At the NSF CE21 Community Meeting: We have such a long way to go
February 1, 2011 at 8:08 am 21 comments
I’m in New Orleans at the first NSF Computing Education in the 21st Century (CE21) community meeting. Three communities have been invited to this meeting:
- People funded under the old CPATH (CISE Pathways to reinvigorate undergraduate education) program. These are people that you typically see at the SIGCSE Symposium, and others interested in CS education invited by NSF program officer Harriet Taylor. I walked into the hotel to see Mary Beth Rosson, Jack Carroll, and Margaret Burnett, whom I think of as CHI (Computer-Human Interaction) researchers who focus on end-user programming.
- People funded by the old BPC (Broadening Participation in Computing) program. Alex Reppening (of AgentSheets) was chatting with Mary Beth and company when I came in. I sat on a panel with Jane Margolis and Lucy Sanders, and co-facilitated a session with Joanna Goode yesterday.
- Education researchers, people who have been or who are now funded by the NSF EHR (Education and Human Resources) Directorate. So, these are people I might see at AERA (American Educational Research Association Annual Meeting) or ICLS (International Conference of the Learning Sciences).
The day for me was enormously intellectually stimulating, but also constantly confusing. There are 400 people here! I know many people here, but from these different communities. I went for a run yesterday morning with Tom McKlin (who is our external evaluator on “Georgia Computes!”) and Cameron Wilson (with ACM on the Educational Policy Committee). I go to NCWIT and BPC meetings with Tom all the time, and I go to SIGCSE and ACM Ed Board meetings with Cameron all the time, and it never occurred to me that the two of them didn’t know each other previously. Don’t we all go to the same meetings? But CE21 is merging communities.
There were so many great moments yesterday, and I don’t have much time now before today’s sessions start. But three people yesterday told me that I had to blog on the meeting, so I want to keep some of that promise this morning.
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 yesterday. 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.
The rest of the day was a dynamic schedule of meetings being defined, and rooms shifting as the crowds grew large, and fascinating chats everywhere. There was so much more that I could blog about, from BPC Alliances meetings; meetings with the folks of the CAITE alliance on how we support statewide pipelines in the future (I think “California Computes!” got its start in the hallways yesterday); great discussions with researchers like Jill Denner, Christy McGuire, Jeff Forbes, and Betsy DiSalvo on their work directions; explaining the dissertation studies of Allison Elliott Tew, Brian Dorn, and Mike Hewner over-and-over (Lijun is here, so she could explain her own); so many people coming to our “Georgia Computes!” poster; and so much discussion about CS10K and the new APCS effort. (I got teased at dinner last night, for talking in a session about the great work of Barbara Ericson, co-PI on “Georgia Computes!” and Director of CS Outreach at Georgia Tech…and not mentioning that she’s my wife. The ladies at the dinner said that they thought I did the right thing, but the guys thought I was withholding information.) And day two starts in 30 minutes.
Entry filed under: Uncategorized. Tags: ACM Ed Board, BPC, CE21, CPATH, GaComputes, NCWIT, SIGCSE.
Computing Ed Research - Guzdial's Take 
1.
Victor Eijkhout | February 1, 2011 at 8:15 pm
I know you’re quoting but I really have to ask
Math education advanced? Having figured out how to do engineering education? Physics being “up there”?
Where did the US score again in that international survey of math/science skills of school children?
The quoted statement sounds like blowing smoke to me.
2.
Mark Guzdial | February 1, 2011 at 8:57 pm
Of course, there’s a difference between knowing how to do something (e.g., through research studies) and making it common practice. And if we look at performance in international studies, CS is FAR behind math and physics.
3.
Andy Kuemmel | February 2, 2011 at 10:22 am
Thanks for a great post Mark! Please, when you get time, post on day 2 !
One question: Do you see a future demand for more graduate students to research CS Education? Are there any programs out there in this area right now?
4.
Mark Guzdial | February 2, 2011 at 10:35 am
CE21 clearly creates a demand for more graduate students in CS Education research — who’s going to do all the work? Will blog more on CE21 Meeting as soon as I clear up the backlog of email, classes (teaching Senior Design and MediaComp Data Structures), and home.
5.
gasstationwithoutpumps | February 2, 2011 at 11:06 am
I think that the “demand” is going to remain small, though the “need” will grow. Computer science researchers will continue to turn up their noses at education research, and education researchers will continue to ignore computer science. The niche of computer science education research will remain tiny for the next decade or two, despite one short-term push by one agency to “do something”.
6.
Mark Guzdial | February 2, 2011 at 2:46 pm
The CE21 criteria for funding say “CS Education Research + Education Research + Broadening Participation” — not one of those, all three. Nobody who gets CE21 funding will be able to avoid doing CS Education Research. If you want graduate students to work on your CE21 project, they will be doing some CS Education Research.
