The Evangelical Blackout of Sexual Orientation Research, Part 2

Last week, I commented on what I see as an evangelical blackout of sexual orientation research by Christian media and organizations. While I stand by that viewpoint, the situation is actually worse than a blackout. The blackout is selective; some new research is reported. However, the studies reported and the way they are reported seem designed to create a slanted picture.

A case in point. Currently, on the NARTH (National Association for the Research and Therapy of Homosexuality) website, scientific advisory board member, Chris Rosik, reviews a new report from Gartrell, Bos and Goldeberg about lesbian parenting recently published in the Archives of Sexual Behavior. The headline for the review is

New Study: Daughters of Lesbian Parents More Likely to Engage in Same-Sex Behavior and Identify as Bisexual

This is definitely a new study. The blackout is not total, but as I will demonstrate, it is selective. NARTH ignores the hard science involved in the brain scan studies but finds one aspect of a small longitudinal study of lesbian parenting to report. Now that you read the headline, read what Rosik says about how the study can be used.

While this small study is valuable as a starting point for longitudinal research into same-sex parenting, professionals and policy makers should be very wary of making any meaningful conclusions from its findings.  Serious methodological limitations also argue against making sweeping generalizations.  As is the case for the vast majority of studies in this area, the sample size is quite small, constituting only 78 adolescents.  The sample of lesbian parents is self-selected and appears to be different from the general population on important demographics such as socioeconomic status and educational attainment.  Demand characteristics (i.e., external influences such as political goals that might motivate study participants to respond in a particular manner) are not considered or assessed by the study’s authors with respect to the lesbian mothers or their adolescent children.

And then…

Certainly the Gatrell, et al. (2011) study provides some intriguing though entirely non-generalizable findings that are consistent with the hypothesis that non-heterosexual experiences and identities are more common among daughters of lesbian families than those raised in heterosexual families.

First, Rosik reports, via headline, the finding that would be of concern to religious conservatives but then in the article says one cannot make such generalizations. If one cannot generalize beyond the sample, then why report the finding as if one could?

The study also found that no children were abused in lesbian homes. This finding is in contrast to heterosexual families where abuse is reported (26% of teens report physical abuse by a parent or caregiver according to national surveys). Since NARTH is commonly represented in cases against same-sex parenting, and such information is relevant to their membership, why was that fact not a part of the headline?

Another interesting finding in the study was that boys were less likely to have been sexual involved with girls in lesbian families than in straight families. Isn’t that what abstinence educators want to promote?

My point here is that NARTH leaders do keep an eye out for new research, however, their reporting of them is selective. And then when they choose to review a study, their review is selective.

I have established that NARTH is a key source of information for Christian right organizations. When some relevant studies are ignored, and others are selectively reported, it seems clear to me evangelicals are poorly served by the organizations they count on for information.

  • David Blakeslee

    My point here is that NARTH leaders do keep an eye out for new research, however, their reporting of them is selective. And then when they choose to review a study, their review is selective.

    Headline should read?

    “Interesting results from a Small Sample of Lesbian parents: less violent and sexually diverse”

    It sounds like the body of the review has the appropriate temperance, but the headline screams: “Nicolosi-Byrd”

    • http://wthrockmorton.com Warren

      David – Your headline would be a much fairer one. Your analysis of the current headline seems right.

  • David Blakeslee

    oops…more sexually diverse

  • David Blakeslee

    …and I think your point in the previous post is excellent that the lay community and “cuts and pastes” this stuff to their talking points are poorly served.

    “You shall know the truth and the truth shall set you free”

  • StraightGrandmother

    What I think will be interesting in the research is to follow these young women into adulthood and see what happens with them. This very well might be the “I want to be a teacher like my mom” syndrome. But then when the child grows up and goes to college they change their mind and pursue a different field of study.

    It is also interesting that none of the boys are any more gay identifying than boys raised in a heterosexually headed family. I can’t put my finger on it exactly, but in here there is something about women’s sexual orientation being more fluid than men’s. Somehow that must apply to this research, doesn’t it?

    Of course the political counterpoint headline to NaARTH’s would be-


    New Study: Sons of Lesbian Parents No More Likely to Engage in Same-Sex Behavior and Identify as Bisexual than Sons Raised by Heterosexual Parents

    This should make the conservative bigots happy shouldn’t it? This would be a good study for them. They always seem like they are more worried about boys turning gay than girls turning lesbian. Hey the boys are fine, no problems there.

  • StraightGrandmother

    Warren or David B, I am working on a project and I need an answer to this question. Just as Rosik stated in this article,

    As is the case for the vast majority of studies in this area, the sample size is quite small, constituting only 78 adolescents. The sample of lesbian parents is self-selected and appears to be different from the general population on important demographics such as socioeconomic status and educational attainment.

