Perceived Efficacy and Risk of Sexual Orientation Change Efforts (SOCE): Evidence from a US Sample of 125 Male Participants

For those who may be interested in more detail on the F1000 retraction that preceded Researchers.One publication, here below is a copy, in pertinent part, of our correspondence with F1000. We believe this was an ideological retraction, the reasons for it were pretextual, and the journal did not engage with us in good faith. I am reporting the correspondence as it happened, without comment; the reader can form his or her own conclusions.

On 13 June 2024, the authors of this article received the following message from the F1000 Editorial Team:

Sullins DP, Rosik CH and Santero P. Efficacy and risk of sexual orientation change efforts: a retrospective analysis of 125 exposed men [version 2; peer review: 2 approved]. F1000Research 2021, 10:222 (https://doi.org/10.12688/f1000research.51209.2).

After the publication of the above article, concerns were raised to the F1000 Editorial Team regarding the data underlying the research presented. As per our policies, we investigated the concerns. Further, we engaged an external reviewer with subject matter expertise to conduct an independent review.

The conclusion of this review and investigation is that the underlying data presented in the above article is the same data used in the article: 'Effects of Therapy on Religious Men Who Have Unwanted Same-Sex Attraction (https://doi.org/10.1177/0024363918788559), written by some of the same authors. This article was retracted by Linacre Quarterly (SAGE Publishing) on 23rd May 2019 due to statistical and methodological concerns. You did not disclose the use of the same data upon submission. Further, both the F1000 Editorial Team and the external reviewer were unable to confirm the integrity of the study design and the article’s conclusions relating to this dataset.

Therefore, in accordance with our editorial policies and to ensure the integrity of the scholarly record, we intend to retract the above article. The reason for the retraction is stated within the retraction notice. The copy we intend to publish is attached. If you would like to provide any further evidence to address our concerns, kindly provide this by 27th June 202

I responded the same day as follows:

We are preparing a response to your decision, which we will send you in a few days. To help us respond more intelligently, I respectfully present a few questions.

1) You state "both the F1000 Editorial Team and the external reviewer were unable to confirm the integrity of the study design and the article’s conclusions relating to this dataset." It is hard to respond to such a vague statement. May I ask what particular features of the study design, and what particular conclusions, you were unable to confirm the integrity of?

2. May we see the report of the external reviewer? If not, why not?

3. Have the original peer reviewers who approved the article for publication been consulted on the decision to retract the study? Have they even been informed of this decision, which contradicts their professional judgment? If not, why not?

As you are aware, the data are publicly available, with the original questionnaire, so any scholar will be able to confirm for themselves the integrity and value of the data. I would be happy to provide for perusal any data structures or statistical models used in our analysis. They are straightforward analyses that are not complicated for any competent data analyst to understand.

We did not receive a response to our questions. On 18 June 2024, we sent our response:

Your stated concern regarding data disclosure is unwarranted, even nonsensical, in our judgement. The data were collected for a dissertation published in 2011, which we have fully disclosed, and included the dissertation's author as a co-author on our paper. How can the use of those data in an unrelated study seven years later (2018), which was subsequently retracted for reasons unrelated to data quality (which you acknowledge), have any bearing whatsoever on our subsequent use of the original data in our study?

To our knowledge, this concern has only been expressed in a single online reader comment. As you know, you have refused to publish our response to this comment, despite several revisions.

To retract an article because it lacks a disclosure of unrelated prior data usage, if such a concern were warranted, is nonetheless overwrought and unnecessary. It is actually trivially easy to rectify. To that end, we have submitted a revised version of our study which adds the following sentence:

“An earlier analysis of the same data (Santero, Whitehead,and Ballesteros 2018) was retracted for reasons unrelated to data quality (Retraction Notice, 2020), followed by a protest from one of the co-authors (Whitehead 2019).”

With the following explanatory note:

“This revision adds a sentence to the end of the first paragraph of the Methods section, which now reads:

"This study presents a secondary analysis of an online survey previously administered to a convenience sample of adults who had undergone therapeutic intervention to alleviate unwanted same-sex attraction. The survey was administered pursuant to the Doctor of Psychology dissertation of Paul Santero (now Psy.D.) at Southern California Seminary (Santero, 2011), which contains a more complete description of the survey methods, administration, and question wording. An earlier analysis of the same data (Santero, Whitehead,and Ballesteros 2018) was retracted for reasons unrelated to data quality (Retraction Notice, 2020), followed by a protest from one of the co-authors (Whitehead 2019)."

