Abstract
Wellness, outness, and sexual orientation identity were examined across multiple demographic categories in a large sample of gay and lesbian parents. No significant differences were found in terms of wellness. However, gay male parents reported higher need for protection, acceptance, internalized homophobia, and more difficult process of identity formation. Older parents reported a more difficult coming out process. Parents of color were less out to their faith communities, had higher identity confusion, but lower need for privacy.
Keywords
An estimated 6–14 million children are being raised by gay or lesbian parents in the United States and an estimated 37% of lesbian, gay, and bisexual identified adults have had a child at some point in their lives (Gates, 2013). In the past decade, there have been a number of influences on gay parenting across the nation and internationally, including increases in media attention, changes in discriminatory fostering and adoption laws, and advances in marriage equality. Increased attention has been due in part to larger numbers of visible gay men and lesbians becoming parents in the context of committed same-sex relationships (Gates & Ost, 2004) and a general increase in diverse family configurations (Hicks & Lee, 2006). In fact, a 2010 Gallup poll indicated that 52% of Americans considered same-sex relationships to be morally acceptable, compared to only 40% in 2001 (Saad, 2010). Despite increased acceptance, gay and lesbian parents continue to face obstacles and prejudice in their abilities to become parents, to protect their children from bias and stigma, and to secure legal and financial protections for their families.
Many adults perceive parenthood as a significant life experience. Research concerning gay and lesbian parenting has focused on two primary areas, on either parental attitudes and behaviors or the child’s well-being based on psychosexual, social, and emotional development. Historically, this research compares gay and lesbian parents with their heterosexual peers. For more than 20 years, research has focused on finding the difference between sexual minority and heterosexual parents (Fulcher, Chan, Raboy, & Patterson, 2002; Patterson, 2006; Wainright, Russell, & Patterson, 2004), not on describing gay and lesbian parents or their needs. Within this line of inquiry, parallels can be drawn to deficit modeling around race. The research questions assume a deficit in the ability of gay and lesbian people to parent adequately. It also places heterosexual parents as the gold standard by which others are to be measured (Riggs, 2006). This type of inquiry persists, despite conclusions from a broad survey of research findings that indicates children raised by gay and lesbian parents fair as well as those raised by their heterosexual counterparts (American Academy of Pediatrics, 2002; Farr, Forssell, & Patterson, 2010; Goldberg & Smith, 2013; Shechner, Slone, Lobel, & Schecter, 2013; van Gelderen, Bos, Gartrell, Hermanns, & Perrin, 2012). Researchers have concluded no difference in the parenting outcomes between same-sex and heterosexual parents (Allen & Burrell, 1996; McNeill, Rienzi, & Kposowa, 1998), with the exception that children raised by same-sex parents tend to be more open to diversity than their heterosexual counterparts (American Academy of Pediatrics, 2002). Additionally, few differences have been found when comparing lesbian and gay parents to heterosexual parents in terms of psychological distress, depression, parenting competence, and perceived parenting stress (Goldberg & Smith, 2009; Shechner et al., 2013). Despite attempts by these researchers to better understand gay and lesbian parents, numerous questions remain including: (a) Who are these parents? (b) What are their coping and wellness practices in the face of continuing discrimination and prejudice? (c) What are the physical, psychological, emotional, and social needs of these families? and (d) how do helping professionals offer support?
Researchers identify wellness as “a way of life oriented toward optimal health and well-being in which the body, mind, and spirit are integrated by the individual to live more fully within the human and natural community” (Myers, Sweeney, & Witmer, 2000, p. 252). As such, stress is in direct opposition to wellness. Stressors such as unemployment, education level, oppression, discrimination due to sexual orientation, unhealthy eating habits, and many more areas can have a deleterious influence on wellness and be a potential source of anxiety for parents. In studies surveyed by Dew & Newton (2005), sexual orientation identity and the influence of internalized homophobia were leading predictors of overall wellness. No studies could be found that examined wellness in gay and lesbian parents and there appears to be a dearth of research about the relationship between the parental wellness and the wellness of children in the general population.
