Focus on the Family does not want its readers to know what the American Psychological Association’s new report says about intentionally flawed studies of ex-gay success stories, and it certainly does not want supporters to know about the organization’s repeated acts of research fraud.
But Focus does want conservative Christians to know what ex-gay lobbyist Joseph Nicolosi, founder and longtime leader of NARTH, thinks of his critics:
Dr. Joe Nicolosi, founder and director of the Thomas Aquinas Psychological Clinic, said the organization has overlooked years of clinical research that shows sexual orientation is changeable through therapy.
“The APA is really failing to not only represent science, which is its primary responsibility,” he said, “but it’s also failing to inform people.”
Good As You, however, points out that the APA report text didn’t overlook Nicolosi’s research; it referenced Nicolosi 60 times and found his work to be seriously flawed. If not for Nicolosi’s legacy of intentionally distorted research, self-promotion, and maltreatment of clients, the APA task force might have had far less reason to spend two years methodically repudiating the less-than-professional behavior toward patients and the bogus pseudoscience of the movement that he led.
Addendum: The APA report criticizes specific examples of Nicolosi’s sloppy research and unfounded assertions.
On page 32:
For instance, to assess whether sexual orientation had changed,
Nicolosi et al. (2000) performed a chi-square test of association on
individuals’ prior and current self-rated sexual orientation. Several
features of the analysis are problematic. Specifically, the nature of
the data and research question are inappropriate to a chi-square test
of association, and it does not appear that the tests were properly
performed. Chi-square tests of association assume that data are
independent, yet these data are not independent because the row
and column scores represent an individual’ rating of his or her past
and present self. Chi-square tests ought not to be performed if a cell
in the contingency table includes fewer than five cases. Other tests,
such as the nonparametric McNemar’ test for dichotomous variables (McNemar, 1969) or the sign (Conover, 1980) or Wilcoxon signed-rank tests (Wilcoxon, 1945) for nominal and ordinal data, respectively, are used to assess whether there are significant differences between an individual’ before and after score and are appropriate when data fail to meet the assumptions of independence and normality, as these data do and would have been more appropriate choices. Paired t-tests for mean differences could also have been performed on these data. There are procedural problems in performing the chi-square test such as missing data, and the analyses are conducted without adjustment for chance, with different numbers of subjects responding to each item, and without corrections to the gain scores to address regression artifacts. Taken together, however, the problems associated with running so many tests without adjusting for chance associations or correcting for regression artifacts and having different respondents in nearly every test make it difficult to assess what changes in scores across these items actually reflect.
Some SOCE teach men how to adopt traditional masculine behaviors as a means of altering their sexual orientation (e.g., Nicolosi, 1991, 1993) despite the absence of evidence that such interventions affect sexual orientation. Such theoretical
positions have been characterized as products of stigma and bias that are without an evidentiary basis and may increase distress (American Psychoanalytic Association, 2000; Isay, 1987, 1999; Drescher, 1998a; Haldeman, 1994, 2001). For instance, Haldeman (2001) emphasized in his clinical work with men who had
participated in SOCE that some men were taught that their homosexuality made them less masculine‚Äîa belief that was ultimately damaging to their self-esteem. Research on the impact of heterosexism and traditional gender roles indicates that an individual’ adoption of traditional masculine norms increases sexual selfstigma and decreases self-esteem and emotional connection with others, thus negatively affecting mental health (Szymanski & Carr, 2008).
The first finding from our review is that there is insufficient evidence that SOCE are efficacious for changing sexual orientation. Furthermore, there is some evidence that such efforts cause harm. On the basis of this evidence, we consider it inappropriate for psychologists and other LMHP [licensed mental health professionals] to foster or support in clients the expectation that they will change their sexual orientation if they participate in SOCE. We believe that among the various types of SOCE, the greatest level of ethical concern is raised by SOCE that presuppose that same-sex sexual orientation is a disorder or a symptom of a disorder. (Footnote: See, e.g., Socarides (1968), Hallman (2008), and Nicolosi (1991); these theories assume homosexuality is always a sign of developmental defect or mental disorder.) Treatments based on such assumptions raise the greatest level of ethical scrutiny by LMHP because they are inconsistent with the scientific and professional consensus that homosexuality per se is not a mental disorder. Instead, we counsel LMHP to consider other treatment options when clients present with requests for sexual orientation change.
