I participated in the NSF "Computing Education for the 21st Century" (CE21) Community Meeting in New Orleans at the beginning of February. The strong message that we got was, "The rest of STEM is way ahead of you folks in CS -- here's your chance to catch up."
Jan Cuny started the session with the basic premise of CE21. Only 1/3 of computing-related jobs are fillable by 2018 with the students currently studying computing, but 70% of the population (women, under-represented minorities, disabled) are missing from computing. Now, Engineering is as bad as Computing in diversity, but all their trends are positive. All our trends are getting even worse.
She then made the point that was perhaps the most startling to many of the attendees: CE21 requires CS Education Research, a focus on Broadening Participation, and real Education Research. Without all three pieces, proposals will be returned un-reviewed. Now, everyone knew why everyone was in the room. I heard several people complain yesterday, “I’m in education and I don’t know anything about CS” or “I’m doing great things in my CS classes, but I don’t have the time to write it up and I don’t know Education methods.” That was the point of the meeting.
Two moments really stood out for me on the first day of the meeting. Jim Hamos is the CE21 program officer from the Education-side of the house (with Jan Cuny of the CS-side of NSF, and Joan Peckham from the Office of Cyberinfrastructure). He gave a frank talk about how behind CS is in the STEM education game. He said that Engineers are just now figuring out how to do Engineering education in K-12, and they’ve been at it for 20 years. He said that mathematics education is by far the most advanced, in terms of having cognitive developmental models and knowing what makes for effective pedagogical methods for their discipline. Physics is way up there. CS is not yet even on the map. He was clearly speaking to the CPATH/SIGCSE audience when he pointed out all the benefits to higher-education faculty within the STEM disciplines of working with education faculty and researchers and with high school teachers. He told us that we had to improve our own higher-ed classes (and that these partnerships will help), that we had to engage in qualitative methods to study the partnerships, and that we had to revise our standards for promotion and tenure to value scholarly contributions to advancing STEM education. He didn’t pull his punches: CS has not yet been in the game, and CE21 is providing the resources and motivation to start.
There were then two parallel plenary sessions. One was aimed at the CS Education audience, and featured Joan Ferrini-Mundy. Joan is a mathematics education researcher, and she was able to explain what they have in math education that we need in CS Ed.
I was in a panel in the other plenary session, aimed at the education researchers. Lucy Sanders convened the panel of Valerie Barr, Jane Margolis, Owen Astrachan, and me, to explain the issues of CS education to education researchers. My favorite moment on that panel was when Jane answered the question (which I’m paraphrasing, describing the question that Jane answered, not necessarily what she was asked), “So why is CS education in such a bad shape with regards to diversity?” She said that it’s culture. She said that we are in such a “Male Day” today, especially with regard to technology and computer science. She described CS as “pumped-up” and “testosterone-filled.” It’s all about keeping up, working huge numbers of hours, always trying out the latest and greatest. We always emphasize to our students about how they have to be constantly working to learn the new things, to be on top of the latest developments. She asked, “How do you make long-term, family-oriented, stable life decisions in that culture?” It isn’t an inviting culture if you are thinking about those values. She really made her point for me when she pointed out that medicine is also long hours, go-go-go, and always about staying on top of the latest advances — but for the purpose of caring and supporting the community. Computer science doesn’t advance those values. It’s there, but it’s not front-and-center like it is in medicine. Here’s how I interpreted Jane’s comment. In computing, the primary motivator is the start-up and the IPO. In medicine, it’s about people. That makes the effort worth it, and changes the culture equation.
I had a lot of enlightening, and sometimes scary, meetings at CE21. I got a chance to talk to Chris Hundhausen who told me about his SIGCSE 2011 paper on building programming environments for blind students. Susan Gerhart has challenged our community of computing educators to think about how our pedagogical tools can be used with visually disabled students. She’s completely right — we tend to use graphical notations (as in Alice, Scratch, and Squeak eToys) to improve students’ ability to get started with computing, but those are useless for a blind student.Chris is actually working on several different ideas including audio debuggers and manipulatives (physical artifacts) for representing the programs. Chris said that his colllaborator, Andreas Stefik (Chris’ former student) is excellent at empirical methods, so all his design ideas are carefully developed with lots of trials. The paper includes results from a test of the whole suite of tools.
I met Aman Yadav, an assistant professor in educational studies at Purdue. He’s actually teaching a CS methods course (how to teach CS effectively), in a program that teaches pre-service high school teachers! How exciting! He only has one student. Aman says that he doesn’t know how many semester that they can afford to offer the class with so few students. The one teacher he has is a math education major, who is taking a minor in CS education. Nobody there is going after CS education as their main focus. We were sitting at breakfast on the second day with Wayne Summers, my collleague at Columbus State University where they have a program to give teachers an “endorsement” (a kind of certification that comes after a teacher’s initial certificate in teaching math, science, business, or whatever) to teach high school computer science. He had one student, but she dropped out in the first semester. In my Computing Education Blog, I mentioned that UTeach has been in existence for 14 years, but only has had 7 graduates who focused on teaching computer science.
I believe that this is our greatest challenge to CS10K, the NSF goal to have 10,000 teachers in 10,000 high schools able to teach quality computer science by 2015. (We have about 2,000 APCS teachers today.) Where will we get the teachers? What’s our argument for becoming a computer science teacher in high school? CE21 might help us create more in-service and pre-service teacher education programs, but it’s not like we are filling our current capacity. Where are we going to get those 8,000 additional people who want to become computer science high school teachers?
My students have been studying some of these issues. Brian Dorn found in his work that graphics designers have all kinds of unfavorable opinions about computer scientists. ”They’re old and nerdy and boring,” said one study participant. Why should high school teachers have more favorable impression of computer scientists? Lijun Ni has pointed out that high school teachers worry (correctly) about taking on a field that requires constant updating and new learning. Why would they want to become one?
For me, this is the biggest challenge of CS10K. If we build it (a CS teacher development program), they likely won’t come. How do we change the odds?
Materials from the meeting are now available. Overall: It was a Wow meeting. I learned a lot, met interesting people, had fascinating discussions. Looking forward to seeing what comes out of CE21!