Like families headed by heterosexual parents, lesbian and gay parents and their children are a diverse group (Martin, 1993). Unlike heterosexual parents and their children, however, lesbian and gay parents and their children are often subject to prejudice because of sexual orientation that turns judges, legislators, professionals, and the public against them, frequently resulting in negative outcomes such as loss of physical custody, restrictions on visitation, and prohibitions against adoption (Falk, 1989; Editors of the Harvard Law Review, 1990).
As with all socially stigmatized groups, the beliefs held generally in society about lesbians and gay men are often not based in personal experience, but are instead culturally transmitted (Herek, 1991). The purpose of this summary of research findings on lesbian and gay parents and their children is to assist psychologists and other professionals to evaluate widespread beliefs in the light of empirical data and in this way ameliorate the negative effects of unwarranted prejudice.
Because many beliefs about lesbian and gay parents and their children are open to empirical test, psychological research can evaluate their accuracy. Systematic research comparing lesbian and gay adults to heterosexual adults only began in the late 1950s, and research comparing children of gay and lesbian parents with those of heterosexual parents is of a more recent vintage. Research on lesbian and gay adults began with Evelyn Hooker’s landmark study (1957) and culminated with the declassification of homosexuality as a mental disorder in 1973 (Gonsiorek, 1991).
Case reports on children of gay and lesbian parents began to appear in the psychiatric literature in the early 1970s (e.g., Osman, 1972; Weeks, Derdeyn, ; Langman, 1975) and have continued to appear (e.g., Agbayewa, 1984). Beginning with the pioneering work of Martin and Lyon (1972), first person and fictionalized descriptions of life in lesbian mother families have also become available (e.g., Alpert, 1988; Clausen, 1985; Jullion, 1985; Mager, 1975; Perreault, 1975; Pollock ; Vaughn, 1987; Rafkin, 1990). Systematic research on the children of lesbian and gay parents did not, however, begin to appear in major professional journals until 1978, and most of the available research has been published more recently.
As this summary will show, the results of existing research comparing gay and lesbian parents to heterosexual parents and children of gay or lesbian parents to children of heterosexual parents are quite uniform: common sterotypes are not supported by the data.
Without denying the clarity of results to date, it is important also for psychologists and other professionals to be aware that research in this area has presented a variety of methodological challenges, not all of which have been surmounted in every study. As is true in any area of research, questions have been raised with regard to sampling issues, statistical power, and other technical matters (e.g., Belcastro, Gramlich, Nicholson, Price, ; Wilson, 1993); no individual study is entirely invincible to such criticism.
One criticism of this body of research (Belcastro et al., 1993) has been that the research lacks external validity because it may not be representative of the larger population of lesbian and gay parents. This criticism is not justified, because nobody knows the actual composition of the entire population of lesbian mothers, gay fathers, or their children (many of whom choose to remain hidden) and hence researchers cannot possible evaluate the degree to which particular samples do or do not represent the population. In the long run, it is not the results obtained from any one specific sample, but the accumulation of findings from many different samples that will be most meaningful.
Research in this area has also been criticized for using poorly matched or no control groups in designs that call for such controls. Particularly notable in this category has been the tendency in some studies to compare development among children of a group of divorced lesbian mothers, many of whom are living with lesbian partners, to that among children of a group of divorced heterosexual mothers who are not currently living with heterosexual partners. It will be important for future research to disentangle maternal sexual orientation from maternal status as partnered or unpartnered.
Other criticisms have been that most studies have involved relatively small samples, that there have been inadequacies in assessment procedures employed in some studies, and that the classification of parents as lesbian, gay, or heterosexual has sometimes been problematic (e.g., some women classified by researchers as lesbian might be regarded as bisexual by other observers). It is significant, however, that even with all the questions and/or limitations that may characterize research in the area, none of the published research suggests conclusions different from those that will be summarized below.
This summary consists of four sections. In the first, results of research on lesbian and gay adults (and parents) are summarized. In the second section, a summary of results from research comparing children of lesbian and gay parents with those of heterosexual parents or with established norms is presented. The third section summarizes research on heterogeneity among lesbian and gay families with children. The fourth section provides a brief conclusion.
