An undergraduate degree in Computer Science Education

Barb and I visited the University of Maryland at College Park yesterday, where they are developing an undergraduate degree in computer science education.  They plan to help generate Jan Cuny’s “10K teachers in 10K schools by 2015.”  It’s a joint effort between Jim Purtilo, Associate Dean for Undergraduate Education in the College of Computer, Mathematical and Physical Sciences, and Linda Valli, Chair of the Department of Curriculum & Instruction in the College of Education.  All the right people were at the table: teachers, representatives of local school districts, state Department of Education officials, and state certification officials.

They invited me to give a talk about what the research questions are in computing education — what do we know, and what don’t we know?  (My slides are available.)  They asked Barb to talk about the efforts in Georgia, which have led to an endorsement (an “add-on” certification to an existing certification), a state-wide CS curriculum for high schools, and large professional development effort. The rest of the day was discussion about what they’re trying to do, what haven’t they thought about yet, what they’re trying to figure out.  It was really interesting.

Barb pointed out that Maryland is one of the few states that could pull this off, because they do have certification for teaching computer science.  They can graduate students who are certified already.  In Georgia, you’d have to be certified in something else, then add-on computer science.  However, to do that, they need to have students take not one but two semester-long “methods” courses, on how to teach computer science well.  I’ve heard of a few schools that have CS Methods classes, but only single-semester courses.  Maryland needs to build two semester-long classes.  The content exists, I think — we know a good bit about how to teach computing well.  However, it requires someone who knows the literature well to build such classes from scratch.  Maryland knows that it will have to hire someone to make this new degree program work. They need to find someone who works in computing education research AND can get tenured at a top research institution like Maryland.  That’s a tall order.

I was critiqued at the meeting for not doing enough work in computing education, or maybe, not doing the right work.  One of the state officials asked us how computer science classes in high school correlate to national standards in technology education, since such standards exist.  What technology skills would one develop in taking a computer science course?  I responded with information about ACM’s Education Policy Committee and said that they were looking at those kinds of questions.  She asked why I wasn’t doing that.  I pointed out that I have other things that I’m doing, that also need to be done.  She got really annoyed that I didn’t see this question as critically important, and I overheard her telling others that they have to “make me” develop these matches to technology standards.  (What does that mean?)  I do understand that establishing a match to standards is very important, and I understand that there are many policy issues that are critically important for the advancement of computing education.  It’s also important to figure out how to teaching computing better and to understand what’s going on when someone learns computing.  Not everybody has to do everything.

One of the most interesting open questions still on the table when the meeting ended was who the students will be in this degree program.  They are designing the program so that students can start in computer science, then decide whether to continue in computer science or switch to computer science education. That’s a reasonable plan, but it’s not obvious to me that that scenario is how they’ll find the students who will be the best computer science teachers.  I can imagine students in education, especially science education, who discover a love for computing who would be terrific computer science teachers.  The current plan requires at least nine classes in computer science.  Does a high school computer science teacher need that much computer science?  (These will be the same classes as for the normal computer science major.)  Certainly, those teachers will really know their computer science.  But just as certainly, such a long and rigorous sequence will scare away most education students who might want to try that path.  Who makes the best computer science teacher, the person with a passion for teaching who discovers computing, or the person with a passion for computing who decides to share it?  I don’t know — it’s an open and interesting question.