Homosexual vs. Heterosexual Parenting: Is There Really “No Difference”?
James R. Aist
(Note: numbers given in parentheses refer to specific references listed in the “References Cited” section at the end of the article)
“Children of homosexual parents are about 12-15 times more likely to be homosexual than are children of heterosexual parents.”
Gay activists have insisted for years that there is no difference in the outcomes of parenting by homosexuals when compared to heterosexual parenting. Indeed, The American Psychological Association (APA) officially supports this claim. And the homosexual movement has used this claim to influence court decisions in favor of allowing homosexuals to legally adopt children. But, is this claim actually supported by the scientific facts? In other words, are the outcomes of homosexual parenting really equivalent to the outcomes of heterosexual parenting?
The APA Brief on Homosexual Parenting
In 2005, the Lesbian, Gay, and Bisexual Concerns Office of the APA published a brief (i.e., literature review) on this topic (1). In this brief, they cited 59 published articles in support of their summary claim that “Not a single study has found children of lesbian or gay parents to be disadvantaged in any significant respect relative to children of heterosexual parents.” However, in 2012, Loren Marks published a detailed analysis and critique of the scientific merit of that brief and the literature upon which it was based (2). Marks found that the studies cited in support of the APA summary claim are woefully lacking in sound scientific principles: 1) 77% of the studies are based on small, non-representative, biased samples of fewer than 100 subjects each; 2) 44% of the studies did not include a control group (i.e., a heterosexual comparison group), which is an absolute necessity for properly designed scientific studies of this nature; 3) 13 of the 33 studies that did include a control group used single parents, instead of two-biological-married parents, for the comparison; 4) the remaining 20 of these 33 studies with control groups ambiguously specified the make-up of the heterosexual control groups as “mothers” or “couples”; 5) the studies evaluated in the brief focused selectively on “gender-related outcomes” (such as, sexual orientation, gender identity, self esteem and self concepts) while societal concerns (such as excessive drinking, drug use, truancy and criminal offenses) were usually ignored; 6) none of the studies tracked societally significant long-term outcomes into adulthood, thus leaving the critical issue of parenting outcomes essentially unaddressed; 7) the brief seems to draw inferences of sameness of parenting outcomes based on analyses of small, non-representative samples lacking necessary statistical power; and 8) although the brief claims that “Not a single study has found children of lesbian or gay parents to be disadvantaged in any significant respect relative to children of heterosexual parents”, it ignores or dismisses two, scientifically sound studies published years earlier that did find evidence suggesting that children of lesbian or gay parents are disadvantaged in several significant respects relative to children of heterosexual parents. [This selective omission of two articles presenting contrary findings, even though the articles have considerable scientific merit (2), belies the strong pro-gay bias of the Lesbian, Gay, and Bisexual Concerns Office of the APA, which can also be easily detected in the mission statement (1) of that office.]
Noting that “Not one of the 59 studies referenced in the 2005 APA brief compares a large, random, representative sample of lesbian or gay parents and their children with a large, random, representative sample of married parents and their children”, Marks concluded that the strong assertions made in the APA brief were not substantiated by the published studies used, and were, therefore, unwarranted. In other words, as of 2005, scientific research had failed to prove that there is no difference between homosexual and heterosexual parenting, contrary to the conclusions in the (strongly biased) APA brief.