This previous retracted secondary use, which is a matter of public record, is superfluous to our study, and can have no reasonable bearing on the quality of the data, which had been collected seven years earlier. However, some critics felt that it should be explicitly mentioned, so we have done so.”

We invite you, or any critics, to use the resources of your own journal to state any particular concerns or critiques you have, to which we may respond, in open, constructive scientific debate. The secretive process in which you form opinions and judgements in private, which results in a vague general denunciation of our study after no communication with the authors or the evidence, is not a process that will lead to better understanding or scientific truth. It is also not a process that is consistent with COPE guidelines or F1000's own stated goals. We think you can do much better, and hereby offer you an opportunity to do so.

Yours truly, with best regards,

Paul Sullins, with Christopher Rosik and Paul Santero

On 3 October 2024 we received the following response from F1000:

Following your rebuttal in June, we engaged a further external reviewer with subject matter expertise to conduct a further independent review. A summary of their comments are below:

“I took a look at the paper, and can conclude that the data and methods are not robust enough to support their conclusions. The most significant problem I see is that the authors use words like “efficacy” and “effectiveness” to describe their results, neither of which can be tested with their methods, which are cross-sectional (single time point) methods. There is no randomization here (as required for efficacy studies, and most effectiveness studies in order to draw conclusions about efficacy and effectiveness), and all assessments were done at a single time point. They do a poor job of describing methods and refer to an unpublished dissertation which does not appear to be available via Proquest, but the methods of single time point data collection and retrospective scoring of past experience are within the publication. The methodological implications of cross-sectional data are that they can’t make any claims about efficacy or effectiveness, or change over time (all change measured at the same time).

Unfortunately, the statistics run are only as good as the data and methods that led to them. Given the problem with the methods here, it is not possible that the statistics could be solid.

Perhaps this is part of the ethical complaint, but discussing same sex behavior and/or orientation as something that goes into remission is also problematic.

My primary concerns are methodological-- that their methods do not support the words they use—words like efficacy and effectiveness. There are many more problems with the article, but the efficacy and effectiveness ones are so central.”

Therefore, in accordance with our editorial policies and to ensure the integrity of the scholarly record, our intention to retract the above article remains.

On 24 November 2024 I responded on behalf of the authors:

This concern can easily be addressed by simple wording change: revise the use of the words "efficacy" and "effectiveness" to "perceived efficacy" or "perceived helpfulness."

If (as your reviewer states) "the most significant problem ... is that the authors use words like "efficacy" and "effectiveness" to describe their results," we can easily change our language to conform to the results we did find.

I respectfully note that, other than this language issue, your reviewer did not note a single specific error or falsehood in any of the analysis or findings of our paper.

We have made no claim that this study is a random controlled trial, but only a report of retrospective self-reports. We state this limitation clearly: "Our findings cannot be definitive regarding any assertion that sexual orientation can change, only that some highly religious men report such changes, the pursuit of which they generally do not perceive to have been harmful."

If your reviewer were indeed a subject matter expert, they would know that this is the current standard of evidence in this field, and characterizes most comparison studies to ours. Here is a partial list.

Blosnich, J. R., Henderson, E. R., Coulter, R. W., Goldbach, J. T., & Meyer, I. H. (2020). Sexual orientation change efforts, adverse childhood experiences, and suicide ideation and attempt among sexual minority adults, United States, 2016–2018. American Journal of Public Health, 110, 1024–1030. https://doi.org/10.2105/AJPH.2020.305637

Ryan, C., Toomey, R. B., Diaz, R. M., & Russell, S. T. (2020). Parent-initiated sexual orientation change efforts with LGBT adolescents: Implications for young adult mental health and adjustment. Journal of Homosexuality, 67(2), 159–173.

Salway, T., Ferlatte, O., Gesink, D., & Lachowsky, N. J. (2020). Prevalence of exposure to sexual orientation change efforts and associated sociodemographic characteristics and psychosocial health outcomes among Canadian sexual minority men. Canadian Journal of Psychiatry, 65(7), 502–509.

Bradshaw, K., Dehlin, J. P., Crowell, K. A., Galliher, R. V., & Bradshaw, W. S. (2015). Sexual orientation change efforts through psychotherapy for LGBQ individuals affiliated with the Church of Jesus Christ of Latter-day Saints. Journal of Sex & Marital Therapy, 41(4), 391–412.