Lesbian and Gay Parents
One belief that often underlies both judicial decision-making in custody litigation and public policies governing foster care and adoption has been the belief that lesbians and gay men are not fit to be parents. In particular, courts have sometimes assumed that gay men and lesbians are mentally ill, that lesbians are less maternal than heterosexual women, and that lesbians’ and gay men’s relationships with sexual partners leave little time for ongoing parent-child interactions (Editors of the Harvard Law Review, 1990). Results of research to date have failed to confirm any of these beliefs (Falk, 1989, 1994; Patterson, 1994b, 1995b, 1996).
Mental Health of Lesbians and Gay Men
The psychiatric, psychological, and social-work professions do not consider homosexual orientation to be a mental disorder. More than 20 years ago, the American Psychiatric Association removed “homosexuality” from its list of mental disorders, stating that “homosexuality per se implies no impairment in judgment, stability, reliability, or general social or vocational capabilities” (American Psychiatric Association, 1980).
In 1975, the American Psychological Association took the same position and urged all mental health professionals to help dispel the stigma of mental illness that had long been associated with homosexual orientation (American Psychological Association, 1975). The National Association of Social Workers has a similar policy (National Association of Social Workers, 1994).
The decision to remove homosexual orientation from the list of mental disorders reflects the results of extensive research, conducted over three decades, showing that homosexual orientation is not a psychological maladjustment (Gonsiorek, 1991; Reiss, 1980; Hart, Roback, Tittler, Weitz, Walston, ; McKee, 1978).
The social and other circumstances in which lesbians and gay men live, including exposure to widespread prejudice and discrimination, often cause acute distress; but there is no reliable evidence that homosexual orientation per se impairs psychological functioning (Freedman, 1971; Gonsiorek, 1991; Hart et al., 1978; Hooker, 1957; Reiss, 1980).
Fitness of Lesbians and Gay Men as Parents
Beliefs that gay and lesbian adults are not fit parents likewise have no empirical foundation (Cramer, 1986; Falk, 1989; Gibbs, 1988; Patterson, 1996). Lesbian and heterosexual women have not been found to differ markedly either in their overall mental health or in their approaches to child rearing (Kweskin ; Cook, 1982; Lyons, 1983; Miller, Jacobsen, ; Bigner, 1981; Mucklow ; Phelan, 1979; Pagelow, 1980; Rand, Graham, ; Rawlings, 1982; Thompson, McCandless, ; Strickland, 1971), nor have lesbians’ romantic and sexual relationships with other women been found to detract from their ability to care for their children (Pagelow, 1980). Recent evidence suggests that lesbian couples who are parenting together tend to divide household and family labor relatively evenly (Hand, 1991; Patterson, 1995a) and to report satisfaction with their couple relationships (Koepke, Hare, & Moran, 1992; Patterson, 1995a). Research on gay fathers has similarly found no reason to believe them unfit as parents (Barret & Robinson, 1990; Bigner and Bozett, 1990; Bozett, 1980, 1989).
Children of Lesbian and Gay Parents
In addition to judicial concerns about gay and lesbian parents themselves, courts have voiced three major kinds of fears about effects of lesbian or gay parents on children.
The first general concern is that development of sexual identity will be impair among children of lesbian or gay parents¬for instance, that children brought up by gay fathers or lesbian mothers will show disturbances in gender identity and/or in gender role behavior (Falk, 1989; Hitchens & Kirkpatrick, 1985; Kleber, Howell, & Tibbits-Kleber, 1986). It has also been suggested that children brought up by lesbian mothers or gay fathers will themselves become gay or lesbian (Falk, 1989; Kleber et al., 1986).
A second category of concerns involves aspects of children’s personal development other than sexual identity (Falk, 1989; Editors of the Harvard Law Review, 1990; Kleber et al., 1986). For example, courts have expressed fears that children in the custody of gay or lesbian parents will be more vulnerable to mental breakdown, will exhibit more adjustment difficulties and behavior problems, and will be less psychologically healthy than children growing up in homes with heterosexual parents.
