually more flexible than their peers who play other roles (top and bottom). It is noteworthy that internalized homophobia actually refers to the misconceptions internalized by homosexuals. Such misconceptions are the result of living in a society that values only heteronormative standards and undervalues the experiences of people with different sexual orientations. This exposes such people to negative feelings about themselves and their sexual orientation and even makes them reject their sexual identity and orientation (Frost and Meyer, 2009; Herek and Mclemore, 2013).Since internalized homophobia has a significant relationship with self-esteem, emotional stability, and self-acceptance (Ross and Rosser, 1996; Rowen and Malcolm, 2003), the low level of internalized homophobia in versatile gay men can be attributed to their increased self-acceptance and self-esteem. Therefore, versatile gay men enjoy a higher level of mental health compared to their peers who play other sex roles. Since versatile gay men do not have stubborn preferences in their sexual behaviors and are more flexible than their peers playing other sexual roles (Hart, 2003), it can be assumed that this group of gay men has more opportunities to interact with the surrounding world, both sexually and socially, resulting in their higher flexibility and adaptation to societal adversity. Iran is a patriarchal society while Gilligan (2018) and Chu (2014) explain patriarchy as an order of living that privileges some men over men, for instance, straights over gays. In patriarchy, from a young age, men learn the codes of masculinity contingent on the suppression of empathy and hiding of their vulnerability necessary for claiming superiority, and by shielding their relational desires and sensitivities, they wish to become a part of the boys’ community. Otherwise, they would not be accepted due to being seen as girly or gay. In other words, there is the internalization of the masculine taboo on tenderness which encourages men to cover their emotional vulnerabilities. So, patriarchy has roots in the separation of the self from the relationship and paradoxically men have to sacrifice their relationships with self and emotions in order to have “relationships”. The price of acceptance into patriarchal order is “The Loss” and the only way to guarantee security toward it is by sacrificing the freedom of intimacy. Furthermore, in this world, being a man means being self-reliant, emotionally stoic, and independent. However, Bowlby's observation depicts that this independence not only isn’t manhood but also is a kind of detachment that can be mistaken for maturity, because it mirrors the pseudo-independent of manhood which in patriarchy is synonymous with being fully human. It can show why heterosexual groups achieve higher mean scores in avoidant attachment style in the current study. This study showed that there was a significant difference between gay men of all three sex roles (top, bottom, and versatile) and heterosexual men in the anxious-ambivalent attachment style. However, the gay men playing the bottom sex role obtained higher scores. The higher mean score of gay men of all three sex roles in the anxious-ambivalent attachment style compared to their heterosexual peers indicates that such individuals greatly need physical-emotional intimacy and usually experience a high level of anxiety in establishing and maintaining intimate relationships, while their relationships have no or negligible effect on reducing their anxiety. Members of this sexual minority are often in turmoil between approaching and avoiding and usually experience no two-way communication; such people experience a complex mix of negative emotions such as sadness, fear, self-criticism, and disability. This finding is consistent with the results of Nematy (2016) who reported the high level of the anxious-ambivalent attachment style in Iranian homosexual men and women and bisexuals compared to their heterosexual peers. However, this result is not consistent with the findings of Ridge (1998) who stated that there was no significant difference between homosexuals and heterosexuals in the frequency of attachment styles and also the findings of Mohr (2008) who reported that there was no significant difference between gay men and their heterosexual peers in attachment styles and intimate relationships. On the one hand, previous studies have shown that most of gay men exhibit greater childhood gender nonconformity than their heterosexual peers do, indicating the relative dominance of feminine behaviors over masculine behaviors. There is a correlation between gender nonconforming in gay men and a poor father-child relationship. The above-mentioned correlation can be attributed to the fact that it is difficult for fathers to accept the gender-nonconforming of their gay children. It should be noted that gay men are more likely than their heterosexual peers to be rejected by and isolated from their fathers in childhood. It is a factor that can independently predict the emergence of the anxious-ambivalent attachment style in adulthood (Bradley, 1989; Lytton, 1991; Bailey & Zucker, 1995; Beard & Bakeman, 2000; Landolt, 2004).Given that Iranian gay grow up in a society dominated by the traditional culture of patriarchy and based on the common parenting styles in Iranian society, parents especially fathers play a very important role in making major decisions about their children's lives and children may live with their parents until marriage or even middle age. So, Iranian gay men are more prone to parental rejection (Nematy, 2016). Under these circumstances, gender nonconformity can increases the chance of losing family support and leads to emerge the anxious attachment style in such people.