Dehlin, J. P., Galliher, R. V., Bradshaw, W. S., & Crowell, K. A. (2015). Navigating sexual and religious identity conflict: A Mormon perspective. Identity, 15(1), 1–22. https://doi.org/10.1080/15283488.2014.989440

76

Dehlin, J. P., Galliher, R. V., Bradshaw, W. S., Hyde, D. C., & Crowell, K. A. (2015). Sexual orientation change efforts among current or former LDS church members. Journal of Counseling Psychology, 62(2), 95–105. https://doi.org/10.1037/cou0000011

Karten, E. Y., & Wade, J. C. (2010). Sexual Orientation Change Efforts in Men: A Client Perspective. Journal of Men’s Studies, 18(1), 84–102. https://doi.org/10.3149/jms.1801.84

All of the above seven studies in this field use exactly the same method as our study, that is, retrospective cross-sectional self-reports, and discuss the reported helpfulness, effectiveness, or efficacy, or not, of the therapeutic experience reported. Why is this standard of evidence not also acceptable for our study?

Attached to this response (in both Word and pdf) is a revision of our study that fully addresses the language issue raised by your reviewer. [Note: This is the version now posted at Researcher. One] We have clarified that all outcome references are to perceived outcomes, not to clinically demonstrated outcomes. Just to be sure, we have removed all instances of the words "efficacy," "effectiveness," and "remission." These changes in sum are minor and do not affect any of the reported results in the paper.

Your journal claims to feature the policy to revise and amend articles in response to critique after initial publication. Consistent with that policy and in that spirit, will you please consider the attached revision, which easily fully addresses the expressed concerns identified by your reviewer (and the published critique)? If you are not willing to consider this minor revision to address these concerns, can you please explain why?

Thank you for your consideration.

With best regards,

Paul Sullins

On 23 January 2025 F1000 responded briefly, without addressing the questions raised in our rebuttal, to confirm that their decision to retract, for the reasons vaguely noted in the retraction notice, was final.

As one of the original reviewers of the manuscript titled “Perceived Efficacy and Risk of Sexual Orientation Change Efforts (SOCE): Evidence from a US Sample of 125 Male Participants” submitted by Sullins and Rosik to F1000, and as is clearly evident from the F1000 webpage, I am disturbed to discover the unprofessional, unethical, unacademic efforts of the journal to retract/cancel this article based on the comments of a single academic without providing 1. Any notice to the reviewers and 2. Without allowing the authors to respond. Science does not answer to professional organizations, political entities or agendas. Our obligation as scientists is solely to investigate the realities that exist around us – in nature and in human behavior, with the sole goal of seeking to improve human well-being. No journal, no academic, no editor, no reviewer and no politician has any right to block the publication of a legitimate study/investigation, most especially when the standards applied to the “cancelation” of articles is one-sided. That is not academic, and it is a violation of the various codes of ethics that seek to ensure that research conducted does not harm but rather benefits and informs the public. It violates the efforts to ensure the respect of the human person (see the Nuremberg Code, Belmont Report), and is a violation of the autonomy that the public deserves and is entitled to in making personal life decisions that relate to science and medicine. Below, I include my modified review that I submitted to F1000 – modified only to add comments pertaining to the corrections that the authors made in response to my comments. I remain in strong support of the publication of this article in order to address the gap in knowledge that the APA itself admits exists.

----

In this manuscript the authors address the topic of voluntary therapeutic interventions to reduce same-sex sexuality, addressing an important and significant gap in research pertaining to a therapy that has been insufficiently researched (APA, 2009), yet prohibited in various states. They report on the findings from a convenience sample of 125 men who had participated in SOCE (Sexual Orientation Change Efforts). The authors seek to examine two questions, investigating whether participation in SOCE was perceived by the consumers “to be helpful in alleviating unwanted same-sex attraction, identification and behavior” and secondly to address the degree that SOCE exposure was “perceived to be psychologically harmful or beneficial”.

The authors conclude that their findings indicate a significant positive net change “for all problem domains” (i.e. the study considered both positive and negative change) following exposure to SOCE. This included a reduction in same sex attraction, identification and sexual activity, with a significant portion of the participants (higher rates in married participants) achieving some level of remission in unwanted same-sex sexuality.

