June 12th, 2012
Whoever invented the term “crowd-sourcing” was onto something, and the BTB crowd is among the best. Regular commenter StraightGrandmother had an interesting email exchange which she posted in the comments section. I want to bring up those email responses here.
Her first email began: “You can’t just force all the Children of Parents who had a same sex relationship into one bucket. it is not representative of what that respondent grew up in for a home life,” She then peppered him with several questions: How many LM/GF had biological parents who were married and stay married, how many were married then divorced, were never married, had a “fling”, etc. In other words, how many LM/GF people fit the more homogenous categories that he created for adult children of heterosexual couples. You can read her full email here. Regnerus replied:
Dear Ms. _______,
I will do my best to get answers to most if not all of your questions, hopefully in the next few days. However, there is not data on “flings,” only the presence or absence of relationships, and whether the respondent lived with the parent and their same-sex partner, and at what age (plumbing the calendar data is time-consuming work, however.)
I believe the article should be publicly available for free on Monday, from the publisher’s website. That is my understanding. I’m sorry you paid for it. I could’ve sent you a copy upon request.
People of good will (and some without) have and will continue to have lots of comments on measurement decisions, etc. Is understandable. Your comments are well-taken. A key priority, however, was always sample size. Curb it too much by slicing groups (wisely, even) into different categories and statistical power drops precipitously. With a much larger sample size, I would’ve done that. Was a judgment call with which some disagree. I maxed what Knowledge Networks could do with their panel, and no research firm out there is in a position to generate a larger N. Perhaps I could’ve left it in the field for another year, but that is quite awhile, and wouldn’t have doubled the sample size of LMs or GFs.
The study was reviewed the regular way, with multiple blind reviews to which I was required to respond.
I added some commentary about the study background, context, at the place where I blog once a week:http://www.patheos.com/blogs/blackwhiteandgray/2012/06/q-a-with-mark-regnerus-about-the-background-of-his-new-study/.
You can cut, paste, and post whatever you wish…
Sincerely yours,
Mark Regnerus
StraightGrandmother thanked him and clarified why she asked about “flings” :
There is something much much different about a home environment where the parents are in a Mixed Orientation Marriage and one spouse has an extra marital affair with someone of their same sex. The troubling part is the extra marital affair not just the fact that it was with a person of the same sex.
This situation is much different than a sole lesbian who establishes a same sex relationship. See the difference between an extra marital fling and a normal two person relationship?
Regnerus responded:
Please post…
By the way, one of the key methodological criticisms circulating is that–basically–in a population-based sample, I haven’t really evaluated how the adult children of stably-intact coupled self-identified lesbians have fared. Right? Right. And I’m telling you that it cannot be feasibly accomplished. It is a methodological (practical) impossibility at present, for reasons I describe: they really didn’t exist in numbers that could be amply obtained *randomly*. It may well be a flaw–limitation, I think–but it is unavoidable. We maxxed Knowledge Networks’ ability, and no firm is positioned to do better. It would have cost untold millions of dollars, and still may not generate the number of cases needed for statistical analyses. If randomness wasn’t the key priority, then we could’ve done it. And we’d have had a nonrandom sample that was no better than anything before it. So, while critics are taking potshots, they should remember that there’s a (low) ceiling to what’s possible here. My team of consultants elected to go with the screener questions (including the one about same-sex relationships) that we did, anticipating–accurately, too–that there would be no way of generating ample sample size if we narrowed the criteria (for who counts as a lesbian parent) to the sort that critics are calling for. We figured that, with the household roster/calendar offering the opportunity to identify who you lived with, we’d comfortably get enough cases wherein the respondent reported living with mom and her partner for many consecutive years. But few did.
Ergo, in contrast to some impressions, I didn’t construct the study to tell the sort of story I wanted.
I hope that people read the three comments that were also published with the study. They voice confidence in the data, while asserting ample concerns about its use.
Sincerely yours,
Mark
You can read an overview of those three commentaries published in Social Science Review here.
In the third email, StraightGrandmother emphasized her concerns about mixed orientation marriages being included in the LM and GF categories:
If your gay/lesbian population primarily produced children in a Mixed Orientation marriage then I feel you should have clearly said that. Because then your research is showing what I believe to be true, Mixed Orientation Marriages are very hard on children. We know that only 1/3 of Mixed Orientation Marriages attempt to stay together after disclosure and of that 1/3 only half manage to stay together for 3 years or more (and it goes really down hill after 7 years).
IF your gay/lesbian population primarily produced children in a Mixed Orientation marriage then I feel you should have clearly said that. Because then your research is showing what I believe to be true, Mixed Orientation Marriages are very hard on children. We know that only 1/3 of Mixed Orientation Marriages attempt to stay together after disclosure and of that 1/3 only half manage to stay together for 3 years or more (and it goes really down hill after 7 years).