Studies Reporting Differences in Parenting Outcomes
The first two of these studies were published by S. Sarantakos. His 1996 paper (3) was a comparative analysis of 58 children of heterosexual married parents, 58 children of heterosexual cohabiting couples and 58 children living with homosexual couples, all matched according to socially significant criteria (e.g., age, number of children, number of parents in the household, education, occupation and socio-economic status). This study has some possible methodological weaknesses and confounding factors, but it also has several strong points of scientific design not present in the studies used for the APA brief (2). [It is significant that the comments of Marks (2) on this paper represent, in effect, a very favorable, post-publication, peer review.] Sarantakos (3) found several important criteria related to the children’s schooling in which homosexual parenting was apparently inferior to heterosexual parenting, including language, math, sports, sociability, learning attitude, parent-school relationships, support with homework and parental aspirations. Based on his results, he concluded that “…in the majority of cases, the most successful are children of married couples, followed by children of cohabiting couples and finally by children of homosexual couples.” Then in 2000, he published a book entitled “Same-sex Couples” (10). According to Marks (2), in this book, Sarantakos published the results of another research project in which he, once again, used two comparison groups, a married couple sample and a cohabiting couple sample, examined several outcomes of societal concern, and, very significantly, reported long-term outcomes in adults 18 years of age or older. Based on his results, which he obtained from statements made by the adult children of the parents, he concluded that adult children of homosexual parents report drug and alcohol abuse, education truancy, sexual activity and criminality in higher proportions than adult children of (married or cohabiting) heterosexual couples. Additionally, Marks (2) noted that Sarantakos (10) reported that “the number of children who were labeled by their parents as gay, or identified themselves as gay, is much higher than the generally expected proportion.” I will return to this finding later in the article.
Now let’s turn to several, more-recent studies, all of which were published in reputable, peer-reviewed scientific journals.
Regnerus (4) used a large (nearly 3,000), random sample of American young adults (ages 18-39) called “The New Family Structures Study.” This large, random sample avoided some of the major flaws of the earlier studies on this topic, such as very small sample sizes, biased sampling approaches (i.e., “convenience sampling”, where the data set is obtained by placing ads in homosexual publications and soliciting volunteers) and non-representative data sets (usually including only lesbian parents) (2). The study found that numerous, consistent differences do exist between children of parents who have had same-sex relationships and those with married, heterosexual parents. More specifically, he found that children of homosexual parents are more likely than those raised by heterosexual parents to suffer from poor impulse control, depression, suicidal thought, require mental health therapy, identify themselves as homosexual, choose cohabitation, be unfaithful to partners, contract sexually transmitted diseases, be sexually molested, have lower income levels, drink to get drunk, and smoke tobacco and marijuana. The study used a cross-sectional design (like a snapshot in time), and so the author was quick to point out that, although many differences were found, the results do not prove that the negative outcomes were caused by homosexual parenting itself. Nonetheless, this research clearly indicates that the claim that there are no differences in parenting outcomes must be re-evaluated with further research. And it also raises the possibility that homosexual parenting may, in fact, produce numerous, negative outcomes in adulthood, when compared to heterosexual parenting.
This publication (4) generated a firestorm of criticism and condemnation, some of which came from other researchers in social science. Accordingly, Regnerus answered the critics with new analyses (12). In this follow-up study, he discussed six of the most common and/or important criticisms and made several changes in response to the criticisms: an important change in the way he referred to homosexual parents and three major adjustments to groups and group assignments within the data set. The results were similar to those in the original article (4), but the magnitude of some of the differences declined somewhat. Thus, this new analysis of the data confirmed the conclusions of the original article; namely, that real differences do, in fact, exist in many outcomes that may be related to homosexual parenting vs. heterosexual parenting, including, but not limited to, sexuality, sexual behavior, educational attainment, smoking and arrests (12). And, once again, Regnerus emphasized that his results do not prove a cause-and-effect relationship between homosexuality and the negative outcomes. The onus now lies with his critics to prove him wrong.
