The core of the argument is that a child's best interest is to be with his/her biological parents in a stable, low-conflict marriage. Now while the stable, low-conflict marriage is important, what the extensive body of research has shown over the last 40 years or so is that the sexual orientation of the parents has no detrimental impact on the outcome of children.
A. Positive research
There are many many studies which show that there is no difference between same-sex and opposite-sex families.
There are small-scale targeted studies:
Biblarz, T., Stacey, J. (2010). How does the gender of parents matter? Journal of Marriage and Family, 72,3-22.http://www.famigliearcobaleno.org/pu...nts-Matter.pdf
Bos, H. M. W., Gartrell, N. K., van Balen, F., Peyser, H. and Sandfort, T. G. M. (2008), Children in Planned Lesbian Families: A Cross-Cultural Comparison Between the United States and the Netherlands. American Journal of Orthopsychiatry, 78: 211–219http://www.nllfs.org/images/uploads/...ilies-2008.pdf
Henny M. W. Bos, Frank van Balen, Children in planned lesbian families: Stigmatisation, psychological adjustment and protective factors,Culture, Health &Sexuality, Vol. 10, Iss. 3, 2008.http://www.narcis.nl/publication/Rec...:uva.nl:307079
Bos, Henny M. W., Hakvoort, Esther M.,Child adjustment and parenting in planned lesbian families with known and as-yet unknown donors (2007) Journal of Psychosomatic Obstetrics & Gynecology, 28, 121-129http://home.medewerker.uva.nl/e.m.ha...voort_2007.pdf
Henny Bos and Theo G. M. Sandfort, (2010) Children’s Gender Identity in Lesbian and Heterosexual Two-Parent Families, Sex Roles, 62, 114-126http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2807026/
Farr, R. H., & Patterson, C. J. (2009). Transracial adoption among lesbian, gay, and
heterosexual couples: Who completes transracial adoptions and with what results? Adoption Quarterly, 12, 187–204.http://people.virginia.edu/~cjp/articles/fp09.pdf
Biblarz, Timothy J., Savci, Evren (2010) Lesbian, Gay, Bisexual, and Transgender Families, Journal of Marriage and Family, 72, 480-497http://onlinelibrary.wiley.com/doi/1...0.00714.x/full
Bos, H. M. W., van Balen, F., & Van den Boom, D. C. (2007). Childadjustment and parenting in planned lesbian-parent families.American Journal of Orthopsychiatry, 77, 38–48.http://www.meerdangewenst.nl/documenten/AJOP.pdf
Perrin,E.C.,&AmericanAcademyof Pediatrics,Committee on PsychosocialAspects of Child, Family Health. (2002). Technical report:Coparent or second-parent adoption by same-sex parents. Pediatrics,109, 341–344.http://pediatrics.aappublications.or...6c302e03b8d796
Vanfraussen, K., Ponjaert-Kristoffersen, I., & Brewaeys, A. (2002).What does it mean for youngsters to grow up in a lesbian familycreated by means of donor insemination? Journal of Reproductiveand Infant Psychology, 20, 237–252.http://psycnet.apa.org/?fa=main.doiL...2002-11380-003
Parenting and Child Development in Adoptive Families: Does Parental Sexual Orientation Matter?Rachel H. Farr, Stephen L. Forssell, Charlotte J. Patterson Applied Developmental Science Vol. 14, Iss. 3, 2010
Then there are longitudinal studies:
US National Longitudinal Lesbian Family Study: Psychological Adjustment of 17-Year-Old Adolescents , Nanette Gartrell and Henny Bos Pediatrics 2010; peds.2009-3153; published ahead of print June 7, 2010
http://pediatrics.aappublications.or....full.pdf+html
Gartrell, Nanette, Bos, Henny, Goldberg, Naomi, (2010) Adolescents of the U.S. National Longitudinal Lesbian Family Study: Sexual Orientation, Sexual Behavior, and Sexual Risk Exposure, Archives of Sexual Behavior.
http://www.familieslg.org/_comun/bib..._Sex_Behav.pdf
Gartrell, N., Deck, A., Rodas, C., Peyser, H., & Banks, A. (2005). The
National Lesbian Family Study: 4. Interviews with the 10-year-old
children. American Journal of Orthopsychiatry, 75, 518–524.
http://www.nllfs.org/images/uploads/...-olds-2005.pdf
Finally, there are large scale nationally representative studies which also show no difference:
Blackwell DL. Family structure and children’s health in the United States: Findings from the National Health Interview Survey, 2001–2007. National Center for Health Statistics. Vital Health Stat 10(246). 2010.