She mentioned Zach Wahls, the young man who was raised by two mothers as the only family he ever knew and who testified about his family before an Iowa statehouse hearing to consider rescinding same-sex marriages in that state, as an example of someone who was really the product of Lesbian Mothers. StraightGrandmother castigated Regnerus for what she saw as carelessness with how he described his “lesbian mothers” category. “You were not up front saying, ‘99.5% of the respondents with a gay or lesbian parent were raised in a home where the parents had a Mixed Orientation Marriage.’ Just that simple sentence would have made everything right. …Because of that one missing sentence clearly stating the Mixed Orientation Status you are being attacked instead of being praised.”
Regnerus replied:
Well, it was quite unlikely that I would be praised in the scholarly community. Nor did I expect to be. Trust me, I’m not surprised by the antagonism, but nor does it feel like a badge of honor to me.
The Daddy donor study, with which I am familiar, is not a population-based random sample. It’s an opt-in sample. I even inquired about this study with Abt SRBI and it was apparent that they couldn’t handle it like KN could. They didn’t randomly survey 1,000,000 people. Only the Census–or some precious few deep-pocketed federal studies–contacts that many. The research team discussed an opt-in sample supplement to boost N, and advised against it because the probability of respondents’ inclusion could not be ascertained. Been there, discussed that, elected not to go there in order to preserve the quality of the sample.
I don’t go into orientation of parents in this study. That is a deeper well, and in the cohort that I’m studying, I don’t presume that either they or their parents would confidently call their parent gay or lesbian (or something else). Ergo, we made it about behavior–and not discreet behavior but something their child would be aware of.
Moreover, plenty of scholars, like Lisa Diamond, assert that women’s sexuality is more fluid than men’s. (I make no claim or assertion either way.) If she’s right, your claim about mixed-orientation marriages seems more fixed, especially for women, than it need be in social and marital reality. But I elected NOT to make this about orientation or self-identity. You suggest more ominous motivation, but I assure you that was not true.
Your accusations are getting more heated, and I’m afraid unless we can correspond civilly, I may have to call a conclusion to this.
Again, please post.
Mark
Latest Posts
Featured Reports
In this original BTB Investigation, we unveil the tragic story of Kirk Murphy, a four-year-old boy who was treated for “cross-gender disturbance” in 1970 by a young grad student by the name of George Rekers. This story is a stark reminder that there are severe and damaging consequences when therapists try to ensure that boys will be boys.
When we first reported on three American anti-gay activists traveling to Kampala for a three-day conference, we had no idea that it would be the first report of a long string of events leading to a proposal to institute the death penalty for LGBT people. But that is exactly what happened. In this report, we review our collection of more than 500 posts to tell the story of one nation’s embrace of hatred toward gay people. This report will be updated continuously as events continue to unfold. Check here for the latest updates.
In 2005, the Southern Poverty Law Center wrote that “[Paul] Cameron’s ‘science’ echoes Nazi Germany.” What the SPLC didn”t know was Cameron doesn’t just “echo” Nazi Germany. He quoted extensively from one of the Final Solution’s architects. This puts his fascination with quarantines, mandatory tattoos, and extermination being a “plausible idea” in a whole new and deeply disturbing light.
On February 10, I attended an all-day “Love Won Out” ex-gay conference in Phoenix, put on by Focus on the Family and Exodus International. In this series of reports, I talk about what I learned there: the people who go to these conferences, the things that they hear, and what this all means for them, their families and for the rest of us.
Prologue: Why I Went To “Love Won Out”
Part 1: What’s Love Got To Do With It?
Part 2: Parents Struggle With “No Exceptions”
Part 3: A Whole New Dialect
Part 4: It Depends On How The Meaning of the Word "Change" Changes
Part 5: A Candid Explanation For "Change"
At last, the truth can now be told.
Using the same research methods employed by most anti-gay political pressure groups, we examine the statistics and the case studies that dispel many of the myths about heterosexuality. Download your copy today!
And don‘t miss our companion report, How To Write An Anti-Gay Tract In Fifteen Easy Steps.
Anti-gay activists often charge that gay men and women pose a threat to children. In this report, we explore the supposed connection between homosexuality and child sexual abuse, the conclusions reached by the most knowledgeable professionals in the field, and how anti-gay activists continue to ignore their findings. This has tremendous consequences, not just for gay men and women, but more importantly for the safety of all our children.
Anti-gay activists often cite the “Dutch Study” to claim that gay unions last only about 1½ years and that the these men have an average of eight additional partners per year outside of their steady relationship. In this report, we will take you step by step into the study to see whether the claims are true.
Tony Perkins’ Family Research Council submitted an Amicus Brief to the Maryland Court of Appeals as that court prepared to consider the issue of gay marriage. We examine just one small section of that brief to reveal the junk science and fraudulent claims of the Family “Research” Council.
The FBI’s annual Hate Crime Statistics aren’t as complete as they ought to be, and their report for 2004 was no exception. In fact, their most recent report has quite a few glaring holes. Holes big enough for Daniel Fetty to fall through.
Boo
June 12th, 2012
“By the way, one of the key methodological criticisms circulating is that–basically–in a population-based sample, I haven’t really evaluated how the adult children of stably-intact coupled self-identified lesbians have fared. Right? Right. And I’m telling you that it cannot be feasibly accomplished. It is a methodological (practical) impossibility at present, for reasons I describe: they really didn’t exist in numbers that could be amply obtained *randomly*. It may well be a flaw–limitation, I think–but it is unavoidable.”