The Introduction clearly addresses the background and knowledge gaps, and clearly highlights the aims of the study. In the Methods the authors likewise clearly outline the measures, subjects and how they were reached, the exclusion criteria, ethical approval of the study, the analyses and reasoning behind the use of specific analysis or combination of responses. The Results are also well written and clear with appropriate reference to the tables and data contained therein. However, the images in Figure 1 and 2 appear to be switched, with Figure 1 representing the post-SOCE and Figure 2 the Pre-SOCE data, in order to correspond with what is reported in the text [this was corrected in the revision]. Given that these graphs represent the same measure Pre- and Post-, I would recommend that the authors combine the graphs into one figure as Figure 1 A and B [authors chose to keep the graphs separate. However, this was not a detrimental decision to the integrity of the information being presented]. Finally, the Discussion is extensive and balanced without any strong and unfounded assertions. It provides a careful interpretation of the results in the context of the current literature, integrating both supporting and opposing literature/findings. Furthermore, it addresses the potential implications of the findings while recognizing and clearly outlining the limitations, necessity for caution in the interpretation of the findings and their application in the clinic and addressing the necessity for further research.

Overall, the study highlights the urgent necessity for further research that avoids the ideological and political polarization surrounding SOCE. Consensus at the therapeutic level (not political or ideological) in the case of SOCE, as per any other therapy, and for the sake of the client, can and should only be reached with continued objective research, in order to provide an evidence-based assessment of the risk/benefits that ultimately should realistically inform professional policies for treatment. To this end, as indicated earlier, significant gaps in knowledge remain. This study provides an important and significant contribution to such an effort to address the existing knowledge-gap.

On 19 July 2023 the journal F1000Research posted a lengthy online critique of the original draft of our study by Dr. Lorenzo Lorenzo-Luaces, which is available at https://f1000research.com/articles/10-222#article-comments . On or about 25 July 2023 we submitted a response, and subsequently submitted the slightly revised version of the paper presented above. As of 19 September 2024, the journal has refused to publish either our response or the revised version. We are pleased to have finally been able to publish the revised version in this forum, and we now publish our response to Dr. Lorenzo Lorenzo-Luaces’s comment:

Almost every study of SOCE has selected a sample of currently-identified LGB persons, from whose reports it finds that SOCE is ineffective and may be psychologically harmful. Our study, which examined a sample of 125 male SOCE alumni without regard to current sexual orientation, found that for many SOCE reportedly was effective and psychologically beneficial. We argue that the difference may be due to an artifact of sampling, in that samples that select for currently-identified LGB persons thereby artifactually exclude any currently-identified heterosexual SOCE alumni, that is, those for whom SOCE was, by definition, most successful. This is akin to evaluating marriage counselling by sampling only persons who subsequently divorced, or a risky life-saving surgery by examining only mortalities, excluding any survivors.

The comment by Dr. Lorenzo Lorenzo-Luaces’ (hereafter LL) seems not to comprehend this point, as well as to misread, misstate or simply miss most of the points it purports to discuss. Perhaps the best way to refute it is simply to actually read our paper. Such a reading would find no “assertion by the authors that they consist of an ideologically diverse team”, as the comment falsely claims exists, and an extensive discussion of the possible effects of high religiosity on the outcomes we find compared to those of other studies, as the comment falsely claims does not exist.

Our conclusion alone (p.14) refutes the latter misstatement, as well as show the citations for the religious differences that the comment claims we do not provide: “Fully 88% of our participants reported attendance at religious services weekly or more often, and only 2.4% reported attending rarely or never. By contrast, in a recent population sample of LGB-identified sexual minorities only 9% reported at least weekly religious service attendance and 69% reported attending seldom or never (Meyer, 2020, p. 324). The former are far more religiously active and the latter far less religiously active than are Americans in general, of whom 33% reported attending religious services at least weekly and 31% seldom or never in 2016 (Pew Research Center, 2019). It is possible that the prospect of change or stability in sexual orientation is linked to the notably high religiousness of the change-oriented sample and the notably low religiousness of the LGB-identified sample. Future research that incorporates both populations could help to clarify this possibility.” (p. 14)

It is quite right to be concerned about possible researcher bias on our part in this very polarized area of research. LL characterized our inferences regarding the possible effect of religiousness on the SOCE outcomes reported by our sample as invalid projections (“big leaps”) of what he perceives to be our own religious bias, stating that “unless I am missing something extremely obvious, the authors report no analysis linking religiosity to the success of SOCEs.”