The effects of oppression and discrimination on gay and lesbian people have been well-documented (Hendren & Blank, 2009; Hicks & Lee, 2006). The consequences can be seen in higher rates of substance abuse and higher rates of suicidality among lesbian, gay, bisexual, transgender, queer, and questioning (LGBTQQ) youth. Emotional consequences can include struggles with anxiety and depression (Frost & Meyer, 2009), internalized shame, fear for physical safety, and fear of rejection, which over a lifetime can have a major influence on self-acceptance and self-esteem.
Moreover, researchers suggest that stress is associated with parental perceptions of the child’s outcomes. For example, Semke, Garbacz, Kwon, Sheridan, and Woods (2010) explained the correlation between parental stress and the parents’ perceptions as a belief that one is lacking support in meeting specific needs of the child as well as being an effective role model. However, when parents perceive themselves as supporters in their children’s outcomes, have parental efficacy, and are more optimistic, the effects of stress are decreased (Lassiter, et al., 2006; Semke, Garbacz, Kwon, Sheridan, & Woods, 2010). These factors also appear to become contributing factors to higher levels of wellness in their children (Gibbons, Gerrard, Cleveland, Wills, & Brody, 2004). In addition to various parental stressors, oppression adds a supplementary obstacle for parents to overcome. Researchers suggest the effects of oppression, such as discrimination and power differential, are felt by both the child and the parent (Gibbons et al., 2004). Although minority parents are faced with the strains of oppression, many continue to strive to free their children from the distressing effects (Flouri, 2004).
Goldberg and Smith (2011) explored the longitudinal effects of accepting and nonaccepting environments (workplace, state laws, family, friends, and neighborhoods) on new gay and lesbian parents and found profound impacts on overall mental health, depression, and anxiety related to the perceived support. Many gay and lesbian parents are concerned about the possibility of their children being persecuted or harmed because of their family structure (Bennett, 2003; Johnson & O’Connor, 2002). In one study, over 75% of lesbian mothers worried that their children might be harassed or teased because of having same-sex parents (van Dam, 2004). Becoming a parent as a lesbian or gay person requires higher levels of outness. Living within the realities of a heterosexist society, gay parents are keenly aware of the need to be strong role models of self-acceptance (Lassiter, Dew, Newton, Hays, & Yarbrough, 2006) and to exhibit a sense of pride in their family configuration. At the same time, they are well aware of the need to prepare their children to handle possible discrimination. Many parents purposefully model openness and pride in their interactions with people outside of the family and are careful not to ignore assumptions of parental heterosexuality and homophobic remarks (Schacher, Auerbach, & Silverstein, 2005). Gay parents must balance this need to empower their children with their own level of outness, self-acceptance, and comfort in a given situation.
Gay and lesbian parents are in essence constantly engaging in stigma management on parallel levels—for themselves as gay or lesbian people, for their same-sex relationship, and for their children. More than half (53%) of lesbian and gay parents surveyed (Kosciw & Diaz, 2008) reported various forms of exclusion from their children’s school communities in formal and informal ways (school policies, events, and activities) and 26% reported being mistreated by other parents. Given their particular community context, gay and lesbian parents must balance the advantages of being out and proud with concern for their children’s safety and well-being. The impact of this pressure is unknown. Surely, these conditions create unique patterns of coping and parenting that are important to understand.
Although an extensive amount of literature exists for child outcomes, little research has been conducted to understand the factors that relate to wellness in lesbian and gay parents. Previous research found that sexual orientation identity and ones level of outness (Coleman, 1982; Fassinger & Miller, 1996) impact the degrees of wellness in sexual minorities. However, nothing is known about the general profiles of lesbian and gay parents and what demographic factors are associated with their overall wellness or which factors make them most at risk. The purpose of this study is to examine factors of wellness, levels of outness, and sexual orientation identity across multiple demographic categories in a large population of gay and lesbian parents. Examining levels of self-acceptance, wellness, and degrees of outness may help us understand more about the overall functioning of gay and lesbian parents under the pressures of parenting within a heterosexist society. Specifically, our research question regarding gay and lesbian parents was as follows: Are there significant differences in wellness, sexual identity and outness based on race, age, income, or community setting?
