New Study on Lesbian Parenting

Fresh from Warren Throckmorton.

Adolescents of the U.S. National Longitudinal Lesbian Family Study: Sexual Orientation, Sexual Behavior, and Sexual Risk Exposure.

Gartrell NK, Bos HM, Goldberg NG.

Department of Psychiatry and Center of Excellence in Women’s Health, University of California, 3570 Clay St., San Francisco, CA, 94118, USA,

This study assessed Kinsey self-ratings and lifetime sexual experiences of 17-year-olds whose lesbian mothers enrolled before these offspring were born in the longest-running, prospective study of same-sex parented families, with a 93% retention rate to date. Data for the current report were gathered through online questionnaires completed by 78 adolescent offspring (39 girls and 39 boys). The adolescents were asked if they had ever been abused and, if so, to specify by whom and the type of abuse (verbal, emotional, physical, or sexual). They were also asked to specify their sexual identity on the Kinsey scale, between exclusively heterosexual and exclusively homosexual. Lifetime sexual behavior was assessed through questions about heterosexual and same-sex contact, age of first sexual experience, contraception use, and pregnancy. The results revealed that there were no reports of physical or sexual victimization by a parent or other caregiver. Regarding sexual orientation, 18.9% of the adolescent girls and 2.7% of the adolescent boys self-rated in the bisexual spectrum, and 0% of girls and 5.4% of boys self-rated as predominantly-to-exclusively homosexual. When compared with age- and gender-matched adolescents of the National Survey of Family Growth, the study offspring were significantly older at the time of their first heterosexual contact, and the daughters of lesbian mothers were significantly more likely to have had same-sex contact. These findings suggest that adolescents reared in lesbian families are less likely than their peers to be victimized by a parent or other caregiver, and that daughters of lesbian mothers are more likely to engage in same-sex behavior and to identify as bisexual.

From the abstract it looks as though this has something for all parties. For the pro-GLBT lobby a clear(ish – 78 isn’t a perfect sample size) indication that lesbians are not more likely to abuse their children then the average for the population. On the other hand, nurture advocates will be excited by the evidence that parental sexual practice might help shape children’s sexual practice.

14 Comments on “New Study on Lesbian Parenting

  1. Hi Peter,

    thank you for this – another thing you’ve posted that I wouldn’t otherwise have been aware of. Have one small quibble though: you comment that “lesbians are not more likely to abuse their children then the average for the population”, but that’s not really an accurate reflection of the abstract, which says “These findings suggest that adolescents reared in lesbian families are less likely than their peers to be victimized by a parent or other caregiver”. “Not more likely” isn’t a synonym for “less likely”…

    Will reply to your other replies to me when it’s not so late!

    in friendship, Blair

  2. I generally think that this study is a great thing, but it’s worth noting that it has been criticized as at least potentially inaccurate/misleading because 1) people likely to abuse their children do not sign up for longitudinal studies 2) the children in the study are not having their sexualities develop typically because they have grown up having it studied. Just food for thought.

  3. The main conclusion is that the Archives of Sexual Behavior, epublication Nov. 6, 2010: is willing to throw out academic integrity out the window by publishing a study that is horribly compromised. This is from the same the same lesbian author who published the widely publicized study that was purported in the popular press to show that lesbians make “better parents” because kids are “better adjusted”. Like the previous study, the new one is supported by the Gill Foundation – a homosexual advocacy PAC.

    Fortunately, the present journal is only some e-publication and not the Journal of Pediatrics – a previously respected journal.

    As BTCarolus points out, the study group is hardly a random sample but one chosen at book fairs and gay pride political rallies and participants had agreed to a very long study time (17 years). The previous control group used unpublished data from a heterosexual sample that were very different in terms of demographics (college education being the main one). This new study uses a data from U.S. Department of Health, Human Services. (2006). National survey of
    family growth. Cycle 6: Sample design, weighting, imputation, and
    variance estimation. Vital Health Statistics, 142, 1–82. There is no discussion of demographics of the “control group” whatsoever.

    The study the control “NSFG data were weighted to ensure that the sample was similar to the United States population in terms of gender, age, and race/ethnicity.” In contrast, the study data of lesbian families most definitely did NOT reflect US population – being much more white and college educated.

    What is the first sentence in the discussion section?:

    “A key finding in the current study was that none of the NLLFS adolescents reported physical or sexual abuse by a parent or other caregiver. This finding contradicts the notion, offered in opposition to parenting by gay and lesbian people, that same-sex parents are likely to abuse their offspring sexually”

    This is outrageous to use the descriptor “key” with the observation. To highlight it by making it the lead sentence is simply embarrassing. It is not a “key” finding in that the sample size of lesbian families is very small and very non-representative. And what is is the rate of victimization in the heterosexual population? Not given! So obviously no p-value analysis was performed. The victimization question was not part of the control group survey.

    This study says nothing about 1) homosexuals are more likely than the general population to have suffered from child abuse and that 2) victims of child abuse are more likely to grow up to be child abusers, thus one should be wary of society giving blanket approval to same sex union families.

