Iona Institute misrepresented research

David Quinn insists that change be driven only by evidence, but when it suits him he cites a discredited author who ignores recent research and relies on distortions – Peter Ferguson 

The Iona Institute has recently been accused of deliberately misrepresenting research, conducted by Child Trends, in its submission to the Constitutional Convention. Senator David Norris raised the issue in the Seanad and has requested an inquiry.

In response David Quinn authored a blog post on Iona’s website defending the use of Child Trends’ research in their submission. In the post Quinn claims that the research was used to show that marriage between two biological parents “is the most beneficial family form that we know of from the point of view of children”. Quinn further states that “the available data does not allow us to say how well children raised by same-sex couples fare compared with the biological married family”. Such a statement slights over 30 years of academic research which concludes that children raised by same-sex parents experience the same outcomes as those raised by two biological parents. Quinn explains that he rejects the current body of research as he believes it to be flawed: “To this day it remains the case that there are no large national surveys that allow us to draw reliable conclusions about the children of same-sex couples.

The question which then arises is why, back in 2005, the American Psychological Association (APA) was so quick to come to the conclusion that ‘the kids are alright’ given the lack of large national surveys examining how the children of same-sex couples actually fare?”.

This sentiment was also stressed in Iona’s submission to the Constitutional Convention: “Some research seems to indicate that children do just as well when raised by a loving same-sex couple as they do when raised by a loving mother and father. Some of this research is attested to by the American Psychological Association. However, this research is invariably flawed in some way. For example, the samples relied on are small, and they are usually non-random”.

In fact, the American Sociological Association states: Decades of methodologically sound social science confirm that whether a child is raised by same-sex or opposite-sex parents has no bearing on a child’s wellbeing

To evidence these assertions Quinn links to a paper written by Loren Marks of the Louisiana State University published in 2011 that critiqued an APA brief. This APA brief catalogued and analysed the results of over 130 publications relating to lesbian and gay parenting and reported that: “there is no evidence to suggest that lesbian women or gay men are unfit to be parents or that psychosocial development among children of lesbian women or gay men is compromised relative to that among offspring of heterosexual parents. 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. Indeed, the evidence to date suggests that home environments provided by lesbian and gay parents are as likely as those provided by heterosexual parents to support and enable children’s psychosocial growth”.

Marks’ article claims that there are insufficient data to make such an assertion as the sample sizes are too small. Quinn argues similarly: we cannot know if children raised by same-sex parents have similar outcomes to those raised by biological married parents; therefore, there should be no legislation for same-sex marriage. Such logic is quite spurious and not reason enough to object to same-sex marriage; however, what I shall show is how David Quinn, so loth to prostitute himself to allegedly ambiguous research here, has no qualms linking to hugely flawed research of questionable academic integrity providing it agrees with his prejudice.

In 2010 Loren Marks was expected to testify on behalf of the defendants in the Proposition 8 court case, then known as Perry v Schwarzenegger. He was called to testify that the two-biological-parent family structure was the most beneficial to children, however, he was dropped after making some egregious admissions in cross-examination:

• He admitted to not actually having read all the research he cited; in fact, he had read “just portions of it”;
• He admitted cherrypicking only data that were relevant to his argument;
• He admitted that his religious convictions may have influenced him when he conducted the study;
• He admitted that his belief that the ideal family structure is marriage between a man and a woman predates his work as a social scientist and does not stem from his research;
• He admitted the reports he used didn’t define “biological” in the genetic sense: it also encompassed adoptive parents. He was forced to remove the word “biological” in the report he prepared;
• He admitted he never actually researched any same-sex couples.

After such a shocking deposition he was summarily dropped as a witness.

Loren Marks had a pre-set conclusion and endeavoured to write a report which would agree with this conclusion regardless of what the evidence and studies actually showed. Luckily he was revealed to be a charlatan. Indicatively, Marks has ties to the National Organisation for Marriage (NOM). The Southern Poverty Law Center which fights intolerance, says the National Organization for Marriage pushes the line of being labelled a hate group because it “continues to spread lies about gays” and uses its website to link to debunked research.

NOM was also described by Michael Cole, spokesperson for Human Rights Campaign, the largest LGBT equality-rights advocacy group and political lobbying organization in the US, as “a secretive player in anti-gay politics, which is posing as an offshore company for anti-gay religious money”. Marks’ paper was published in Social Science Research, the editor of which, James Wright, also has ties to NOM.

Not a single study has found children of lesbian or gay parents to be disadvantaged in any significant respect

It is clear Marks has an agenda. It is no surprise that somebody who is willing to doctor research to ensure he arrives at his predetermined conclusion later authors another paper in which he tries to do the same.

Marks’ main contention (repeated by Quinn) with the brief is that the sample sizes in the publications cited by the APA are too small. However, this is not a sufficient reason to negate the studies. William Meezan and Jonathan Rauch carried out a similar review of same-sex parenting in 2005 and took sample sizes into greater consideration, detailing the problems of small sample sizes. They found that gathering data regarding same-sex parents was quite difficult because same-sex couples only represent a tiny minority of the population (1.8%) and are geographically sparse. Despite this, Meezan and Rauch concur with the APA brief’s conclusion. Meezan and Rauch further identify four more papers which they deem “methodically rigorous” and which have large sample sizes. These studies also concur with the APA.

Marks strangely only analyses 59 out of the 130+ publications cited by the APA: less than 50%.

Even stranger is the fact that Marks published this in 2011: the APA brief was prepared in 2005. There is a six-year gap and Marks fails to analyse any of the papers which were published in the intervening years. In fact, the APA brief only included research conducted up until 2003, so that is, in reality, an eight-year gap where dozens of papers were published, some of which were nationally-representative studies containing large samples, and Marks fails to take them into consideration. This makes Marks’ review of APA’s brief obsolete and therefore irrelevant to today’s debate about same-sex marriage.

Marks purports to cite a study by Dr Sotirios Sarantakos which claims that children of same-sex couples had worse outcomes than children of married heterosexual couples. However, Sarantakos’ research was dismissed by the APA as his methodology skewed the results. He almost exclusively studied children of same-sex parents who had experienced a divorce and when his results were compared to children of heterosexual parents who also experienced a divorce they were almost identical. Sarantakos himself admitted the results were not due to any fault in the parenting but more likely due to anti-gay bigotry in others. Yet Marks attempts to cite this as evidence that not all research shows that same-sex parents benefit children in the same manner as biological married parents.

Also, Marks’ paper only reviewed the research conducted by the APA which contained exclusively research completed before 2002. Quinn fails to acknowledge the dozens of papers published in the past 11 years and the numerous organisations which have all issued public statements in support of same-sex marriage: the Canadian Psychological Association, American Academy of Pediatrics, Australian Psychological Society, American Psychoanalytic Association, American Psychiatric Association, North American Council on Adoptable Children, Royal College of Psychiatrists, American Academy of Child & Adolescent Pscyhiatry, American National Association of Social Workers and the Child Welfare League of America.

David Quinn has misrepresented one research paper; and claims that there have been no large national surveys completed to gauge accurately the outcome of children of same-sex parents and that this is an adequate reason to oppose same-sex marriage. Not only is this untrue but to evidence his claim Quinn cites a report authored by Loren Marks who has a record of manipulating research. Quinn and Marks both amateurishly ignore the scientific techniques now deployed in the field of child developmental psychology. Large-scale studies are not required to ascertain accurate results. Quinn himself has no social science credentials, and Marks’ research is in the field of marriage and religious faith. There is a scientific consensus among psychologists on the best research methodology and it is these methods that are used in the same-sex parenting studies.

In essence, David Quinn has attempted to negate decades of social science research conducted by hundreds of scientists by citing one paper written by a discredited researcher who only reviews the APA brief from 2005, ignoring all other research. In fact, the American Sociological Association, in an amicus curiae prepared this year, stated: “Decades of methodologically sound social science research, especially multiple nationally-representative studies and the expert evidence introduced in the district courts below, confirm that child wellbeing is the product of stability in the relationship between the two parents, stability in the relationship between the parents and child, and greater parental socioeconomic resources. Whether a child is raised by same-sex or opposite-sex parents has no bearing on a child’s wellbeing. The clear and consistent consensus in the social science profession is that across a wide range of indicators, children fare just as well when they are raised by same-sex parents when compared to children raised by opposite-sex parents”.

It would be disingenuous of David Quinn to continue to decry the numerosity and quality of research. In fact, such persistent rejection of such a vast body of research borders on social science denialism akin to evolution- and climate-change -denialism.

The Constitutional Convention ended with 79% in favour of legislating for same-sex marriage, so there may well be a referendum on the issue in the near future. It is, therefore, imperative that the public are fully informed and facts treated, unlike opinions – as sacred.

No, Ferguson misrepresents …

The reason Marks reviewed only 59 of the APA-cited 130 paper is that eight are ‘unpublished dissertations’, 26 don’t compare gay families with other families, and 54 fall short of AA guidelines on use of statistics – David Quinn 

Having to respond to the accusation that you have ‘misrepresented’ research is a bit like having to respond to the accusation that you beat your wife. It automatically creates an insinuation against you that probably never quite goes away.

‘Morning Ireland’ made a similar allegation against the Iona Institute recently when presenter Cathal MacCoille said that New Zealand academic, Dr David Fergusson was “unhappy” at how the Iona Institute summarised a research paper by him which says there is no evidence that abortion improves the mental health of women.

In fact, Dr Fergusson never said he was unhappy and MacCoille had to admit this on air two days after the original false allegation was made.

For its part, Peter Ferguson’s article amounts to little more than an extended rant against me, and even more so against Loren Marks of Louisiana State University.

But I am going to completely ignore the ad hominem attacks against both myself and Marks and instead go straight to the research and what it has to say about the effects of different family structures on child wellbeing.

Ferguson puts together what he imagines is an impressive list of research into the effects of same-sex parenting on children. He quotes the American Psychological Association’s summary of some of this research which says, “there is no evidence that lesbian women or gay men are unfit to be parents or that psychosocial development among children of lesbian women or gay me is compromised relative to that among offspring of heterosexual parents”.