Method
Participants and Procedures
Participants were gay or lesbian parents (N = 305) who were surveyed at a large gay pride event in a southern urban setting. Of the participants, 220 (72%) were female, 77 (25%) were male, and 8 (3%) did not identify by sex; 224 (73%) were Caucasian, 27 (9%) African American, 21 (7%) Native American, 9 (3%) Hispanic, 9 (3%) Alaskan Native, 1 Pacific Islander, and 14 (5%) reported other. Additionally, 57 (19%) were age 30 or younger, 67 (22%) were age 31–37, 77 (25%) were age 38–45, and 48 (16%) were age 46 or above. In terms of annual income, 28 (9%) reported earning less than US$20,000, 78 (26%) reported earning US$20,001–40,000, 73 (24%) reported earning US$40,001–60,000, 48 (16%) reported earning US$60,001–80,000, 31 (10%) reported earning US$80,001–100,000, and 43 (14%) reported earning more than US$100,000 per year. In terms of where they lived, 111 (36%) reported living in an urban setting, 121 (40%) reported living in a suburban setting, and 64 (21%) reported living in a rural setting. Nine (3%) did not choose to report their setting.
The data for this study were collected utilizing a convenience sample at a gay pride festival in a large southern city. A large sign was placed at a booth inviting people who self-identified as gay or lesbian parents to complete a survey. Participants were asked to complete the three instruments and a demographic form. The demographic sheet included questions regarding ethnic/racial identity, sex, age, sexual identity, annual income, parenting status, and region of residence. Of the 400 research packets that were distributed, 89 were not usable in the analysis because they were either incomplete (82) or the demographic form indicated they did not identify as a gay or lesbian parent (7). Prior to initiating our analysis, we determined that the missing data were missing completely at random (MCAR) as evidenced by a nonsignificant finding for Little’s MCAR test (p = .621). Therefore, we continued our analysis by removing the missing cases. We then examined the data using least-squared regression to produce a Mahalanobis distance for each participant for the 5F-WEL-A, the Lesbian, Gay, and Bisexual Identity Scale (LGBIS; Mohr & Fassinger, 2000), and Outness scales. Subsequently, we investigated if the Mahalanobis distance value had χ2 p values of .05 or greater, and we removed those cases. This was done to identify outliers in the data and to reduce the likelihood of rogue influences biasing the outcomes. We removed six cases due to being outliers, resulting in a final sample of 305. A power analysis conducted using G* Power software (Faul et al., 2007) indicated that 305 was a significant sample size to detect differences between two groups with a medium effect size and power set at .8.
Instrumentation
The constructs of interest in this study are level of general wellness, level of outness or openness regarding sexual orientation, and sexual orientation identity. To that end, the participants were assessed using three instruments: (a) the Five-Factor Wellness Inventory (5F-WEL-A; Myers & Sweeney, 2005), (b) Outness Inventory (OI; Mohr & Fassinger, 2000), and the LGBIS (Mohr & Fassinger, 2000).
5F-WEL-A
The 5F-WEL-A is a 106-item inventory that assesses for wellness along five factors of wellness including (a) essential self, (b) social self, (c) creative self, (d) physical self, and (e) coping. Essential self is composed of spirituality, self-care, gender identity, and cultural identity. The social factor consists of the third-order factors of friendship and love. Creative self consists of factors that parallel the Adlerian creative self factors. Those include cognitions, emotions, control, constructive humor, and vocational engagement. The physical factor is made up of the third-order factor of exercise and nutrition. The coping factor is comprised of four third-order factors that include realistic beliefs, stress management, self-worth, and leisure. Hattie, Myers, and Sweeney (2004) reported an α coefficient of .94. The coefficients for second-order factors ranged from .90 to .94. In our sample, the 5F-WEL-A demonstrated high internal consistency with a Cronbach’s α of .92.
