Yeah, I laughed when I saw I was plotted closest to Gandhi... (of course, I told the wife about it!)... but that's not what I meant by "representative"... just wasn't surprised at all that I was "squarely" in the lower left quadrant. When I first looked at the result page I didn't realize right away that the graph at the top was a sample graph and I thought to myself, "I'm centrist?" That would have been news to me.


>Well, I thought it was amusing - nothing more. As for "representative", I'm not sure what the axes really mean (again, no explanation of the methodology). I find it interesting that people such as Gandhi, Stalin, Thatcher etc. are shown as sample results when it's extremely unlikely they actually took the test. If your result came out near Gandhi, say, how do you know you really are similar to Gandhi on these two axes? And why are there hardly any people shown in the lower right quadrant?
>
>Slashdot Polls says it best: "If you're using these numbers to do anything important, you're insane."
>
>******************
>
>Actually, without knowing the methodology there is an experiment that could be performed to see if there is any inherent bias in that test. Simply (?) perform the test 1,000 or 1,000,000 times, randomly choosing answers each time. With no bias, the expected result would be a circle, uniformly filled with test-result X, Y points. If the results were other than that, it would indicate bias.
>
>As any good university professor would say, "this is left as an exercise for the student" < g >