Next, I want to summarize two related articles that compared the academic achievement of children with homosexual parents to those with married, heterosexual parents. Allen et al. (5) reexamined a previous study by Rosenfeld (11) that used a restricted sub-sample of a large, U.S. data base, the U.S. Public-Use Microdata Sample of the 2000 census. Rosenfeld concluded that “When one controls for parental SES and characteristics of the students, children of same-sex couples cannot be distinguished with statistical certainty from children of heterosexual married couples.” Using the same data set, but alternative comparison groups, an unrestricted sample and incorporating controls for the subgroups omitted in the Rosenfeld study, Allen et al., found that children being raised by same-sex couples are 35% less likely to make normal progress through school compared to children of heterosexual, married parents. The second of these two studies (6) used a much larger, random sample of the 2006 Canada census to examine high school graduation rates. The results showed that children living with gay and lesbian families were only about 65% as likely to graduate high school as were children living in heterosexual married families. Moreover, daughters of homosexual parents did considerably worse than sons in this study. This paper confirms the findings of Allen et al. (5), and these two studies, taken together, cast doubt on the ubiquitous claim that no difference exists; children living with same-sex parents do, in fact, perform poorer in school when compared to children from married, opposite sex families (6). However, these two studies also had a cross-sectional design; therefore, cause-and-effect inferences, or conclusions, cannot be made as to why these differences exist. Yet the results do raise the possibility that homosexual parenting itself may, in fact, result in poorer performance in school, when compared to heterosexual parenting.
The last study that I want to summarize addresses a difference found also in the Sarantakos (10) and Regnerus (4) studies: namely, that children of homosexual parents are much more apt to become homosexual themselves. The general consensus among researchers in this field of inquiry has been that there is no such difference. Schumm (7) conducted two meta-analyses of the results from previous studies concerning this issue (a meta-analysis combines and analyzes data from selected published studies in order to increase the sample size sufficiently to detect smaller differences and increase statistical power, thereby compensating for the small sample sizes and lack of statistically significant differences in the selected studies). The main thrust of his report deals with statistical analyses of the results of ten studies involving family histories of adult children with homosexual parents. The results of these analyses showed that 45% of the adult children of homosexual parents were homosexual. Using a verifiable figure of 3% as the prevalence of homosexuals (gay, lesbian and bisexual) in the general population, I calculated that adult children of homosexual parents are about 15 times more likely to be homosexual than are adult children of heterosexual parents. This difference is not only statistically significant, it is also clearly of a very large magnitude. But Schumm didn’t stop there. He proceeded to take data from 26 other studies that had concluded that there is no difference and to analyze them in a similar manner. When he restricted the data to those children who were 17 years old or older at the time the data were collected (in order to address the issue of adult outcomes per se), he found that 28% of the adult children of homosexual parents were homosexual, whereas only 2.3% of the adult children of heterosexual parents were homosexual. Thus, in this meta-analysis, adult children of homosexual parents were about 12 times more likely to be homosexual than were adult children of heterosexual parents. How could these extremely large and statistically significant differences (12-15 fold) come about? Schumm discussed five of the possibilities: 1) parental modeling of sexual orientation; 2) parental preference for the child’s sexual orientation; 3) the child’s greater questioning of their own sexual orientation; 4) parental desire for grandchildren; and 5) non-parental modeling of sexual orientation by homosexual friends of the homosexual parents. Regardless of the mechanism(s) involved, these results, taken together, strongly suggest that the post-natal environment of children with homosexual parents has a powerful influence on the development of homosexuality in the children, and that, in turn, argues persuasively against the popular notion that homosexual people are “born gay.” The strength of this argument is easily appreciated when one considers the fact that identical twin studies have demonstrated conclusively that the maximum contribution of all pre-natal influences (genetics, hormones, etc.) on the subsequent development of homosexuality can be no more than about 10%-15% (8). Some of the remaining 85%-90% post-natal influence could very well be a result of environmental and experiential factors inherent in the homosexual parenting context.
Possible Confounding Factors
Are there other factors commonly experienced by children of homosexual parents that may contribute to the negative outcomes of homosexual parenting reported in these studies? Osborne (9) identified several such factors, including divorce of the biological parents, social and structural stigmas targeting homosexual parents and their children, and multiple family forms (i.e., various combinations of remarriage, single parenting, cohabitation, adoption and step-parenting). The presence of these untested confounding factors makes it impossible, at present, to conclude that the sexual orientation of homosexual parents causes most of the negative outcomes in their children; nor does it rule out such a conclusion. That said, it appears to me that the most likely negative outcome of homosexual parenting that might be heavily influenced by the homosexual orientation of the parents is the homosexual orientation of their adult children, because 1) the effect has been reported in at least three separate studies using different data sets, and 2) the effect is so large (up to 12-15 fold) that it would not be expected to be attributable to unaccounted for, confounding factors. Perhaps the courts should not be so quick and eager to legalize homosexual adoption after all.