OI
The OI is an 11-item inventory that was developed to assess the degree to which lesbian, gay, or bisexual individual discloses their sexual orientation to others. There are three subscales within this instrument (a) Out to Family, (b) Out to World, and (c) Out to Religious community. Together these scores comprise the Overall Outness total score. Mohr and Fassinger examined the factor structure and the psychometrics of the OI with a group of lesbians (n = 590) and gay men (n = 414) between the ages of 18 and 69. They reported that it had an adequate internal consistency, with Cronbach’s α ranging from .74 to .97 and confirmed the three-factor structure (Mohr & Fassinger, 2000). In our sample, the OI demonstrated high internal consistency with a Cronbach’s α of .81.
LGBIS
The researchers used the LGBIS to collect the data for this study which was available on the author’s website prior to the publication of the revision (Mohr & Kendra, 2011). The author of this version stated, “No published data are available on the LGBIS, but unpublished analyses on LGBIS data from a large, national sample suggest that the psychometric properties of the measure are comparable to those of the LGIS. This was expected, given that the two measures are virtually identical in wording” (J. Mohr, personal communication, February 9, 2010). The LGBIS is a 27-item Likert-type inventory that was designed to assess dimensions of lesbian, gay, and bisexual identity. In effect, this model of identity appraises for the degree of self-acceptance in an environment of oppression. This model, unlike other marginalized identity models, does not attempt to place the individual’s temporal place in the identity development process. It is intended to assess multiple dimensions of sexual orientation identity. The subscales of this instrument include the following: (a) Need for Privacy, (b) Need for Acceptance, (c) Homonegativity, (d) Difficult Process, (e) Identity Confusion, and (f) Superiority. The Need for Privacy subscale, which is composed of 6 items of the assessment, assessed the level of comfort and control about level of disclosure of sexual orientation. Need for Acceptance is the level of concern an lesbian, gay, or bisexual (LGB) individual has about the acceptance heterosexuals have toward sexual minorities. Homonegativity is the level of internalized oppression endorsed by the individual. Difficult Process is an assessment of the self-perceived adversity of being a sexual minority. Identity Confusion is the degree of resolution that one has with her or his orientation. Superiority is the level of supremacy a sexual minority feels toward heterosexuals. A second-order factor, negative identity, reflects the degree to which individuals have overall difficulties related to their sexual orientation identity. This measure of Negative Identity is calculated by averaging scores on Homonegativity, Need for Privacy, Need for Acceptance, and Difficult Process subscales. Previous research on the LGBIS indicated content and construct validity for the scale, as well as an adequate internal consistency with Cronbach’s α ranging from .65 to .81 with a sample of lesbians (n = 590) and gay men (n = 414; Mohr & Fassinger, 2000). In our sample, the LGBIS demonstrated high internal consistency with a Cronbach’s α of .92.
Data Screening and Analysis
In our analysis, we compared the wellness, sexual identity, and outness scores of a sample of lesbian and gay parents between various demographic variables. The demographic variables of interest included sex, race, age, and community setting. Due to sample size limitations and for the sake of efficiency in interpretation, we selected to group race into two categories: (a) Caucasian, which represented the majority culture, and (b) people of color (POC), which represented racial and ethnic minorities who may experience other forms of prejudice. We employed a multivariate analysis of variance (MANOVA). The MANOVA is an extension of the traditional analysis of variance and is suited for comparisons between groups when using multiple dependent variables (Tabachnick & Fidell, 2013). Additionally, because the negative identity score and total outness score is a composite of several scales and nonnormal, we elected to use a nonparametric test, the Kruskal–Wallis.
Prior to beginning any analysis, we assessed our data for normality, through visual examination of various plots and conducting the Kolmogorov–Smirnov test of normality. We discovered several subscales on the 5F-WEL-A were nonnormal, including the essential self, the social self, and physical self. Additionally, all of the data on the LGBIS and the OI were nonnormal. Fortunately, the analysis being conducted (i.e., MANOVA) is robust to violations of normality with large sample sizes (Tabachnick & Fidell, 2013). Before conducting each MANOVA, we also examined the data using Box’s M test. This analysis examines the homogeneity of the variance–covariance matrices. In each of the following MANOVAs, when Box’s M test was found to be nonsignificant, we utilized the Wilks’s λ statistic. However, if Box’s M test was found nonsignificant, we utilized the more conservative Pillai’s Trace (Tabachnick & Fidell, 2013).
