Bisexuality & statistics: Twice as many dates?

[emanix]

[Edit, 8th of October 2015: I reworked this essay into an illustrated 'comic' format, which you can check out below the original text!]

I've seen it in all sorts of places, the assumption that bisexuality has to mean that a person is equally attracted to both sexes, not notice the difference between, and so forth... Marcus Morgan has a lovely

Comments

[joreth]

This is exactly the problem I have with "research" that asks the participants to give their orientation or gender by *identity*, not by practical application, and those that do not make allowances for the differences between desire & practice.

In other words, IMO, a bisexual is someone who is *attracted* to both/all genders to some degree, not necessarily someone who has had sexual relationships with both/all genders. That leaves out all the bisexuals who just haven't had a chance yet, and counts unfairly all the people who have experimented or had other reasons for having a sexual relationship with a gender they ultimately decided was not a gender they were attracted to (i.e. gay men who "pass", particularly in the more fatal years of discrimination).

This was actually the basis of some of my tweets recently about being called "transphobic". I objected to a survey's use of personal identity as the marker for either gender or orientation on the basis that it does not accurately describe what people *really* do or think because there are cultural reasons for choosing labels, including the misunderstanding of what the label means, like the ones you're talking about here.

But suggesting that a survey ask what people *like* rather than what they *do*, and refusing to let them fill in their own identity label instead of describe themselves without an identity label was considered "transphobic" because trans people want to be able to give their own identities and not be forced into boxes by people who don't understand the trans community. Or something.

Wanting accurate results in research instead of catering to personal identity labels that are rife with misunderstandings and that contradict reality apparently makes me insensitive to marginalized people (yes, they said that).

[emanix]

Ooh that complexity... My answer to this when designing studies is to include both identity/preference and genetics/practice if it's at all relevant, though it's a long time since I last threw one out. Not much use if you're limited on the number of questions, I suppose.
I start to get the feeling lately that I may get drawn into doing research again, whether I like it or not. Hanging out with too many academics to not feel the pull.

By the way was it you that linked to the survey studying bisexual women on surveymonkey? I sent a critique via email and got a really nice reply from the author. Lovely when that happens :)

[joreth]

No, I don't think that was me :-)

I agree, if the study is not limited in some way, having both preference/identity or genetics/practice options would be ideal. It would also highlight the discrepancy between "identity" and practice or genetics. But sometimes researchers have to make choices, and it is my opinion that the choice that more accurately reflects reality is always the better option. Giving people the option to just identify as "lesbian" when they are, in a behavioural and/or attracted-to sense, bisexual, hurts the data and the communities (as one example).