Advocates have gotten Blanchard's model recongized in the DSM-IV-TR and HB-SOC despite a lack of evidence. They do this by treating as evidence for Blanchard's model, studies that link sexual orientation and transsexuality. This is not appropriate because they neglect studies with null-results and fail to meet a basic standard of science, falsifiability.

This is page 5 of 13.          [1]   [2]   [3]   [4]   [5]   [6]   [7]   [8]   [9]   [10]   [11]   [12]   [13]  

So Little Evidence for Blanchard's Theory?

You may be very surprised to read that there is only one peer-reviewed empirical article supporting Blanchard's key theoretical points. First, to the extent you care about scientific conclusions you might be disconcerted that a theory would be so heavily endorsed with such little evidence. Are the findings at the foundation of this theory replicable? Second, you may have heard advocates of Blanchard's model talk about the enormous amount of evidence supporting his model. Why this discrepancy?

A basic standard in science is that a good finding is replicable. If one person conducts and experiment and another person conducts the same experiment, they should both get the same results. Are Blanchard's basic findings replicable? As mentioned previously, only one empirical study has been published to support Blanchard's basic findings. He published three additional empirical papers that extend upon this basic finding. Yet these papers do not also replicate the original result. Perhaps no one has ever tried to replicate Blanchard's result? But having been published in 1989, this certainly is not for lack of sufficient time. Could it be that nobody knows about the theory? This is very unlikely given its endorsement in the DSM and HB-SOC. Furthermore, people with considerable influence among either transsexuals or sexuality researchers have endorsed Blanchard's model. Anne Lawrence published about the topic on her web-site years ago. J Michael Bailey runs SEXNET which is a mailing list for professionals studying transsexuality and related topics. Another possibility is that others have attempted to replicate Blanchard's results and failed. In our field, it is very difficult to publish what are called "null results." That is, if you conduct a study and don't find a difference between groups (e.g., those attracted exclusively to their target sex versus others), academic journals are unlikely to publish it. This is called the "file-drawer phenomena." Social scientists can spend considerable effort trying to replicate results when they are unaware others have failed to do so in the past. The problem is usually less severe in terms of overall scientific progress. Unreplicable findings and the theories they support fall by the way-side as they are cited less often. In this case, however, the publicity by advocates of Blanchard's theory keep it prominent in the DSM, HB-SOC, and now even textbooks.

Advocates of Blanchard's model say there is considerable evidence supporting it. But then why would I say that there is only 1 publication supporting Blanchard's key points and 3 other publications supporting various corollaries? To me, a study only supports a theory if it was possible that the results of the study could have falsified (provided evidence inconsistent with) the theory, or at least some of its parts. The standard I suggest comes from a historically common belief about the importance of falsifiability in science (e.g. Karl Popper). Advocates of Blanchard's model seem to note as evidence numerous studies consistent with Blanchard's model even if they would still believe Blanchard's model had the results been different. In particular, the studies most commonly cited by advocates are those suggesting sexual orientation is predictive of differences among transsexuals. For example, Blanchard (1995) found that MtF transsexuals attracted exclusively to men ask for SRS at a younger age ["A structural equation model for age at clinical presentation in nonhomosexual male gender dysphorics." in "Archives of Sexual Behavior", v.23(3), 311-320.]. Though I agree this finding and many others are interesting, I do not see it as relevant to testing Blanchard's model because even if MtF transsexuals attracted exclusively to men asked for SRS at later ages, Blanchard's model could still be correct.

Even if it were consistent with historical scientific standards to consider many of these studies as evidence for Blanchard's model, it should also be noted that many studies have attempted but failed to use sexuality as predictive of differences among transsexuals. These null-results are published because something else in the study was more predictive. Fruend et al. (1991) noted that among several factors only gender dysphoria was predictive of 3 different classes of gender identity disorder in male-bodied persons attracted to women. A scale of androphilia did not differentiate the groups.. Using an exploratory factor analysis in his dissertation Smith (1997) found a stable feminine gender identity was predictive of psychological adjustment for originally male-bodied transsexuals attracted to women. A sexual arousal factor, including items about autogynephilia, failed to be as useful. Johnson and Hunt (1990) found that androphilia was not linked to any MMPI measures of psychological disturbance.

This is page 5 of 13.          [1]   [2]   [3]   [4]   [5]   [6]   [7]   [8]   [9]   [10]   [11]   [12]   [13]