I wanted to see what all the possible results were and figure out why i got the results i did. Unfortunatly this is not one of the tests that lets you look at all possibilities straight from the results page.
I took a look at the source, and thankfully it's one where you can see the values of all possible answers. However the result itself is calculated in some otehr script so you can't just directly input the results, so I had to go through an do each of them by hand, however doing so _did_ enable me to figure out the weird results =)
The questions have values ranging from 0 to 10. In fact, every question has ten possible answers, with one answer for each value except for 1. There are no answers with a value of 1 at all.
I found that giving the same value answer for every question would give you different results (posted below) as you would expect. However i also noticed that i needed to change more than 50% of the questions to the new value to get the new result sometimes. I finally figured out by setting all questions to 0 and then switching them to 10 one at a time that it's not which value you answer the majority of the time, it's the _average_!
The script adds up all your answers numerically, and then looks at a list of ranges to figure out what your final result is. Which means that if you answer _every_ question "appropriatly" then you'll get the "correct" answer. Other than that though things quickly start getting random.
If you set the test to all "No Sex" answers and then change three of them to "Gay Sex" answers, you get "Gang Bangs", change another, and it's "Asians," with none of the individual answers indicating a preference for Gangbangs or Asians.
And the results are: