Independent Studies for One and All

I haven't blogged any this week because like anything that goes up, it must also come down. I finished last week on a decent high and Saturday was my birthday so I thought it was all going well. And then NanoGEARS caught some daycare baby plague and promptly passed it on to me. She seems to be fine now though, thanks for asking.

I'm somewhat disappointed that I didn't post earlier this week because writing usually clears my head and gives me some time to relax. And I know all of you were fretting that my daily post wasn't showing up in your RSS reader so I want to apologize to my ten(s) of fans out there.

Now, on to the topic at hand.

There was an Op/Ed in the New York Times on letting kids make up their own high school curriculum by Susan Engel that had some very interesting things in it. Essentially, 8 students who spanned the spectrum of high school success got to plan their own curriculum for a semester. It talked about the increased student motivation and how they set goals that were at least as high as a typical AP class. It's an interesting study and I think has some potential as an educational method.

One place that I would like to see this curriculum model applied to is at the graduate level. The reason I say this is because I took traditional courses (and wrote a MS thesis) at UGU whereas, at OldEuropeU there are no defined classes for your PhD. So I've seen both types of systems personally which both have some definite pros and cons. I'll tackle the traditional system first.

I ended up taking 8 classes at UGU during my MS work to satisfy the coursework part. Of those 8 courses, I only really learned anything in 3 of them and none of that stuff I learned is knowledge I really use now. I still remember the stuff from those 3 should I happen to need it but I really couldn't tell you what was useful in those other 5 classes. I remember doing work for them (and a lot of work for that one stupid graduate math course that was the biggest waste of....) and taking tests and getting good grades but nothing really stuck with me.

Taking 5 classes that were essentially useless for my previous research and my current research (I made a big left turn when starting my PhD) seems like a lot of wasted time and effort on my part and the part of those professors. It seems like going through the motions for the sake of going through the motions. I read all these blogs and hear all of these professors discuss how much work their doing for teaching, research, and proposals. Maybe not teaching unnecessary classes is the way to go.

On the contrary, at OldEuropeU, I didn't have any classes for my PhD work. This is a blessing and a curse. The good part about that is you don't have to take any unnecessary courses. You can spent all of your time doing research, which is what you and your advisors really want anyway. Without classes your first two years, you can really make headway on your research. This will hopefully translate to more papers, more conferences, and a competitive advantage for you when you finish your PhD. Also, you really focus on the things you need to learn for your research. You don't get sidetracked by "the new shiny" that occurs when you do take a class you like.

There are some extreme downsides too. You need to be 100% self motivated to learn something. Just going through the motions so you didn't ruin your graduate GPA isn't enough. You end up lacking in something. In some cases, this is diversity, i.e. you weren't exposed to different concepts and different ideas because you had to take new classes. In other cases, you lack a basic and fundamental understand of the field. This is where I lack because of not taking PhD classes. I started out with some basic assumptions which allowed me to skip over the whole fundamental background in the field where my PhD subject rests and now it's hard to go back over to learn that.

I think graduate classes should largely be based around independent studies in predominately your research field. For instance, I took a controls class that I couldn't tell you a thing from it. Yet, during both my MS and PhD work, I had to build working control systems. In both cases, I managed to get them working but that was mostly though trial and error. Rather than taking a book-heavy course on controls, it would have been much more useful for my class to be a study on this particular control system. In both cases, I didn't need an advanced understanding to get it working nor for papers and such, but it would have been to my betterment to gone through it more thoroughly. Once I had a real system that I was using, then I could apply all of those textbook things like loop shaping, and overshoot minimizing, etc etc. It wouldn't have been entirely necessary for my research but it would have expanded my actual knowledge in an area that I should have known something about.

I think most graduate students could pick about 4-6 topics in their research and do independent studies more thoroughly on the subject than if they took a class on it and just went over the theory.

What do you think? Would you rather have defined your own series of independent studies than taken classes? How many classes during your graduate work have been useful for you and your research?