Box Turtle Bulletin

Box Turtle BulletinNews, analysis and fact-checking of anti-gay rhetoric
“Now you must raise your children up in a world where that union of man and box turtle is on the same legal footing as man and wife…”
This article can be found at:
Latest Posts

Tools of the Trade

Jim Burroway

June 21st, 2012

Many, many years ago, a co-worker decided to take up woodworking as a hobby. He moved his cars out of his suburban garage — there would no longer be any room for them there — and purchased some of the finest sets of tools and machinery available: band saw, jig saw, router, drill press, and a few others I couldn’t tell you the names of because I’m not a woodworker. He went top of the line, too (or so he assured me, and I believe him), making sure he had the best bells and whistles. That was just his machinery. His array of hand tools was equally impressive lined up on his shiny new cabinets where, in each drawer, he laid a piece of foam with cut-outs in the shape of his tools. This way, there would be a place for every tool and a tool in every place. It was, all in all, a weekend warrior’s dream.

And then he built his book cases and media center. They weren’t awful by any means, but they weren’t spectacular either. I had a hard time seeing how they were much better than what I could have made with my far humbler collection of mismatched and somewhat rusty tools laying around the house. Assuming I could find them.

My co-worker’s dazzling workshop and my mess of a garage come to mind as I read some Mark Regnerus’s postings defending — no, not defending, but bragging about — his purported study on gay and lesbian parenting I reviewed last week. I say purported because with most of the adult children studied spending much less than three years with their gay or lesbian parent, it seems to me to be a study that is not about gay parenting but about children of divorce, separation, parental infidelity, and a host of other sources of instability. Gay parents, it turns out, are barely in the picture at all.

Regnerus blithely ignores this criticism. Instead, he distinguishes his study by the shiny $785,000 worth of tools that he assembled to produce it. It’s often the first point he mentions when he talks about it, like in his defense posted on the Baptist-affiliated Baylor University web site:

By contrast, Regnerus relies on a large, random, and representative sample of more than 200 children raised by parents who have had same-sex relationships, comparing them to a random sample of more than 2,000 children raised in heterosexual families, to reach his conclusions. This is why sociology professor Paul Amato, chair of the Family section of the American Sociological Association and president of the National Council on Family Relations, wrote that the Regnerus study was “better situated than virtually all previous studies to detect differences between these [different family] groups in the population.”

National random samples are inordinately expensive. It is why they are so rare in social science research. It’s every social scientist’s wet dream to be able to assemble a garage like Regnerus’s, but for almost every one of them such a tool set is way out of their reach. So having a tool as shiny and impressive as a nationally random sample is a really big deal. I’d say that it would probably very difficult for most people to resist showing it off.

But what I see happening is that Regnerus wants us to be dazzled by his tools — and they are truly dazzling — and not look too closely at the book case he built with them. And if that book case looks a little off kilter, maybe its because he used his tools incorrectly. If you use a drill press when you really needed a router, you’re going get results that fall short of what you promised. If you ask adults between 18 and 39 if either of their parents ever had a same-sex relationship and leave it up to the responder to ponder whether a short tryst counts, then you are going to get a sample that doesn’t resemble anything like the kind of relationship that is comparable to any kind of marriage, gay or straight.

And as his study has little to say about anything close to resembling marriage, it also says almost nothing about parenting. In his study, only 57% said they had lived with their mother and her partner for at least four months before the age of 18, and only 23% reported living with their father and his partner for the same length of time. Only 23% of children of so-called “Lesbian Mothers” and 2% of children of so-called “gay fathers” — both loosely defined — reported living with their parents and their parents’ same-sex partners for three years or more. If this is any kind of a study about gay parenting, it is even more so a study about the fifteen-plus years these kids spent outside of that so-called relationship.

And so after constructed a sample that was guaranteed to look nothing like a stable family environment, and then compared it to a sample that was specifically constructed to be a stable as possible, he found differences. That happens when you compare apples to orangutangs. And what else would you expect? When you build an unlevel shelf, you shouldn’t be surprised to see all of the results sliding off to one end. Sure, he may have started with a nationally representative sample, but that doesn’t give him license to move populations around within his samples with the purpose of achieving his aims, or to make claims about what his sample represents when his sample wasn’t constructed to represent what he claims.

And these issues point to a much bigger problem with Regnerus’s study. He wanted to examine the impact of gay parenting on children, but his tools, as impressive and expensive as they were, turned out to be the wrong tools for the job. As powerful as they otherwise would have been, they lacked the critical features that he needed to do the job correctly. And so lacking those important features, he does what a lot of weekend hobbyists in a hurry end up doing: he kludges it and hopes nobody notices. Except they have, and those kludges have been the crux of the criticisms against his study.

