Archive for February, 2010
Here at the AP CS Advisory Group meeting this weekend, the first five curriculum developers, teachers, and pilot testers of the “Computer Science: Principles” course definition were named.
- Beth Simon of University of California at San Diego will be teaching 900 students (!) in Fall 2010 using the new AP CS definition. She’ll use Alice with the book by Wanda Dann, Steve Cooper, and Randy Pausch. She’s also planning to use Excel. She’s planning to use a peer instruction model.
- Jody Paul will be teaching this class at Metropolitan State College of Denver. His is an open-enrollment school, so he has no control on pre-requisites of students. He’s planning to focus on connecting students’ life experiences with the learning objectives about computing. He is going to use Scratch and visualization tools.
- Larry Snyder of University of Washington, Seattle, is going to create a new course to parallel his successful fluency with information technology course. His new course will be in Python and will have a heavy emphasis on the Web, to relate computing concepts to a computational phenomenon that students care about.
- Dan Garcia is going to continue develop his course on “Beauty, Joy, and Awe of Computer Science.” His course uses a new version of Scratch called “BYOB” for “Build Your Own Blocks.” BYOB-Scratch uses a Lisp-like computational metaphor, e.g., where lists can contain blocks, and a “Run” block can execute a piece of block/data in a list. Dan’s course already hits most of the items in the new AP CS requirements.
- The fifth pilot tester is Tiffany Barnes of University of North Carolina at Charlotte who wasn’t able to attend the meeting, so I can’t report on her plans. (She’s on leave this semester.)
It’s exciting that the five pilot-testers are going in such different directions, which in itself emphasizes the flexibility in the new requirements. The overall curricular definition is up around 70 pages now — there’s a lot of definition to live up to. What happens next with the AP CS depends a lot on these five. God and the devil are both in the details.
I commented a few weeks ago about Dick Lipton’s interesting blog post about the extinction of universities. The thread has continued there, and the most recent comment is absolutely fascinating — especially the bottom line. This isn’t a new problem. I’m presenting a shortened version here:
…excerpts from Everett Dean Martin’s “The Meaning of Liberal Education”:
“The motives which lead people to seek college education divide the students into three types. First there are the few who love learning. …
A second type of student attends college and university in large numbers. The motive is preparation for a professional career. Many of the best students belong to this type. …
“The third type, the majority of undergraduate students, are for the most part pleasant young men and women of the upper middle class. Their parents are “putting them through college” because it is the expected thing to do. Students of this type enjoy four happy years, largely at public expense, with other young people of their own age in an environment designed to keep them out of mischief. I have no doubt this grown-up kindergarten life is good for them; most of them seem to appreciate it. In later years they remain enthusiastically loyal to Alma Mater, coming back to football games and class reunions and contributing to the support of the college. As alumni their influence is not always on the side of progress in education, but perhaps they make up for this failure in other ways.”
Now about the only leisure class we have in America is the undergraduate student body. It is bad for the morale of any institution to sail under false colors, and colleges are popularly supposed to be educational institutions. The college faculties themselves must to some extent share this popular delusion, or else they would not permit the public to go on believing it. ….”
Martin’s book was published in 1926.
Alan Kay just asked in another comment thread about the motivations for getting more students through CS education courses, perhaps with less than threshold understanding of computing. I’ve had this piece queued up to post in my blog for some time — it’s an article arguing that there are enormous economic impacts of getting more kids through high school. David Brooks has argued similarly that there are enormous impacts for getting more kids through undergraduate. I think Alan’s point is well-taken — what is the threshold below which graduating someone is just dumping someone with a sophomoric level of knowledge on society? Or is graduating someone with some level of knowledge better than graduating fewer (who are above threshold) and flunking out those who are below?
Few people realize the impact that high school dropouts have on a community’s economic, social, and civic health. Business owners and residents—in particular, those without school-aged children—may not be aware that they have much at stake in the success of their local high schools. Indeed, everyone—from car dealers and realtors to bank managers and local business owners—benefits when more students graduate from high school.