    My question is- Is the vast amount of psychological research on other psychological topics

    NOT Small sample size

    NOT self reported

    Thank you.

  • David Blakeslee

    SG: as sample size increases and as the sample is based upon objective criteria (rather than self-report) the results are more reliable and typical of the general population.

    Therefore more likely to be true, generally.

  • StraightGrandmother

    David B, thank you very much. I figured that. Could you please re-red my question again and try to answer it. Many thanks.

  • Teresa

    Warren,

    I certainly agree with the thesis you’re pointing out here; but, I’m mystified as to calling this “a study”; something even being considered as noteworthy. Yes, it’s new; and, yes, NARTH would knee-jerk it, as usual; but, where’s any of the scientific method here?

    I need to go back and look at what credentials the authors of the study have. This sounds like a grad student, needing a topic for a paper, putting out an ad for volunteers for the ‘research’ … asking 10 questions … and, wa-la, here’s a Research Paper.

    I’m sorry to be so ornery about this; but, this is quite over-the-top in its Headline, as well as considering it something of value. Rosik points out huge flaws in the study; but, then calls it valuable as a starting point in this area. The only value I can see … is how not to do research. And, any group who latches onto this “as a study” should be known by the company it keeps.

  • David Blakeslee

    The vast amount of research is very jumbled and unable to be characterized in a typical manner.

    The study the Rosik quotes is typical of a number of studies, which are not very useful, except for encouraging other scientists to study the specific issues raised in a more scientific manner (objective measures rather than self-reports and larger sample sizes).

    There is a huge problem in using studies like Gartrell et.al. (and Rosik’s interpretation) to guide social policy…as they are not really scientific at all.

    Unfortunately, these kinds of studies are often tossed about in public policy debates as if they are settled science…and they are not.

    Women’s sexuality appears more fluid than men’s base upon a growing and consistent body of research.

    I don’t know what makes conservative bigots happy, or liberal bigots happy. :)

  • http://funfrotfacts.blogspot.com Throbert McGee

    Among the sons from lesbian families, 2.7% reported a bisexual identity and 5.4% indicated they were predominantly or exclusively homosexual.

    This point ought to be of particular interest to Nicolosi and his NARTH cronies, since 100% of the sons from lesbian households had “absent fathers”!

  • http://wthrockmorton.com Warren

    TMcGee – Thanks for pointing that out. I overlooked that. That might be worth a post in itself.

  • http://funfrotfacts.blogspot.com Throbert McGee

    Just following up on my previous comment: remember, Nicolosi’s claim is that even in two-parent heterosexual households where the father is always present, mere “aloofness” or “emotional detachment” on the dad’s part may be sufficient to turn the son gay.

    Yet, in this survey of boys who grew up in lesbian-headed families where there was never a masculine paternal authority figure living in the household at all, the reported rate of male homosexuality is about the same as the national average.

    ¡Muy misterioso!

  • Teresa

    @Throbert, much more mysterious is that 1/2 of America is being raised in single-headed households … majority without fathers. As you’ve reckoned, you don’t get much more distant than being without. Are 1/2 of the young men in America gay?

    I’ve tried bringing this up to parents involved with Encourage, the Catholic family/friends group, who seem not to want to grasp this fact. Why they would continue to latch onto being the ‘guilty’ party; and, beat themselves up over something so not true … I have no idea. It is, what it is.

  • Teresa

    Women’s sexuality appears more fluid than men’s base upon a growing and consistent body of research.

    David B., most often it is cited as going from gay to str8. The big elephant in the room in this area is the large number of women (large in comparison to a small segment) moving from str8 to gay relationships.

    I’d like to see a study on this, particularly. The number of older women, divorced or widowed, who are ending up in same-gender relationships, I think (no evidence) is far greater than the movement from gay to str8. Just my anecdotal evidence.

  • http://funfrotfacts.blogspot.com Throbert McGee

    @Teresa: I’ve also brought up the case of male “military brats”, who would seem to provide a good test demographic for Nicolosi’s idea that the “risk of male homosexuality” is increased by the absence of a father during a crucial window when a boy is under 3 years old.

    If Nicolosi’s hypothesis is correct, we should expect to see a spike of homosexuality rates among male “military brats” who were toddlers during the Korean, Vietnam, and First Gulf Wars — and perhaps also an elevated rate of male homosexuality among Navy brats, since Naval personnel often have extended deployments away from their families even during peacetime periods.

    However, I don’t imagine that “Men — join the US Navy and triple the odds that your sons will be homosexual!” would be a very popular slogan with NARTH’s cultural-conservative fans, or with the Pentagon…

  • Lynn David

    Throbert McGee……

    Among the sons from lesbian families, 2.7% reported a bisexual identity and 5.4% indicated they were predominantly or exclusively homosexual.