7.
gasstationwithoutpumps | February 2, 2011 at 4:55 pm
There is $25 million (spread over 2 years) in CE21, which is not a negligible amount of money, but it is only 2% of the approx $1300 million for computer and information science and engineering at NSF for the same period, so don’t expect everybody to suddenly jump on the education badwagon.
8.
The Revolution Will Not Be Tweeted « Computing Education Blog | February 3, 2011 at 10:24 am
[…] blog more on CE21 as I can. After Jane Margolis made her impassioned plea for a change in culture in CS and CS education, Owen Astrachan said that he has been teaching that computing is about […]
9.
Mark Miller | February 8, 2011 at 5:15 pm
I agree with Jane that it’s about culture, but if we delve deeper, I think the problem becomes broader than that.
One of the things I used to reflect on is how the languages that are often considered “esoteric” and “irrelevant” (I’m referring Lisp, Smalltalk, etc.) have had features for decades which “modern” languages (at least ones developers have become more conscious of because they’re “new”) have only recently brought to light for many people. Interesting how you can take something old and make it “new” again by just changing its presentation, and the features that surround it… I use this as an analogy for the problem Jane cited about “always having to keep up.” The reason people always have to keep up in this industry is because the industry works incrementally. Nothing is ever fleshed out, thought through completely. It’s based on functionality that sells in the moment, often based on how to “leverage” existing technology to adapt and cope with new demands. It’s not designed to solve problems 20, 50, 100 years into the future. The expectation is that you’ll be changing your language (and now changing your VM), changing your hardware, changing your operating system, and changing your database system every 5-10 years. There’s a “throw away” mentality that is rampant. This goes back decades, and was highlighted in the worry about Y2K. It showed the contradiction in this thinking, because lots of places tried to make their old systems hang on for as long as possible, despite the expectations of programmers (if they even thought about this issue) that their systems would be replaced as a matter of course before the threshold was reached:
The one industry I remember hearing was well prepared for this was the financial sector, because they *had* to plan decades into the future, with 30-year bonds, and 30-year mortgages, etc.
Maybe Jane is on the right track when she suggests there needs to be more thought devoted to how to contribute to the community. That should be elaborated upon: “What does that mean, though?” In my view, the field could go a long way in that direction if it would focus more attention on what human beings are, and how to create systems that help us solve problems that involve using computers. I’m trying to word this explicitly, because I don’t just mean, “Set up a web site that promotes community involvement.” That to me is just continuing the problems that Jane complains about in some key areas. I mean a system that promotes its own improvement so that new problems can be solved with it in the future, and people won’t have to spend months learning new systems every year or so. The idea being that you can build on what’s already been created in more powerful ways, because it wasn’t designed for just one purpose at the start. It involves getting into what a lot of CS doesn’t focus on, which is having experiences that deal with thinking about “What is computing?”, and creating models that try to answer that question. The tools used being engineering, mathematics, and science. It involves thinking more broadly about what we’re creating; not just an app., but a new way to think about solving problems, at very least, or a new way of seeing what you have.
This is a perspective I’m continuing to learn about myself. It’s not something I claim expertise in, but now that I see the deficiencies in the field as it is, I can try to contrast it with something better.
Having said this, this is not even close to a view that’s held by industry, nor academia, by and large. The only thing most of industry knows is the engineering perspective, because the sole goal is to build things that do something practical in the moment. The building material and the method for building the thing is only based on other pragmatic or ideological objectives. This is driven by the “black box” model of computing, where it’s “just a box that does something.” It’s derived from an Industrial Age model of what makes up a machine, and what its purpose should be. Like I was saying, I think CS could go a long way to addressing Jane’s complaints if it would focus on reorienting the thinking of people about this, but what I’m realizing as I’m writing this is it can’t do it alone. The problems with CS education are in good part a measure of the deficiencies in how the other technical disciplines are taught.
What’s complained about in your post is not just endemic to start-ups, but most of IT generally: Everything is a project. It’s often coping with something that’s already been created, and sometimes what’s “already been created,” in the context of what I’ve discussed here, are new technologies. “We need to take this whatsit, whosit, and fitsit, and make it do X, Y, and Z.” The technology is static for the project. You just need to learn how to take advantage of its features to work towards a singular goal, which is the project’s objective. In other words, cobble it together and add more to the pile…
10.
Pierre Bierre | February 20, 2011 at 4:34 pm
Mark – You showed me around the MIT AI lab in 1973 just because I walked in off the street and was curious. Those 20 min. you spent with me were life-changing, as it showed me the human side of CS…welcoming, open, filled with possibilities.