Got into a fight yesterday with an engineer. We were both at an advisory board meeting for a new project here at Georgia Tech about teaching middle school science with Lego robotics. It’s a great project — very careful in assessment, using Janet Kolodner’s Learning By Design classroom structures and rituals, with a national advisory board including several engineering faculty who work with Lego robots in outreach programs. I raised the issue of girls and Lego at several points during the day, and at the end of the day, I asked, “Is it too late to consider something other than Lego? Or are we bought in now? Could we use another form of robotics?”
At that point, one of the engineers said, “Wait, I have to call him on this.” Then turned to me, “You seem to be under the misperception that girls don’t like Lego robots. You’re wrong!”
Something in “misperception” and “wrong” got my dander up. I snapped back, “And I’ve got an n of 1500 that says that you’re wrong!” At this point, people between us literally stepped back to avoid the cross-fire.
Janet inserted a comment here, “Well, wait a minute, it all depends on what you do and your approach…” I interrupted her and explained. “We’ve been doing Lego Robots with girls for four years now with Georgia Computes! We find that Lego Robots don’t change girls attitudes about computing, doesn’t make them more interested in STEM fields, and doesn’t make them see themselves as doing computing. We find that PICO Crickets and Scratch do. ” It’s all about increasing the odds of success.
After the exchange, several engineers came up to me to tell me how their outreach program really works, and how their FIRST Robotics team has 30% females. I asked how they knew it was working. “You can just tell!” or “The girls keep coming back!” How can you tell? Why do they keep coming back? Maybe it has nothing to do with the robots?
As I mentioned earlier, I’m the School of Interactive Computing. We do a lot of qualitative analysis around here. A commonly heard phrase in our hallways is, “The plural of anecdote is not data.“ We’re pretty careful with our data collection in “Georgia Computes!” We do surveys before and after each activity. We use an external evaluator, to try to avoid the kids telling us what they think we want to hear. We regularly use focus groups to spot check our findings.
Janet’s right, of course. Some of the engineers are probably right, and their programs are really working. From what I can tell, none of them that I spoke with know that for sure because they are not trying to evaluate things carefully. They have a bunch of anecdotes.
I do believe that one can create an excellent Lego Robotics outreach program that engages girls. Our data suggest that it doesn’t happen automatically with the default curriculum. You have to do something special, and it would be great if someone would do the studies to figure out what that something special is so that we could all be doing it. Let’s change the default curriculum.
The question I’m raising here about Lego Robotics for girls is the same one that I raised earlier about CS1. How would you know if it’s a bad idea? How do you know that it’s just not working, given any random set of students, teacher, classroom, textbook, and curriculum? The data on CS1 support the claim that any particular offering of CS1, with any popular and well-used textbook and curriculum, will probably result in a 30-50% failure rate. You have to do something unusual to beat those numbers. Further, I think that the data support the claim: “Given the number of studies, and the large number of subjects, the probability of engaging girls with Lego Robotics is low.” Is it possible? Sure! It would be great to do it right at scale! But any one person’s experience with their outreach program or FIRST Robotics team is not going to change that statement.
The March 2010 Communications of the ACM (CACM) includes publication of two Blog@CACM pieces, a sort of point-counterpoint. CACM published my piece about “How we teach computer science is wrong,” where I argued that dumping students in front of a speeding compiler is not the best way to ramp students up into computing, and that we might think about instructional design mechanisms liked worked examples. CACM also published Judy Robertson’s piece “Introductory Computer Science Lessons — Take Heart!” where she argued that what we actually do introductory computing actually has the right pieces that good instructional design research recommends. When I heard that they were going to publish these two pieces together, I thought it was a great idea.
