Spitzer’s ‘retraction’ of his sexual orientation change study: what does it really mean?
May 31, 2012 (NARTH.com) - A great deal of attention is currently being given to the recent “retraction” by Robert Spitzer, M.D., of his important study of sexual-orientation change (Spitzer, 2003a). The quotation marks around “retraction” are purposeful, for what has happened should not be characterized as a retraction. While this turn of events has now become a favorite talking point for those opposed to sexual orientation-change efforts (SOCE), the language of retraction reflects politically motivated speech rather than scientific analysis. What follows is intended to help those confused by Spitzer’s actions and the subsequent media feeding frenzy to understand what has really occurred. I have outlined below some key points that seem to have been lost in the partisan utilization of this turn of events.
1. Spitzer has not retracted his study. The proper term for what Spitzer has done is provided in the title to his recent letter of apology: He has reassessed his interpretation (Spitzer, 2012). It appears that he may have originally wished to retract the 2003 study, but the editor of the journal in which the study was published, Kenneth Zucker, Ph.D., denied this request. Zucker has been quoted regarding his exchange with Spitzer as observing:
You can retract data incorrectly analyzed; to do that, you publish an erratum. You can retract an article if the data were falsified-or the journal retracts it if the editor knows of it. As I understand it, he’s [Spitzer] just saying ten years later that he wants to retract his interpretation of the data. Well, we’d probably have to retract hundreds of scientific papers with regard to interpretation, and we don’t do that. (Dreger, 2012)
What Zucker is essentially saying is that there is nothing in the science of the study that warrants retraction, so all that is left for one to change is his interpretation of the findings, which is what Spitzer appears to have done.
2. Spitzer’s change of interpretation hinges on his new belief that reports of change in his research were not credible. Instead, he now asserts that participants’ accounts of change were “self-deception or outright lying” (Spitzer, 2012). In taking this position, Spitzer has aligned himself with original critics of the study. When the original study was published, peer commentaries about the study had been solicited and were published in the same issue. Among those who questioned the reliability of the self-reports of change were many familiar opponents of SOCE: A. Lee Beckstead, Helena Carlson, Kenneth Cohen, Ritch Savin-Williams, Gregory Herek, Bruce Rind, and Roger Worthington.
3. The case for the credibility of participants’ account of change still remains. Remember that nothing about the science of Spitzer’s research was flawed. Like all research pursuits, the methodology had limitations, but a reasonable case for accepting the validity of these accounts was made at the time, and still stands today. At the time his study was published, Spitzer (2003a) reported, “...there was a marked reduction on all change measures. This was not only on the three measures of overt behavior and sexual orientation self-identity…but also on the seven variables assessing sexual orientation itself” (p. 410). In addition, 119 of his sample of 200 participants reported achieving “Good Heterosexual Functioning,” which was defined in terms of increasing satisfaction in opposite sex sexual behaviors and decreased same-sex fantasy.
Among the peer commentaries that agreed with Spitzer’s original interpretation, Wakefield (2003) noted that, “...to assume without evidence that reports of changes must be deceptions begs the question of whether change sometimes occurs” (p. 457). Spitzer (2003b) himself responded to the critics by noting:
Therefore, the critics are correct in claiming that significant response bias could have been present but they certainly have not proved that it was present. They also did not point to anything in the study results that suggests response bias. I acknowledge that some response bias could certainly have occurred, but I find it hard to believe that it can explain all of the reported changes…Surely if bias were present, one would expect that subjects (as well as their spouses) would be motivated to give particularly glowing accounts of marital functioning. They did not. (p. 471)
It is curious that Spitzer’s (2012) apology seems to imply that he earlier claimed his research proved the efficacy of SOCE. As was understood at the time, the design of Spizter’s study ensured his research would not definitively prove that SOCE can be effective. Certainly it did not prove that all gays and lesbians can change their sexual orientation or that sexual orientation is simply a choice. The fact that some people inappropriately drew such conclusions appears to be a factor in Spitzer’s reassessment. Yet the fundamental interpretive question did and still does boil down to one of plausibility: Given the study limitations, is it plausible that some participants in SOCE reported actual change?
