Reader Comments

Post a new comment on this article

Methodology Question

Posted by jturban on 16 Aug 2018 at 18:49 GMT

"A 90-question survey, consisting of multiple-choice, Likert-type and open-ended questions, was placed on three websites where parents had reported rapid onsets of gender dysphoria."

What were these websites?

No competing interests declared.

RE: Methodology Question

llittman replied to jturban on 16 Aug 2018 at 23:29 GMT

Thank you for your question. The three sites are listed in the materials and methods section (under “procedure”) and are: 4thwavenow, transgender trend, and youthtranscriticalprofessionals. In the discussion section of the paper, you will find more explanation of targeted recruitment for first descriptive studies, the websites that I used for initial recruitment, and the use of snowball sampling technique in this study to widen the reach of potentially eligible respondents.

The following passage from the discussion section provides the relevant context about parents viewing the websites used for recruitment.

“The websites that were used for recruitment are sites which specifically offer to support parents worried about their transgender-identifying children and the population viewing these websites may be different from populations viewing websites that promote a “gender-affirming” perspective and both populations may differ from a broader general population in their attitudes about transgender-identified individuals.

It would be most accurate to characterize the differences between parents viewing the websites used for recruitment and parents viewing websites that promote a gender-affirmative perspective as parents with a difference of opinion about how best to evaluate and treat gender dysphoric children and adolescents, with the former favoring judicious use of transition and the latter favoring a liberal use of transition. However, some may argue that the parents recruited from the websites used might be more oppositional to transgender-identified individuals in general. To address this potential concern, respondents were asked specifically whether they believe that transgender people deserve the same rights and protections as others and 88.2% of respondents gave affirmative answers to the question which is consistent with the 89% affirmative response reported in a US national poll [63]. All self-reported results have the potential limitation of social desirability bias. However, comparing this self-report sample to the national self-report samples [63], the results show similar rates of support. Therefore, there is no evidence that the study sample is appreciably different in their support of the rights of transgender people than the general American population. It is also important to note that recruitment was not limited to the websites where the information about the study was first posted. Snowball sampling was also used so that any person viewing the recruitment information was encouraged to share the information with any person or community where they thought there could be potentially eligible participants, thus substantially widening the reach of potential respondents. In follow up studies on this topic, an even wider variety of recruitment sources should be attempted.”

And the following passage describes the use of targeted recruitment:

“It is not uncommon for first, descriptive studies, especially when studying a population or phenomenon where the prevalence is unknown, to use targeted recruiting. To maximize the possibility of finding cases meeting eligibility criteria, recruitment is directed towards communities that are likely to have eligible participants. For example, in the first descriptive study about children who had been socially transitioned, the authors recruited potential subjects from gender expansive camps and gender conferences where parents who supported social transition for young children might be present and the authors did not seek out communities where parents might be less inclined to find social transition for young children appropriate [62]. In the same way, for the current study, recruitment was targeted primarily to sites where parents had described the phenomenon of a rapid onset of gender dysphoria because those might be communities where such cases could be found. The generalizability of the study must be carefully delineated based on the recruitment methods, and, like all first descriptive studies, additional studies will be needed to replicate the findings.”

I hope that this explanation was helpful. Thank you for your interest.

No competing interests declared.

RE: RE: Methodology Question

glein replied to llittman on 25 Aug 2018 at 19:37 GMT

Targeted recruitment seems almost necessary, however I wonder what impacts the specific dynamics of the chosen sites had on the results.

From transgendertrend.com's home page:
"This site is for everyone who is concerned about the social and medical ‘transition’ of children, the introduction of ‘gender identity’ teaching into schools and new policies and legislation based on subjective ideas of ‘gender’ rather than the biological reality of sex."
...
"It is for everyone who questions the medicalisation of childhood feelings, the invasive and life-changing treatment of minors and the aggressive tactics of transgender and LGBT organisations to promote and normalise this medical experiment."

From 4thWaveNow's About page:
"After much research and fruitless searching for an alternative online viewpoint, [the site founder] began writing about her deepening skepticism of the ever-accelerating medical and media fascination with the phenomenon of “transgender children.”"

youthtranscriticalprofessionals.org is a private wordpress site and my only insight into it is a posting on transgendertrend.com: https://www.transgendertr....

These sites are catered not just to a particular "conditional" demographic (parents with children who identified as transgender during adolescence) but to a particular mindset demographic as well (those parents who disbelieve the validity of their child's transgender claims and believe they are brought on by external societal dynamics). The survey result that only 2.4% of parents believing their child's assertion is correct supports this distinction and suggests result aggregation grouped by response to that question would be a valuable insight.

