Dramatic Learning Experience

Well, I'm a little behind in my posting, so the next couple of posts are going to be out of order with respect to what we covered (or will cover) in class, so please bear with me.

The most dramatic learning experience for me was while I was at Pitt working on my masters. It was in the first semester of the second year of the two year program, and I was all ready to get out of there with my masters and call myself done with school FOREVER. After two semesters of taking classes that I only marginally understood what was going on, I was forced to take a Statistical Consulting class as part of my degree requirement.

This class is set up as follows: For the first two weeks, you talk about consulting and meeting clients in broad terms, and then the instructors says "You have a meeting on this date, and at this time". The client then lays out his statistical problem and you attempt to solve it using whatever knowledge you may have accumulated over your brief life in statistics (which was about a year for me). The class, which had a size of 6, then presented their project to the class and we were all able to comment or ask questions of the consultant to not only further his knowledge but help them with their project.

My first client had an interesting problem, but no big lights went on in my mind and I struggled to get through the project (it was a reliability problem, and I don't care about that topic that much). However, I received a second client a short time later and the project that I was given KICKED ASS.

To analyze the data correctly, I had to do a test of equivalence on a longitudinal, random coefficient model. For all the non-statisticians, it's a difficult model to conceptualize let alone analyze, and admittedly I did not analyze it correctly (I did a test of difference, not equivalence, but spun it well ;-) ).

Now, I'd be lying if I said this particular topic was "the most dramatic learning experience", but as part of this class it was. The whole class was that experience because it showed me a bunch of things. First of all, it showed me how much I actually learned in the first year of grad school. But it also showed me that I actually like statistics, and love being able to put real-life problems into mathematical solutions and use "hard-science" constraints to come to and objective decision**.

This realization didn't come into play just from the aforementioned project, but from all of the topics presented in class. It was clear to me that I was not only grasping the projects assigned to me, but the projects' of the other students as well. This allowed my confidence to grow to a point that I was able and willing to submit my analysis plan to my professor, who just happened to be the most intimidating man on earth (I'm not even being sarcastic, I'm still scared of him). And it was this project that showed me that I may actually be good at this statistics thing, and I may actually like it.

So what did I learn from this class. Two things: The first, as I mentioned before, is that I actually like statistics, especially the applied part, and I want to continue in the field. That's important because that'll be my job.

The second is much more important. I learned that for me to truly enjoy something, and to really excel in it, I need to immerse myself in that topic. I found out that my personality is such that I can't window shop for things that I like. I need to go in, take everything for a test drive, and discard the things that I don't like. And once I do find something I like, I need to work to be able to incorporate as an important part of my life.



**The distinction between "squishy-science" and "hard-science" was made by a colleague in her blog. I have a whole posting on how it applies to my field in my head, but have yet been able to sit down and work through some of the philosophical nuances that I need to complete some arguments. Catch her blog here.