A third category of specific fears expressed by the courts is that children of lesbian and gay parents may experience difficulties in social relationships (Editors of the Harvard Law Review, 1990; Falk, 1989; Hitchens ; Kirkpatrick, 1985). For example, judges have repeatedly expressed concern that children living with lesbian mothers may be stigmatized, teased, or otherwise traumatized by peers. Another common fear is that children living with gay or lesbian parents may be more likely to be sexually abused by the parent or by the parent’s friends or acquaintances.
Three aspects of sexual identity are considered in the research: gender identity concerns a person’s self-identification as male or female; gender-role behavior concerns the extent to which a person’s activities, occupations, and the like are regarded by the culture as masculine, feminine, or both; sexual orientation refers to a person’s choice of sexual partners–i.e., heterosexual, homosexual, or bisexual (Money & Earhardt, 1972; Stein, 1993). To examine the possibility that children in the custody of lesbian mothers or gay fathers experience disruptions of sexual identity, research relevant to each of these three major areas of concern is summarized below.
Gender identity. In studies of children ranging in age from 5 to 14, results of projective testing and related interview procedures have revealed normal development of gender identity among children of lesbian mothers (Green, 1978; Green, Mandel, Hotvedt, Gray, ; Smith, 1986; Kirkpatrick, Smith, ; Roy, 1981).
More direct assessment techniques to assess gender identity have been used by Golombok, Spencer, and Rutter (1983) with the same result; all children in this study reported that they were happy with their gender, and that they had no wish to be a member of the opposite sex. There was no evidence in any of the studies of gender identity difficulties among children of lesbian mothers. No data have been reported in this area for children of gay fathers.
Gender-Role Behavior. A number of studies have examined gender-role behavior among the offspring of lesbian mothers (Golombok et al., 1983; Gottman, 1990; Green, 1978; Hoeffer, 1981; Kirkpatrick et al., 1981; Patterson, 1994a). These studies reported that such behavior among children of lesbian mothers fell within typical limits for conventional sex roles. For instance, Kirkpatrick and her colleagues (1981) found no differences between children of lesbian versus heterosexual mothers in toy preferences, activities, interests, or occupational choices.
Rees (1979) administered the Bem Sex Role Inventory (BSRI) to 24 adolescents, half of whom had divorced lesbian and half of whom had divorced heterosexual mothers. The BSRI yields scores on masculinity and femininity as independent factors and an androgyny score from the ratio of masculinity to femininity. Children of lesbian and heterosexual mothers did not differ on masculinity or on androgyny, but children of lesbian mothers reported greater psychological femininity than did those of heterosexual mothers. This result would seem to run counter to expectations based on stereotypes of lesbians as lacking in femininity, both in their own demeanor and in their likely influences on children.
Sex role behavior of children was also assessed by Green and his colleagues (1986). In interviews with the children, no differences between 56 children of lesbian and 48 children of heterosexual mothers were found with respect to favorite television programs, favorite television characters, or favorite games or toys. There was some indication in interviews with children themselves that the offspring of lesbian mothers had less sex-typed preferences for activities at school and in their neighborhoods than did children of heterosexual mothers.
Consistent with this result, lesbian mothers were also more likely than heterosexual mothers to report that their daughters often participated in rough-and-tumble play or occasionally played with “masculine” toys such as trucks or guns; however, they reported no differences in these areas for sons. Lesbian mothers were no more or less likely than heterosexual mothers to report that their children often played with “feminine” toys such as dolls. In both family types, however, children’s sex-role behavior was seen as falling within normal limits.
In summary, the research suggests that children of lesbian mothers develop patterns of gender-role behavior that are much like those of other children. No data are available as yet in this area for children of gay fathers. Sexual Orientation. A number of investigators have also studied a third component of sexual identity: sexual orientation (Bailey, Bobrow, Wolfe, & Mikach, 1995; Bozett, 1980, 1982, 1987, 1989; Gottman, 1990; Golombok et al., 1983; Green, 1978; Huggins, 1989; Miller, 1979; Paul, 1986; Rees, 1979).