Nontraditional Families and Childhood Progress Through School; Michael J. Rosenfeld; Demography, Volume 47, Number 3, August 2010, pp. 755-775
Finally there is another large scale study (actually touted as the largest of its kind yet completed) which is nearing publication. The interim report of the project found that:
"On measures of general health and family cohesion children aged 5 to 17 years with same-sex attracted parents showed a significantly better score when compared to Australian children from all backgrounds and family contexts. For all other health measures there were no statistically significant differences."
The Australian Study of Child Health In Same-Sex Families (ACHESS) - Interim report
In fact, the volume of research in this field has lead to a robust consensus being adopted by almost every major professional body.
Consensus Positions
American Psychological Assocation
Position Statement in Support of Legal Recognition of Same-Sex Civil Marriage
Amicus brief submitted in support of 9th Circuit Court of Appeals challenge to California Prop 8
Lesbian and Gay Parenting Resource Publication
Canadian Psychological Association
Brief presented to the Legislative House of Commons Committee on Bill
C38
American Academy of Pediatrics
Policy statement - Coparent or second-parent adoption by same-sex parents
Australian Psychological Society
Lesbian, Gay, Bisexual and Transgender (LGBT) parented families - A literature review
American Psychoanalytic Association
Position statement on Gay and Lesbian Parenting
American Psychiatric Association
Adoption and co-parenting by same-sex couples
North American Council on Adoptable Children
Gay and Lesbian Adoptions and Foster Care
Royal College of Psychiatrists
Submission to the Church of England's Listening Exercise on Human Sexuality
American Academy of Child & Adolescent Pscyhiatry
Children with lesbian, gay, bisexual and transgender parents
American National Association of Social Workers
Amicus brief - California Supreme Court - Case No. S147999
Child Welfare League of America
Position statement on parenting of children by lesbian, gay and bisexual adults
Furthermore, the quality and quantity of the research in this field has affected court decisions on the topic.
Legal Decisions
Third District Court of Appeal, State of Florida, Docket No. 3D08-3044
"As a result, based on the robust nature of the evidence available in the field, this Court is satisfied that the issue is so far beyond dispute that it would be irrational to hold otherwise; the best interests of children are not preserved by prohibiting homosexual adoption."
B. Negative research
Over the last year or two, a small number of "studies" have appeared in the media which have been seized upon by those on the NO side as if a) they were the most robust and conclusive study ever published and b) there were no contrary studies out there. I thought that in the interest of completeness and sheer bloody-mindedness I would list them. Oh, and I also thought it would be nice to show why they're all wrong.
1. The New Family Structures Study (Mark Regnerus)
1.1. Methodological Problems
The first problem is that the Regnerus study contains several serious flaws in its experiment design.
1.1.1. Population sample
The first problem is the sample of individuals chosen to undertake the questionnaire. The study interviewed individuals only between the ages of 18-39 and questions them on their childhood. This means that the author is only looking at a snapshot of time between 1972 and 2011. There are two problems with this. Firstly, it means that the study has no depth. There is no longitudinality which is something that usually makes for the most robust conclusions. The author even makes an acknowledgement of this fact:
"There are several things the NFSS is not. The NFSS is not a longitudinal study, and therefore cannot attempt to broach questions of causation."Secondly, this snapshot covers an area of massive flux with regard to same-sex parenting. It moves from a time when homosexual parenting was entirely unusual and mostly prohibited to a time when it is becoming an accepted family form and yet seeks to make broad conclusions about the snapshot as a whole.