On the other hand, the early parent-child relationship is not the only source of information to determine the attachment style of gay men (Allan, 2018). The results of Landolt (2004) showed that the rejection of gender nonconforming children not only from their parents, but also from their peers can lead to the formation of the anxiety attachment style in adulthood. Gay men also have a strong desire to explore and nurture their identity in a social context that allows them the opportunity to do so. Community and peers can serve gay men as a caregiver who helps a child to develop a lovely and secure self. In addition to relationships with parents, relationships with peers can independently affect the attachment style of gay men. Childhood gender nonconformity can affect a gay man’s relationships with his peers. The boys who exhibit cross-gendered behaviors from early childhood are usually punished by their peers. Many gay men are brutally harassed because of this gender nonconformity; some of them have reported that they were usually rejected by their peers from childhood to adolescence. Moreover, peer rejection can mediate childhood gender nonconformity and the anxious attachment style. It can be hence concluded that gender nonconforming is closely related to anxiety in intimate relationships because it can lead to peer rejection and, thereby, increase the anxiety level (Saghir, 1973; Fagot, 1977; Landolt, 2004; Allan, 2018).In another study conducted by Sherry (2007), it was shown that the preoccupied-fearful (anxious-ambivalent) attachment style exhibited the strongest relationship with internalized homophobia, resulting in poor communication performance, less satisfaction with relationships, and establishment of less intimate relationships. In Iranian patriarchal culture, the male gender is manifested by two rules: (1) A man must be attracted to the opposite sex, and (2) a man must be homophobic (Eslen-Ziya, 2016). Furthermore, this culture not only relates masculinity to the exhibition of behaviors conforming to gender-specific stereotypes but also considers a lower social position for women than men. So, Iranian gay men are perceived more feminine and are accused of being “less of a man.” To compensate for this social view, gay men internalize homophobic behaviors, such as negative feelings towards their feminine side and that of the other gay men, to create a safe haven for their masculinity. The imitation of the generally accepted masculinity criteria by Iranian gay men in an attempt to avoid rejection from parent and peers increases their anxiety levels and leads to the emergence of the anxious attachment style in them.

Research limitations and recommendations

The main strength of this study was that it was the first research about the comparison of gay men playing different sex roles with their heterosexual peers in attachment styles. However, due to the methodological and theoretical limitations, the study findings should be generalized very cautiously. The first research limitation was non-random sampling; due to the existing social, cultural, and political conditions of Iran regarding issues related to sexual minorities, it is almost impossible to use random sampling in this sexual minority. The second research limitation was that an online questionnaire was used to collect data in order to ensure the safety of the participants; considering the great fear of the members of this sexual minority about being identified and punished, those who have filled out and sent the questionnaire probably enjoy specific personal features such as higher educational attainment, better socioeconomic status, higher levels of self-disclosure, and less avoidance. This is a hypothesis that needs to be further examined in the future. The third research limitation was related to online surveys in the above-mentioned apps (i.e. Hornet, Telegram, WhatsApp, and Instagram); the false self-expression of people about their sexual orientation and identity in cyberspace may affect the study results. This is especially true for people who introduce themselves as full-top. However, a preliminary interview was conducted in this study to identify non-false profiles to increase the accuracy of the results. Since the author used the Internet and dating software applications for purposive sampling, the study sample included only a small proportion of people belonging to this sexual minority, and it was not possible to access a large number of gay men who were not active in social networks for various reasons such as old age or lack of Internet access. The demographic questionnaire used in this study for self-label sex roles did not include options for identifying and separating versatile-top and versatile-bottom gay men. Therefore, these two groups of gay men were categorized as a single group named “versatile”. However, future studies are recommended to develop a measurement tool to classify and prioritize different sex roles of gay men. This study opens the door for a number future exploration. For instance, comparing parent-child relationship of three main sex roles of gay men with heterosexual men. Furthermore, comparing internalized homophobia of three main sex roles of gay men in different populations and various religions. Finally, comparing psychological flexibility of three main sex roles of gay men in different populations and in comparison, with heterosexual men.

Acknowledgments

This paper was extracted from a certified master's thesis presented by Seyed Mohsen Pourmohseni Shakib at the University of Guilan (No. 136292, January 12, 2021). The authors would like to thank all the participants who helped us in this research.

Conflict of Interest

The authors declare no conflicts of interest. This study did not receive any funding.

Disclosure statement

No financial interest or benefit was derived from the application of this research

Ethical approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. Ethical approval for the study was obtained from the Council of postgraduate courses at university of Guilan (No. 136292, January 12, 2021) and the Local Research Ethics Committees. Informed consent was obtained from the participants.