US National Longitudinal Lesbian Family Study: Psychological Adjustment of 17-Year-Old Adolescents

To the Editor

We applaud the paper by Gartrell et. al. who reported that adolescents (age 17), who were born by donor insemination and were raised by lesbian couples, were rated significantly higher in social, school/academic, and total competence and significantly lower in social problems, rule-breaking, aggressive, and externalizing problem behavior than their age-matched counterparts in Achenbach’s Child Behavioral Check List (CBCL), a normative sample of American youth. (1) The importance of this study is that it is longitudinal (17 years) reporting on 78 adolescents with a sample retention of 93%. The researchers have published other data about this group of children and report that more is to come from this data set. In a more recent excellent review of the literature, Biblarz noted the essential normality of growth and development of children raised by lesbian parents (2) There are only a few long term studies of children with lesbian parents with a focus on child development and the Gartrell cohort, with its high retention rate and relative diversity of parents places it among other longitudinal studies in this manner.

We agree with the Commentary by Dr. Joseph Hagan, one of the editors of the AAP Publication: Bright Futures, “and when we see these moms or dads with their kids in our practice, we call them families� (3). The authors recognize the limitations of their study: a non random study and the lack of self report by the subjects or by teacher report in this specific analysis. In addition, the study group is compared to the historical normative comparison group with a much higher geographic diversity and a higher sampling of minorities. This study is what it is: a study of lesbian mothers who tend to have higher education and socioeconomic status than the norm. By definition there are methodological issues which have been raised by other letters and addressed by the authors. The value of this study remains: a naturalistic longitudinal observation of a particular group of parents raising children to the best of their ability. The children turn out OK, even when a significant percentage of the parents of study participants have separated from their significant others just as many married couples experience divorce in a life time.

The CBCL is essentially a research tool- it grapples with many of the issues that arise naturally in the development of a child but it is nonetheless merely a tool to tap parental concerns about a child. From that standpoint, it is the gold standard of research studies and that this paper stood up under the scrutiny of such a tool is significant.

What are the take home messages of this study from our clinical perspectives as a primary care pediatrician and developmental and behavioral pediatrician with more than 50 practice years between us? Parents who love and nurture their children generally raise them to be competent adults. Families differ (4). Our society nurtures and supports families and provides enhanced social and legal supports to those family members raising children as married couples. Some states have legalized gay marriage while others have not. A number of states legally limit non-married couples or single adults from adopting or becoming foster parents. From this study, we learn of the essential “normality� of children raised in lesbian couples born through donor insemination. People have children in and out of marriage and live their lives in all sorts of family arrangements with or without partners and with or without other support systems. What we do know, as validated by this study, is that effective parenting in a loving and nurturing support system is critical to the development of competent adults and future citizens of this county. The sooner policy makers and legislators realize this, the better off we and especially our children will be.

1. Gartrell N Bos H. US National Longitudinal Lesbian Family Study: Psychological Adjustment of 17-Year-Old Adolescents Pediatrics 2010;126;28-36;

2. Biblarz TJ, Stacey J. How Does the Gender of

Parents Matter? Journal of Marriage and

the Family. 2010;72(1):3–22

3. Hagan J F. What Shall We Call Them? Pediatrics 2010;126;175-176;

4. Schor EL. Family Pediatrics: Report of the Task Force on the Family

Pediatrics 2003;111;1541-1571

Ben Siegel, M.D. Primary Care Pediatrician, Marilyn Augustyn, M.D. Developmental - Behavioral Pediatrician Boston Medical Center, Boston University School of Medicine We have nothing to disclose

Conflict of Interest:

None declared

Gartrell and Bos concluded that adolescents raised in lesbian-mother families since birth demonstrate more competencies and fewer behavioral problems than a comparator group of adolescents. Their interpretation is subject to two additional limitations, however, neither of which is noted by the authors. First, the differences they observed were not adjusted for observed (or unobserved) differences in socioeconomic factors. It is unclear why conventional statistical adjustments were not performed, as Table 1 indicates clearly that some rudimentary data were available: the comparator adolescents were more likely to have black or Latino/a parents and more likely to be from the Midwest or South. Second, and perhaps more importantly, the differences they observed may be due to social desirability bias. The potential tendency for study participants to "present themselves and their families in the best possible light" [1] (p.279) was acknowledged in the first report of data from the NLLFS. Given that the present study makes use of a comparator group to draw inferences about same-sex parenting, I would expect this limitation to be even more strongly emphasized. The effect of social desirability bias is known to be intensified when study subjects are similar, in terms of social distance, to those administering the study [2]. Therefore, more information about how study participants were recruited, samples of the language used in follow up letters, and examples of how prior findings from the NLLFS were presented to the lesbian community would be helpful for readers attempting to assess the potential effects of social desirability bias. Prior studies of homonegativity [3,4] and abstinence-only sex education [5] have employed methodologies to address potential social desirability bias, but use of these techniques is not described in the present study.

Alexander C. Tsai, MD, PhD
Department of Psychiatry
University of California at San Francisco

1. Gartrell N, Hamilton J, Banks A, Mosbacher D, Reed N, Sparks CH, Bishop H. The national lesbian family study: 1--interviews with prospective mothers. Am J Orthopsychiatry. 1996;66(2):272–281.
2. Nederhof AJ. Methods of coping with social desirability bias: a review. Eur J Social Psychol. 1985;15(3):263-280.
3. Morrison MA, Morrison TG. Development and validation of a scale measuring modern prejudice toward gay men and lesbian women. J Homosex. 2003;43(2):15-37.
4. Morrison TG, Parriag AV, Morrison MA. The psychometric properties of the Homonegativity Scale. J Homosex. 1999;37(4):111-126.
5. Jemmott JB, Jemmott LS, Fong GT. Efficacy of a theory-based abstinence-only intervention over 24 months: a randomized controlled trial with young adolescents. Arch Pediatr Adolesc Med 2010;164(2):152-159.

Conflict of Interest:

None declared

This article is interesting and well-written. The conclusions, however, that children of lesbian mothers demonstrate superior psychological adjustment compared to children of traditional families, even when the parents separate before the children are fully grown, are, on their face, a bit fantastic. Is the implication, that fathers are an undesirable component of the family, to be taken at face value? Such a conclusion, notwithstanding the caveats acknowledged by the authors in their discussion, begs for a better study with randomly selected subjects, objective measurement and followup, and appropriate control groups.

Conflict of Interest:

None declared

I can certainly appreciate the need to undertake longitudenal studies concerning the psychological outcomes of children of lesbian families. In the 80's the stigmitization of families and children with lesbian or gay parents was significant and at times severe. However, I must take issue with the interpretation and conclusions of the authors as well as the decision by Pediatrics to publish the article.

The study conclusions were based solely on the parental responses to the Child Behavior Checklist. Parents who complete CBCL's on their own children for a study that could potentially report negative findings on the outcomes of children raised in lesbian homes have a clear, self- serving bias. The fact that the study chose not to include the self reported CBCL or the teacher CBCL is mentioned, but it begs the point? Why? Were the results contradictory? On the surface it appears that the study authors are only reporting data that supports a specific, predetermined view-point. I will not be referencing this article or results as valid until ALL of the data is made public for review.

Conflict of Interest:

None declared

I write regarding the article by Nanette Gartrell and Henny Bos entitled "US National Longitudinal Lesbian Family Study: Psychological Adjustment of 17-Year-Old Adolescents" published online.

This study is important for several reasons. First, the NLLFS is a prospective, longitudinal study that has followed a cohort of American lesbian families enrolled between 1986 and 1992 with a remarkable retention rate of 93%. The authors point out that the study population differed from the comparison population in a number of ways (had less racial/ethnic diversity, the parents volunteered to be in the study and were drawn from three major metropolitan centers at the beginning of the study, and the comparison group was overall of higher SES status), but this limitation does not negate the importance of the findings. One of the strengths of this study is that it used the CBCL parent report, an instrument that has been well-validated on a variety of populations, to assess adolescent psychological adjustment. This article also represents a major contribution to the literature as it is the first time we have data on older adolescents who have been raised long-term by lesbian mothers.

The authors found that adolescents raised since birth in lesbian- mother families are overall well-adjusted compared with a normative, but not matched, population of American adolescents.

