Let me take a break from my review of the APA task force report on appropriate therapeutic responses to sexual orientation to point out the new on-line Discover article on efforts to change sexual orientation.
There is an interesting quote from the chair of the task force on the Exodus study:
“Everything was wrong with that study,” Glassgold says. “[Yarhouse and Stanton] chose the wrong statistics to evaluate, they violated statistical laws, and they didn’t have a control group—just a small sample of people recruited from religious groups. They followed the individuals over a couple of years, but didn’t specify that the subjects should only try one intervention at a time, so they tried many at the same time. So we aren’t sure which, if any, intervention was causal.”
This critique of our study (as it is reported in this article) is overstated and simply fails to acknowledge ways in which the study was an improvement in design to previous studies (e.g., longitudinal, prospective [Phase 1 participants], use of multiple measures of sexual orientation and sexual identity, symptom distress, etc.). The study also had a good retention rate (~75% over a three year span). The comment on preferred statistical analyses is incorrect. More sophisticated analyses would have been important had we been comparing different groups of participants to see if they responded to the ministry differently over time. That was not the point of the study; rather, we were interested in falsifying a theory that there would be a uniformly negative effect, which would have shown up with the statistical analyses we used.
While this kind of dismissal of the study is unwarranted, I think the potential misuse of our study comes from those who want to see in it categorical change (completely gay to completely straight). We reported meaningful shifts (for some) along a continuum, and this should be nuanced in light of the retrospective and prospective aspects of the study. It would make sense to me to have ministries begin there (the possibility of meaningful shifts along a continuum for some) and keep expectations realistic.
Mark Yarhouse 