His response so far has boiled down to this: don’t pay any attention to how he used his tools. Just look at how shiny and expensive they are. But as every craftsman will tell you, it’s not the tools that make the better product, but the care of the craftsman who uses them wisely.

Comments

POST COMMENT | COMMENT RSS 2.0

StraightGrandmother
June 21st, 2012 | LINK

Jim let me restate the data in a different way.
So Let’s add up the numbers-
(2)- Stayed with mom and her partner 18 years
(6)-Stayed with mom and her partner 10+ years
(18) Stayed with mom and her partner 5 years
Total = 26
———
(81) Stayed with mom and her partner a good share of a year or more (but he deliberately does not define the “or More”
Grand Total = 107

He “branded” 163 women as Lesbian Mothers
But we see that 56 (34%) never even lived with their “branded” Lesbian Mother. We don’t know that they are Lesbians AT ALL. For all we know “romance” could have been a single kiss.

Keep in mind that he only keeps a calendar of where the *kids lived.* He does not keep a calendar of where the parents lived. So these missing 56 or 34% of “Lesbians” Mothers very well could have run off and gotten married to a man.

The child believes mom kissed a woman (we don’t know because we never ask Mom) kid moves out of mother’s home and into a home with anybody *other than* her mother and never lives with her mother again. Mother marries a man and lives happily ever after for 35 years. NOW Shall we say that the kid had a Lesbian Mother?

He needs to get off of that Lesbian Word and start using the scientific term WSW, Women who have Sex with Women. And really we don’t even know if there was ANY Sex involved at ALL for 34%. It could have been a kiss. It is probably just more messed up heterosexuals skewing the results.

Blake
June 21st, 2012 | LINK

Great points both of ya. Keep up the good work. & thanks!!

Ben In Oakland
June 21st, 2012 | LINK

Are you trying to say that Regnerus is a tool?

Muscat
June 21st, 2012 | LINK

Good analogy. The fetishism and hand-waiving is quite worrisome. Random sampling is only useful if you use the data to actually speak about what you randomly sampled. Not if you pretend it’s a random sample of something you didn’t actually sample.

TampaZeke
June 21st, 2012 | LINK

At this point the anti-gays wouldn’t care if Jesus himself came down and declared this “study” bogus. They’ve already gotten EXACTLY what they wanted out of it. They got their money’s worth, that’s for sure!

grantdale
June 21st, 2012 | LINK

Regnerus becomes more and more obvious with each interview.

The sample may be random, but what he did with it is far from random. It was a deliberate and faulty re-categorisation of part of the data set, leading to a deliberate and faulty conclusion.

StraightGrandmother
June 21st, 2012 | LINK

I continue to read and research which requires me to go back and re-read some articles. Something I missed earlier. ON Dr. Regnerus’ personal blog he talks about only one other large study, “A: Most have not, as I elaborate in the literature review section of the study. That’s what’s unique about this study. Only Michael Rosenfeld’s 2010 article in Demography utilized a large population-based sample to compare one outcome among same-sex and other types of households. Others have worked with existing population-based samples, but rather small ones. But apart from Rosenfeld’s study, this is the largest nationally-representative sample of same-sex households, and I looked at 40 different outcomes, not just one or two.
http://www.patheos.com/blogs/blackwhiteandgray/2012/06/q-a-with-mark-regnerus-about-the-background-of-his-new-study/

Alright so I went and read Rosenfeld’s research which was incidentally used during the Prop 8 Trial.

HERE is FINALLY one LARGE SAMPLE research paper that compares how children do for these catagories. (Remember same gender civil marriage was not legal in any State in 2000)

Cohabiting Lesbian Couples
Cohabiting Gay Couples
Cohabiting Heterosexual Couples

Guess what? The children of Lesbian and Gay Cohabiting couples did BETTER than the children of Cohabiting Heterosexual Couples.

The study looked at what ages the children were for their grade. They were looking for kids who had flunked and were thus older.

http://www.stanford.edu/~mrosenfe/Rosenfeld_Nontraditional_Families_Demography.pdf

Actually that was a really good study so I see why Regnerus references it. The study is based on the US 2000 Census

I use U.S. census data to perform the first large-sample, nationally representative tests of outcomes for children raised by same-sex couples. The results show that children of same-sex couples are as likely to make normal progress through school as the children of most other family structures. Heterosexual married couples are the family type whose children have the lowest rates of grade retention, but the advantage of heterosexual married couples is mostly due to their higher socio economic status. Children of all family types (including children of same-sex couples) are far more likely to make normal progress through school than are children living in group quarters (such as orphanages and shelters).