    This point ought to be of particular interest to Nicolosi and his NARTH cronies, since 100% of the sons from lesbian households had “absent fathers”!

    Nicolosi could use that to his advantage as the 5.4% represents perhaps double what the rate of homosexuality is among males.

    But then again it could be an indication that male homosexuality and female homosexuality are somehow genetically connected – if the children of the lesbian parents were indeed born to them. Or did they use gay men for sperm donors, indicating again some genetic connection? Lots of possibilities to consider.

  • StraightGrandmother

    I asked this question-

    My question is- Is the vast amount of psychological research on other psychological topics

    NOT Small sample size

    NOT self reported

    Let me ask it a different way.

    Please remove the research and studies on sexual minorities from your mind. Think of all the other research and studies (not about sexual minorities) does that research use small sample sizes and are NOT self reported. Maybe things like drug and alcohol addiction for example.

    Is small sample sizes and self reporting quite unique compared to all other areas of the study of psychological issues?

    Thx!

  • Lynn David

    Ok… now I see your next post, so nevermind….

  • http://aebrain.blogspot.com Zoe Brain

    I don’t care who uses what to which advantage. The point is, is it good science, and if it is, is it just one of many data points that only when analysed together build a picture we can have confidence in, or does it stand alone?

    Sample size <500 means I treat it as a data point. 20 such studies,even if differing in methodology, all with samples of ~30 and saying the same kind of thing, I'd find pursuasive.

    When dealing with biological studies of gender identity, sample sizes are often around 20. There just aren't that many brains of cadavers of provably transsexual people that we are able to disect. It's only MRI and PET scans that we have reasonable sample sizes in the hundreds.

  • Pingback: New study: Lesbian parents not associated with homosexual behavior in sons — Warren Throckmorton

  • David Blakeslee

    SG,

    Trying to answer that question:

    The vast amount of research is very jumbled and unable to be characterized in a typical manner.

    Teresa is right about the study, it is low grade “survey” research. This is common in all fields and interests in psychology. It is the easiest to do. As I said earlier, is it is often the prelude to a more scientific study as it identifies issues which are worth studying more scientifically.

    Research on sexual minorities is more prone to this kind of study (Warren and others may differ on this and I welcome their perspective).

    This is because they are already a very small group of people (so large sample sizes are much harder to get).

    This is because many prefer anonymity due to social stigma.

    It is easier to study groups which are larger and less fearful of social stigma, therefore it is easier to construct more scientific studies.

    Using surveys is quite common in all research, because it is the easiest way to gather information…and it is the least accurate and reliable.

  • David Blakeslee

    SG,

    Also, I am thinking, this is a longitudinal (life span) study. Which makes it more scientifically interesting than a simple survey which takes a “snapshot” (one time) assessment of the children in these homes.

    So, in that regard it is more unusual and more meaningful than a simple survey.

    Good scientific longitudinal studies collect data from a large sample of participants who are interviewed or assessed using objective criteria and their results are compared to a matched sample of participants who differ from them in a core characteristic (sexual identity, age, intelligence, religion, ethnicity or Socioeconomic class).

  • StraightGrandmother

    David B am I seeing a contradiction? I am making an honest attempt to figure this out.

    You said this

    This is common in all fields and interests in psychology. It is the easiest to do. As I said earlier, is it is often the prelude to a more scientific study as it identifies issues which are worth studying more scientifically.

    But then you said this

    Research on sexual minorities is more prone to this kind of study (Warren and others may differ on this and I welcome their perspective).

    So which way is it? Here please help me craft an appropriate statement to use at the end of a power point presentation. Say I am giving a presentation of the Jones & Yarhouse study on religiously based sexual orientation change efforts.

    At the end, on the last slide can I say

    This study it just one data point of many other data points from other studies. Only when analyzed together, will they reveal a scientific fact that we can have confidence in. Studies with sample size of less than 500 is a data point that gets added to the heap.

    Good, Better Best.

    When it comes to reliability of psychological studies and research

    Good = an example

    Contacting organizations who sponsor support groups for drug addiction and being put in touch with members of a support group. Then surveying the individuals who attend these support meetings. (Note people who have dropped out are not included in the survey)

    Better = Contacting psychologists and asking them to to answer questions relating to their patients who are or were, under their care for drug addiction. Contacting those clients of the psychologist and surveying them. These clients would b both those who succeeded in their addiction to drugs and those who failed. Less than 500 people are studied.

    Best = Finding a random sample of over 500 people the general population usually via phone surveys. The study members are interviewed by professionally trained psychologists.

    I have a reason for asking this. I am working on a project that will be talking with an audience without a college degree. Therefore I need short sentences and easy to understand concepts and terms.