In spite of all the recent media hoopla, nothing has really changed regarding the interpretive choice one faces regarding the limitations of self-report in this study. Either all of the accounts across all of the measures of change across participant and spousal reports are self-deceptions and/or deliberate fabrications, or they suggest it is possible that some individuals actually do experience change in the dimensions of sexual orientation. Good people can disagree about which of these interpretive conclusions they favor, but assuredly it is not unscientific or unreasonable to continue to believe the study supports the plausibility of change.
4. There is an unspoken double standard in the reports of Spitzer’s reassessment. The probable influence of political and other non-scientific factors in how Spitzer’s reassessment is being portrayed can be seen in which interpretations of self-report data receive favored notoriety and which are relegated to unfavored exile. Yarhouse (2003) observed this lack of consistency at the time of the study:
Memory recall of this sort can be unreliable. To be fair, however, much of what we know about LGB experiences, including theories for the etiology of sexual orientation and studies of sexual identity development and synthesis, is based upon retrospective studies utilizing memory recall. Any time proponents of the biological hypothesis for the etiology of homosexuality cite the Bell et al. (1981) study they are referencing a study that utilized retrospective memory recall. The Shidlo and Schroeder (2002) study also relied upon memory recall and is subject to the same criticism. (p. 462).
Spitzer (2003b) had similar observations in defending his findings, implying that demand characteristics could have influenced the self-reports of participants in other related research:
This study had essentially the same design and a similar recruitment strategy of ex-gay subjects as in the Beckstead (2001) and Shidlo and Schroeder (2002) studies. This raises the question of why so very few of their subjects gave answers consistent with a change in sexual orientation whereas the majority of my subjects did. The possibility of researcher bias must be considered. (p. 471).
A triumphal embrace often accompanies self-report data that suggests harm from SOCE, the equivalence of gay and heterosexual parenting, and other foci that fit with the preferred narrative of gay activists. It is unfortunate but not surprising that reports of sexual-orientation change are subject to unrelenting skepticism while other self-report data such as that of Shidlo and Schroeder (2002) seem to be reified as universal fact even though they suffer from similar limitations. If Spitzer’s study is to be rejected for its use of self-report data, should not methodologically equivalent research against SOCE receive a similarly skeptical reception? While scientific fairness would seem to demand this, political interests clearly do not.
5. Personal and sociopolitical contexts may provide insights into Spitzer’s reassessment. I once spoke briefly with Dr. Spitzer by phone years ago following the publication of his research. He seemed to be a kind and compassionate man who exemplified the spirit of genuine scientific curiosity. No doubt he was grieved that some used his work to make unsupportable claims of SOCE efficacy and this may have resulted in unfulfilled expectations by some gay and lesbian consumers. Yet it is certainly possible that other needs beyond his concern for human welfare were at play in his apology.
It is hard to imagine the fall from professional grace that Spitzer took due to this study. In a very short period of time, his status within his profession changed from that of a heroic pioneer of gay rights to that of an unwitting mouthpiece for practitioners of SOCE, whom many of his colleagues deem morally reprehensible. Before and after the study was published, Spitzer confirmed that he was getting a high volume of hate mail and anger directed at him (Spitzer, 2003b; Vonholdt, 2000). A decade of being hammered by your friends, colleagues, and the gay community that once revered you would surely take a toll on any of us.
Spitzer currently suffers from Parkinson’s disease and is in the twilight of his life, which makes it understandable that he would reflect on what sort of legacy he wants to leave. Hero or villain, icon or pariah-which legacy would anyone prefer to have? I can not say for sure that these non-scientific considerations influenced Spitzer’s decision to “retract” his study, but I can say that it is hard for me to conceive how they would not. Spitzer likely knows infinitely more gay and lesbian persons than he does individuals who report change in sexual orientation. This may have made it difficult for him to see that in trying to atone for the harm gay men and lesbians in his professional network claimed resulted from the study, he simultaneously caused harm to participants in his study who experienced change and now are told they were deceived or lying. All of this serves to underscore how personal and subjective the practice of social scientific discourse can be when the subject matter is entangled in a major sociopolitical debate.