Shorter version: I know I’m making claims I can’t actually support, but I don’t actually care.
Brennin Statistician
June 12th, 2012
As a real statistician, and not someone playing one on the internet, I find Dr. Regnerus’ replies entirely reasonable. In fact, I would not have been nearly as polite to the gadfly “StraightGrandmother,” whom I have observed touting studies when she agrees with the conclusions, even though she manifestly does not understand statistics.
I think Burroway’s criticisms have some merit but this study appears to be (I have not read the article yet but intend to soon) far superior to the junk cited by the APA brief. I have had to lump groups together that, ideally, I would like to keep separate in my analyses for the reason Dr. Regnerus mentioned to “StraightGrandmother.” For example, I recently ran analyses with race-ethnicity as a factor where it was NH Black vs. all other race-ethnicity combinations. Now, one difference between those analyses and Dr. Regnerus’ is that the group difference we were really interested in was HIV+ vs. non-HIV, not race-ethnicity but the underlying principle is the same, i.e., statistical power to detect differences.
Jim Burroway
June 12th, 2012
But this is the fundamental problem with statistics when they are divorced from their meaning. It sounds like some statisticians are more interested in making the numbers behave than in making the logic behind the numbers meaningful. If your numbers don’t behave — i.e. there aren’t enough of them for statistical power — just add more numbers whether they are related to the underlying meaning or not. You can beat numbers into submission that way, but that doesn’t mean that you will then be able to correctly interpret what those numbers mean when you are through with the flogging.
But the problem is that when you do that the goals became confused. The goal now is not to study people, but to increase statistical power. I understand that this distinction may be lost on statisticians, but in the real world this is still important.
I think SG is correct. If he had presented his study the way it was but framed it according to the data that was actually available — children of broken and unstable mixed-orientation relationships — we would have a very different story. And I think a more valuable and interesting one.
Pliny
June 12th, 2012
You don’t have to understand statistics to recognize that a statistician chose not to control for variables that directly affect his thesis.
Comparing apples to oranges only works when you’re interested in the properties of roughly spherical fruits.
Boo
June 12th, 2012
“I have had to lump groups together that, ideally, I would like to keep separate in my analyses for the reason Dr. Regnerus mentioned to “StraightGrandmother.—
But that’s just it, for what reason did he “have” to lump the groups together? To justify the money that was spent? If you don’t have a good enough sample to get a meaningful comparison, then, well, you don’t have a good enough sample to get a meaningful comparison. Science doesn’t care how hard you tried.
Muscat
June 12th, 2012
Jim – This is exactly right. If the story had been framed according to the data, I don’t think people would have a problem with the study. But instead, we have data scavenged together to (apparently) tell a story it then no longer tells. And the cherry on top is that this scavenged data is treated as superior to other forms of scavenged data because it is “random.” Sort of like taking a random sample of people who have committed adultery and then publishing a study based on that data to make claims about polyamory. Truly laughable.
Mark Miner
June 12th, 2012
First, Straight Grandmother rocks, for her persistence and willingness to cut through the statistical BS and hold Regnerus accoutable.
Second, what interests me here is way the study (and particularly Regnerus’ discourse about it) makes clear the conceptual blinders of the investigator. What is homosexuality, to Robert George, & co, but failed heterosexuality? So when we study “Homosexual Parents,” we’re naturally going to have our eyes WIDE OPEN for people who have failed at heterosexality, as characterized by Broken Marriages, Flings, Mixed Orientations, etc. Couples who OWN the homosexuality, OWN the choice to be gay married, OWN the choice to have children, where are they? It’s a mountain of stats & verbiage to plow through, but that’s the question that Straight Grandmother and the rest of us gadflies should be throwing at this study: of your “Homosexual Parents,” what was the general level of intentionality about their homosexuality and parenting? If we had them march to opposite sides of the room, how many of the “Homosexual Parents” would be closer to Dan & Terry, who wrote a book about their very deliberate choice to have a kid, and how many would be closer to some loser of a straight guy who jacked off with his buddy in a hotel room, and his wife found out about it, and she divorced him and he lost his job, etc.? I mean, the number of “Homosexual Parents” cited is not large. In statistics they talk about “cherry-picking.” Everything in Regnerus’s rhetoric smells like turd-picking
Brennin Statistician
June 12th, 2012
And the cherry on top is that this scavenged data is treated as superior to other forms of scavenged data because it is “random.â€
Statistics according to Muscat: Don’t bother with random sampling when conducting a poll re: the presidential race, just take a stroll around your block and ask those you encounter along the way whom they’ll vote for come November.
I think Mr. Burroway’s criticisms are pretty fair, on the whole, and he at least has a decent grasp of basic statistics. The same cannot be said of you.
cowboy
June 12th, 2012
I’m with Mark Miner, I’m almost ready to send $31 directly to StraightGrandMother.
Boo
June 12th, 2012
“Statistics according to Muscat: Don’t bother with random sampling when conducting a poll re: the presidential race, just take a stroll around your block and ask those you encounter along the way whom they’ll vote for come November.”