Results
Our first MANOVA examined the differences in 5F-WEL-A scores between male and female participants. The findings were nonsignificant with Wilks’s λ = .006, F(5, 291) = 2.07, p = .069, η = .03. Thus, there were no differences in scores based on sex. We then examined the difference in LGBIS scores between sexes. A MANOVA indicated significant differences in the comparison, Pillai’s Trace = .068, F(6, 288) = 3.508, p = .002, η = .07. Examination and application of Bonferroni’s adjustment on the tests of between-subject effects revealed significant differences between sex on Need for Acceptance (p = .005, η = .03), Difficult Process (p = .001, η = .04), and Internalized Homonegativity and Binegatvity (p = .001, η = .04). There were no significant differences between sex on the subscales Identity Confusion and Superiority. The other subscales did not meet the cutoff criteria of Bonferroni’s adjustment. A comparison of mean scores demonstrated that men scored significantly higher than females on the Need for Acceptance Scale, the Difficult Process Scale, and the Internalized Homophobia Scale. Specifically, on Need for Acceptance, males had a mean score of 2.67 (SD = 1.15), whereas the mean score for females was 2.25 (SD = 1.13). On the Difficult Process Scale, males scored 3.15 on average (SD = 1.41) and females scored 2.56 (SD = 1.25). On Internalized Homophobia Scale, males scored 2.16 on average (SD = 1.01) and females scored 1.74 (SD = .91). We conducted a separate Kruskal–Wallis test to examine the Negative Identity subscale scores; and there was also a significant difference between sexes on the negative identity scores, χ2(2, N = 295) = 14.99, p < .01, F(1, 293) =16.94, p < .001, η = .05. On the Negative Identity Scale, males scored a significantly higher mean rank (Mdn = 180.36) than females (Mdn = 136.57); for the Outness subscales, there was no significant difference between groups, Wilks’s λ = .971, F(3, 221) = 2.239, p = .09. However, on the total outness scores, males scored a significantly lower mean rank (Mdn = 95.78) than females (Mdn = 120), χ2(1, N = 225) = 6.41, p ≤ .02.
We then moved our analysis to comparisons on race. We separated race into two groups—White and POC. In this analysis, the MANOVA revealed no differences on the 5F-WEL-A scores, F(5, 299) = .971, p = .113, η = .03. As for the LGBIS, the data revealed significant differences, Pillai’s Trace = .094, F(6, 296) = 5.092, p < .01. Further examination of the data revealed that Whites produced a significantly higher mean score (M = 3.88, SD = 1.18) on the Need for Privacy Scale and a significantly lower mean score on Identity Confusion (M = 1.433, SD = .88). Likewise, negative identity scores also did not differ between groups, χ2(2, N = 303) = .707, p = .40, nor on total outness, χ2(1, N = 231) = .263, p = .61. However, one of the Outness subscale scores did differ by race. Specifically, POCs scored significantly lower (M = 4.29, SD = 2.27) than Whites (M = 5.08, SD = 2.31) on Out to Religious Community, Wilks’s λ = .960, F(3, 227) = 3.12, p = .03, η = .04.