Also with regard to sample selection, Regnerus opens his study with an attack on previous studies such as the NLLFS (National Longitudinal Lesbian Family Study) for their poor sample selection. However, Regnerus used Knowledge Networks to gather his sample data despite having prior knowledge that a) KN had a limited ability to connect him with people who fell under the criteria of the study and b) there were other companies available who could have supplied more rigorous data.
1.1.2. Classification problems
There are several problems in this section in the design of the experiment.
The first problem is the classification of family form. The survey uses the following categories for analysis:
1.IBF: Lived in intact biological family (with mother and father) from 0 to 18, and parents are still married at present (N = 919).
2.LM: R reported R’s mother had a same-sex romantic (lesbian) relationship with a woman, regardless of any other household transitions (N = 163).
3.GF: R reported R’s father had a same-sex romantic (gay) relationship with a man, regardless of any other household transitions (N= 73).
4.Adopted: R was adopted by one or two strangers at birth or before age 2 (N = 101).
5.Divorced later or had joint custody: R reported living with biological mother and father from birth to age 18, but parents are not married at present (N = 116).
6.Stepfamily: Biological parents were either never married or else divorced, and R’s primary custodial parent was married to someone else before R turned 18 (N = 394).
7.Single parent: Biological parents were either never married or else divorced, and R’s primary custodial parent did not marry (or remarry) before R turned 18 (N = 816).
8.All others: Includes all other family structure/event combinations, such as respondents with a deceased parent (N = 406).
The problem here is that while broken homes are included in both types of same-sex families, they are definitionally excluded from heterosexual couples so that the comparisons made later on are fundamentally flawed.
The second problem with the classification above is that any instance of infidelity with a same-sex partner moves the respondent into either LM or GF regardless of the nature of the infidelity (i.e. one night stand vs. long term relationship).
The final problem with the classification above is that a number of respondents were found to fit into more than one category. To resolve this overlap Regnerus makes arbitrary decisions in order to achieve the results he wants:
"That is, a small minority of respondents might fit more than one group. I have, however, forced their mutual exclusivity here for analytic purposes. For example, a respondent whose mother had a same-sex relationship might also qualify in Group 5 or Group 7, but in this case my analytical interest is in maximizing the sample size of Groups 2 and 3 so the respondent would be placed in Group 2 (LMs). Since Group 3 (GFs) is the smallest and most difficult to locate randomly in the population, its composition trumped that of others, even LMs. (There were 12 cases of respondents who reported both a mother and a father having a same-sex relationship; all are analyzed here as GFs, after ancillary analyses revealed comparable exposure to both their mother and father)."That, btw, is as good an example of bad science as you're ever likely to see.
In classifying the data this way, Regnerus creates a homogeneous IBF category and compares it against LM and GF categories which are made up of a number of different relationship types both stable and unstable. That dog won't hunt Monsignor.
1.1.3. Analytical problems
The first problem in this category is Regnerus' idea of what constitutes a bad outcome for children. In particular Regnerus classifies the "current" receipt of public assistance (i.e. social welfare) as a bad outcome.
The problem with this analysis is that America is currently in the depths of one of the deepest economic crises it has faced, yet Regnerus makes no attempt to analyses the other socio-economic factors that play into this in analysing the data.
The second and bigger problem is that the classification errors made by Regnerus in his experiment design leads him to making the wrong comparison. Children in a family where one or both parents were in a same-sex relationship were found to have similar outcomes to those in divorced, cohabiting and step families and markedly different from intact heterosexual families. This is because he designed the study this way. What he actually ended up doing was comparing stable families against non-stable ones and the conclusions don't offer any new insight into the field which is what research is supposed to do.
To put it simply he analysed bad data with a badly designed experiment to get bad results and ended up with bad conclusions.1.2. Ethical problems
Herein lies the truly repugnant aspect of this study. Regnerus made some bad choices in even opting to start this study which not only impacts on his conclusions but on science as a whole. Bad science is in and of itself unethical but Regnerus' actions were even more so.
