In order to find out if transsexuals of different sexual orientations really have different sexual fantasies, we need to make sure we are looking for a difference between transsexuals instead of a difference across sexual orientation that we happen to find among transsexuals. Necesssary control groups were left out of Blanchard's analyses.

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

Need of Control Groups to Separate Sexual Orientation from Transsexuality

Control Groups are an especially important part of many social science and natural science studies. A control group is a group of participants who resemble the group you would like to study in all relevant ways aside from the element you would like to study. For example, let's say you would like to test a new medicine to see if it helps people. You give it to a group of people and a month later they are more healthy. Was it the medicine? Maybe. But it's also possible that just giving people the medicine made people think they would get better and so many of them did. Some might just get better even without medicine. To make sure the medicine is what really matters, you would give another group of participants a placebo, something that looks like the medicine but really is nothing medicinal at all. To find out if the medicine worked, you look to see if the group who got medicine improved more than the control group.

Blanchard's hypothesis is that the sexual orientation of transsexuals is related to having sexual fantasies of being their target sex. To test this hypothesis he compares MtF transsexuals oriented toward men to MtF transsexuals with a sexual orientation not exclusively towards men. In a sense, that seems like a control group: how much more common are autogynephilic fantasies in transsexuals not exclusively attracted to men than those who are exclusively attracted to men. But what this control group fails to distinguish is the role of sexual orientation, separate from gender incongruence, in predicting fantasies about being a women. There should really be 4 groups of participants: the two transsexual groups, typically-gendered women attracted exclusively to men, and typically-gendered women not exclusively attracted to men. That is, maybe lesbian women find thinking about themselves as women more of an erotic charge than straight women do. In accordance with Blanchard's model, we would predict that MtF transsexuals not exclusively attracted to men would have more sexual fantasies about being a woman than any of the other three groups.

If you are a scientist, you probably see the need for a control group as obvious. Scientists know 'common sense' can be misleading and it's simply standard practice to include control groups for rigorous scientific conclusions. But a less scientific reader might question, with all the sincerity of their common-sense, if we really need a control group. "Don't we just know typically-gendered women don't have fantasies like this?" There are two perspectives from which this common-sense argument is inadequate: we don't know what goes on in the minds of our participants and the minds of researchers can be influenced by subtle biases.

I can understand the common-sense point if we assume autogynephilic fantasies are like those described by J. Michael Bailey in his book, "The Man who would be Queen: The Science of Gender Bending and Transsexualism", or the way Ray Blanchard describes them in his clinical intuition articles. But Blanchard's empirical studies did not score transsexuals for having such 'extreme' fantasies. Remember that transsexuals completed this survey from their own perspective about what the questions meant. They didn't necessary interpret their answers as meaning fantasies nearly as elaborate or as extreme as what Blanchard envisioned. This (Blanchard, 1989) was the first paper I read by Blanchard. Before reading the article (& not knowing the theory), I completed his scales because they were in the appendix. It's like the super-intellectual version of taking quizzes in Mademoiselle magazine!!!! Here is an example of a question I said "yes" to. It's from the "Autogynephilic Interpersonal Fantasy Scale", "Have you ever become sexually aroused while picturing yourself as a fully dressed woman being admired by another person?" To make the question appropriate to administer to both male and female-bodied persons, the question might be modified to, "Have you ever become sexually aroused while thinking about yourself, as a women who is fully dressed, being admired by another person?" When I said "yes" to this question, I was thinking about something very different from the sorts of things I later read in Blanchard's other work. Blanchard should not be faulted for failing to consider every way his questions might be interpreted. That is exceptionally difficult. Knowing that no-one can ever be fully aware of every way your participants might construe your questions, it is always best to include relevant control groups.

A second reason control groups are necessary is to prevent the biases of the researcher from influencing their view of their participants. Previously I mentioned that research into transsexuality assumes transsexuality is something deviant, a disorder. This assumption may influence how researchers think about their transsexual participants. Consider the following example from a social psychology study of prejudice. A participant in this study is told either a construction worker or housewife hit somebody. When asked to elaborate, the construction worker is characterized as having been more violent than the housewife (Kunda & Sherman-Williams, 1993; Krueger & Rothbart, 1988). How is this relevant? Replace housewife with women, construction worker with transsexual, and hit (violent act) with a sexual act. MtF transsexuals may be viewed as being more extreme in their sexual acts than typically-gendered women are in their sexual acts because the stereotypes of transgenderism include a more kinky sexuality.

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