Chapter 8 begins with a criticism of Nicolosi, among others, for endorsement of involuntary ex-gay therapy for youths:
Publications by LMHP directed at parents and outreach from religious organizations advocate SOCE for children and youth as interventions to prevent adult same-sex sexual orientation (Cianciotto & Cahill, 2006; Kennedy & Cianciotto, 2006; Nicolosi & Nicolosi, 2002; Rekers, 1982; Sanchez, 2007). Reports by LGB advocacy groups (e.g., Cianciotto & Cahill, 2006; Kennedy & Cianciotto, 2006) have claimed that there has been an increase in attention to youths by religious organizations that believe that homosexuality is a mental illness or an adverse developmental outcome. These reports further suggest that there has been an increasing in outreach to youths that portrays homosexuality in an extremely negative light and uses fear and shame to fuel this message. These reports expressed concern that such efforts have a negative impact on adolescents’ and their parents’ perceptions of their sexual orientation or potential sexual orientation, increase the perception that homosexuality and religion are incompatible, and increase the likelihood that some adolescents will be exposed to SOCE without information about evidencebased treatments.
The report adds on page 72:
Childhood interventions to prevent homosexuality have been presented in non-peer-reviewed literature (see Nicolosi & Nicolosi, 2002; Rekers, 1982).57 These interventions are based on theories of gender and sexual orientation that conflate stereotypic gender roles or interests with heterosexuality and homosexuality or that assume that certain patterns of family relationships cause same-sex sexual orientation. These treatments focus on proxy symptoms (such
as nonconforming gender behaviors), since sexual orientation as it is usually conceptualized does not emerge until puberty with the onset of sexual desires and drives (see APA, 2002a; Perrin, 2002). These interventions assume a same-sex sexual orientation is caused by certain family relationships that form gender identity and assume that encouraging gender stereotypic behaviors and certain family relationships will alter sexual orientation (Burack & Josephson, 2005; see, e.g., Nicolosi & Nicolosi, 2002; Rekers, 1979, 1982).
The theories on which these interventions are based have not been confirmed by empirical study (Perrin, 2002; Zucker, 2008; Zucker & Bradley, 1995). Although retrospective research indicates that some gay men and lesbians recall gender nonconformity in childhood (Bailey & Zucker, 1995; Bem, 1996; Mathy & Drescher, 2008), there is no research evidence that childhood gender nonconformity and adult homosexuality are identical or are necessarily sequential developmental phenomena (Bradley & Zucker, 1998; Zucker, 2008). Theories that certain patterns of family relationships cause same-sex sexual orientation have been discredited (Bell et al., 1981; Freund & Blanchard, 1983; R. R. Green, 1987; D. K. Peters & Cantrell, 1991).
The recent nonreligious interventions are based on the assumption that homosexuality and bisexuality are mental disorders or deficits and are based on older discredited psychoanalytic theories (e.g., Socarides, 1968; see American Psychoanalytic Association, 1991, 1992, 2000; Drescher, 1998a; Mitchell, 1978, 1981). Some focus on increasing behavioral consistency with gender norms and stereotypes (e.g., Nicolosi, 1991). None of these approaches is based on a credible scientific theory, as these ideas have been directly discredited through evidence or rendered obsolete. There is longstanding scientific evidence that homosexuality per se is not a mental disorder (American Psychiatric Association, 1973; Bell & Weinberg, 1978; Bell et al., 1981; Conger, 1975; Gonsiorek, 1991; Hooker, 1957), and there are a number of alternate theories of sexual orientation and gender consistent with this evidence (Bem, 1996; Butler, 2004; Chivers et al., 2007; Corbett, 1996, 1998, 2001; Diamond, 1998, 2006; Drescher, 1998a; Enns, 2008; Heppner & Heppner, 2008; Levant & Silverstein, 2006; Mustanksi et al., 2002; O’Neil, 2008; Peplau & Garnets, 2000; Pleck, 1995; Rahman & Wilson, 2005; Wester, 2008).