We think we made it obvious that the men of our sample were extremely religious. We stated this most fully in the paragraph on page 6 that begins: "Regardless of affiliation, the members of the sample indicated a very high level of religious observance," and ends "These characteristics correspond to a group that strictly observes religious norms regarding worship, marriage and fertility within marriage." In between, we reported that "(88%) reported attending religious services at least once a week, a proportion at least four times higher than the national average," that very few were divorced, and that almost all the married men had children and none of the unmarried men did. We also presented the data showing their high religious attendance and affiliation in Table 1 on page 4. Elsewhere we referred to the study participants as a "sample of highly religious men," (p. 7) and noted under "Limitations" that "our findings should not be generalized, especially in reference to sexual minority women and those who are not highly religious."

Furthermore, our conclusion stated: “Fully 88% of our participants reported attendance at religious services weekly or more often, and only 2.4% reported attending rarely or never. By contrast, in a recent population sample of LGB-identified sexual minorities only 9% reported at least weekly religious service attendance and 69% reported attending seldom or never (Meyer, 2020, p. 324). The former are far more religiously active and the latter far less religiously active than are Americans in general, of whom 33% reported attending religious services at least weekly and 31% seldom or never in 2016 (Pew Research Center, 2019). It is possible that the prospect of change or stability in sexual orientation is linked to the notably high religiousness of the change-oriented sample and the notably low religiousness of the LGB-identified sample. Future research that incorporates both populations could help to clarify this possibility.” (p. 14)

There is even more consideration of religiosity in our paper, but we think that the above material in our study about the possible relationship of religiosity to the sample outcomes is enough to demonstrate that LL’s statement that “the authors report no analysis linking religiosity to the success of SOCEs” is emphatically false.

LL does offer some helpful advice about a citation and language which we have tried to accommodate in the current revision. He advises that “the fact that the data were used in a prior publication [that was retracted] is worth noting in the service of transparency.” We do not agree that declining to cite a retracted study constitutes some sort of nondisclosure, and note that our study already provides enough information that other scholars have not had any trouble noting this history (Schumm and Crawford 2023). However, we have no objection to including more detail on this point for those who may find it helpful. In response to LL’s comment have added the following sentence to the Methods section: “An earlier analysis of the same data (Santero, Whitehead,and Ballesteros 2018) was retracted for reasons unrelated to data quality (Retraction Notice, 2020), followed by a protest from one of the co-authors (Whitehead 2019).”

LL also objects to our use of the term “remission” for the reduction of same-sex orientation elements as possibly connoting “that the behavior in question is pathological.” We note that the Oxford English Dictionary lists “lessening the severity of a disease” as the fifth most likely meaning of the word “remission”, following forgiveness of sins, release from debt and “diminution of effect or force”, which is the sense in which we used it (Oxford English Dictionary, 2023). We did not intend by this usage to suggest, nor do we think it connotes on an objective reading, that homosexual attractions were a psychological problem for the men in our sample. Notwithstanding this, we appreciate LL's candid remarks on how he interpreted the word, and for the sake of readers like him have revised to use language which may potentially be less misunderstood.

The concern of LL’s comment regarding the “assertion by the authors that they consist of an ideologically diverse team” falsely attributes to us an assertion we never made. While our paper advocates for such research teams, precisely in order to overcome the ideological polarization present in this area of research, we (regrettably) made no claim to represent such a team ourselves. The last three paragraphs of LL’s comment are premised on this misunderstanding on his part and stray, in our opinion, into pejorative ad hominem critique that is not pertinent to the argument of our paper, to which we accordingly decline to respond.