In the Algorithmic Geometry HS course my team is piloting, we are definitely interested in laying a long-term foundation for spatial informatics(math/CS problem-solving), one that will be broadly applicable across most science and engineering fields for several decades.
But does teaching 21st century HS mathematics from an information-processing framework qualify as CE21?? Well, maybe not in 2011. Hooray for Jan C. getting $25M at CISE.
Joan Ferrini-Mundy…where is EHR k12 funding for teaching software math? All those Education PhDs who are experts on “how math is learned” are stuck in the pre-computational paradigm (paper and pencil analysis). They will probably rule the roost until a next generation of Ed PhDs comes along who learned in Math class how to automate their solution in software.
Or, people at NSF might decide to actually take some risks and get on-board paradigm shifts earlier in the game, to help the U.S. remain competitive.
Mark Miller is right, the CS Ed field needs to concentrate on presenting welcoming, human-oriented career options that won’t fizzle out when women leave to have their first baby, nor for the majority when they hit 45. The focus of content wants to be on 1st principles with a long shelf-life. If I were managing CE21, everyone would have to write a 1-page essay outlining the 1st principles of our field of science.
I’ve worked in the software industry for 25 years, and don’t perceive it as being “testosterone-laced” as framed by Jane. It’s much more well-suited to the androgenous. My perspective from Algorithmic Geometry teaching is that girls have completely closed the Math gap (this year we have 8 girls and 3 boys). Hopefully, this trend will register on the radar at NSF when deciding how best to get young people involved with computing.
11.
Ontological Categories in Computer Science « Computing Education Blog | March 10, 2011 at 8:57 am
[…] reminded of Jim Hamos’ talk at the CE21 meeting: We have such a long way to go. Physics is at the point where they can develop ontologies that […]
12.
A Definition of Computational Thinking from Jeanette Wing « Computing Education Blog | March 22, 2011 at 6:50 am
[…] the need for a good, authoritative definition of computational thinking. I told her about the CE21 Community Meeting where I saw K-12 evaluators looking for a definition that they could use to develop an assessment […]
13.
CE21 Advice: Go beyond your classroom « Computing Education Blog | March 29, 2011 at 9:16 am
[…] the CE21 Community Meeting, I heard people talking about the great things that they’re doing in their classroom, and how […]
14.
Show Me The Code « Computing Education Blog | April 18, 2011 at 10:32 am
[…] problems, as opposed to reading solutions and learning to learn from reading. I’m preparing my CE21 proposal now, and spending a lot of time learning what educational psychologists know about how to write […]
15.
We get the chance to beat the book: NSF CE21 funded CSLearning4U « Computing Education Blog | September 23, 2011 at 10:39 am
[…] get the chance to beat the book for CS learning! Our NSF CE21 (Computing Education in the 21st Century) proposal was funded for about $990K from October 1, 2011 to September 30, 2013. The goal of this […]
16.
The long tail may not hit a target: High school teachers « Computing Education Blog | January 31, 2012 at 10:00 am
[…] of all, was the audience for the Stanford CS classes like the audience of potential CS10K teachers? I’m not convinced. First, when I read the comments to posts about the the Stanford […]
17.
Doubts of my students: Expert teaching is no better than good-enough teaching « Computing Education Blog | February 24, 2012 at 10:27 am
[…] knowledge of the discipline and knowledge of learning sciences?” Since I don’t know how to measure knowledge of CS well, nor how to measure CS PCK, I have two unknowns, so I can’t really answer the […]
18.
BBC News – Google funds computer teachers and Raspberry Pis in England « Computing Education Blog | June 7, 2012 at 6:28 am
[…] like NSF’s CS10K, so Google is trying to play that role. (Maybe the UK should try to clone Jan Cuny, who has done more to improve computing education in the United States than anyone I can think of?) […]
19.
Brief Trip Report on ICER 2012: Answering the global needs for computing education research « Computing Education Blog | September 18, 2012 at 11:44 am
[…] perspective. There was little from ICER and the computing education research community. The problems that we need solved will require work from both communities/disciplines/fields. How do we get there? Share this:EmailDiggRedditFacebookPrintStumbleUponTwitterLike this:LikeBe […]
20.
What is the current state of high school computer science professional development? The results of the UChicago Landscape Study « Computing Education Blog | January 14, 2013 at 8:00 am
[…] I find the results a little depressing. The folks at UChicago who do the study compare us to professional development in Science or Mathematics, and we don’t much look like that. We have such a long way to go. […]
21.
Visit from Farnam Jahanian, AD for CISE at NSF | Computing Education Blog | May 22, 2013 at 1:05 am
[…] his talk, he made an explicit argument which I’ve heard Jan Cuny make, but hadn’t heard an NSF AD make […]