Wednesday, September 07, 2011

Thinking Again about Group Work

Sorry about the very lame sparse posting. I blame poor time management on everyone's parts -- including the people who invite me to meetings so dull that they are unworthy of blogging.

Did I mention that I am taking a graduate-level operations research class this semester? So far class has met six times. We have covered: Why linear programming is an important tool in business (prof talked about when Real Businesses have had problems that have been solved by linear programming), how to solve linear programming problems with two variables by graphing in the plane, and how to solve linear programming problems with an Excel plug-in. Nothing yet about the nuts-and-bolts of the method of solution; just "put it into Excel and click here." All this in just six class periods! The math department teaches this to freshmen faster and in more depth. I totally went to graduate school in the wrong field.

So today we spent the last 15 minutes of class working in groups on a linear programming problem isomorphic to the ones we'd seen in class and easier than some of them from the homework. It was one of these, "You are making product A and product B, which have contributions of profit of X_A and X_B. It takes two machines to make these products. Machine 1 takes t1_A to make product A and t1_B to make product B. Machine 2 takes t2_A to make product A and t2_B to make product B. How many of each product should you make?" The sort of thing that the math department would expect freshmen to be able to do. And we had to work in groups.

I have figured out why working in groups in classes bugs me so much. There is no good reason for me to work with other people on this problem. I can solve it faster myself than dealing with my groupmates being wrong and confused. More generally, this problem does not require any specialized skills beyond what we have been taught in class, and all of us in class are assumed to have roughly the same skill level relating to the course material. Also, this was a small problem; there was not more work than one person had time for.

In contrast, I work in groups at work all the time and no one tells me that I should be working in groups. If I need to do something using TAU to profile some code, I will work with someone who knows about TAU. If I need to work on gathering a lot of information, I'll work with someone else who also needs the information, and we'll split up the list of people to contact. We work in groups at work because that is the best way to get the work done.