The title they chose was, “Too much programming too soon.” I think it’s really about the definition of “programming.” I do think a novice facing an empty edit buffer in an IDE is an awful and scary way to get started with computing. However, I deeply believe that programming is a wonderful part of computer science, but programming more broadly than “Debugging a blank sheet of paper.” It’s creative, powerful, awesome, and often surprising. There are lots of ways of getting started with programming that are much less scary, such as Squeak Etoys, Alice, and Scratch. I also think that we should explore reading examples, modifying existing code, debugging code, and new kinds of activities where students do limited text programming, some form of “reduced cognitive load” activities. We need broader definitions of what “programming” means.
I first learned that information technology in Malaysia is female-dominated at an NCWIT meeting last year, which I blogged on last May. (A blog post about which Skud was pretty unhappy.) Now, a new book is out describing this phenomenon: “Masculinity, Power and Technology: A Malaysian Ethnography,” What I find most interesting about this is the sharp contrast with prevailing attitudes here. We in the West often see technology as obviously masculine — even our pre-teen Girl Scouts tell us that. Yet, the Malaysian experience points out that the relationship is constructed, not innate. This gives me hope that we can correct the misperception. A relationship constructed can also be re-constructed.
“In the U.S., technology and masculinity are very connected, which is not the case in Malaysia,” said Ulf Mellstrom, a professor of gender and technology at Luleå University of Technology in Sweden and a Clayman faculty research fellow, who discussed the topic at a presentation called The Intersection of Gender, Race and Cultural Boundaries: or Why is Computer Science in Malaysia Dominated by Women? “In a short time, booming industrialization has created new opportunities for women while transforming and reforming established society.”
Network World News had an article that several people sent to me, about the enormous growth of jobs for graduates with CS degrees.
“I think the job market is what’s driving the growth,” says Professor Bruce Porter, Chair of the Department of Computer Sciences at the University of Texas at Austin, which has seen its enrollment increase more than 5% this year. “The government has made it clear that computer science is a growth field, and I think that message is getting back to students and their parents.”
Corporate recruitment of top computer science grads has remained steady throughout the economic downturn. Last spring, at the height of the recession, Georgia Tech’s College of Computing had the highest job placement rate of any major on campus and the highest starting salary.
This morning, my colleague Aaron Lanterman sent me a link to a blog he follows, by an engineer working for NASA, whose job it is to make sure that the fuel tank doesn’t blow up the shuttle when the tank drops away. He’s about to lose his job, due to the end of the shuttle program, and he’s upset with the focus on engineer education.
Coming down from on high today, in a different but nonetheless highly-public venue/context, is the word from Corporate:
It’s terribly important that something be done about the dire state of engineering education. This country (and thus, by extension this engineering-heavy enormous American corporation) is suffering from a terrible engineering shortage.
Huh? Run that by me again, please? And this time try to spin it so it makes sense using Earth Logic(tm) rather than Corporate Logic?
Are there jobs out there, or aren’t there? Or is it all a bit of spin? Note that the Network World News article talks about enrollment being up. It’s up for us, too — in comparison. Today, we have 800-900 BS in CS majors (depending on how you count), and about 300 BS in CM majors, for a total of (at most) 1200. In 2001, we had 1500 BS in CS majors. We’re “up” in comparison with how “down” we were. I believe that new graduates are getting jobs, and those jobs have good salaries. All graduates? And what happens after that?
A couple of really interesting pieces on the Glitch Project appeared yesterday, based on a conference paper that Betsy DiSalvo was lead author on. Glitch, as you may recall, is about getting African-American teen men interested in computing, by teaching them to be and hiring them as game testers. Betsy raises some challenging ideas in her paper, such that the geek image of masculinity is at odds with the African-American image of masculinity.
One of the blog posts is at Black Digerati, and the second (cited below) is at Racialicious. The second one is intriguing for its comments. In particular, the claim that white geeks are actually the product of class privilege. Those class distinctions relate to recent discussions here about Blacks, Latinos, and Women in Silicon Valley and the non-open culture of open source software development.