    • I think the sample size is the key criticism (if any) to be made of this study. There are, as you point out, issues in the composition of the sample as well.

      That said, anybody who chooses to accept the findings of this survey must also accept that Yarhouse’s longitudinal study of ex-gays is equally valid (same-ish sample size / sample composition issues).

      • "That said, anybody who chooses to accept the findings of this survey must also accept that Yarhouse’s longitudinal study of ex-gays is equally valid (same-ish sample size / sample composition issues)."

        Not true. The problem is wildly irresponsible overreach in the lesbian family studies. The conclusions of the two lesbian family studies are 1) children of lesbian parents are "better" adapted (where part of the term "better" is more accepting of lesbianism) and 2) lesbians are less abusive to their children. Both conclusions are in no way supported by the respective "studies". The first was doomed by non-equivalent (and unpublished) control group whose demographic differences easily explains away any difference in "well adjustedness". In the second, we have a cohort that doesn't have any self-reported abuse and concludes that abuse is therefore is lower among lesbian families. Simply appalling. The control group wasn't even asked about abuse. I can certainly find a group of smokers that doesn't develop cancer and heart disease. If I then conclude, that smoking doesn't increase the risk of cancer and heart disease but rather decreases it, I would be rightly tossed out of the medical research community.

        A decent editor should have rejected outright the abuse portion the second paper (which the authors repeatedly place in the most prominent spot) and had it only include the portion about the sexuality of children of lesbian partners. What do we find here? Daughters of lesbian couples more likely to engage in lesbian activity. Reasonable (but flies in the face of born gay). Age of first sexual contact is delayed in children of lesbian parents. This later conclusion is doomed by lack of control with respect to socio-economic and education levels between the non-random lesbian parent cohort and the general sample. Certainly, socio-economics factors have been shown to play a large role in age of first sexual activity, so one must reject any conclusion about this issue.

        In contrast, there is no overreach in the conclusions of the Yarhouse study. It simply says, here was our protocol and here were our results – mileages may vary.

  4. Hello,

    just to say ‘hear, hear’ Peter to your reply to Eusebius – would have made the same point myself about the sample size of the Jones/Yarhouse longitudinal study.

    Eusebius, can you cite any evidence for assertions 1 and 2 in your last paragraph?

    in friendship, Blair

  5. Homosexuality and history of childhood sexual abuse:
    "…one might expect that homosexual males would be more likely to have been sexually molested as children. Several studies confirm this hypothesis.

    Shrier and Johnson interviewed 300 intercity boys as part of a general intake health history at an adolescent outpatient clinic. Forty of the adolescents disclosed an experience of being sexually assaulted by a male before puberty. These 40 were compared to an age-matched control group from the same sample of boys who do not report abuse. Shrier and Johnson found that while 90% of the control group reported they were heterosexual, only 42.5% of those reporting sexual victimization said they were heterosexual – 47.5% said they were homosexual, and 10% said they bisexual. Only 6 of the 40 victimized boys had revealed the molestation to anyone prior to participation in the study interview.(Johnson 1985)

    Shrier and Johnson feel that the 40 cases found may represent only a faction of the actual incidence, because although nearly half of the clinic patients are under 15, not one boy under 15 admitted molestation. In addition only 6 of the 40 who admitted being molested were under age 17. This suggests that younger adolescents may be extremely unwilling to admit previous molestation and may do only because they are suffering from some sexually related problem.(Johnson, 1985)

    In a study by Remafedi, 239 men aged 13 to 21 who identified themselves as homosexual or bisexual were queried. Of these 42% had been sexually abused or assaulted. In addition 30% had attempted suicide; 23% used cocaine; 66% used marijuana regularly; 29% had been arrested; 11% had accepted money for sex.. Based on the number of unprotected sexual acts and number of sexual partners, the author concluded that 63% of these young men were at extreme risk for HIV infection.(Remafedi, 1994) It was not clear from the study if the 42% who had been abused were among those who attempted suicide, used drugs, or engaged in unsafe sex.

    In 1992 Lynda Doll and associates published a study of "self-reported childhood and adolescent sexual abuse among adult homosexual and bisexual men." They identified 1,001 men who had engaged in sexual activity with another man in the previous five years and had attended a STD clinic. …Among the 1,001 participants 360 (37%) reported sexual contact before the age of 19 with a partner whom they perceived as being older or more powerful than themselves (94% of these the first such contact was with a male). In 53% of the cases the abusers were over 19, in 43% of the cases they were family members. Force was used in 49% of all contacts.
    …(Huges 2000) reported in the Journal of Gay and Lesbian Social Services by Hughes and team in 2000. In this study, 41% of lesbians reported CSA while among straight women the number was 24%.
    (Wilson and Widom 2009) Childhood physical abuse and neglect were not significantly associated with same-sex cohabitation or sexual partners. Individuals with documented histories of childhood sexual abuse were significantly more likely than controls to report ever having had same-sex sexual partners (OR = 2.81, 95% CI = 1.16-6.80, p = .05); however, only men with histories of childhood sexual abuse were significantly more likely than controls to report same-sex sexual partners (OR = 6.75, 95% CI = 1.53-29.86, p = .01). These prospective findings provide tentative evidence of a link between childhood sexual abuse and same-sex sexual partnerships among men, although further research is needed to explore this relationship and to examine potential underlying mechanisms.