A purely scientific approach to the limitations of Spitzer’s research would be to conduct more rigorous outcome research, something that he along with others have been calling for all along (Spitzer, 2003a, 2003b; Jones, Rosik, Williams, & Byrd, 2010). Even the APA Task Force’s Report on Appropriate Therapeutic Responses to Sexual Orientation (American Psychological Association, 2009) issued a call for such studies to be undertaken. Unfortunately, the reality appears to be that the APA and other institutions in a position to fund and conduct outcome research on SOCE in conjunction with NARTH and other SOCE practitioners have no real interest in doing so. They have nothing to gain by such research, as outcomes unfavorable to SOCE would not meaningfully change their current skepticism, while outcomes favorable to SOCE would be a public relations and public policy disaster for them.
I doubt that Spitzer would “retract” his assessment of the likelihood that needed follow up studies would be conducted (Spitzer, 2003b):
Given the cost and complexity of such a study and the current view in the mental health professions of the benefits and risks of reorientation therapy, such a study is not going to happen in the near future. This is unfortunate because of the real questions raised, albeit admittedly not resolved, by this study (p. 472).
So instead of more and better research on SOCE, we find activists and their supporters in the media pouncing on a change of interpretation in an effort to preempt legitimate scientific debate. Nuance, context, and balanced analysis all be damned. What seems to be foremost is the use of Spitzer’s reassessment to bludgeon SOCE supporters into submission and silence. Is it really far-fetched to suspect science is being held hostage to political agendas here?
I sincerely hope that this brief analysis helps clarify what did and did not happen when Spitzer “retracted” his earlier study. No new scientific finding was discovered that discredited SOCE. No egregious methodological flaw was identified. The same arguments forwarded in favor or against the study a decade ago still stand. Legitimate debate about the study’s significance can and should still take place. Nothing has changed other than Spitzer has revised his earlier interpretation for what are likely to be a host of understandable but inherently non-scientific reasons. This is his right to do, but let no one tell you that in doing so he has discredited his research or alternative interpretations more favorable to those who report change in their same-sex attractions and behavior.
American Psychological Association (2009). Report of the APA task force on appropriate therapeutic responses to sexual orientation. Retrieved from http://www.apa.org/pi/lgbt/resources/therapeutic-response.pdf
Beckstead, A. L. (2001). Cures versus choice: Agendas in sexual reorientation therapy. Journal of Gay and Lesbian Psychotherapy, 5(3/4), 87-115.
Bell, A. P., Weinberg, M. S., & Hammersmith, S. K. (1981). Sexual preference: Its development in
men and women. Bloomington, IN: Indiana University Press.
Dreger, A. (2012, April 11). How to ex an “ex-gay” study. [Web log post]. Retrieved from http://psychologytoday.com/blog/fetishes-i-dont-get/201204/how-ex-ex-gay-study
Jones, S. L., Rosik, C. H., Williams, R. N., & Byrd, A. D. (2010). A Scientific, Conceptual, and Ethical Critique of the Report of the APA Task Force on Sexual Orientation. The General Psychologist, 45(2), 7-18. Retrieved from http://www.apa.org/divisions/div1/news/fall2010/Fall%202010%20TGP.pdf
Shidlo, A., & Schroeder, M. (2002). Changing sexual orientation: A consumers’ report. Professional Psychology: Research and Practice, 33, 249-259.
Spitzer, R. L. (2003a). Can some gay men and lesbians change their sexual orientation? 200 participants reporting a change from homosexual to heterosexual orientation. Archives of Sexual Behavior, 32(5), 403-417.
Spitzer, R. L. (2003b). Reply: Study results should not be dismissed and justify further research on the efficacy of sexual reorientation therapy. Archives of Sexual Behavior, 32(5), 469-472.
Spitzer, R. L. (2012). Spitzer reassesses his 2003 study of reparative therapy of homosexuality [Letter to the editor]. Archives of Sexual Behavior. Advance online publication. doi: 10.1007/s10508-012-9966-y
Wakefield, J. C. (2003). Sexual reorientation therapy: Is it ever ethical? Can it ever change sexual orientation? Archives of Sexual Behavior, 32(5), p. 457-459.
Vonholdt, C. R. (2001). Interview with Robert L. Spitzer: Homosexuality and the reality of change. Bulletin of the German Institute for Youth and Society, 1, 33-36. Retrieved from http://www.dijg.de/english/homosexuality-reality-of-change/
Yarhouse, M. A. (2003). How Spitzer’s study gives a voice to the disenfranchised within a minority group. Archives of Sexual Behavior, 32(5), 462-463.