Regnerus claims, “Oh my study is better I measured 40 different outcomes and I asked a lot more questions, such as have you ever been inappropriately sexually touched by a parent”

BUT Regnerus does a poor (I was going to use a different word there) job identifyling Lesbain and gay headed homes. He divines this by his famous question of “While you were growing up until age 18 did either one of your parents have a romantic relationship with someone of thier same sex?” Then he followes this up with his calendar where the adult kids write down who they lived with from birth to age 18. Regnerus is asking the *kids* to make a judgment call if they thought their parents ever had a gay fling. He never asks the parents.

The beauty of the census is that the adults, the parents are directly asked if you are gay or lesbian or not. So there is no guessing or I would say Regnerus *divines* if the kid grew up in a same sex headed home.

While it IS true that the Rosenfeld research simply measures if a kid is age appropriate for his class and not anything else, at LEAST we get to compare apples to apples and oranges to oranges and in fact the gay parents DID BETTER, thank you very much!!! My take away is that the Rosenfeld data is more limited but a hell of a lot more accurate.

chiMaxx
June 21st, 2012 | LINK

The back-and-forth at Slate between William Saletan and Regnerus is interesting–and starting to shed some light, I think: http://www.slate.com/articles/double_x/doublex/features/2012/gay_parents_study/gay_parents_study_mark_regnerus_and_william_saletan_debate_new_research_.html

In the most recent entry, as I write this, #4: Have I convinced you that you’re wrong? http://www.slate.com/articles/double_x/doublex/features/2012/gay_parents_study/the_gay_parents_study_what_does_no_difference_really_mean_.html Saletan hits on these central issues (read the whole thing for the evaluation on his answers:

“1. Does the New Family Structures Study show any difference in child outcomes between IBFs (intact biological families) and same-sex households of similar duration? No.

“2. Does the NFSS show that gay parents, per se, are more prone to unstable relationships than straight parents are? No. Because… all the kids in the sample were born between 1971 and 1994, [when]gay sex was illegal in many states, and gay marriage was illegal everywhere.

“3. Does the NFSS provide a more statistically valid sample of same-sex households than prior studies have offered? No…. We end up with a bunch of “no difference” samples from prior studies, skewed by education and socioeconomic status, versus an NFSS sample skewed by a focus on relationship behavior rather than family structure.

“4. Does the NFSS shift the burden of proof to those who claim “no difference”? You argue that it does, based on the size and randomness of your sample, combined with the known effects of instability in studies of adoption and stepfamilies. But both of your arguments are confounded by other factors.

Finally (and the “two questions” referenced here are long and hard to summarize), Saletan makes this observation: “The NFSS, with its preconceived focus on comparing IBFs to parents who had gay relationships—that is, comparing a structure to an orientation—seems designed to conflate the two questions. Unlike your harsher critics, I think this conflation was a blind spot, not a strategy.”

I think Saletan’s last point is well-taken. Clearly, Regnerus’s previous work suggests he has a bias that intact biological families are superior to other family structures. That is part of why Witherspoon would choose him rather than one of the authors of the 59 previous studies to lead this research. That that could have led him to confuse a sexual orientation with a family structure in the pursuit of this research is quite possible.

It will be interesting to see Regnerus’s response to this. If Saletan is right, it will be an interesting response, one the Witherspoon institute will be unlikely to link to.

(I would also be unsurprised to learn that Saletan has been reading the discussion here at BTB.)

jutta
June 21st, 2012 | LINK

> the wrong tools for the job

That’s what I wanted to say all the time (but didn’t because I was trying to keep up with reading everything that was written about this study).

If you want to study something that only is present in a tiny minority, a random population sample just will not work for you.

If you want to study the health implicatons of extensive running (50 miles and more), a random sample won’t work either, because chances are that you will find less than one such runner in your sample of 15000. You will have to contact sports clubs, running events, (ultra-)running magazines etc. to recruit your participants.

Same if you want to study a rare disease: You will contact hospitals, doctors, selfhelp groups but not look in a random sample just to find that your searched feature is not present there.

BTW: The children of the longitudinal study (NLLFS) are now around 22 to 24. http://www.nllfs.org/ So data about planned samesex families are available – and Mark Regnerus knows it, as he mentions the NLLFS and other studies several times in his paper.

Timothy Kincaid
June 22nd, 2012 | LINK

It is true that “better situated than virtually all previous studies to detect differences between these [different family] groups in the population.”

However, having the best seat in the house does not make one a useful critic, especially if one can’t distinguish between an opera and a musical.

Leave A Comment

All comments reflect the opinions of commenters only. They are not necessarily those of anyone associated with Box Turtle Bulletin. Comments are subject to our Comments Policy.

(Required)
(Required, never shared)

PLEASE NOTE: All comments are subject to our Comments Policy.