In all studies, the great majority of offspring of both gay fathers and lesbian mothers described themselves as heterosexual. Taken together, the data do not suggest elevated rates of homosexuality among the offspring of lesbian or gay parents. For instance, Huggins (1989) interviewed 36 teenagers, half of whom were offspring of lesbian mothers and half of heterosexual mothers. No children of lesbian mothers identified themselves as lesbian or gay, but one child of a heterosexual mother did; this difference was not statistically significant.
In a recent study, Bailey and his colleagues (1995) studied adult sons of gay fathers and found more than 90% of the sons to be heterosexual. Because the heterosexual and nonheterosexual sons did not differ in the length of time they had resided with their fathers, the effects of the exposure to the fathers’ sexual orientation on the sons’ sexual orientation must have been either very small or nonexistent.
Other Aspects of Personal Development
Studies of other aspects of personal development among children of gay and lesbian parents have assessed a broad array of characteristics. Among these have been separation-individuation (Steckel, 1985, 1987), psychiatric evaluations (Golombok et al., 1983; Kirkpatrick et al., 1981), assessments of behavior problems (Flaks, Ficher, Masterpasqua and Joseph, 1995; Golombok et al., 1983; Patterson, 1994a), personality (Gottman, 1990), self-concept (Gottman, 1990; Huggins, 1989; Patterson, 1994a; Puryear, 1983), locus of control (Puryear, 1983; Rees, 1979), moral judgment (Rees, 1979), and intelligence (Green et al., 1986).
Research has shown that concerns about difficulties in personal development in these areas among children of lesbian mothers are unwarranted. As was the case for sexual identity, studies of these other aspects of personal development have revealed no major differences between children of lesbian versus heterosexual mothers. One statistically significant difference in self-concept emerged in Patterson’s (1994a) study: children of lesbian mothers reported greater symptoms of stress but also a greater overall sense of well-being than did children in a comparison group of heterosexual families. The responses of both groups were, however, within a normal range (Patterson, 1994a). Overall, the belief that children of gay and lesbian parents suffer deficits in personal development has no empirical foundation.
Studies assessing potential differences between children of gay and lesbian versus heterosexual parents have sometimes included assessments of children’s social relationships. The most common focus of attention has been on peer relations, but some information on children’s relationships with adults has also been collected. Research findings that address the likelihood of sexual abuse are also summarized in this section.
Research on peer relations among children of lesbian mothers has been reported by Golombok and her colleagues (1983), Green (1978), and by Green and his colleagues (1986). Reports by both parents and children suggest normal development of peer relationships. For example, as would be expected, most school-aged children reported same-sex best friends and predominantly same-sex peer groups (Golombok et al., 1983; Green, 1978). The quality of children’s peer relations was described, on average, in positive terms by researchers (Golombok et al., 1983) as well as by lesbian mothers and their children (Green et al., 1986).
No data on the children of gay fathers have been reported in this area. Studies of relationships with adults among the offspring of lesbian and gay parents have also yielded a generally positive picture (Golombok et al., 1983; Harris & Turner, 1985/86; Kirkpatrick et al., 1981). For example, Golombok and her colleagues (1983) found that children of divorced lesbian mothers were more likely to have had recent contact with their fathers than were children of divorced heterosexual mothers. Another study, however, found no differences in this regard (Kirkpatrick et al., 1981).
Harris and Turner (1985/86) studied the offspring of gay fathers as well as those of lesbian mothers; parent-child relationships were described in positive terms by parents in their sample. One significant difference between lesbian and gay parents, on the one hand, and heterosexual parents, on the other, was that heterosexual parents were more likely to say that their children’s visits with the other parent presented problems for them (Harris ; Turner, 1985/86).
In the Golombok et al. (1983) study, children’s contacts with adult friends of their lesbian mothers were also assessed. All of the children were reported to have contact with adult friends of their mothers, and the majority of lesbian mothers reported that their adult friends were a mixture of homosexual and heterosexual adults.
Concerns that children of gay or lesbian parents are more likely than children of heterosexual parents to be sexually abused have also been addressed. Results of work in this area reveal that the great majority of adults who perpetrate sexual abuse are male; sexual abuse of children by adult women is extremely rare (Finkelhor & Russell, 1984; Jones & MacFarlane, 1980; Sarafino, 1979).