Unfortunately, computer geeks are generally really oblivious to their class privilege, mistaking their class privilege for how much geekier they are than other people. For example, there is a common sentiment among programmers who hire other programmers that it is bad to hire programmers who only started programming at the beginning of college. Their logic is that if you only started programming in college, instead of when you were a child, then it means that you are not really interested in programming, and you only went into CS to make money/get a job. There is zero recognition of class privilege and class differences.
Warning! The following post is even more self-serving (or at least, student-serving) than usual! I have two PhD students on the market this year. A former student is on the market (just finishing his post-doc). I know a very good new PhD in computing education from another institution who is also now on the market. I was bemoaning their luck this weekend, at going on the market at such an awful time for US academia. But as I started thinking about the points against hiring in computing education, I suddenly realized that I had it backwards. There could hardly be a better time to hire computing education faculty!
For the first time perhaps ever, there are multiple sources of funding for computing education. NSF has announced not one but two new funding programs that will fund work in computing education: One focused on the whole pipeline from middle school up, and another focused on tools. DARPA has a solicitation out now in computing education, and rumor has it that more is in the pipeline.
The journal ACM Transactions on Computing Education is in its second year. The conference International Computing Education Research Workshop is entering its fifth year. There are now good publication venues, appropriate for earning tenure for junior faculty in computing education.
It would be a great thing for the field of computing education for some school to develop strength in computing education. To have some big wins, to scale up some efforts to go after big goals (like CS 10K) would be a benefit for everybody. Who should go after this win? What would be the criteria for a school that should go after some serious depth in computing education?
- More than one faculty member — a more senior person and a new hire, at least. There are few institutions with more than one research faculty member in computing education (by some measure, e.g. funded by NSF and publishing at ICER and ACM ToCE). To have that depth would be a great start. That gives the junior person someone to help with those first proposals to NSF and DARPA. Two (or three or four) can do more than one.
- Have a strong Computer Science Department. The funding programs I mentioned are coming out of CISE (Computer and Information Sciences and Engineering) Division. It’s going to be hard for an Education school, or maybe even an IS/IT department, to get CISE reviewers to buy into a big effort. You’ll frankly need a CS department that reviewers will recognize as being high-quality. There are a bunch of interesting research questions in computing education which can only be answered from computing expertise. What is pedagogical content knowledge for computing education? How do we use new media (from Kindles, and video games, even to YouTube) to teach computing well? (Where are today’s Rocky’s Boots and the Schoolhouse Rock of computing education?)
- Finally, and this part may not be obvious: A strong Education school. Cameron Wilson and I have a piece coming out in CACM about funding for computing education. The budget in the EHR (Education and Human Resources) Directorate at NSF is some 50 times larger than what CISE can put into computing education. EHR is necessary to roll out nationwide scale-up. It’s going to be hard to convince EHR reviewers to fund really big grants (e.g., tens of millions) in scaling up computing education efforts without serious education involvement. CS-based researchers can figure out the basic computing education mechanisms, but it’s Education that knows how to scale. A CS department will be hard-pressed to make a strong case for, say, a large-scale teacher education effort if it has never done pre-service or in-service teacher development, and has no existing courses or instructors in teacher professional development. There are some great education research questions in scaling up computing education efforts. How do you grow computing education when its identity is confused? How do you create a community of CS teachers, and create a strong identity as a CS teacher? How do we get to 10K CS teachers in only five years?
Just to identify the limit of my self-serving nature, I don’t see Georgia Tech making these hires. We’re a state school, so we’re in very tight times. We have no education program. (By the way, I’m not looking to move to help someplace build up this CS Ed depth — especially with our son starting Georgia Tech in the Fall!) However, I do think that this is a great time for somebody to put these pieces together, make some new hires, and further computing education.
Ian Bogost has been reading old Apple Computer ads (from back when they were Apple Computer), and noting how users were encouraged to actually program their Apple II computers.