Questions 37 and 38 of the survey quantify (approximately) the number of friends in the group to identify as transgender, while 39 simply asks whether a "majority" of the friend group identified as transgender. This question seem particularly susceptible to the availability heuristic, since instances of a child's friends coming out as transgender are already going to be front and center in the parent's mind. Question 35's addendum partially accounts for this, however limits itself to only a single group rather than the child's entire (active) social network, which may be quite larger, and may not account for dynamics within that group itself. Specific quantifiable questions like "how many members were in the group when the first friend came out?" and "how many members are in the group now?", along with questions targeting the child's superset of friend groups might suss that out a bit more (majority indicator could then be calculated).

No competing interests declared.

RE: RE: RE: Methodology Question

llittman replied to glein on 27 Aug 2018 at 15:35 GMT

Thank you for your questions.I have placed >>> before your comments and answered in the paragraphs below.

>>>Targeted recruitment seems almost necessary, however I wonder what impacts the specific dynamics of the chosen sites had on the results.

Thank you for acknowledging the necessity of targeted recruitment.

>>>These sites are catered not just to a particular "conditional" demographic (parents with children who identified as transgender during adolescence) but to a particular mindset demographic as well (those parents who disbelieve the validity of their child's transgender claims and believe they are brought on by external societal dynamics). The survey result that only 2.4% of parents believing their child's assertion is correct supports this distinction and suggests result aggregation grouped by response to that question would be a valuable insight.

I agree that the study population is a specific population, and they may have unique characteristics. This is true also of Dr. Kristina Olson’s work on social transition of young children—parents who are attending gender conferences, having their kids attend gender expansive camps, and have allowed their children to socially transition before this was common—may also have unique characteristics. Both of these studies are first descriptive studies and both areas will greatly benefit from follow up studies with recruitment from a wider selection of sources.

>>>Questions 37 and 38 of the survey quantify (approximately) the number of friends in the group to identify as transgender, while 39 simply asks whether a "majority" of the friend group identified as transgender. This question seem particularly susceptible to the availability heuristic, since instances of a child's friends coming out as transgender are already going to be front and center in the parent's mind. Question 35's addendum partially accounts for this, however limits itself to only a single group rather than the child's entire (active) social network, which may be quite larger, and may not account for dynamics within that group itself. Specific quantifiable questions like "how many members were in the group when the first friend came out?" and "how many members are in the group now?", along with questions targeting the child's superset of friend groups might suss that out a bit more (majority indicator could then be calculated).

I agree with you that additional studies will be needed to better explore these specifics. You are correct that by asking about the number of friends coming out before and after the target child--does not provide a denominator. Four kids out of four is different than four kids out of 100. In order to get a sense of percentages, I asked the question about majority. And I hope that there are additional studies conducted to explore the timing of disclosures, the timing of friendship group formations, the sizes of the friendship groups at different times, movement of friends into and out of friendship groups, smaller social networks vs larger social networks, and even the spread of specific concepts over social media. My study is meant to be a first study so that additional studies can be conducted. There is much to learn in this area and I hope that other researchers will consider exploring this as well.

No competing interests declared.

RE: RE: Methodology Question

dhart922 replied to llittman on 30 Aug 2018 at 19:41 GMT

4thWaveNow should NEVER have been sourced. This site is comprised of parents who desperately want to "prove" that gender dysphoria is an expression of sexual orientation. On information and belief, it was started by someone with a religious objection. Many of the postings are from one person with multiple personalities which could be the case with completed docs.

Competing interests declared: LGBT Advocacy - I write the Slowly Boiled Frog

RE: RE: Methodology Question

dhart922 replied to llittman on 30 Aug 2018 at 19:41 GMT

This comment has been removed.

Competing interests declared: LGBT Advocacy - I write the Slowly Boiled Frog

RE: RE: RE: Methodology Question

Deleted_Account replied to dhart922 on 31 Aug 2018 at 00:11 GMT

Very true: I'm sure that everyone can agree that 4thwavenow when it tags all of its stories as such "https://4thwavenow.com/pa..." is probably going to have a lot of bias and politically bent views that might skew the polling data.

But still, the most important thing is that while this is claimed to be a "descriptive study", no case studies or any other form of assessment of the purportedly ROGD youth and young adults was ever carried out. We have no confirmation if ROGD really exists or we are just getting the poll answers you'd expect from a politically partisan website that tags everything with slurs. This is why all other studies that claim to have discovered a new disease assessed alleged cases at minimum. A political poll is no substitute and this is a shining example of why.

No competing interests declared.