Firstly, Regnerus in a video interview admitted that his study does not work "to the long-term benefit of science". If ever there was a reason for not doing a study, that is it.
Secondly, Regnerus accepted a grant of $35000 from the Witherspoon Instiute to produce this study, hoping that Regnerus' reputation as a social scientist (which isn't looking great now) would give it an air of authority. Regnerus admitted that because of the grant he rushed it into publication for use in their 2012 election material.
Finally, Regnerus didn't approach the NIH for funding because in his words:
"I had a feeling when we started this project that it would not survive the politics of, in my opinion, the peer review system at the National Institute of Health"I think that sentence speaks for itself.
As a result of this information coming to light Scott Rose, a freelance writer and novelist wrote this open letter to the University of Texas. Consequently, the university has launched an investigation of Regnerus for scientific misconduct.
2. Loren Marks
The next to enter the rogue's gallery is Loren Marks, author of "Same-sex parenting and children’s outcomes: A closer examination of the American psychological association’s brief on lesbian and gay parenting", the study which took a swipe at the APA's amicus curiae brief in support of LGBT parenting. The study was flawed in multiple ways as I have outlined below.
1. In 2010, Loren Marks was called as an expert witness for the defence in a case challenging the legality of California's Proposition 8. Marks testified that children did best with their biological or genetic parents. Under cross-exmaniation Marks admitted that he had cherry picked data from the studies and he had not in fact read most of the studies which he cited in his report (which he later replicated as the study you linked to).
Two other defense experts (social science researchers from McGill University) also admitted during depositions that:
"equal marriage would increase family stability, improve the lives of children, and that gay men and lesbians have faced a long history of discrimination including violence. They also acknowledge broad scientific and professional consensus in favor of equal marriage."
The deposition transcripts and case report can be found here. Marks' involvement is described thusly:
"Plaintiffs’ attorneys last week introduced video of the deposition of Loren Marks of Louisiana State University, who had been expected to testify for the defendants that the ideal family structure is for children to be raised by two married “biological” parents, which Marks said meant the genetic parents. Marks admitted that he only read parts of the studies he relied upon in making his conclusion. It was then pointed out that those studies actually defined “biological” parents in a way that included adoptive parents — not just genetic parents. Marks then stated that the word “biological” should be deleted from the report he prepared for this case, and also admitted he considered no research on gay and lesbian parents, effectively revealing his research as fatally flawed."
2. Mark's main point of criticism is the small sample sizes used in the studies cited by the APA. While Marks portrays this as a problem for the conclusions drawn by the studies, this is not the opinion of other researchers in the field. Meezan and Rauch's literature review of same-sex parenting in 2005 points out that same-sex couples represent a small and geographically diverse population and that gathering a large sample size is a methodological problem rather than an analytical one. Rosenfeld also notes this in his census study in 2010, pointing out that all same-sex couples taken together represent just 1.8% of family forms in the United States.
3. Despite publishing his paper seven years after the brief (and nine years after the latest study included in the report) which he criticizes, Marks makes no attempt to incorporate studies outside those cited in the APA brief to show whether or not they support his claims. There have been large sample-size nationally representative studies conducted subsequent to the APA's brief. Two noteworthy examples of these include the 2010 Rosenfeld study mentioned earlier and the US Dept. of Health & Human Services study in 2010.
4. Marks also, as noted above, fails to account for any research conducted in to lesbian and gay parenting which fatally unhinges any valid analysis.
5. Marks makes no acknowledgement of the many other medical and social work bodies which have issued position statements in favour of equal marriage which makes the entire paper rather redundant.
6. Despite reviewing 59 papers cited by the APA in his paper, Marks fails to point out that there are in fact 65 empirical studies specifically related to gay and lesbian parents and their children cited in the report. He makes no acknowledgement of why excludes the remaining six. Furthermore, the report also cites empirical studies related to the general fitness of lesbian women and gay men as parents as well as many literature reivews, meta-analyses, legal reviews and individual case studies in support of its conclusions. All told there are over 130 publications cited in the report over half of which are ignored by Marks.