Moreover, the overwhelming majority of child sexual abuse cases involve an adult male abusing a young female (Jenny, Roesler, & Poyer, 1994; Jones & MacFarlane, 1980). Available evidence reveals that gay men are no more likely than heterosexual men to perpetrate child sexual abuse (Groth & Birnbaum, 1978; Jenny et al., 1994; Sarafino, 1979). Fears that children in custody of gay or lesbian parents might be at heightened risk for sexual abuse are thus without basis in the research literature.
Overall, then, results of research to date suggest that children of lesbian and gay parents have normal relationships with peers and that their relationships with adults of both sexes are also satisfactory. The picture of lesbian mothers’ children that emerges from results of existing research is thus one of general engagement in social life with peers, with fathers, and with mothers’ adult friends–both male and female, both heterosexual and homosexual. Studies in this area to date are few, and the data emerging from them are sketchy. On the basis of existing research findings, however, fears about children of lesbians and gay men being sexually abused by adults, ostracized by peers, or isolated in single-sex lesbian or gay communities are unfounded.
Diversity Among Gay and Lesbian Families
Despite the tremendous diversity evident within gay and lesbian communities, research on differences among lesbian and gay families with children is as yet quite sparse. One particularly important kind of heterogeneity involves the circumstances of children’s birth or adoption. Some men and women have had children in the context of heterosexual relationships that split up after one or both parents assumed lesbian or gay identities.
Much of the existing research on lesbian mothers, gay fathers, and their children was initiated to address concerns that arose for such families in the context of child custody disputes, and it was often designed at least in part to examine the veracity of common stereotypes that have been voiced in legal proceedings. A growing number of men and women have also had children after assuming lesbian or gay identities. Recently, a small body of research (e.g., Flaks, et al., 1995; McCandlish, 1987; Patterson, 1994a, 1995a; Steckel, 1987) has begun to address issues relevant to families of this type. Parents and children in these two kinds of families are likely to have experiences that differ from one another in many respects.
Many issues (for example, residential versus nonresidential parenting)have yet to be addressed directly by research. In this section, research findings on the impact of parental psychological and relationship status and on the influence of other stresses and supports are described. One dimension of difference among gay and lesbian families concerns whether or not the custodial parent is involved in a couple relationship, and if so what implications this may have for children.
Pagelow (1980), Kirkpatrick et al. (1981), and Golombok et al. (1983) all reported that, in their samples, divorced lesbian mothers were more likely than divorced heterosexual mothers to be living with a romantic partner; however, none of these investigators examined connections between this variable and children’s adjustment or development in lesbian mother families.
Huggins (1989) reported that self-esteem among daughters of lesbian mothers whose lesbian partners lived with them was higher than that among daughters of lesbian mothers who did not live with a partner. Because of the small sample size and absence of statistical tests, this finding should be seen as suggestive rather than conclusive.
On the basis of impressions from her own work, Kirkpatrick has also stated her view that “contrary to the fears expressed in court, children in households that included the mother’s lesbian lover had a richer, more open and stable family life” than did those in single parent lesbian mother households (Kirkpatrick, 1987, p. 204).
Issues related to division of family and household labor have also been studied. In families headed by lesbian couples, Patterson (1995a) found that, although mothers did not differ in their reported involvement in household and family decision-making tasks, biological mothers reported more time spent in child care and nonbiological mothers reported more time spent in paid employment. In families where mothers reported sharing child care duties relatively evenly between themselves, parents were more satisfied and children were better adjusted. Thus, equal sharing of child care duties was associated with more advantageous outcomes both for parents and for children in this study.
Another aspect of diversity among gay and lesbian families relates to the psychological status and well-being of the parent. Research on parent-child relations in heterosexual families has consistently revealed that children’s adjustment is often related to indices of maternal mental health. One might therefore expect factors that enhance mental health among lesbian mothers or gay fathers also to benefit their children.