Except that’s not what he’s saying at all. He can correct me if I’m wrong, but what I believe he’s saying is that randomizing your sample is useless if you go on to make an invalid comparison. Since the large majority of gay familes experienced divorce, the proper comparison group would be straight families that experienced divorce. The author chose not to do this. As a result he did not make a valid comparison.
The ultimate irony,of course, is that the conservative groups that are touting this study are doing everything they can to create more of those families by attempting to force gay people back into the closet, which inevitably means more gay people marrying straight people and eventually divorcing.
Priya Lynn
June 12th, 2012
Brennin, the only one impressed with you is yourself.
Lightning Baltimore
June 12th, 2012
“I haven’t really evaluated how the adult children of stably-intact coupled self-identified lesbians have fared. Right? Right. And I’m telling you that it cannot be feasibly accomplished.”
This seems the crux to me. He admits he’s not evaluated the very families his “study” deems unsatisfactory.
Steve
June 12th, 2012
In theory, it’s of course true that random sampling is superior. But due the shenanigans he pulled with the biased questions and the creation of the sub-groups it can hardly be called truly random anymore. He took a random sample and then ordered it to his achieve desired outcome.
Steve
June 12th, 2012
Even if we cede his point that a random sampling of a high number of planned gay families were impossible, he still fails *spectacularly* at what he actually did.
Why did he not structure the people he sampled to find children that were born to a straight marriage, but then one parent came out and they were then raised in a same-sex relationship? I don’t buy his point that there were allegedly few of those. There are tons such people.
If he couldn’t do that, then he should have packed up and not drawn any conclusions. But instead he manipulated the samples to throw such families together with families where someone may have had a same-sex relationship in college. For all we know that parent may be bisexual and they divorced for wholly different reasons. And he threw in people who are closeted and cheated on their spouse. Again, in of itself it may be worthwhile to study how such behavior negatively affects families, but that’s not what he studied.
ohgeez
June 12th, 2012
You all need to take statistics 101. RANDOMNESS matters. It really does.
Lightning Baltimore
June 12th, 2012
Yes, ohgeez, randomness matters quite a bit. On the other hand, so does honesty.
Priya Lynn
June 12th, 2012
We’re all familiar with randomness and its importance ohgeez but it doesn’t mean squat if you beat up your numbers to gain it.
StraightGrandmother
June 12th, 2012
Brennin Statistician – Yeah when you can’t come up with a good argument whip out an ad hominem insult “gadfly StraightGrandmother”
Trouble for you IS I totally understand the statistics. I knew what he was doing right from the get go before Dr. Regenerus even confirmed it. I knew it HAD to be MIXED ORIENTATION MARRIAGES with MIXED ORIENTATION SEXUAL UNIONS because I AM a grandmother and KNOW VERY WELL the years the survey covered. You are probably to young to know but 1977 was the Anita Bryant Years.
I KNEW that there was no way on God’s Green Earth that he found THAT many people who were born to a Lesbian Mother using a sperm donor OR to a gay father using a surrogate who were born from 1972 to 1993, and who raised that child in a same gender home. No way. The only remaining alternative is exactly what I said it would be, people who were born to parents in a MIXED ORIENTATION MARRIAGE and/or in a MIXED ORIENTATION SEXUAL UNION.
As I said before I don’t mind that data, I think it is correct. MIXED ORIENTATION SEXUAL RELATIONSHIPS that produce children are very bad for the children. This data proves that. And it proves nothing more than that.
Sonny you wanna insult StraightGranmother you better have something to back that up. Because I was right, that is exactly what the data showed, what I said it would. (pardon my feistiness I have not been publicly ridiculed in a long time)
Steve
June 12th, 2012
The only was his sample was random was in the way he gathered it. He took random people and then asked them highly biased and leading questions. That destroyed any semblance of randomness he tried to achieve. The way he sorted people into “gay parents” and “straight parents” is NOT random and designed explicitly to achieve a preconceived outcome
Stephen
June 12th, 2012
“Well, it was quite unlikely that I would be praised in the scholarly community. Nor did I expect to be. Trust me, I’m not surprised by the antagonism, but nor does it feel like a badge of honor to me.”
Why would he expect the scholarly community to be hostile? I find this puzzling. A note of self-righteous paranoia seems to suddenly appear for no perceptible reason.
Stella J
June 12th, 2012
In order to be honest about findings, one needs a random sample. I agree Regnerus should have emphasis the shortcomings of the data in his article – but this is not uncommon in quantitative social science articles. He’s completely honest in his emails and responded politely to a very rude email. Makes those who critique this look bad. I have my issues with this article and don’t believe we can say gay parents are bad for kids if raised in stable families, but we need to have statistically informed critiques.
Alec
June 12th, 2012
As I indicated in a previous post, this guy is the new Paul Cameron. His responses are disingenuous. He wants to play the victim card while telling everyone that he isn’t interested in playing victim.