On his blog, Ferguson has quoted other bodies which summarise the research in much the same way.

For good measure, Ferguson then falls back on that hoary old favourite, the argument from authority, by saying that to contradict the above statement is to “slight over 30 years of academic research” on the topic, and by extension the researchers themselves.

This is not an argument at all of course. We must look at the research itself, rather than at the researchers’ credentials, and assess how good it is. The basic criticism to be made of the research is that it does not use large, random samples to come to its conclusions.

Loren Marks points this out in ‘Same-sex parenting and children’s outcomes: A closer examination of the American psychological association’s brief on lesbian and gay parenting’, published in the journal, Social Science Research.

Ferguson tries to discredit Marks by cherry-picking quotes against him from hostile websites. One of the accusations is that Marks reviews only 59 out of the 130 or so papers the APA cited.

However, the reason for this is that only a portion of those 130 plus papers deal with empirical studies specifically related to lesbian and gay parents and their children. The APA lists 67 such studies but eight are “unpublished dissertations” reducing the number to 59.

Most of these tiny samples, commonly quoted by supporters of gay marriage such as those referred to by Ferguson, don’t even have the benefit of being randomly selected. They are often self-selected

Marks’ article (and I invite readers to look it up and judge it for themselves) includes a table summarising the 59 published studies cited by the APA. It speaks for itself. Marks shows that 26 of the 59 studies use no comparison groups, that is, they don’t compare the gay and lesbian families they are studying with any other families at all.

He shows that 54 out of the 59 studies fail to follow the APA’s own guidelines by telling readers what their statistical power is, which means we can’t say how reliable their findings are.

Many of the studies didn’t even look at child outcomes, let alone make comparisons with children in other family types.

So it seems it is Peter Ferguson who is doing the misrepresenting here. He should present the Marks paper more fairly and accurately.

Hedging his bets, Ferguson himself more or less acknowledges the methodological problems at one point in his article when he quotes a 2005 paper by William Meezan and Jonathan Rauch called ‘Gay marriage, same-sex parenting and America’s children’.

As Ferguson says, in that paper Meezan and Rauch acknowledge the deficiencies of much of the research – some of which the APA used   – into same-sex parenting to date, including the very small sample sizes.

But he then points to the fact that Meezan and Rauch identity four more papers “which have large sample sizes”. But do they? A read of the Meezan/Rauch brief says otherwise.

One of the studies looks at 44 children raised by lesbian couples. Another looks at 39 lesbian-headed families. A third looks at 34 lesbian couples and 21 lesbian single mothers. The fourth looks at 30 lesbian couple families.

These studies aren’t large. They’re tiny. As a result their statistical power is very small. It is a bit like telling us that we can rely on the finding of an opinion poll consisting of a few dozen people. Even if the few dozen were randomly selected, it would still lack statistical power and would not be truly representative of the general population.

But most of these tiny samples, commonly quoted by supporters of gay marriage such as those referred to by Ferguson, don’t even have the benefit of being randomly selected. They are often self-selected.

So we have more misrepresentation by Ferguson.

When large scale-studies based on nationally-representative samples of different kinds of families are conducted, what do they tell us?

Child Trends, a US-based think tank summarises it very well in its 2002 briefing paper, ‘Marriage from a Child’s Perspective: How Does Family Structure Affect Children, and What Can We Do About it?’.

It says: “Research clearly demonstrates that family structure matters for children, and the family structure that helps the most is a family headed by two biological parents in a low-conflict marriage…There is thus value in promoting strong, stable marriage between biological parents”.

I have been accused of misrepresenting this finding because Child Trends points out that the paper says nothing about same-sex families. But who said it did? Not me.

In fact, on this score the quote Ferguson tries to use against me in this respect backfires on him. Badly.

The quote says: “This Child Trends brief summarizes research conducted in 2002, when neither same-sex parents nor adoptive parents were identified in large national surveys. Therefore, no conclusions can be drawn from this research about the well-being of children raised by same-sex parents or adoptive parents.”

Precisely the point. Child Trends drew on research that allows us to properly compare children from different family structures and didn’t include children with same-sex parents because they are not to be found in “large, national surveys.”

When children from married, biological families can be properly compared with children from other family structures, the former generally have the best outcomes.

The bottom line is that at this point we can say little about children from same-sex families given the absence of large-scale studies of those families.  Meanwhile the burden of proof is firmly on those who insist that having two parents of the same sex is just the same from the point of view of child outcomes as being raised by their own biological parents.

In one of his blogs, Ferguson lists a number of studies which have been published since that Child Trends brief in 2002 which he insists are large, national studies. Look them up. He’s wrong. In fact, Ferguson misrepresents them by pretending they are large, national studies.

All this shows is that he is so keen to accuse his ideological opponents of ‘misrepresenting’ data that he can’t see it when he is doing that very thing himself.

Ultimately all we have is Ferguson flailing about angrily, throwing dust in the air, and relying on personalised attacks and arguments from authority to press home his point, meaning the only one doing the misrepresenting is him.