Lott-Whitehead and Tully (1993) reported considerable variability in the amounts of stress described by lesbian mothers, but did not describe sources of stress nor their relations to child adjustment. Rand, Graham, and Rawlings (1982) found that lesbian mothers’ sense of psychological well-being was associated with their degree of openness about their lesbian identity with employers, ex-husbands, and children; mothers who felt more able to disclose their lesbian identity were more likely to express a positive sense of well-being. Unfortunately, no information about the relations of these findings to adjustment or development among children of these women has been reported to date.
Another area of great diversity among families with a gay or lesbian parent concerns the degree to which a parent’s gay or lesbian identity is accepted by other significant people in a child’s life. Huggins (1989) found a tendency for children whose fathers were rejecting of maternal lesbian identities to report lower self-esteem than those whose fathers were neutral or positive. Due to small sample size and absence of significance tests, this finding should be regarded as preliminary and suggestive rather than definitive. Huggins’ (1989) finding does, however, raise questions about the extent to which reactions of important adults in a child’s environment can influence responses to discovery of a parent’s gay or lesbian identity.
Effects of the age at which children learn of parental homosexuality have also been a topic of study. Paul (1986) found that offspring who were told of parental gay, lesbian, or bisexual identity either in childhood or in late adolescence found the news easier to cope with than those who first learned of it during early to middle adolescence. Huggins (1989) also reported that those who learned of maternal lesbianism in childhood had higher self-esteem than did those who were not informed of it until they were adolescents. From a clinical perspective, it is widely agreed that early adolescence is a particularly difficult time for children to learn that a father is gay or that a mother is lesbian (Bozett, 1980; Pennington, 1987; Schulenberg, 1985).
Some investigators have also raised questions about the potential role of peer support in helping children to deal with issues raised by having a gay or lesbian parent. Lewis (1980) was the first to suggest that children’s silence on the topic of parental sexual orientation with peers and siblings might add to their feelings of isolation from other children. All of the 11 adolescents studied by O’Connell (1993) reported exercising selectivity about when they disclosed information about their mothers’ lesbian identities. Paul (1986) found that 29% of his young adult respondents had never known anyone else with a gay, lesbian, or bisexual parent, suggesting that the possibility of isolation is very real for some young people.
Potentially negative effects of any such isolation have not, however, been uncovered in research to date. Lewis (1980) suggested that children would benefit from support groups consisting of other children of gay or lesbian parents, and young people interviewed by O’Connell (1993) agreed, but systematic evaluations of such groups have not been reported.
In summary, research on diversity among families with gay and lesbian parents and on the potential effects of such diversity on children is only beginning (Martin 1989; Patterson, 1992, 1995b) Existing data on children of lesbian mothers suggest that children may fare better when mothers are in good psychological health and living with a lesbian partner with whom they share child care.
Children may find it easier to deal with issues raised by having lesbian or gay parents if they learn of parental sexual orientation during childhood rather than during adolescence. Existing data also suggest the value of a supportive milieu, in which parental sexual orientation is accepted by other significant adults and in which children have contact with peers in similar circumstances. The existing data are, however, still very sparse, and any conclusions must be seen as tentative.
It is clear, however, that existing research provides no basis for believing that children’s best interests are served by family conflict or secrecy about a parent’s gay or lesbian identity, or by requirements that a lesbian or gay parent maintain a household separate from that of a same-sex partner.
In summary, there is no evidence to suggest that lesbians and gay men are unfit to be parents or that psychosocial development among children of gay men or lesbians is compromised in any respect relative to that among offspring of heterosexual parents. Not a single study has found children of gay or lesbian parents to be disadvantaged in any significant respect relative to children of heterosexual parents. Indeed, the evidence to date suggests that home environments provided by gay and lesbian parents are as likely as those provided by heterosexual parents to support and enable children’s psychosocial growth.
It should be acknowledged that research on lesbian and gay parents and their children is still very new and relatively scarce. Less is known about children of gay fathers than about children of lesbian mothers. Little is known about development of the offspring of gay or lesbian parents during adolescence or adulthood. Sources of heterogeneity have yet to be systematically investigated. Longitudinal studies that follow lesbian and gay families over time are badly needed.