How can randomness possibly matter if your comparison is useless? It cannot. Combine that deliberate decision with his political agenda and the political agenda of the groups funding him, and this is clearly a hit and run study designed to give cover to bullshit, bigoted arguments.
And his apologists are here crying “randomness” as though it carries meaning under the circumstances.
If I was cross-examining this guy and he was posing as an expert witness, I’d have a field day. He can’t handle benign inquiries from a concerned grandmother, he’d be crushed under cross-examination and, as we shall soon see, through peer review.
Bernie
June 12th, 2012
Brennin, Jim makes a such a strong point on this that it can’t be denied. I firmly believe that the report is jaded, and that this is just one instrument that was produced for the upcoming elections.
Peter Sprigg pulled the same crap in 2010 when he stated that gays were 5x more likely to commit sexual assaults in the military. His numbers didn’t jibe, so I read the entire DOD report. The little Baptist minister had found numbers to fudge and make the results he wanted. The imp lied.
I do not for one single moment trust any of this. The timing is uncanny.
Regan DuCasse
June 12th, 2012
@Brennin Statistician:
How long before people stop using gay people as lab rats? It wouldn’t surprise me if such information became even tougher to obtain because the researchers can’t be trusted with the data.
Gay adults are still extremely vulnerable to systemic discrimination that adversely affects them and their families. And clearly, the aim of certain organizations is to continue to do so by force of gov’t.
Did Regenerus put in any disclaimers that his results not be used to political detriment of the subject?
Did he assure those in his sample of the same?
He didn’t include how such issues affect gays with children or who married op sex spouses. Especially considering the age group of the children of his study, when two decades ago, the socio/political climate was far less supportive.
And such an atmosphere would have affected these gay parents greatly when THEY were growing up.
These are stressors that time and time again are excluded from research such as his.
His conclusions, are the same as any other anti gay conclusion, that it’s the sexual orientation of the parents and not the adverse conditions in which to be gay, that might cause generational social maladjustment.
I’ve noticed that very little is said about the straight parents who have raised their gay child badly and it’s sometimes had deadly results.
No one considers what happens when misinformed straight parents follow doctrine, rather than their own instincts to protect their child.
Why is it the gay folks are expected to open their lives to such scrutiny, and then are betrayed?
The census, is still skewed by the closet and considering a betrayal such as Regenerus’s. Were I a gay person, except with some intense vetting, I would tell the next researcher coming along to kiss my ass.
fannie
June 12th, 2012
Right on, StraightGrandmother!
And ditto this:
“Brennin, the only one impressed with you is yourself.”
If one actually reads the critiques people here are making, many people here do understand statistics and the importance of randomness.
What many of us object to is the framing of this study as being about “same-sex parenting.” A more accurate framing would be that it’s a study about “parents who have ever had a same-sex relationship in their lives.”
grantdale
June 12th, 2012
@Jim If he had presented his study the way it was but framed it according to the data that was actually available — children of broken and unstable mixed-orientation relationships — we would have a very different story.
Knowing, as Regnerus did, that his sample of “lesbian” and “gay” parents were overwhelmingly similarly situated to heterosexual single parents and heterosexual divorced parents he could have made the comparisons between those groups.
Could. But he chose not to. Instead he’s gone well beyond where his data sample could take him.
An eyeballing of his tables suspects he would have found no statically significant difference between those two properly compared groups. (Leading to a legit conclusion that the sexual orientation of parents — of itself — is unrelated to outcomes for children).
Instead, Regnerus has controlled for the one variable he appears to be very certain influences outcomes (family instability)… but only for his heterosexual parents.
He failed to control that variable for his GLB sample, and for a very weak reason. As Boo said: if you don’t have a good sample (for the purpose), you don’t have a good sample (for the purpose).
But what do we have? A crippling methodological flaw, and publication of wrongly controlled and compared sample groups, and wording in the paper (and, importantly, elsewhere) that fails to make any of this crystal clear; indeed wording that often serves to mislead. (read the opening of his Slate article, as example).
At the risk of breaking some recent blogger’s rule: I’m calling it as Cameronesque.
His peer reviewers have really let Regnerus down.
grantdale
June 12th, 2012
ps: and three hearty cheers for StraightGrandmother
Palmer
June 12th, 2012
I’d still like to know what the “harm” is. Is that covered in this study at all?
If the control group, the straights had been of the same nature, I mean caught cheating or from broken marriages, then this research would have some value. As it stands it just appears to be another hit piece.
Helen in Ireland
June 12th, 2012
I have been following this with great interest, and kudos to StraightGrandmother for her persistence in running Mr Regnerus.
May I make a suggestion from left-field, one which might make lemonade from this lemon of a study?
PRAISE the study from the point of how it demonstrates the societal problems that the gay and lesbian parents experienced during the closeted era of 1972-1993. POINT out the similarities between the GF and LM groups and those of broken heterosexual families with result effects on the children.
Then EMPHASISE STRONGLY just how IMPORTANT it is that FULL marriage equality be extended to LGBT citizens so that the children of LGBT familes are given the stability and positive support that children of heterosexual couples receive.
