Okay, this is the first research survey I’ve ever done, and I notice a few minor flaws in my test, but otherwise, I think I did pretty well. The test is supposed to determine political persuasion, and me and my partner are looking to see if their is any noticeable deviation between the data of males and females.
After we get answers from a subject, we average them together (I still want to keep the data for specific questions though. My partner gave me some data with just the total scores, and I think I’m going to have to toss it). The average will fall on a scale of 1 - 5. In short, 1 means extreme authoritarian, that you believe the government should have the right to do whatever it wants, and 5 means extreme libertarian, that you believe the people should have the right to do whatever they want.
I see now that I made a few mistakes, and the data won’t be as pure as I wished it would be. I can’t revise the test for you guys, because I’ve already have the results of around 30 subjects recorded. As it turns out, I let liberal / conservative and republican / democrat stances sort of leak in. In all fairness though, as long as I give everyone the same test, the data will be consistently relative, and I can still determine the variance between the results of the genders, if on a slightly skewed scale. As of now, there is a deviation of slightly less than .2, with the average for one gender being 2.9, and the average being 2.7 (I forget which is which though because I left my papers at school). So, the difference is negligible.
On to the errors. What I should’ve done for each of the ‘1‘ options was phrase it so it if asked if the government should have the right to do something. However, since each answer set follows the same pattern (1 = very strict, restrictive government, 5 = citizens have the option to do whatever) the scale is relatively accurate. In all honesty, figuring out how liberals / conservatives line up to libertarians / conservatives is making my head hurt.
After I compile the data, I’ll post the results, a histogram, broken down by gender, and some conclusions. I’d like to be able to add another dimension; deviation between the results of two communities; Ridley High School students, faculty / members of the Straight Dope Message Board (I think I’ll change the latter to ‘internet community’. My teacher might find the name odd.) In stats class, my teacher said 30 subjects is a healthy amount. I hope to get either double or triple that.
Please, do give me any comments and recommendations. Thanks everyone.