I gave up on wading through familystructurestudies.com/articles/ to find the real analysis. The first link seems biased.
Regardless, this changes my beliefs regarding lesbian marriages. My personal experience has been that long-term lesbian couples produced the most successful children (my sample pool was a whopping 2 samples deep).
Thoughts, criticisms, or arguments against this result?
Not really, as that was regarding lesbian parents, not specifically marriages. I thought perhaps you were hinting that you now believed lesbians should not be allowed to get married because of this study, which is why I asked for clarification.
That would be a nitpick over a word which has no real meaning: we both live in the US and you know I live in Oklahoma.
I used "marriage" out of respect for the 2 couples I do know that cannot get married but wish to (being in Oklahoma kind of makes it impossible to get married).
They consider themselves married and I'll be damned if I let the law get in the way of two adults who are in love with each other. They refer to each other as "wife", so I will too.
I don't know how you could have gotten that out of my post, really. I think that's a pretty far stretch for interpretation on my words.
Gotcha. Was just curious was all, as obviously I had no way of knowing you were using the word 'marriage' in such a way, you understand. Thanks for clarifying, though.
As far as the studies specifically, though, I don't really care. Good parents can belong to any social/ethnic group, as can bad parents. Studies like this don't mean much to me.
My complaint regarding this result is that it, at least partially, is measuring the social pressures such couples create for their children (by simply existing). To me, this is showing the result the prejudice and discrimination are still faced and it takes a toll on the children.
But that criticism does not fit well into the results of the study because it should show the same result for gay-men, as well: it doesn't (at least, the results show that the lesbian couples, in particular, are creating very statistically significant results). So, clearly, there is something else going on that is particular to lesbian couples and their children. I want to know more.
From my own personal experience, lesbian couples get less bullshit than the males. Obviously, that is anecdotal at best and completely opposite of reality, at worst. But if my experience is generally true, then shouldn't the male couples with children have poorer performing children?
Personally, I'm always very cynical when it comes to these studies, it seems like they are very often intentionally slanted and done by groups with a predetermined bias and prejudice so they can say "look, see gay marriage is bad, think of the children, ban it."
Obviously I can't say if that's true in this particular case, wouldn't surprise me, though.
I didn't look through that entire website but it looks like male-male couples are better at producing successful children than single mothers. So it would appear that we should ban single-motherhood before male-male marriages!!!
And here BBC was telling us that the Y gene could go very soon
Lesbian couples with children represent such a fraction of any given group of people, even at a state level, so any failures that they have academically or successes would be magnified in percentage because the sample size is necessarily small. The other aspect is that a study may not account for all variables, which could include bullying from school peers, especially if the child's parents' sexual orientation is widely known. Single mothers had similar social integration issues and faced persecution and their children suffered years ago because of social exclusion, bullying, and even issues from staff who were judgmental or unsympathetic.
Also, I've seen studies that prove just the opposite; that same-sex couples are no more or less better at parenting successful children than anyone else, but I don't recall where this study took place or when.
It was a very large study: "The NFSS, though, is most noteworthy for being the second-largest probability sample of children of parents with gay or lesbian relationships."
Here were the motivations of their research, which does explore your idea regarding this topic:
"Prior to the NFSS, the academy had come to believe that children raised by gay or lesbian parents fared, on average, no worse, and in some cases better, than children raised by heterosexual parents. Dr. Regnerus and a group of social scientists decided to evaluate this claim empirically by studying the responses of children who were raised in a variety of family structures. The goal of the NFSS, from the beginning, was to gather the best social science possible to address the question of what family arrangements were best for children. The results of the NFSS research revealed that the “no differences” claim—the claim that children raised by parents in gay or lesbian relationships fared no worse and in some cases better than children raised by intact biological parents—was not true. On the contrary, the children of these households, on average, did worse than children raised by their biological, still-married parents."
What I mean is, it would be a pretty big stretch to reject the results because of an "error magnification". With samples this large, errors are generally minimized/reduced to paltry results. The results could have been created in error or the particular analysis could have been malformed.
I could name more issues: it could be an error in the statistics due to a sampling bias against lesbians (I believe this would be a type I error (I would call into question, with such a large sample, even mentioning "Type I Error" anywhere near this study result), if true(for example, lesbian couples with children are more forthcoming about their children's failures than other couple demographics)), and a research bias (which I believe we all agree is a possibility).
Sometimes, organizations with pro-LGBT motives find a result that does not favor their political agenda. When they release their results that would diminish some of their positions, that should be chalked up to amazing academic honesty and it would be the "true spirit of science." However, that does not seem to be the case in this study because the lead researcher supposedly put together a diverse team: from conservative to liberal.
Also, the work was peer-reviewed and credibly published.
Lastly, they said this of their methods:
"The NFSS studied a statistically large, random sample of the general U.S. population. By using a sample that is randomly selected and statistically large, scientists can be highly confident that the sample represents the makeup and behavior of the bigger population with almost as much certainty as if they had surveyed every individual in the population. In all prior surveys of gay parenting, except one by Dr. Michael J. Rosenfeld of Stanford University, the sample size of children was either too small to make meaningful claims or the participants were not selected randomly but by other methods like “snowball” or convenience sampling. These other methods tend to select people that are more similar to each other than is the case in the larger population that they are meant to represent. They inherently lack the representative accuracy and strength of random samples.