As a pediatrician who specializes in and advocates for children and adolescents in foster care, the results presented in this study reinforce my clinical experience with same sex couples who foster and adopt. We have a shortage of foster parents nationally, and there are 120,000 children in foster care awaiting adoption in the United States. Yet, three states legally prevent same-sex couples from adopting children, even when they foster those same children, and other states are considering this more restrictive policy (1). Some private agencies refuse to recruit, train and certify same-sex couples to be foster parents, effectively eliminating them as a resource for children. This study which shows that adolescents raised since birth in lesbian-mother families are faring well has great relevance for informing policy makers and judges who decide which adults should be allowed to foster and adopt children.

1. Gates G, Badgett LMV, Macomber JE, Chambers K. Adoption and Foster Care by Lesbian and Gay Parents in the United States. March 23, 2007. Available at: http://www.urban.org/publications/411437.html

Conflict of Interest:

No competing interests

Prof Shum responds to the reassuring data about the adolescent children of lesbian parents who participated in the NLLFS presented in the important recent article by Gartrell and Bos and to comparable evidence about the negligible effects of parental gender on children that Timothy Biblarz and I discuss in “How Does the Gender of Parents Matter,” (Journal of Marriage and Family Feb 2010), with the suggestion that “academic debate may be shifting from whether lesbian parents are as fit as heterosexual parents to the extent to which heterosexual parents (and fathers, in particular) may be less fit than lesbian parents.” Similarly, the title of a recent article in the Atlantic magazine (July-Aug 2010) asks, “Are Fathers Necessary?”

Such responses to these studies, however, are misguided. The crucial findings in both publications support the broader implications of cumulative research on effective parenting. The positive outcomes associated with planned lesbian parenthood do not depend on sexual orientation per se, but on selection effects associated with the degree of deliberation and effort involved in choosing to become a lesbian parent and with the shared and supportive child-rearing values that are particularly prevalent among the lesbian parents in these studies.

Although popular discourse inevitably will indulge in rhetorical and polarizing questions that pit lesbian parents against heterosexual (or gay male) parents, scholars should resist emulating such unconstructive debates. These recent studies underscore the comparative irrelevance of sexual orientation and gender to effective parenting. This should help to shift academic debate away from questions that compare parents by sexual orientation or gender to those that investigate ways to promote effective parenting by women and/or men within the full range of family structures in contemporary society.

Conflict of Interest:

I co-authored "How Does the Gender of Parents Matter?" Journal of Marriage and Family 72, n.1 (Feb 2010).

We are very pleased that so many researchers internationally have taken an interest in this largest, longest-running, prospective, in-depth study of American lesbian families in which the children were conceived by donor insemination. In response to e-Letters on the online forum, we would like to clarify several points. The socioeconomic status (a composite of parental education and income) of the lesbian family sample was very similar to that of the normative sample: 82% of the lesbian families are middle—upper-middle/upper class, and 88% of the normative sample are middle—upper-middle/upper class, so in fact the normative group is slightly higher in socioeconomic status (see Table 1). Of course every scientific investigation has shortcomings, and we acknowledged in the manuscript that the lesbian sample is not representative, that other factors could be taken into account, and that the mothers volunteered to participate. The concern that the data in our article are based only on parental reports is certainly valid. However, we have an extensive data base of information (including numerous standardized instruments) that was obtained through self-reports from the 17-year-old adolescents themselves that are currently being analyzed and will be the subject of future publications. The US National Longitudinal Lesbian Family Study represents the first truly longitudinal look at adolescents reared from birth in planned lesbian families, and as such provides an important window into their lives.

Nanette Gartrell, MD Henny Bos, PhD

Conflict of Interest:

Authors of the article 'US National Longitudinal Lesbian Family Study: Psychological Adjustment of 17-year-Old Adolescents

The article raises an unusual methodological problem for surveys. There is a well known strong tendency for subjects to volunteer for surveys which show them in a good light. In this case there is an enormous political incentive for mothers to volunteer for such a survey and ensure upbringing is exemplary. However the measures presented are tests on the children and the interesting question arising is: is an indirect volunteer effect possible – the attitudes or even instructions of the mothers to the children affecting the test results? I sympathise greatly with the researchers who imply a genuine random survey even today would be very difficult, but no definitive answer can really be possible in its absence.

