Thursday, July 28, 2011

Preparing for a new semester

My strength and fitness are improving daily with my rides to campus. Last monday night, I went to watch my summer league team play. They had 1 sub to begin with but then he had to leave and another player may have broken a bone in his hand, leaving them with only six players. It was so windy as a storm blew through at the beginning of the game, that I thought about trying to play a little. If I could safely huck (which I can't yet), I was thinking that I could walk around the field, and be available to huck the disk and then let my team play a 6-person zone. Fortunately, the wind calmed down and those crazy thoughts passed. I can't waste 7 weeks of work just because I feel a little better.

I got the slightly frustrating news today that I will be in the boot for another week. Really a week and a half since I see the surgeon again on Monday of the following week at exactly 8 weeks after surgery. I was hoping to be ahead of schedule enough to get some boot-free walking in, but the PT is not cleared to make that sort of decision. As I note above, I have to avoid being stupid at this point. I am at 6 weeks and 3 days post surgery and that is just not enough time to get out of the boot...

In between the hectic preparations of a poster for a conference, an accelerated search for a position for this fall, and trying to get a new course proposal ready for submission, I have been thinking about my fall class. It is a senior level undergraduate course in mixed models. Basically it involves statistical methods for correlated data that might come from measuring the same person repeatedly or students in the same classroom or spatially, temporally, or spatio-temporally correlated observations.

I try to let my teaching be influenced by recent experiences, be they research or life experiences. I got a push in sharing my life with my students by Tim Robinson at the University of Wyoming. We team-taught an engineering stats course and he had an almost folksy presentation style (in a good way) with his Virginia accent and stories about his family and his research projects permeating lectures. This made a class 80 students feel like a much smaller class setting. When I took over the class, I wasn't ready to bring that familiarity into my teaching but have slowly gotten more relaxed when I am teaching and more willing to share what I am up to with my students. My classes are heavy on my research projects as examples of hard learned lessons in various areas. But I am also willing to let students see that we all try to balance work, family, and fun - letting them know that sometimes when I am not in my office I am out hiking, biking or skiing. I am in my office enough hours with my door open to also try to set a strong example for how much work is required to do this job. This fall, my achilles tendon rupture and rehab will be a prominent part of my class and discussions.

During my surgery, I discovered the "new" infra-red based technology to get body temperature readings non-invasively. It uses the temperature of the blood vessels in the temple area. I looked around and discovered that you can get them relatively cheaply ($30 for a decently rated one, $10 for one "fresh" from China) and even more cheap is the hardware version of the tool - a laser-based temperature gauge ($10). I am yet to get a reasonable body temperature reading from the laser temperature gauge but it may be possible; the "real" gauge seems to work well although I haven't bothered with the rectal temperature calibration test. Regardless, I am going to try to integrate the real and alternative temperature gauges into my class in the fall. I am hoping to use them and the idea of doing a study of body temperatures to motivate some of the big issues in the class during the first few lectures. I plan to have them design, collect, and analyze a study using the different gauges. I have a few different ideas that would be fun but am trying to remain open to whatever ideas they might come up with in class.

I also starting to find some interesting articles related to achilles tendon problems for my class in the fall. One is here which contains a summary of possibly more interesting studies but with statistical methods that might not be as good a fit with the class. I like this one because it has nice graphics and is a simple repeated measures analysis with a single(sort-of) quantitative explanatory variable. It also illustrates why researchers might be interested in taking repeated measures instead of one measure per "subject" - they are working with cadaver's legs. There are a couple of issues with how things are reported and the model but I'll leave that for my students to sort out on their homeworks...

I was also entertained to find an achilles related data set in R in the repolr package. It is a repeated measures, ordinal response data set with a 3-level response of level of recovery. I have been working in this area and looked at that package last spring, prior to my injury, and didn't think too much of the data set back then. I will definitely be revisiting it later.

I find that I am struggling a bit to write this to whatever audience I have - maybe I should split my posts into different topic areas. I mix my life and work up together, but maybe my posts should have more separation. I am beginning to think that blogging is as much for yourself as it is for other people... it certainly was in my first days post-surgery.

4 comments:

  1. Are they still looking for a 216 supervisor?

    ReplyDelete
  2. Interesting, though the penultimate paragraph could use some lay translation

    ReplyDelete
  3. Rory - we hired a new 216 supervisor.
    Sam - After looking up what penultimate was to find out which paragraph you were talking about... R is a statistical software platform, repolr is a suite of functions and data that someone wrote to add to R, and ordinal data is slightly more complicated to analyze than "regular" quantitative data. I might use the data set they provided as we develop some graphical techniques to explore the results from those models.

    ReplyDelete
  4. I agree with the last sentence, Mark!

    ReplyDelete