Sunday, October 04, 2009

Things I've Been Doing Instead of Statistics

  1. I've spent the past few days cleaning the house. By the end of the weekend it will be normal-people dirty. Yes, cleaning products whose labels say, "IT IS A VIOLATION OF FEDERAL LAW TO USE THIS PRODUCT IN A MANNER INCONSISTENT WITH ITS LABELING" do a lot better job of cleaning than the ones that claim to be good for the environment. My only complaint about the packaging of the heavy-duty acid etching bathroom cleanser is that the strength is given as a percentage, not in molarity (or molality). I'd like to be able to calculate how much oven cleaner or drain clearer to use to neutralize it.

  2. I now sort of have a cat. Meet Cayley.

    I have a former student who tends to the feral cat colony in the student slum neighborhood across the street from campus. I have sponsored the mom cat, who has been fixed since this photo was taken (and the kittens have been adopted out). My cat is back living on the streets near campus. I am way, way, way, way too allergic to have a cat live in my house. Yes, that's right. In addition to having a Homeless Relative (who is now living near Rutgers, I think), I also have a Homeless Cat.

  3. Working on planning for my summer job for 2010. Since the group will have a large number of young men from all over the world (Europe, Middle East, Asia, maybe Africa), I'm trying to find a good all-ages venue for viewing a sporting event on TV on July 11, 2010. If you don't know what I'm talking about, then you probably don't deal with large groups of non-Americans during the summers of even-numbered years.

  4. Just got an email from a group of honors students who were supposed to meet with me earlier this week. Trying to figure out how to avoid the train wreck that they are driving themselves towards.

  5. And the title of this post lies. I did a bit of the statistics homework. Since I've stopped paying attention, I don't really know what I'm doing. Especially since I didn't come into this class knowing much of anything at all about statistics. One question asked something along the lines of "what sort of test would you use to determine whether or not the following groups were all drawn from the same population and all have the same mean?" Since the data was crazily, ridiculously, exaggeratedly, obviously non-normal in the extreme (and not amenable to transformation), I googled "non-parametric ANOVA" and answered the question "Kruskal-Wallis." Stuck on one of the other questions though, as it asks why the p value is so high for the data provided in the question, and the software happily chirps that F is like a bazillion and p is miniscule.