TampaZeke
June 12th, 2012
It seems to me that Brennin Statistician, for all his/her arrogance and condescension, is being schooled by Straight Grandmother. He’s insulted her with an ad hominem attack. He’s talked down to her and everyone else. He’s bloviated ad nauseam about this statistical rule and that, but he has YET to actually address her point and that is his statistical methodology was just fine to prove the point that it proves but it doesn’t prove what the author claims it proves.
Boo
June 12th, 2012
I have a suspicion that StraightGrandmother is actually a team of highly trained specialists. Perhaps they all wear capes.
olterigo
June 12th, 2012
“I don’t go into orientation of parents in this study. That is a deeper well, and in the cohort that I’m studying, I don’t presume that either they or their parents would confidently call their parent gay or lesbian (or something else). Ergo, we made it about behavior–and not discreet behavior but something their child would be aware of.”
Whatever he was trying to make it about, it is about “gay fathers” and “lesbian mothers”. And if he wants to pretend that he didn’t think this would happen, I have a bridge to sell in Brooklyn.
It’s also interesting how “your accusations are getting more heated.” Why wouldn’t they, if one side pretends like the criticism does not cripple this million-dollar study?
Hyhybt
June 12th, 2012
It does no good to start with the best possible sample if you misuse it. If you can’t get a sample large enough to make the comparison you’re trying to make, what good does it do to throw in garbage?
The thing to do, if all who had a parent with a gay relationship must be lumped in together to get a large enough sample, would be either to drop that aspect of the study and only compare the categories you do have enough for without poisoning your sample, or else (since that one area was the point in the first place) to throw out entirely those who had the same parents all along (because they didn’t have enough gay equivalents) and compare everything-else-lumped-together-gay with everything-else-lumped-together-straight. You don’t get as many numbers that way, but then, all the others generated are worse than worthless anyway.
chiMaxx
June 12th, 2012
Right. The analogy that comes to mind is a study that claims to compare Fred Karger supporters with Romney supporters based on a random sampling of Republican primary voters. But they discover that there aren’t enough Fred Karger supporters in the sample to draw meaningful conclusions, so they lump Perry, Santorum, Bachmnn and Paul supporters in with the Karger supporters in order to increase the sample size–but they still refer to them as Karger supporters in the analysis, because that’s what they set out to study, and these other groups are all similar to the Karger supporters in that they didn’t support Romney.
jerry
June 12th, 2012
If I had any doubts about the honesty of Regneurus he settled it with the following remark to SGM:
“Your accusations are getting more heated, and I’m afraid unless we can correspond civilly, I may have to call a conclusion to this.”
I will tip my hat as well to SGM for a well done questioning of this study.
CPT_Doom
June 12th, 2012
I am a trained economist and have worked in the healthcare quality arena for more than 2 decades. I live my life dealing with statistics and think both Jim’s original critique and the questions raised by StraightGrandmother are dead on. Although Regnerus is to be praised for using a random sampling approach to identify his subjects for this research, that sampling system alone cannot and does not, in this instance, guarantee that the results will actually answer the question being asked.
It appears that Regnerus’ null hypothesis was that the children of gay or lesbian parents would have the same outcomes of those with straight parents. Interestingly, Regnerus could reject that null hypothesis ONLY when dealing with children of “intact biological families.” The children with heterosexual parents not married to their other biological parent had results remarkably similar to children with gay or lesbian parents who had once been in heterosexual relationships.
However, Regnerus has not “proven” that gay or lesbian parents caused any differences in child outcomes that do exist between the children of “intact biological families” in his research and those that report a parent with at least one same-sex relationships. That is because his sample was still inadequate – as he admits – to answer the question “do children raised in a stable, intact male-male or female-female family fare any differently than those from ‘intact biological families’?” He simply did not have enough adults who had experienced that family structure to make a comparison.
We can see that the sample of adults chosen for this study is not representative of the children currently being raised in male-male or female-female households. You can see that from the census data. The families in Regnerus’ study who had a parent with a same-sex relationship also received government welfare at very high rates – above 50% (for some reason I can’t open his study right now). Yet if you look at the 2010 census data (http://www.census.gov/hhes/samesex/ – look at the excel file with “Characteristics of same-sex households”) you see that at least 66% of same-sex couples own their own home and at least the same percentage have income rates above $50,000/year. Now the census data does not tell us if same-sex households with children are poorer than other same-sex households, but given the costs of adoption, in vitro and surrogacy, the most common ways same-sex couples build their families, it seems unlikely.
Therefore there is a disconnect between the family experiences of respondents to Regnerus’ survey and the status of current families headed by same-sex couples. This makes the results of Regnerus’ work even more suspect.
StraightGrandmother
June 12th, 2012
chiMaxx I think you made a BRILLIANT analogy with the Fred Karger Supporters, very easy to understand and grasp what went on with the Rengerus “research” using your analogy.
StraightGrandmother
June 12th, 2012
CPT_Doom
I LOVE your analysis of the census data. That is one element I think you (or might have been somebody else but I thought it was you) brought up. To look up the census data to see if it could be true that that many so called “LM” “GF” were really reprentitive of woman+ woman Male+male realy true gay families.
