Misleading yourself for fun and profit

    A web site came my way yesterday that captivated my interest for a while: http://www.genderanalyzer.com/. A nod to Kev of Paper Wilderness fame for that one.
    It attempts to detect whether the text at a URL you give it was written by a man or a woman, and its approach of classification by comparison with a large body of known samples should be quite a sound one. This is not a million miles from what I do for a living, so I find it to be a very interesting site.
    Give it this blog, and it says that it is likely to have been written by a woman. Phew, validation, I'm a girl!
    Unfortunately though it's never quite that easy. Just from curiosity I gave it some other text I write in another field, as bloke. The analysis returned:  it was likely to have been written by a bloke. Damn, I'm a bloke after all!
    I'm not writing in a special voice in either place. This is me, this is what you get. Kev expressed disappointment that Paper Wilderness returned a probability of being written by a woman, based on my experience I'd suggest he needn't worry.
    It's tempting to seek validation in software like this. Most of us will probably have encountered the COGIATI test, a questionnaire that is designed to give you an assessment of the level of your transsexuality. When I do that test honestly, it tells me I'm mostly female. No surprises there (It's a while since I did it co I can't remember my exact score). But I'm afraid I can't trust it because with a little conscious effort I can make it tell me I'm completely female, or that I'm barely transsexual at all. In short, I can make it tell me what I want it to, or to put it another way, it tells me what I want to hear. Not good enough, I want something that tells me the truth even if it disappoints me.
    There is another test, one that doesn't involve software. What's your digit ratio? The ratio of the lengths of your index finger to your ring finger is gender dimorphic, if you are male you are more likely to have shorter index finger than ring finger. My digit ratio is 1:1, which of course states that I should be female. Yay, I'm a girl! I mean, that indicates the hormonal balance in the womb, so it can't lie, can it? I asked my wife, after all she's a girl, what would her digit ratio be? Shorter index finger than ring finger. Well clearly she's no bloke, so that serves me right for putting too much store in finger lenghs. The key is in the phrase "More likely" above, it is still quite possible for a woman to have a male digit ratio and vice versa.
     It is always a bit of fun to try these tests, but I think it is dangerous to put too much store by their results. I hope I've demonstrated above that they can not always be trusted, and all the validation we should need should start between our ears.
     There is one more test. It's a traditional British test, and it's guaranteed 100% accurate. Simply answer the following question for an instant result.

Do you understand the offside rule? 

    I'll leave it to you to decide which answer validates your gender identity.