  6. Glasser 2001:

    Results Among 747 males the risk of being a perpetrator was positively correlated with reported sexual abuse victim experiences. The overall rate of having been a victim was 35% for perpetrators and 11% for non-perpetrators. Of the 96 females, 43% had been victims but only one was a perpetrator. A high percentage of male subjects abused in childhood by a female relative became perpetrators. Having been a victim was a strong predictor of becoming a perpetrator, as was an index of parental loss in childhood.

    Conclusions The data support the notion of a victim-to-victimiser cycle in a minority of male perpetrators but not among the female victims studied. Sexual abuse by a female in childhood may be a risk factor for a cycle of abuse in males.

  7. Hi Eusebius,

    OK, I think my question's been answered in abundance :) Not sure what else to say except a few things… I don't have the statistical knowledge to go through all your examples, but I would just note that you don't apply the same hermeneutic of suspicion to the studies you quote, as you did to the study in Peter's original post. It isn't clear from what you say whether there could be concerns about sample size or composition in any of the studies you mention; though in fairness you've quoted enough to show that your 2 assertions above might be too simplistic.

    Also, it's worth following the link to Warren Throckmorton's site – some cogent comment there about the study in the original post I thought.

    in friendship, Blair

  8. I've always been dubious about studies like these. I mean, the only way to know "for a fact" how a person will turn out would be if they could grow up/be raised by each possible parental group- which is impossible. As in, a male & female child first being raised in a mom/dad household, then they are raised next in a lesbian HH, then a gay male HH, etc…I say this because I believe most of who we are is inherent in us when we're born.

    If you look at the finding that more girls than boys reported themselves as bisexual- well, you don't know if those specific girls would have reported that no matter who raised them, you see what I mean? There's also the point that perhaps these girls, being raised in a female, same-sex HH, were more open to bisexuality merely because the rigid str8 path was not being thrust upon them as the only option. Since it didn't affect the same percentage in boys- it appears the outcome had less to do with the mothers own sexuality & more to do with the individual children.

    Anyway- interesting topic…

    • You fundamentally misunderstand the nature of statistical research like this. The data indicates that children in certain parental setups are *more likely* to have certain outcomes. That doesn't mean they all will, but that there are factors which contribute more in certain environments.

      The argument that since boys and girls have different outcomes the answer must be biological doesn't fit the data. If this were so we would expect to see identical outcomes regardless of the parenting. This is not so, so the first obvious line of exploration is what is different about the parenting. Unless of course you want to argue that children of gay parents are somehow biologically fundamentally different than children of straight parents…

      • "You fundamentally misunderstand the nature of statistical research like this."

        No, I get it just fine. I just happen to think that the outcomes seem skewed, therefore I don't place a lot of faith in them.

        "Unless of course you want to argue that children of gay parents are somehow biologically fundamentally different than children of straight parents…"

        No, I don't really like arguing, thanks. And maybe the children of gay parents ARE biologically different, who knows? I may not be a statistician, but I AM a lesbian raising a boy & a girl- so I do have a bit of insight that many do not. I know that my children, compared to their str8 friends, are much more open & non-prejudiced. This is, for me, much more telling about how children raised in a lesbian household actually fare…and much more important than a bunch of forms & figures.

        • It's very likely that your children are very well adjusted and open to modern societal structures. You've obviously done a good job of raising them. But macro-level analysis like the research above is good at telling us what, on average, the differences between two groups (in this case children of gay and straight parents) are. Within each of those groups there will always be a variance of outcomes, but one anecdotal report does not destroy the validity of the large sample.

          If you want to argue that "the outcomes seem skewed", you need to present more evidence then just "because it doesn't fit with one example I have experienced" (to paraphrase). At the moment you come across as someone who is railing against this challenging research, not because you can demonstrate that the macro-level analysis is incorrect, but because you don't like the consequential realities of the research. This is exactly the same as many of my conservative colleagues who reject the clear statistical evidence that there is, on average, a biological component of some degree to an individual's homosexuality.

  9. I am looking at research that makes no sense at all. Mind you, I'd be one of the first ones to bask in the glow that lesbian couples make good parents- but the numbers here don't necessariy reflect any sort of reality…

    You're telling me that 78 teenagers answered questions- honestly. Teenagers honestly answered questions about their sexuality? Yes, of course- teens are known for their honesty! The study showed nearly a fifth of the girls have already concluded that they are bisexual- at 17? Wow! These girls really must have gone through some sort of intense introspection to have arrived at this at such a young age…Or, are they mistaken- or simply think it sounds like something cool to say? And not one girl, not one out of 39, expressed same-sex attraction…even though other studies say that's not possible, percentage-wise?

    At any rate, thanks for the interesting opinions. Cheers!

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.