I really made note of that comment but didn’t know how to track that down, thank you MEGA MUCH MWAH!
StraightGrandmother
June 12th, 2012
Well the bright side is The Witherspoon Instituite and the Bradley Foundation threw $750,000 right down the toilet that could have been used in other projects to hurt sexual minorities. The Bradley Foundation just got their 2011 Annual Reports printed up and they are not yet on line but they will be shortly.
In 2010 The Bradley Foundation gave $50,000 to the Witherspoon Institute for a Marriage Project
Gave $80,000 for Operations
Gave $75,000 for a National Security Project
I can’t wait to see who they funded in 2011.
Steve
June 12th, 2012
In other news, Regnerus admits that he allows his faith to influence his research and is even proud of it:
http://tcc.trnty.edu/alumni/profiles/regnerus/
“I’ve noticed that some Christian professors see a disconnect between their faith and their profession. I believe that if your faith matters, it should inform what you teach and what you research.”
I consider that highly unethical. He is nothing but a fundamentalist religious nutbag.
Timothy Kincaid
June 12th, 2012
In our analysis of Regnerus’ study and it’s publication, I think we need to be aware of a subtlety that might otherwise elude us. Regnerus’ motivations are not anti-gay in the strictest sense. This study was not necessarily to prove that gay families are inferior.
Rather, his motivations can be found in this statement:
In other words, his message is not “gays are inferior”, but rather “heterosexual married families are superior”. Gays are but one of the inferior groups.
(As a comparison: the KKK is not so much anti-Latino as they are white supremist. Latinos are but among blacks, gays, and Jews and anyone else who comes along that is “us”).
Steve
June 12th, 2012
You always expect the best in people and it’s really naive in this area. What he writes isn’t necessarily what he thinks. It’s a typical tactic of anti-gay groups to soften their blows in such language. No, he isn’t a far-right demagogue like Peter Sprigg or Bryan Fischer, but he really has something against gay people. He has a history of publishing articles and op-eds with an anti-gay slant.
Bernie
June 12th, 2012
Tim,
Subtlety is the perfect word. However, his motivations behind this were anything but innocuous and scholarly.
StraightGrandmother
June 12th, 2012
Hey Jim you got a link over at The New Republic. Thank GOD Jim you worked your butt off on this on Saturday and got this to our side early. Now as soon as the study is hitting the airwaves the truth is right there. We are NOT playing catch up, we are toe to toe and his own words damn him.
http://www.tnr.com/blog/plank/104019/its-time-mark-regnerus-get-collectively-dumped
Bernie
June 12th, 2012
Thanks SGM, for the link to NR. I just twittered the heck out of that article.
Timothy Kincaid
June 12th, 2012
chiMaxx –
love it!!
StraightGrandmother
June 12th, 2012
NOW 24 Hours Later, after maybe a few e-mails he might have received, NOW he sounds a bit more honest. What do you think?
“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, adding it’s true, he did not know that relationship. “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.”
Regnerus said it’s entirely possible that instability in the household led to some of the reported negative outcomes in adult children of same-sex parents. He said children of heterosexual couples in an unstable home were also found to fare worse than those in a stable environment.
“People gay or straight should stick with their partners, he said. “I think the study provides evidence of that.”
http://www.cbsnews.com/8301-504763_162-57451777-10391704/kids-of-gay-parents-fare-worse-study-finds-but-draws-fire-from-experts/
I bet it was Rob’s pic that I sent all over creation that may have encouraged him to pause and reconsider. This is a hell of a perfect picture.
http://wakingupnow.com/blog/wp-content/uploads/2012/06/quit-damning5.png
Jay Jonson
June 12th, 2012
Does anyone know anything about the journal Social Science Research and/or its editor and editorial board?
Brennin Statistician
June 12th, 2012
Steve and Alec seem to be trying to one-up each other with their histrionic screeching. (Aside to Alec and Steve: I would recommend avoiding gay stereotypes instead of running into them head on.)
1. Alec’s claim that Dr. Regnerus is the new Paul Cameron and that he should “lose his license” (he is a PhD sociologist, not a mental health professional) is pure flatulence.
2. Steve wrote:
But due the shenanigans he pulled with the biased questions and the creation of the sub-groups it can hardly be called truly random anymore. He took a random sample and then ordered it to his achieve desired outcome.
Steve alleges Dr. Regnerus asked “biased” questions, yet he fails to provide any examples. Also, the rest is incoherent babble.
Quit while you are behind.
Brennin Statistician
June 12th, 2012
I AM a grandmother and KNOW VERY WELL the years the survey covered.
I do not doubt you are a grandmother (or straight, for that matter).
You are probably to young to know but 1977 was the Anita Bryant Years.
I wasn’t around for Anita Bryant’s 1977 socio-political campaign, true.