Finally, we do not intend by our response to suggest that concern about researcher bias regarding our study is misplaced. On the contrary, we agree that LL has raised an important and legitimate concern. Both religious and irreligious secular scholars merit suspicion of possible bias in many areas of research today, and particularly regarding contested topics such as SOCE. An excellent treatment of this issue can be found in the recently published article by Schumm and Crawford titled “Difficulties With Methodology in Social Science Research With Controversial Issues Regarding Human Sexuality” (Schumm & Crawford, 2023). These University of Kansas researchers compared five studies of human sexuality that may be suspected of liberal confirmation bias with three, including our study, that may be suspected of conservative confirmation bias. They raise a number of helpful questions about our study, mostly regarding detailed scales and interactions which we were not able to address, but which they suggest could be addressed in further research. 1 Overall, Schumm and Crawford (2023) suggest that “[s]cientists must be open to seeing their best “sacred cow” ideas or theories disproven or modified with increases in research on such issues” (p. 194). In that spirit—using the same data that we analyzed, which was published along with the study—we commend the exploration of such questions by anyone, of any theoretical or ideological perspective—including LL—whether the results confirm or rebut our findings.

Blosnich, J. R., Henderson, E. R., Coulter, R. W., Goldbach, J. T., & Meyer, I. H. (2020). Sexual orientation change efforts, adverse childhood experiences, and suicide ideation and attempt among sexual minority adults, United States, 2016–2018. American Journal of Public Health, 110, 1024–1030. https://doi.org/10.2105/AJPH.2020.305637

Meyer, I. H. (2020). Generations: A study of the life and health of LGB people in a changing society, United States, 2016-2019: Version 1 (Version v1) [dataset]. Inter-University Consortium for Political and Social Research. https://doi.org/10.3886/ICPSR37166.V1

Oxford English Dictionary. (2023). Remission, n. Oxford University Press; Oxford English Dictionary. https://doi.org/10.1093/OED/3736533463

Pew Research Center. (2019). In U.S., church attendance is declining: Detailed Tables. https://www.pewforum.org/wp-content/uploads/sites/7/2019/10/PF_10.17.19_rdd_update-00-018.png

Schumm, W. R., & Crawford, D. W. (2023). Difficulties With Methodology in Social Science Research With Controversial Issues Regarding Human Sexuality. The Linacre Quarterly, 90(2), 194–216. https://doi.org/10.1177/00243639221082213

Sullins, D. P. (2022). Sexual orientation change efforts do not increase suicide: Correcting a false research narrative. Archives of Sexual Behavior, 51(7), 3377–3393. https://doi.org/10.1007/s10508-022-02408-2

Note from the authors: This study is a revised version of the following study: Sullins DP, Rosik CH and Santero P. Efficacy and risk of sexual orientation change efforts: a retrospective analysis of 125 exposed men [version 2; peer review: 2 approved]. F1000Research 2021, 10:222 (https://doi.org/10.12688/f1000research.51209.2) This version revises some of the language, adds a sentence acknowledging the retraction of a previous study for unrelated reasons, updates the literature review, and corrects a coding error in the reported percentages in Tables 3, 5 and 6, which slightly strengthens the results.

We have chosen to publish the revised study in this journal because the original publishing journal has refused to publish it, instead advising us that it intends to retract the study three years after publication for what we believe is a specious and ideologically motivated reason. In July 2023 Dr. Lorenzo Lorenzo-Luaces posted an online comment suggesting that we include an affirmative disclosure of a previous study that used the same data which had been retracted for reasons unrelated to the data. We responded that the previous study had no material relation to our study, and this history was already clear from the information already in our study (citing other scholars who had already noted it), but that we had no objection to including such a sentence if it would be helpful. Our revision includes a sentence citing this history.

Since July 2023 the journal F1000Research has refused to publish either our response to Dr. Lorenzo-Luaces' critique or the revised version. Instead, in June 2024 they advised us that they intended to retract our study due to the lack of this citation. We have responded that 1) that is a novel, specious and unwarranted reason to retract a study; 2) if applied consistently, such a criterion would lead to the retraction of hundreds of other studies that have happened to use data that had previously been used in a study retracted for other reasons; and 3) however, if they think it is so important, we have no objection to including the desired citation, which consists of a single sentence, in a revised version of the paper. (The journal, F1000Research, is designed to accommodate multiple revisions in response to critical debate. Their published protocol calls for each revision to be re-submitted to the original peer reviewers for further comment and approval. However, when we submitted our revision, instead of following their protocol, the journal summarily removed the revision and all traces of it from their website.)

Rather than permitting the journal F1000Research to suppress evidence and debate on this issue, we hope in this forum to instead enable more debate and consideration of the evidence, from any ideological point of view.

We also believe that the conclusion of our study, which calls for the development of ideologically diverse research teams to overcome ideological polarization in data acquisition and studies of this controversial topic, may be of particular interest to the readers of this journal.