Gay activists have insisted for years that there is no difference in the outcomes of parenting by homosexuals when compared to heterosexual parenting. In 2005, the APA published a brief on this topic. In this brief, they cited 59 published articles in support of their summary claim that “Not a single study has found children of lesbian or gay parents to be disadvantaged in any significant respect relative to children of heterosexual parents.” However, in 2012, Marks published a detailed analysis and critique of the scientific merit of that brief and the literature upon which it was based. Marks found that the studies cited in support of the APA summary claim are woefully lacking in sound scientific design and principles and concluded that the strong assertions made in the APA brief were not substantiated by the published studies used, and were, therefore, unwarranted. Several other, more scientifically sound, studies have provided evidence that, indeed, there are many, often large, and very significant differences in the outcomes, possibly related to homosexual parenting compared to heterosexual parenting. These differences include inferior performance in school, a much lower graduation rate, poor impulse control, depression, suicidal thought, requirement for mental health therapy, cohabitation, unfaithfulness to sexual partners, contraction of sexually transmitted diseases, sexual molestation, lower income levels, drunkenness, tobacco and marijuana use and a very strong tendency for the adult children of homosexuals to self-identify as homosexual. In fact, adult children of homosexual parents are about 12-15 times more likely to be homosexual than are adult children of heterosexual parents. While it is not possible from these studies to conclude that the homosexual orientation of the parents directly or indirectly caused most of the negative outcomes found in their children, the results strongly suggest the possibility that homosexual parenting may not be equivalent to heterosexual parenting after all. The exception may be the homosexual orientation of the adult children of homosexual parents, which, in all likelihood, is heavily influenced by the homosexual orientation of the parents. At the very least, the claim that there are no differences in the outcomes of homosexual vs. heterosexual parenting should be re-evaluated with further research, based on the most scientifically sound research presently available. Perhaps the courts should not be so quick and eager to legalize homosexual adoption after all.
1. Committee on Lesbian, Gay, and Bisexual Concerns of the American Psychological Society. 2005. Lesbian and Gay Parenting. (click HERE)
2. Marks, L. 2012. Same-sex parenting and children’s outcomes: A closer examination of the American Psychological Association’s brief on lesbian and gay parenting. Social Science Research 41:735-751. (click HERE to download)
3. Sarantakos, S. 1996. Children in three contexts: Family, education, and social development. Children Australia 21(3), 23–31.
4. Regnerus, M. 2012. How different are the adult children of parents who have same-sex relationships? Findings from the New Family Structures Study. Social Science Research 41:752-770. (click HERE)
5. Allen, D., et al. 2013. Nontraditional Families and Childhood Progress through School: A Comment on Rosenfeld. Demography 50(3), 955-961.
6. Allen, D. 2013. High school graduation rates among children of same-sex households. Review of Economics of the Household 11:635-658.
7. Schumm, W. 2010. Children of Homosexuals More Apt to Be Homosexual? A Reply to Morrison and to Cameron Based on an Examination of Multiple Sources of Data. Journal of Biosocial Science 42:721-742.
8. Aist, J. 2012. Are Homosexual People Really “Born Gay”? (click HERE)
9. Osborne, C. 2012. Further comments on the papers by Marks and Regnerus. Social Science Research 41:779-783.
10. Sarantakos, S. 2000. Same-sex Couples. Parramatta, N.S.W: Harvard Press
11. Rosenfeld, M. 2010. Nontraditional Families and Childhood Progress Through School. Demography 47(3):755-775.
12. Regnerus, M. 2012. Parental same-sex relationships, family instability, and subsequent life outcomes for adult children: answering critics of the new family structures study with additional analyses. Social Science Research 41:1367-1377. (click HERE)
(For more articles on HOMOSEXUALITY, click HERE)