StraightGrandmother
June 12th, 2012
Jay Jonson
Why yes I do know the contact for the Journal Social Science Research. And I can assure you they put out a list of all their editors on the internet. I may have contacted one, two, three, four, you know I kinda lost track. I did not use their convenient on line form but I went on the internet and looked up their e-mail address so I could include Rob Tisinai’s great picture right inside my e-mail
http://wakingupnow.com/blog/wp-content/uploads/2012/06/quit-damning5.png
Here is the link to the Journal this is published in they have an on line form to contact like 20 different Editors and Advisory Editors.
http://www.journals.elsevier.com/social-science-research/editorial-board/
Boo
June 12th, 2012
Brennin Statistician
It’s okay. If I were you I’d want to run away from the actual issues being raised too.
Or maybe you would care to tell us why it was appropriate to separate out the divorced familes for straight parents but not for gay parents? I mean, I’d hate to think that all you have to offer is more empty bluster. I certainly have reason to think that, but I would hate to.
Marco Luxe
June 12th, 2012
Best comment — from Tafter, another poster
June 12th, 2012 | LINK
The problem I have with all of this is that, despite some great data collection and numerous caveats in the publication, Mark Regnerus is running around selling this study as proof of something sinister about same sex parents. Try to square his measured statements above with his quote in LiveScience:
“The study found “that the scholarly and popular consensus that there are no notable differences between the children who grew up with a mother or father in a same-sex relationship and those whose (heterosexual) mother and father were and are still married is a fiction,†study researcher Mark Regnerus, a sociologist at the University of Texas, told LiveScience. “
Steve
June 12th, 2012
How the questionnaire was biased was already explained by Jim in his articles. Pay some attention. I don’t need to explain it again.
Alec
June 13th, 2012
I did not suggest that Mr. Regnerus lose his license (although prospective graduate sociology students with a quantitative bent interested in family and sexuality studies might think long and hard before accepting an offer of admission by UT). Nor have I disclosed my sexual orientation, and I doubt my “histrionic screeching” as you describe my posts is uncalled for. It is very clear to me that Regnerus wants the role of Paul Cameron, if not the suggestion that he is at all similar to Cameron. He clearly contacted anti-gay groups and outlets to promote his study, I think it is clear that he was looking to establish that gay parents were inferior and he used the easiest method of doing that: Looking to failed mixed orientation marriages, and not even bothering to attempt to find a sufficient sample of long term intact two-parent same-sex families.
Muscat
June 13th, 2012
@Boo – Yes, that’s more or less what I was saying. Let’s say I collect a random sample of families that gets me, say, 5 Asian-American families, and then because the sample size for Asian-American families is so small I lump together other ethnic minorities I also got from random sampling. If I then try to say I can conclude something about Asian-American families based on that lumped-together sample, and on top of that I claim that my study is better than previous studies of Asian-American families because I used random sampling, on both counts I am full of hooey (I believe that’s the technical term ;) ). I think this is a fair analogy for what Regnerus did with “same-sex couples/parents”.
@Brennin – The fact that either you couldn’t figure out that was my critique or else just jumped to the least charitable interpretation probably tells me all I need to know. But, just so you know, I’m a social scientist trained in quantitative methods at one of the best programs for statistical training in the country, so your condescending comments suggesting I don’t understand or appreciate statistics are really off base. Thanks.
Boo
June 13th, 2012
Muscat- I wouldn’t put too much stock in anything Brennan said. Actual experts generally don’t feel the need to constantly trumpet their expertise.
Victor
June 16th, 2012
Dear StraightGrandmother,
Thank you. You’re a Saint! Thank you for your drive, tenaciousness, and expertise AND reposting all of this stuff!
Victor
StraightGrandmother
June 20th, 2012
I think I am going to post this on all the Regnerus topics on Box Turtle. Let’s say we all work together and make a terrific website where we collect all this information on the Regnerus’ Study.
We crowd source this and everybody helps. I think what is missing is a lot of comments on the research by other Sociologists. We can all take a State and then call all the Universities in that State and speak to Sociologists there and ask them to provide their feedback on the research.
Other people can work on collecting up all the direct quotes from Regnerus.
Oh and shouldn’t we collect up all the places the data is being misreported? I saw for example a quote something like, “Well this study proves that pedophilia is rampant with gays”
And what about a website? Should we buy a domain name or should we use Blogger or Word press or something? If we hosted this separately how much would that cost us in bandwidth if a lot of people visited? Does anybody have idea on this?
Let’s set this up and organize this so it will be real easy for lawyers from our side to have a good reference point to jump off from.
What else? Do you like this idea? Who will help? I don’t mind if this is a part of Box Turtle and they are getting recognition for this. But I do think to do this right it is to big for the few guys at Box Turtle we need a lot more helping hands than just those guys. Whadda think?
Chuck
March 4th, 2013
I didn’t read all the comments here. I barely got through all of the text in the original post, so it’s possible that this has already been posted by someone else.
I understand his sample size was too small to work with to have a study on specifically stable same-sex families. As long as that is clearly stated that is fine.
However his study goes on to compare it to stable heterosexual families. It would seem to me a better comparison would be to heterosexual families that have the same proportion of instability as the homosexual families he’s faced. Then he would be able to make a definitive statement about whether sexuality plays into raising of children in unstable families (which would be difficult in either case, but who knows, may be more difficult with parents who have had same-sex relationships).
It’s simply not very good science.
Leave A Comment