When we move past the scientifc flaws in the paper, we see that there are also deeper ethical concerns with Marks and his work. First of all, as noted above Marks admitted under oath that he didn't know any same-sex couples. This is not an area of research in which Marks is actively engaged. In fact since 2002, Marks' research efforts have been on a national qualitative study exploring the links between family and religious faith.Before publishing his study, Marks made a preprint available to the Republican Party committee defending the Defense of Marriage Act. In fact both Marks and Regnerus rushed their research through publication so that it would be available for the 2012 elections. Additionally, Marks, Regnerus and James Wright (editor of the journal where both studies were published) have ties to both the NOM and the Witherspoon Institute and all three are contributors to National Review a conservative political soapbox.
References:
Press release from American Foundation for Equal Rights
Plaintiff Motion to Exclude the expert report and testimony of Katherine Young, Loren Marks and David Blankenhorn
3. No Basis
The final club in the NO side bag is this publication (can't think of a better word for it) from two researchers Robert Lerner and Althea Nagai. Again we find multiple flaws leading to unsupportable conclusions.
1. Mistakes of fact
The first problem with "No Basis" is that the authors in several places make several basic reporting mistakes displaying some remarkably sloppy fact-checking. This, by the way, is the importance of citing peer-reviewed sources so that we don't have to go through such basic mistakes.
a) Brewaeys et al (1997)
This is a surprising and notable inclusion in the list of studies since it has been mentioned on multiple occasions, even by gay marriage opponents such as Steven Nock as "a well defined analysis that attempted to study entire populations rather than samples of them". It is as rigorous as any study in the field of social science gets.
The first mistake that the authors make is that they report the sample size in Table 6 as 72. However, as you can see from the actual paper:
Donor insemination: child development and family functioning in lesbian mother families
the sample size is 98. Furthermore, Lerner and Nagai criticise Brewaeys for its use of ANOVA as its test metric. However, you can see from reading the paper that multiple statistical tests (including Student's t-test) are used. Furthermore, something that Lerner and Nagai fail to point out is that the study controls for demographic differences in the comparison groups and also corrects for multiple comparisons.
b) Harris and Turner (1986)
Once again the sample size reported by Lerner and Nagai is 27, whereas the actual sample size in the paper is 39.
Gay and Lesbian Parents
c) Chan, Raboy and Patterson (1998)
Psychosocial Adjustment among Children Conceived via Donor Insemination by Lesbian and Heterosexual Mothers
What do you know, another sample size reporting error, this time claiming 77 instead of the 80 used in the paper. Furthermore Lerner and Nagai claim that:
"Chan et al use t-tests extensively. The t-test is a statistical proce- dure that compares the means of two groups to see whether the dif- ference between the two means are likely to be due to chance or statistically significant. "
Well, yes but the authors fail to point out that such t-tests are also supported by multiple regression analysis tests.
Furthermore, Lerner and Nagai, make a bold and rather confusing claim by stating:
"The primary problem in using the Bonferroni correction in these studies, however, is that it introduces a bias in favor of the investiga- tors’ own hypotheses."
Such a claim makes it plain that Lerner and Nagai are not researchers in any serious scientific field. For those who don't know the Bonferroni correction is a correction for multiple comparisons. Basically, if you compare groups using multiple characteristics, the odds that one group is better than the other by chance alone increases significantly. While the Bonferroni correction is non-conservative when used to affirm the null hypothesis, it is still in line with best practice in the field.
2. Methodological issues
The first issue with the approach taken by Lerner and Nagai is the failure to detail the effect of the flaws they describe. The overall impression that Lerner and Nagai attempt to give to the reader is that the 49 studies included in their document are flawed to the point that the conclusions of these studies are either wrong or unsupported by their results. However, this simply isn't the case.
Take this paper for example (cited by Lerner and Nagai).
Lesbian mothers who conceived after donor insemination: a follow-up study
This paper does contain a flaw, but is not flawed in the way Lerner and Nagai describe. The issue with the Flaks paper is that while the conclusions hold for the dataset collected, the conclusions are based on a sample size which is too small to be generalised to the entire population. This is a small but important distinction which the authors fail to clarify.