The NFSS also surveyed participants on a broad range of outcomes spanning physical and emotional health, experience in personal relationships, and general social behavior. The one other survey that studied a statistically large, random sample (Rosenfeld’s) studied only one outcome: children’s educational achievement.
Finally, the NFSS surveyed the children of parents who had a same-sex relationship directly and as young adults (after they had reached maturity). Many previous studies surveyed only the children’s parents, and no prior study surveyed the children in their young-adult years."
__________________
Last edited by dadudemon on Nov 26th, 2013 at 04:57 AM
Hundreds of children living with gay or lesbian couples is still a very small sample. Also, I was not immediately aware that this took place in Canada.
EDIT: I just really looked into this source. You realize this Witherspoon Institute is utterly opposed to same-sex marriage, right? And that the guy who conducted the study is an advocate of the same kind of thinking? This is Confirmation Bias: The Study we're talking about here.
Since those early days, Regnerus has signed on to a “friend of the court” brief in both gay-marriage cases recently taken up by the U.S. Supreme Court, urging the court to uphold California’s ban on same-sex marriage and the federal Defense of Marriage Act. He has blogged about his skepticism regarding the health of kids raised by gay parents, and he’s signed on to speak at a National Organization for Marriage-affiliated conference dedicated to arming college-age kids with research that opposes gay marriage.
As The American Independent reported last month, the Witherspoon Institute, the conservative think tank that funded the bulk of the New Family Structures Study, pushed to have the study’s results out before “major decisions of the Supreme Court,” according to documents obtained through a public records request.
And from a cursory examination of the guy's website, here's one "large random sample":
Highlights
► The New Family Structures Study collected data from nearly 3000 adults. ► I compare young adults who grew up with a lesbian mother or gay father. ► Differences exist between children of parents who have had same-sex relationships and those with married parents. ► This probability study suggests considerable diversity among same-sex parents.
Last I checked, Canada's population was approaching 35 million. That study hardly seems large enough to be all-encompassing, and the source seems very very suspect. I'd throw this one out.
2. Confirmation bias: unsubstantiated and probably false.
In the lead researcher's own words:
"People will say I'm irresponsible without weighing in with stronger data," he said. "This is the best quality data we've seen so far. If they don't like the results, I'm sorry."
"There are some valid criticisms that are being made, such as the measurement decision on who should be called a lesbian mother in this study," Regnerus said. "People might say that's irresponsible to do this study without all these stable lesbian couples in the study," he said, adding the random sampling only found two out of the 175 children who said they lived in a home with both same-sex parents throughout all 18 years. "I would have been happy to compare them but they did not exist in large enough numbers."
His idea that collecting a sample that is representative of the population is a more apt comparison is closer to reality because we're not going to get the rare "same lesbian couple for all 18 years of the child's life" scenario, very often.
And this is what he said regarding the instability criticism: "People gay or straight should stick with their partners, he said. "I think the study provides evidence of that." I would note that these are not the words of someone that has confirmation bias in the manner you have accused.
And another researcher stated earlier (which I think was the motivation for Regnerus' research) "...burden of empirical proof is on those who argue that the children of sexual minority parents fare worse than the children of heterosexual parents."
Lastly, he approached 4 pro-gay groups for funding for his research: all 4 of them declined. Why?
"When the NFSS was broadly outlined in late 2010, the Witherspoon Institute approached four different funding sources that were known to be committed to gay rights and also to have an interest in the welfare of children. They were asked to be partners by providing financial support to fund a study (the NFSS) with the proviso that none of the funding sources would have any influence regarding the design, implementation, or interpretation of the data. They were told the study would be conducted at a major research university and that the team of scholars involved in the design of the study would be evenly represented across ideological lines. All four declined."
Isn't that more telling of a "liberal research" slant than confirmation bias from Regnerus?
Lastly, don't you think that the critics could levy better criticisms other than ad hominems since he published his data and methods? He readily acknowledges the criticism that 'broken homes produce worse outcomes for children'.
But don't you think another group could quickly and easily disprove the notion that gay couples have worse outcomes by using his very own data against him?
"Sampling methods were flawed." False: it was a very large sample and it was a true random sample, unlike the other studies (minus one) that he called into question. Another researcher could go through his data and make their own conclusions.
Again, your sample-size criticism is a flawed criticism. It just doesn't apply. "Not large enough of a sample" just doesn't apply. You create a massive slippery slope with your criticism because then we must throw out all other studies regarding gay couples and child-outcomes because none of them, minus 1, even comes close to as large as Regnerus' attempt. On top of that, Regnerus' sampling methodology was more honest than most of the other studies because it was more random and representative of the population (in statistical collection, asking for respondents for a particular study has a bias. You just can't say, "Hey, I need you to answer questions about you children, I need you to be gay, and I'm researching how effective gay parents are": that will create a sampling bias that will screw up the results and that was one of Regnerus' criticisms of the preceding studies).
The source seems solid, peer reviewed, and published. Methods seem solid, as well.
I'll hold onto this one and request more research, of this type and caliber, be conducted.
__________________
Last edited by dadudemon on Nov 26th, 2013 at 06:22 PM
KMC Community Forums
Nov 26th, 2013 03:28 AM
