Growing CS Ed through Schools of Ed, and CT is Unlikely: Report from Oldenburg

I’ve written a couple times now about the workshop I attended at the University of Oldenburg the first week of June. (See the post where I talked about my two weeks in Germany.)  For Blog@CACM, I wrote a post about teaching as collective practice and the workshop I took with Barbara Hofer (see post here). I wrote about learning about teacher beliefs and self-efficacy from Helenrose Fives here (see post).

Before we left for the workshop, I got to spend time with Ira Diethelm at the University of Oldenburg and one of her students.  Ira is one of at least 16 (that Ira could count) CS Education professors in Germany.  Ira works with pre-service teachers, in-service teachers, and graduate students.  Her graduate students build outreach efforts and curricula as part of their research, then roll them out and provide resources to teachers.  It’s remarkable what Ira is doing, and I understand that the other German CS Ed professors do similar things.  I came away with a new insight: If we want to bootstrap and sustain CS Education in the United States, we should fund several endowed chairs of CS Education at top Schools of Education.  Eventually, we have to have pre-service computing education programs if we want to make CS education sustainable (see that post here).  Creating these endowed chairs gives us the opportunity to create positions like Ira’s in the United States.

Overall, the workshop was a terrific experience.  The PhD student work was fascinating, and I enjoyed discussing their research with them. It was great to hear about German research perspectives that I hadn’t previously, like the Model of Educational Reconstruction that informs science education (see paper here). Barbara and Helenrose were only two of a several outstanding international education researchers who attended.  As I mentioned to Pat Alexander (who has a lengthy Wikipedia page of her accomplishments), I enjoyed being able to wallow in educational psychology for a week, because I so rarely get to do that.  I gave a talk on three of our projects related to the theme of developing teachers: on Lijun Ni’s work on teacher identity, on the Disciplinary Commons for Computing Education, and on our ebook for preparing CS teachers.  (See Slideshare here.)

The response to my talk was fascinating.  Some of the German mathematics education researchers are deeply opposed to computing education in schools. (I suspect that more than one of them completely skipped my talk because they are so opposed.)  “Computing education keeps stealing from mathematics teachers, and learning mathematics is more important.”  At my talk, Pat Alexander asked me the same question that Peter Elias asked Alan Perlis in 1961, “Won’t the computer eventually just understand us?  Doesn’t the computer just become invisible and not need to be programmed?”  I told the story about Alan Perlis’s talk and about Michael Mateas’s argument, “There will always be friction.” From the computing educators, I heard a lot of anger. The German computing education researchers feel that other fields squeeze CS out because the they are not willing to allow computing education to take up any time or budget in the curriculum.

Probably the most interesting pushback was against computational thinking.  The educational psychologists thought it was unbelievable that learning computing would in any way impact the way that people think or problem-solve in everyday life.  “Didn’t we believe that once about Latin? and Geometry?” asked Gavin Brown.  The psychologists at the workshop I attended saw a clear argument that we need to introduce computing in high school so that students can see if it’s for them, but not to teach general problem-solving skills. If we really want algorithmic thinking, they can design easier ways to achieve that goal than teaching programming.

We can probably help students to learn about computing in such a way that it might influence problem-solving on the computer. That’s part of Jeanette Wing’s model of Computational Thinking (see her 2010 paper). It’s the “Computational Thinking in Daily Life” part that the psychologists weren’t buying.  That learning about computation helps with computational X is quite reasonable. If you understand what IP addresses are, we can help you to understand DNS problems and to realize that it’s not really that big of a deal if Wikipedia stores your IP address (see story about Erika Poole’s research).  There is evidence that learning one programming language will likely transfer to another one (see Michal Armoni’s paper on transfer from Scratch to a text-based language).  Learning to program is unlikely to influence any problem-solving in everyday life.