Once you’ve unlocked the power of the desk-top computer, you’ll be using Apple in ways you never dreamed of. You don’t want to be limited by the availability of pre-programmed cartridges. You’ll want a computer, like Apple, that you can also program yourself. … The more you and your students learn about computers, the more your imagination will demand. So you’ll want a computer that can grow with you as your skills and experience grow.
Where did that attitude go? Is it because there are so many “pre-programmed cartridges” (there’s an App for that!) that there is little reason to program? Or is it that programming is too hard? Or that knowing about the internals of a computer is now considered too esoteric or unimportant?
Ian and I have been exchanging email about this, and he reports that he is getting a lot of comments along the line: “I don’t have to know how my car works. Why should I have to know how my computer works?“
Ramesh Jain once gave me a good argument in response. Marshall McLuhan said that all tools that humans made were extensions of human facilities. A bicycle extends our legs, to let us move faster and further before tiring. A car extends both our arms and our legs, because now we can carry more while we go faster and further.
A computer extends our mind. While it’s certainly possible to use one’s mind without attempting to know oneself, it’s more powerful to be reflective, to consider how one’s mind works, and to identify one’s strengths and failings in thought. We can use a computer without understanding it, but not as well, and not as effectively. You can use it for typing and counting and talking without reflection, but if you want to use a computer for thinking, you are better off knowing something about how it works and about how to go beyond those “pre-packaged cartridges.” When it comes to thinking, there’s no app for that.
Hooray for Stanford! What they did sounds a lot like what we did with Threads: reduce required curriculum, encourage students to pursue subfields (our “Threads”), and include non-CS classes in the Threads — and for much of the same reasons. Great to hear that it’s working, and that they’re getting more (and more diverse) students!
The revamped curriculum consolidated the core required courses to six, down from 15, and allowed students to not only specialize in subfields like artificial intelligence and computer graphics but also allowed classes in other areas, like studio arts or human computer interaction count toward their computer science requirements.
This is the article that Dan Hickey linked to in his comment to the post on teachers cheating. It’s an interesting but short piece about best practices in assessment. I was really struck by the fact that Cisco Networking Academy is held up as a model for how to do assessment right. Interesting when the for-profit computing education does better at assessment than the non-profit, traditional computing education.
This investigation has revealed both obstacles and opportunities to developing assessments needed for real reform of educational policies and practices.
Like many others, we are concerned that the evaluation criteria for broader state-level RTT proposals may well lock in some of the problematic testing practices of No Child Left Behind. There is a danger that many states will continue to rely on the same narrow tests of basic math and reading skills that have thus far failed to lead to instruction that enhances deep conceptual understanding and innovative problem-solving.
The good news: students hoping to major in STEM fields are growing as a proportion of the overall student population, reaching Cold War-era levels of interest. The bad news: students who start out planning to major in STEM fields graduate at far lower rates than their non-STEM classmates, especially if they’re black, Latino or Native American. “We’re seeing this increase over the last 15 years in students’ interest in STEM fields,” said Eagan, a postdoctoral fellow at UCLA, “but we’re not seeing a corresponding increase in students’ graduation rates.”
In other words, we’re getting a lot more interest in CS-STEM (Yay!), but they’re leaving or flunking out before graduating (Boo!). Why is that? (By the way, “CS-STEM” is the term that DARPA is promoting to represent Computer Science + Science, Technology, Engineering, and Mathematics.) The HERI@UCLA study that this cites explicitly does include computer science.
I found the comment thread really interesting on this piece. Whose fault is it? Some of the commentors suggest that it’s the fault of the high schools. Students are unprepared upon entering a CS-STEM degree program, so they leave. Others say that it’s the College classes. The blame doesn’t lie with the faculty, some say, but the budget which keeps faculty from doing things that engage students (like lab activities and undergraduate research). David Brooks, as mentioned here previously, lays the blame at the feet of the faculty who aren’t doing enough to support the lower-ability student (perhaps underprepared, perhaps without reasonable expectations because they’re the first in their family to go to College).
