Resignation
Today, I hereby abdicate any hope of being a master quantitative statistician. I sat down to catch up on some reading for my statistics class. I flipped through a couple of chapters casually because the words seemed pretty familiar. We have been talking about t-testing, F-distributions, ANOVAs [analyses of variance], and chi-square testing. The latter two are slowly going out of vogue since regression models can give us more precise information about the relationship between two variables, and only a select diehard cadre of psychologists are still messing with those methods. I can tell you that we can use a chi-square distribution if we're only concerned with what a sample is doing. Since I try to make generalizations of a population based on sample data, then I don't use much of the chi-square.
I have one more assignment to do in statistics before the final exam. I have to do an ANOVA and a chi-square test on a data set to determine if religious denomination has anything to do with garnering education.
Last time, I tested whether or not education had anything to do with children. It does a little, but not much. The sample revealed that for every five years of education received, you are likely to have one less child. It was a statistically significant finding for the sample, but I was not apt to make any generalizations for two reasons. One, in a matter of common sense, American education is compulsory up to twelve years, and the next largest proportion of the educated stop at a Bachelor's degree. If it takes four years to earn that, then we might not see much of a difference in having children. Education is bimodal between two values that may not lead to a difference in the dependent variable. Also, the regression showed that education only accounted for 21% of the variation in the number of children, so there may be one or more better explanations (e.g. independent variables) that may better account for the number of children in a family.
Okay, so I get that stuff -- in rather general terms. I have to admit that statistics language is very sticky, even with something simple like hypothesis testing. [clears throat] "At the 5% significance level, we reject the null hypothesis. There is more than a 95% chance that we could get a test statistic of some specific value if the population mean were as indicated in the null hypothesis." That's right. A lot, a lot, a lot of words.
Here's what I'm willing to do. For starters, I'm going to enjoy my Thanksgiving holiday come hell or high water. Dan has graciously invited me to hang out with him and his family, so I will take a drive to Los Angeles this coming Wednesday and I'll return home Friday evening. When I get back, I will get a good night's rest and slowly whittle away at the rest of the term. I have a stats assignment and a short paper due on the second, an exam to proctor on the twelfth, grading, and then a seminar paper and a final exam on the sixteenth.
Come to think of it, if I need to run away from my statistics book, then let it be for a good reason. I need to do some research for my social psychology seminar paper. I'm doing a literature review to see if there is a relationship between group size and positive affect generation. If something pans out there, I imagine it will be easier to write a compelling paper on the differences between working with a densely intraconnected, loosely interconnected grassroots organization than a vast political party in the city.
Yeah. I'll just take this one day at a time. I think it takes quite a man to admit when he's confused. I have no qualms about saying I don't understand this stuff. I suppose it's fair, too. I haven't taken statistics in many, many years and none of the stuff we covered in undergrad got this advanced. I would even say the same thing about sociological theory. It has been a good number of years since I've spent time studying it; thankfully, I feel like I'm learning something new quite often and I find value in it. I feel that way about statistics too. I just need to get the confusion out of my system first in order to appreciate it.
So, let's leave it at this. I'm going to go on an educational bender. I may be a little hard to find, but I don't know want you to worry. I will be sure to get some air on a regular basis. I will eat. I will probably have a beer or two during some downtime. I will call my mother and encourage her to pray for me, and I will not hesitate to solicit care packages from friends and family. I will be happy with whatever grades I wind up with, but I will do my best to get as high of a grade as possible.