Our next analysis examined the impact of age. For the purposes of equal cell distribution, the participants were grouped into four categories: 30 and below, 31–37, 38–45, and 46 and above. There were differences found when comparing the participants 5F-WEL-A scores according to these age-groups, Wilks’s λ = .896, F(15, 665) = 1.797, p = .031, η = .04. However, when examining the data further, the tests of between-subject effects revealed that none of the dependent variables were significant after applying Bonferroni’s adjustment (Tabachnick & Fidell, 2013). When comparing age on the scores of the LGBIS, there was a significant Box Test of equality of covariance, and as a result, we interpreted Pillai’s Trace. In this analysis, there were significant differences between the age-groups, Pillai’s Trace = .200, F(18, 723) = 2.86, p < .001, η = .07. The between-subject effects indicated a significant difference between age-groups on the Difficult Process subscale (p < .001) and the Internalized Homonegativity scale (p < .001). The Tukey’s Post Hoc test indicated that on the Difficult Process subscale (M =3.37, SD = 1.48) and the Internalized Homonegativity Scale (M = 2.27, SD = 1.24), there is a significant difference between those aged 46 and above and all the other age-groups, with those in the 46 and above age-group scoring a significantly higher mean score. As for the negative identity total score, there was a significant difference between groups, χ2(3, N = 248) = 16.20, p < .01. Those aged 46 and above scored a higher mean rank (Mdn = 156.09) than those 38–45 (Mdn = 129.71), 31–37 (Mdn = 114.25), and 30 and below (Mdn = 102.99). A Mann–Whitney U test was conducted as a post hoc examination. The Mann–Whitney revealed that those aged 46 and above differed significantly, from those aged 31–37, U = 1,069.00, z = −3.057, p < .01, and those aged 30 and below, U = 803.00, z = −3.63, p < .01, but not those aged 38– 45, U = 1,726.500, z = −.918, p = .359. In terms of outness, the data revealed a significant difference in ages, Wilks’s λ = .830, F(9, 430) = 3.81, p < .01, η = .06. Participants differed on their level of outness to their religious community, with those aged 46 and above (M = 5.64, SD = 1.92) reporting significantly lower scores of outness from those aged 30 and below (M = 4.05, SD = 2.51). The Tukey’s HSD revealed that the other age-groups did not differ significantly nor did the groups differ on total outness, χ2(3, N = 183) = 3.48, p = .32.
In further analysis of the data, we elected to compare community settings. Namely, we were interested to see if there was a significant difference in 5F-WEL-A, LGBIS, and outness scores based on urban or rural environments. The MANOVA indicated no significant differences across settings on the 5F-WEL-A scores, Wilks’s λ = .965, F(10, 578) = 1.048, p = .401, η = .02. Likewise, on LGBIS scores, there were also no significant differences found on the subscales, Wilks’s λ = .944, F(12, 572) = 1.392, p = .165, nor on the Negative Identity subscale, χ2(2, N = 294) = 3.55, p = .17. On the Outness subscales, the data revealed a significant difference in between groups, Wilks’s λ = .924, F(6, 438) = 2.96, p < .01, η = .04. Like in each of the previous analyses, there was no significant difference in Outness to Work or Outness to Family. However, participants reported a significant difference in their level of outness to their religious community, with those living in an urban settings reporting significantly higher scores on outness (M = 5.58, SD = 1.92) from those in rural (M = 4.08, SD = 4.08) and suburban areas (M = 4.73, SD = 2.39). The Tukey’s HSD revealed that the other groups did not differ significantly. However, the Kruskal–Wallis test did reveal that on total outness scores, groups did differ significantly, χ2(2, N = 224) = 6.32, p < .05, between those in an urban setting (Mdn = 126.93, n = 81), a rural setting (Mdn = 103.19, n = 54), and a suburban setting (Mdn = 105.02, n = 89). A Mann–Whitney U test revealed that those in an urban setting differed significantly from those in a rural, U = 1,722.00, z = 2.09, p < .05, and suburban setting, U = 2,901.00, z = 2.19, p < .05, but there was no significant difference among the other groups. Participants living in an urban setting scored higher on total outness than all other groups.
Discussion
In this sample, gay and lesbian parents did not significantly differ from each other in wellness according to any of the demographic categories examined. Results in other studies that examined demographic differences (e.g., race and gender) have shown both significant and nonsignificant differences in wellness (see Myers & Sweeney, 2005). Similarly, Dew and Newton’s (2005) survey of wellness research across sexual orientation and gender shows only minimal differences in overall wellness among gay, lesbian, and heterosexual men and women.
In terms of sexual orientation identity, gay male parents in this study reported a higher need for protection, acceptance, and internalized homophobia. They also reported a more difficult process in terms of identity formation. This is consistent with other studies (Mohr & Fassinger, 2000; Newcomb & Mustanski, 2010) and not surprising given gay men experience more overt societal discrimination than lesbians (Feinstein, Goldfried, & Davila, 2012; Kite & Whitley, 1996). All of this seems to signal that gay men and gay men who are parents generally have a more difficult time with their identity development process. Perhaps gay men who are parents have lower self-esteem than their heterosexual counterparts due to discrimination. This again may point to higher levels of homophobia experienced by men or perhaps there is a systematic difference in how men and women perceive and respond to homophobia in general.
Higher scores on the Need for Acceptance subscale are correlated with lower self-esteem (Mohr & Fassinger, 2000). Gay men are less safe in the world and suffer greater discrimination when they are out (Golom & Mohr, 2011; Hudepohl, Parrott, & Zeichner, 2010). Becoming parents may make them more visible to the world and less likely to pass as heterosexual. This may have an impact on their overall self-esteem and possibly make them in need of more support when becoming parents. Questions remain about the impact this has on their perceptions of their abilities as a parent. Gay male parents may feel less qualified to be parents due to societal attribution to women being more motherly/nurturing. They may have a lower sense of parental efficacy than lesbian parents. Does sexual identity in heterosexual parents compare in any way to those in this study? What does this mean for parenting self-esteem in a sample of gay fathers?
In a previous study (Lassiter, et al., 2006), gay and lesbian parents connected self-esteem and parenting by stating “you have to believe you have something to offer to your child…and you have to really want to be a parent to be a gay parent.” Johnston and Mash (1989, p. 248) define parenting self-esteem as encompassing both affective and cognitive attitudes toward parenting, including satisfaction derived from parenting and perceived self-efficacy as a parent. Parenting satisfaction is defined as feelings such as pleasure, fulfillment, and motivation experienced in the parenting role (Sabatelli &Waldron, 1995). Parenting self-efficacy refers to parents’ beliefs about their competence as a parent, including parents’ perceived ability to deal effectively with their child’s problems and to influence their child’s behavior and development in a positive way (Coleman & Karraker, 2000; Johnston & Mash, 1989). It is conceivable that gay and lesbian parents who are experiencing anxiety and depression may have lowered parenting self-esteem due to a tendency toward negative evaluations about one’s skills and abilities and a reduced interest in and enjoyment of daily life. This is not to say that gay men do not make good parents, but rather they may need more assistance in counseling and psychotherapy to aid in greater levels of self-acceptance and to protect against the impact of daily oppression. Increasing self-acceptance may also have an impact on the well-being of children raised by gay and lesbian parents. In fact, research with children supports the potential mediating role of parenting self-esteem in the relationship between parent and child well-being (Johnston & Mash, 1989; Ohan, Leung, & Johnston, 2000). In a study examining moderating factors in the parent–child relationship, Hughes and Gullone (2010) found that poor paternal parenting self-esteem may have adverse consequences for adolescent well-being.
Younger lesbian and gay parents in this study reported a less difficult process in identity formation and coming out than their older counterparts. Perhaps the lower mean score for the younger group of parents on the Difficult Process Scale is the result of greater societal acceptance generally for LGBTQ people in recent years than the oppression experienced by their older peers. The higher Internalized Homophobia subscale mean of the older group of parents may also be the result of having lived with a greater amount of cultural stigma and homophobia. In this study, participants who lived in urban settings were more likely to be out to their religious community than those in rural settings. This may be due to greater acceptance experienced by LGBTQ people in urban settings in general and having greater access to more progressive and accepting urban religious organizations. In this study, younger parents were less out than older parents to their religious communities. This is concerning because in previous studies (Lassiter, et al., 2006), gay parents have expressed a strong need for affirming religious/spiritual communities for raising their families. These gay and lesbian parents described finding affirming communities as empowering to them as parents and important to the welfare and self-esteem of their children. They spoke of the need to be strong, out, and confident, so they could prepare their children to face discrimination in the world. This is also concerning because younger parents in our sample are more likely than older parents to have children still living at home.