Conflict of Interest:

None declared

Dear colleagues,

I have read the article of Nanette Gartrell and Henny Bos "US National Longitudinal Lesbian Family Study: Psychological Adjustment of 17-Year-Old Adolescents".

I am surprised with the lack of any attepmt of the authors to address the very important factors in the study.

As it can be easily seen from Table 1, the populations in comparison are very different in race composition, socio-economic status of participants, and region of the country. The population of chidren from the conventional sample (Achenbach Normative CBCL Sample) has many times more minorities and many more children from the South.

It was shown not once that the outcomes of the study are strongly dependent on the above mentioned factors, and exactly in the direction that the study reveals.

Only gender and group (NLLFS vs.Achenbach Normative CBCL Sample) were used as predictors. I can not understand wny the authors did not make proper adjustments for other factors. They do mention it as one of the limitations of the study. It would be very easy to match the study population with the appropriate control population. Other way to treat the problem would be to adjust for the factors of race, region and socio- economic status within the MANOVA analysis (although the sample size becomes critical in this case).

Also, I am surprised that the editorial board and reviewers did not pay attention to such an obvious deficiency of the study.

In my opinion, above mentioned creates a strong doubt in the conclusions of the study.

Respectfully, Alex I. Kartashov, Ph.D.

Conflict of Interest:

None declared

There is no question that the US National Longitudinal Lesbian Family Study (NLLFS) [1] is commendable for its depth, duration, and contemporary relevance. Its results follow a recent report by Biblarz and Stacey [2] which yielded findings that suggested that lesbian parents were better parents than heterosexual parents. Thus, since Gartrell and Bos [1] found the same relative outcome, academic debate may be shifting from whether lesbian parents are as fit as heterosexual parents to the extent to which heterosexual parents (and fathers, in particular) may be less fit than lesbian parents.

However, at least 67% of the mothers in the NLLFS [3] had at least a college education compared to approximately 28% of women of similar age in US Census data (after which the Achenbach comparison data was modeled), yielding an odds ratio = 5.14 (p < .001, 95% CI, 2.71-9.77). The relatively better adjustment of the children of lesbian parents may reflect, at least in part, higher levels of education or different sibling constellations relative to heterosexual parents rather than parental gender or sexual orientation per se.

Gartrell and Bos [1] also reported higher levels of aggressive behavior (effect size = 0.49, p < .07) among adolescents whose lesbian mothers reported that their children had been stigmatized. While it is quite likely that stigmatization could lead to higher aggressiveness, it is not impossible that higher aggressiveness could lead to higher levels of peer rejection, which could be interpreted as homophobic.

In previous reports from the NLLFS, other outcome variables were considered that were not reported in this wave. In addition to child adjustment, other variables of interest may include parental preference and expectations for adolescent's sexual orientation, as well as the adolescent's sexual orientation, level of secure attachment, adherence to delayed gratification principles, level of sexual activity, gender flexibility or nonconformity to traditional gender roles, and plans for cohabitation before marriage, among others.

While the NLLFS results [1] may appear to support policies favoring the rights of gay, lesbian, or bisexual (GLB) parents, they could also be interpreted as evidence against the fitness of male parents, regardless of sexual orientation, placing gay father couples at risk for being seen as even less fit than couples with only one father. However, at present, there is far less research on gay fathering than there has been for lesbian parenting, leaving us with few empirical answers concerning any effects of gay fathering [4].

1. Gartrell N, Bos H. US National Longitudinal Lesbian

Family Study: psychological adjustment of 17-year-old

adolescents. Pediatrics 2010; 126(1): 1-9. 2. Biblarz T, Stacey J. How does the gender of parents

matter? Journal of Marriage and Family 2010; 72(1):

3-22. 3. Gartrell N, Deck A, Rodas C, Peyser H, Banks A.

The National Lesbian Family Study: 4. Interviews with

the 10-year-old children. Am J Orthopsy 2005; 74(4):

518-524. 4. Sirota T. Adult attachment style dimensions in women

who have gay or bisexual fathers. Arch Psychiatr Nurs

2009; 23(4):289-297.

Conflict of Interest:

Dr. Schumm has served as an expert witness for the State of Florida in a previous trial concerning gay adoption.