Secondly, there is the illusion of control. For example, when examining the studies for sample size, Lerner and Nagai point out that they excluded one study with good sample size (Riddle et al.) because it lacked a heterosexual control group. However, papers such as Brewaeys and Chan cited above show that a control group may not always be a prerequisite for a solid paper. To explain this further, let's look at babies. Babies at birth are frequently tested using an Apgar score. These scores measure a baby in several categories against well-established norms, birth weight, for example, to determine whether the baby is healthy or not. Similarly, since many of these studies study child outcomes, they can be measured against established psychological norms using standard tests as outlined in the paper by Chan et al.
3. Relevance
This is the single most important reason why Lerner and Nagai's criticisms are not valid. It's not that they were never valid it's that they're not valid anymore. There are several reasons for this.
3a - Best practice.
The primary resource used by Lerner and Nagai in levelling their criticisms is a set of guidelines established by psychometrician Jacob Cohen in 1988. However, in 1999, two years before No Basis, the American Psychological Association issued updated guidelines for conducting psychology research:
Statistical Methods in Psychology Journals: Guidelines and Explanations
These guidelines form the basis of best practice in the field and serve to undermine the criticisms of some of the papers in Lerner and Nagai's work, in particular Brewaeys et al.
3b - Meta analyses
One of the most powerful modern tools in scientific research is a meta-analysis. Unlike a review article where the conclusions and results are cursorily examined, meta-analyses essentially combines the different datasets and reanalyses them as a single dataset to see if the results hold for the larger set. This meta-analysis:
A Meta-Analysis of Developmental Outcomes for Children of Same-Sex and Heterosexual Parents
(published in 2008) uses 19 eligible studies including 16 cited in the Lerner and Nagai document. It finds that:
"In sum, children raised by same-sex and heterosexual parents were found to not differ significantly in terms of their cognitive development, gender role behavior, gender identity, psychological adjustment, or sexual prefer- ences. For the outcome that was significantly different between children of same-sex and heterosexual parents, the finding was in favor of same-sex parents."
3c - Rigorous studies
At this point I should point out that as I have previously stated, this work is badly out-of-date. Although published in 2001, the most recent paper examined dates from 1998. 16 years is a very long time in any field of research but particularly in one with as much political valence and human rights implications as this one.
In any event, since 1998, there have been several studies which not only satisfy the established best practice in the field but the crazy extremes documented by Lerner and Nagai. Examples include:
Psychosocial Adjustment, School Outcomes, and Romantic Relationships of Adolescents With Same-Sex Parents
Children With Lesbian Parents: A Community Study
3d - Large-scale studies
One of the valid criticisms of social science research prior to the millennium was the lack of suitable large-scale studies so that the results which were established on a small-scale basis would hold on a national level. However since 2000, there have been several such studies.
The first of these is the Wainwright study cited above which studied 12,105 adolescents taken from the National Study of Adolescent Health, a government funded program.
Then of course there are these studies, you know the ones I keep telling you about BB:
- Family Structure and Children's Health in the United States: Findings from the National Health Interview Survey, 2001-2007
- Rosenfeld, M.J., (2010), Nontraditional families and childhood progress through school Demography, 47(3), 755-775
- Crouch, S. R., Waters, E., McNair, R., Power, J., & Davis, E. (2012). ACHESS–The Australian study of child health in same-sex families: background research, design and methodology. BMC public health, 12(1), 646.
- Crouch, S. R., Waters, E., McNair, R., Power, J., & Davis, E. The Australian Study of Child Health In Same-Sex Families (ACHESS) - Interim report
All of these studies represent the gold standard of research in this area. They are large scale and nationally representative. And yet they confirm the same findings documented in the other research previously.
Lerner and Nagai's critisms are flawed and obsolete. However, even had they been valid at the time, we do not need to rely on studies of that type to support the hypothesis. At least not any more.
No comments:
Post a Comment