In this sample, gay or lesbian POC who were parents are less likely to be out to their religious community, which is important because they may tend to rely on religious community support to cope with stressors in their lives in general (Barden, Gutierrez, Gonzalez & Ali, 2016; Gutierrez, Barden, Gonzalez, & Ali, 2016; Boyd-Franklin, 2010). This has implications for gay and lesbian parents who are POC and their ability to get support in parenting. If a religious community is where one might naturally seek personal support and support in parenting and that community is discriminatory or shaming in terms of sexual identity, how do lesbian and gay parents who are POC get support in creating and raising families? The findings also leave questions about how much pressure is placed on these parents to be self-rejecting by nonaccepting religious communities. What impact does this potential lack of support and affirmation have on their parenting self-esteem? What impact does it have on their children? Where can POC get support in parenting and does this put more pressure on them to not be self-accepting while parenting? The general parenting literature examining parental involvement in religious communities shows generally positive effects on children including overall family satisfaction (Strahan & Craig, 1995), absence of substance abuse (Dudley, Mutch, & Cruise, 1987), and children’s social responsibility that includes honesty, empathy, trustworthiness, self-control, and obedience (Gunnoe & Hetherington, 1999). Following a similar positive trend, parental religiosity has been associated with affectionate, loving parenting, clear boundary setting, and autonomy promoting child-rearing practices (Carothers, Borkowski, Lefever Bruke, & Whitman, 2005; Gunnoe & Hetherington, 1999). So it stands to reason that the greater the access and comfort with accepting religious communities for gay and lesbian parents will be better for their children.
White gay and lesbian parents in this sample scored significantly higher on need for privacy and significantly lower on identity confusion than parents who were POC. The higher need for privacy may be reflective of a more individualistic, less collectivist orientation than POC. That is the POC may culturally value the community or group than White gay or lesbian parents. Likewise, lower scores on identity confusion may be attributable to the rejection by POC of identity labels such as gay or lesbian and with the complexities of dual identity development (Crawford, Allison, Zamboni, & Soto, 2002).
Limitations
The limitations of this study include the fact that the population sampled was from a gay pride festival in a large urban setting in the southern United States. It is possible that lesbian and gay parents who would likely attend a pride festival would be more comfortable being out to the world and may feel more empowered as sexual minorities who are parenting. This may have created a type of ceiling effect that limited the potential differences between demographic groups. Furthermore, a sample that came from a more rural region of the country may have given very different results. It is also possible that those parents who agreed to complete the research packet were more likely to be more self-accepting versus those who declined. Another limitation is the underrepresentation of POC in the sample.
Implications and Future Research
Because gay and lesbian people access mental health services at higher rates than their heterosexual counterparts, it is important that counselors, psychologists, and social workers are aware of the needs of this community of current and future parents. The current study implies that older gay or lesbian parents, parents who are POC, and gay male parents may need more support with self-acceptance and sexual orientation identity development. This may have strong implications for development of parental self-esteem, parental confidence, and parental self-efficacy in this growing population of gay and lesbian parents. In order to create affirming and empowering counseling relationships, mental health providers will need to be knowledgeable about the process of self-acceptance and identity development in LBGTQ populations, the influence of parental self-acceptance on children, and the positive attributes of gay and lesbian parents.
Given the lack of research focused on gay and lesbian parents, greater societal acceptance, and changes in laws protecting the rights of lesbian and gay couples and families, more research into the characteristics and needs of lesbian and gay parents is needed. Specifically related to the findings in this study, it would be helpful to examine the differences and similarities in wellness among lesbian and gay parents as compared to their heterosexual counterparts given the impact of discrimination and internalized homonegativity. Unfortunately, there remains lack of research on the wellness practices of heterosexual parents and the effect on child well-being. Likewise, it would be important to examine the effects of wellness practices on wellness in children raised by lesbian and gay parents. In previous studies, gay and lesbian parents articulated the need to empower their children to face discrimination due to their family’s makeup and this seems to have a potential impact on the wellness of both parent and child. A large quantitative examination of the impact of higher levels of identity development and self-acceptance in gay and lesbian parents on child outcomes seems important to understanding the mental health needs of these families.
Researchers also need to examine the important factors that move gay and lesbian parents toward greater levels of self-acceptance given its importance to self-esteem. We know that higher levels of parental self-esteem result in better outcomes for children. Additional studies are needed to examine the relationship between self-acceptance and perceptions of parental efficacy among gay and lesbian parents.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.