The author of the piece blames the Colleges, and points out that it’s the minority students who are hurting the most:
Instead, much of that blame lies with the nation’s colleges and universities for deterring students somewhere between freshman year and the completion of a bachelor’s degree in four or five years, he said. “Something that happens in college – and it goes beyond just preparation – is losing students.”
A third of white students and 42 percent of Asian-American students who started college as intended STEM majors graduated with STEM degrees by the end of five years. For underrepresented minorities, the five-year completion rates were much lower — 22.1 percent for Latino students, 18.4 percent for black students and 18.8 percent for Native American students. Some of those students may have still graduated in four or five years but changed to a non-STEM major or transferred out of the institution they entered as freshmen. Others may still be working on their degrees.
To hit the goals that Barack Obama has set out for graduating students, we need to understand better why we are losing so many students, and how to correct that.
Lucy Sanders, CEO of NCWIT, visited Georgia Tech yesterday as the Center for the Study of Women, Science and Technology’s 2010 Distinguished Lecturer. Her talk was full of some pretty depressing statistics about the declining participation of women in computing. After the talk, she moderated a more-lively-than-usual discussion that started with a question “Why? Why did women’s participation in computing start declining at the end of the 1980′s?”
Lucy was honest. “I don’t know.” She said that she had heard lots of theories, but welcomed suggestions from the audience. Lucy said that her personal opinion was that the dot-com boom didn’t help women. The geeky, work 24/7, take-all-risks image of the male technology entrepreneur helped to make the perception of computing as more male. She also pointed out that history has shown that when wages rise in a field, men fill it and women leave it.
Barb Ericson had a suggestion that I don’t recall hearing before which made a lot of sense to me. When she was a software engineer at Bellcore, she programmed mainframes in large terminal rooms. Few people had terminals at their desks. Those terminal rooms were social spaces where women made up 30% of the workforce.
Then the PC became mainstream. People had PC’s on their desk and could work alone, without going to a terminal room. Jane Margolis and Alan Fisher talked in “Unlocking the Clubhouse” how the PC came into the home through the fathers and was seen mostly as a male device. If there was one in the family, it would most often be in the son’s room. Video games were created for the PC by guys for guys. As the PC displaced the mainframe, the males displaced the females.
It’s an interesting theory, as were several others that were suggested yesterday. Knowing the cause in decline in participation doesn’t necessarily mean that we immediately know a solution. We can’t get rid of PC’s, we don’t want to lower wages, and we can’t make the dot-com era not happen (short of a time machine). It’s still useful, though, to figure out what caused the decline to give us clues as to the way forward.
Hispanics and blacks made up a smaller share of the valley’s computer workers in 2008 than they did in 2000, a Mercury News review of federal data shows, even as their share grew across the nation. Women in computer-related occupations saw declines around the country, but they are an even smaller proportion of the work force here.
The trend is striking in a region where Hispanics are nearly one-quarter of the working-age population — five times their percentage of the computer work force — and when dual-career couples and female MBAs are increasingly the norm.
While it may be true that talent does not distribute evenly, it’s hard to believe that so few Hispanics, blacks, and women have the talent for Silicon Valley tech firms. There’s an explanation later in the piece that speaks to computing education themes.
Other reasons, experts say, include a history of valley companies hiring well-trained tech workers from the Pacific Rim, a weak pipeline of homegrown candidates, and a hypercompetitive business environment that leaves little time to develop workers.
Those latter two points are issues that we’ve discussed previously here. Computing education in the United States is in poor shape. There are few opportunities for working adults to develop themselves and improve their marketability in technology. The rest of the article points out that the problem is getting better nationally, but not in Silicon Valley, but because the Valley serves as the model for the rest of the country (“This is like ‘top gun school’ for Techies,” said one expert quoted in the article), the concern is that the trends there could be a drag on the rest of the country.