    Now i completely made up the Good Better Best Examples. Can anyone here kindly edit these to what is correct? Please. I would do it myself but i don’t know what is correct.

    I would use this same statement regardless of the research. In other words if there is a study that says sexual orientation change does not work, they get the same statement. A study that says sexual orientation change does work, they get the same statement. I want to present unbiased facts. And to correctly show people how they should rate this study and how this study fits in the overall research on the topic. I am a layman but I am trying to do a good job, a correct job.

  • David Blakeslee

    SG,

    Wow…good idea.

    As you have noted on occasion sometimes my words are too many or not common usage.

    Talking to lay friend today, used “cognitive mediation” for “thinking.” :).

    I will look at above and try to be concise, helpful and easy to understand.

  • StraightGrandmother

    Thank you so much David B.

  • David Blakeslee

    SG,

    So which way is it?

    It is both: common in the field generally and more common in work with sexual minorities for the reasons cited above: already a very small group in the population (2-4%) that often seeks anonymity due to social stigma.

    If you are citing Yarhouse and Jones, that group of folks is even smaller (double minorities SSA and not wanting to identify as such) and even more likely to seek anonymity due to social stigma.

    This study it just one data point of many other data points from other studies. Only when analyzed together, will they reveal a scientific fact that we can have confidence in. Studies with sample size of less than 500 is a data point that gets added to the heap.

    Good, Better Best.

    When it comes to reliability and accuracy of psychological studies and research

    Good = an example

    Contacting organizations who sponsor support groups for drug addiction and being put in touch with members of a support group. Then surveying the volunteering individuals who attend these support meetings. (Note people who have dropped out are not included in the survey). The study would not include people who dropped out of treatment or those who did not wish to participate in the study. Some surveys (surveys are often sent home to complete) might be completed by friends and family rather than the alcoholic; contaminating the data. Much important data will be unable to be gathered in this study, or it will be gathered from the wrong person.

    Better = Contacting psychologists and asking them to to answer questions relating to their patients who are or were, under their care for drug addiction. Contacting those clients of the psychologist and surveying them. These clients would b both those who succeeded in their addiction to drugs and those who failed. Less than 500 people are studied.Contacting several inpatient rehabilitation programs throughout the country who agree to assess their clients at intake, upon discharge and at 6 month follow-up using Clinical Interviews performed by Drug and Alcohol Counselors. (Every participant is included in the study, even drop outs and treatment failures, collecting more data and collecting using a trained Clinician means that the information comes directly from the participant). Problems, this research only interviews those who are seeking care and can afford care; it cannot control for demographic distortions in sample collection (if the overwhelming majority of participants are white, rich men; the results cannot be generalized confidently to other groups like women and minorities)

    Best: See Better but add that the study takes all available participants and randomly assigns them to the various demographic groups that are typically present in the general population. In this regard the participants are representative the the overall group we are trying to study (not a subset as above). Administer the same clinical interview over the same period of time. Results from this study are much more likely to be able to accurately describe the general population of alcoholics who seek treatment (although it’s flaw is that it does not account for alcoholics who do not seek treatment in these facilities).

  • StraightGrandmother

    Thank you so much David B!!! You came through for me :) Sweet!

    Now I will try and distill this down further to keep your points but make it shorter. If it is to long people won’t read it. I’ll be back later so please check back.

  • StraightGrandmother

    Just a minute. I do not understand this at all

    Best = See Better but add that the study takes all available participants and randomly assigns them to the various demographic groups that are typically present in the general population. In this regard the participants are representative the the overall group we are trying to study (not a subset as above)

  • David Blakeslee

    When you match study participants to the general population in terms of demographic characteristics, then the results of the study are more easily described as “typical” of the general population.

    In the previous studies, they are “subsets” of the general population (volunteers, for example in the Good group; or disproportionately “White” in the Better group–my example).

    All participants are assigned to the study randomly (for example if the national demographic is 60% whites to 20% Mexican Americans to 10% African American to 10% other; and 53% women to 47% men; then the participants in the study must fit this demographic with participants in each demographic group randomly assigned to that group).

  • StraightGrandmother

    Thank you David B. Keep checking back here if you would be so kind. I’ll have more on this that I will hope you will review.

  • StraightGrandmother

    Warren back to the topic. They just do NOT want to know. All they want to do is write anti gay stories. Here is the Top 10 Anti gay stories from World Net Daily

    http://www.wnd.com/?pageId=380149

  • Pingback: Why do so many people believe being gay is a lifestyle choice, when it's already been proven to NOT be? - Page 15 - City-Data Forum

  • Pingback: The Scandal of the Evangelical Double Mind


CLOSE | X

HIDE | X