Archive for November, 2011
I strongly agree with Alfred Thompson — we do need to think about how to get computer science (as a powerful literacy for thinking and expression) integrated into the curriculum, in the ways that make sense. How do we get there? Seeing cross-curricular influences in contest entries is an interesting way of getting started on that path. If the cross-curricular projects get extra points or even awards, then there is incentive for others to follow. That doesn’t change curriculum yet, but it does get teachers and students creating exemplars and thinking about how to do further integration.
I would argue that involving computer science in multi-disciplinary projects helps to move computer science more into the core of the educational process. It helps show the value to administrators as much as it does to students and parents. He helps create a better more rounded learning experience. It helps to in effect justify the existence of CS in the core to those people who for some reason don’t already buy into the idea.So while I truly would like to see more serious, focused computer science or computer programming projects at next year’s Global Forum it will be in Athens Greece BTW I am pretty happy to see cross disciplinary projects that include some computer science, a little Kinect programming, perhaps some Windows Phone development or SharePoint integration along with math, science, history, social studies, current events, environmental awareness or what ever else teachers are looking for innovative ways to teach.
I completely believe that the “elites” are having a major up-tick in CS applications. Is everyone? I am not hearing the same level of optimism at all colleges and universities. When I visited Melbourne last week, I learned that CS applications for next year’s undergraduate class are down 10% for the entire state of Victoria. Enrollment is a big issue there.
The nation’s best undergraduate computer science programs are bracing for a record number of applications this fall, as more high school seniors are lured by plentiful jobs, six-figure starting salaries and a hipster image fostered by the likes of Steve Jobs and Mark Zuckerberg.
Early admissions are piling up at elite tech schools, including Carnegie Mellon University, Harvey Mudd College and Rose-Hulman Institute of Technology – all of whose undergraduate computer science and engineering programs are rated tops by U.S. News & World Report, the de facto college ranking in the United States.
As I mentioned last week, Barb and I spent the week visiting Australia. We gave talks in Melbourne (keynote at Melbourne Computing Education Conventicle and teacher workshop), Adelaide (presentation on MediaComp at Festival of Teaching and Learning and on CS outreach), and Sydney (joint keynote at Sydney Computing Education Conventicle, then MediaComp talk at alumni end-of-term BBQ). I’ve uploaded all the talk slides at http://coweb.cc.gatech.edu/mediaComp-plan/1, if you’d like to see any of them.
It was a really interesting time to visit Australia. Higher education is going through some dramatic changes there. I don’t understand the current system all that well, so I don’t understand exactly what’s changing. I did hear a lot about what the new system will look like.
Higher education enrollment used to be “capped,” with a certain number of well-ranked students being given access to Universities. There are almost no private Universities in Australia.
Under the new system, there are “no caps.” Universities can take as many students as they wish (where Universities have some say on their goals for their enrollment, e.g., to keep a high test score average for an entering class vs. having increased diversity). Private Universities will be allowed, maybe even encouraged. There will be funding for Universities associated with taking students from lower socioeconomic status (SES) high schools. Funding will be tied to retention and graduation rates. All of these changes start in 2012.
The new system features TEQSA, the Tertiary Education Quality and Standards Agency. I heard a presentation in Adelaide about TEQSA and how the University of Adelaide is responding to it. Currently, there is a huge effort in establishing standards for all their undergraduate and graduate programs, and starting next year, there TEQSA will also be part of accreditation (a “regulatory function,” in the words of their website) starting in January 2012. I was a bit worried about the standards process from the presentation I heard. It’s all “demand-driven” (their emphasis) with interviews with literally hundreds of stakeholders, especially industry. After my talk on Media Computation, someone asked me, “But is industry asking for Liberal Arts majors to learn to program?” Of course not, I explained. Teaching all Liberal Arts majors about programming is about enabling innovation and creating a market differentiation. I told one of my stories about students in Liberal Arts, Architecture, and Management getting interviews and jobs because they have a unique computer science background that peer students might not. I explicitly said that that’s the problem of being “demand-driven” in setting standards — how do you plan for the future where your students are going to live? (Our host in Adelaide, Katrina Faulkner, said that the Dean of Humanities approached her after my talk and asked if the Adelaide CS department could offer a similar course. Someone from their Education school came up to me and said that her eyes were opened — she’d never thought about teaching CS to high school teachers before. A good start for a conversation!)
I suspect that some of the higher education changes were drivers for our visit. Some funding for our visit came from the Australia Council of Deans of ICT programs. I heard a presentation in Sydney from Tony Koppi of ACDICT, where he talked about a study that they had just completed about why students leave computing programs. The most common complaints were the same ones that drove the design of Media Computation, e.g., that students found the CS courses irrelevant and boring. Tony explicitly called for exploring contextualized computing education. Retention is clearly on their minds nowadays, and that’s one of the outcome variables that MediaComp has had the most success influencing. They are also quite concerned with drawing more students into computing, especially among women and members of underrepresented minorities. Barbara was pressed for her lessons learned on CS outreach in all three cities. I don’t think ACDICT is telling Australian computing departments to do as we do, but they are asking computing departments to think about these issues.
One of the most interesting interventions I heard about from ACDICT was their ACDICT Learning and Teaching Academy (ALTA), whose goal is: “To contribute to improvement in the perception of and the actual quality of learning and teaching across the ICT disciplines.” I thought the “perception of and actual quality” combination was particularly realistic! I heard a presentation on ALTA in Melbourne, where they emphasized that they want to address “Grand Challenges” in computing education. I’m interested in watching to see what they identify!
Yes, we found Australia interesting and fun. We took our two daughters. (Our son is a Sophomore at Georgia Tech, and had class through Wednesday before Thanksgiving, so couldn’t afford the time off.) We pet kangaroos in Melbourne; had dinner in an Australian home in Sydney that sat on a peninsula directly between Sydney Harbor and the ocean, with beautiful views to both sides; and saw a play at the Sydney Opera House Friday before heading home. Melbourne was so interesting (Royal Botanic Gardens are a must see), Adelaide was so beautiful (probably the prettiest campus I’ve yet visited), and Sydney is probably really nice when it isn’t raining for four days straight. The visit to the Sydney Hyde Park Barracks was a highlight for me, learning about how Sydney grew into the amazing city it is from a society of convicts. I’m really glad that we got the opportunity to visit and learn about Australia. Thanks to the ACM Distinguished Speakers Program, ACDICT, and funding from our hosts. Our hosts took great care of us, and we’re grateful: Catherine Lang at Swinburne, Katrina Falkner at U. Adelaide, and Judy Kay at U. Sydney.
I’m surprised that this is controversial: A for-profit university, that provides e-textbooks for free to its students, is going to start producing its books in-house. The professors are concerned that they might not be fairly compensated, or that the quality won’t be high. Both of those are factors that can be easily controlled — and this isn’t even a novel practice. The Open University in the UK has been doing this for decades. It was (at one time) the fourth largest publishing house in the UK (says Gordon Davies, past CS Chair at Open U. UK). This is their common practice. Why is it uncomfortable for American faculty?
APUS occupies a relatively comfortable niche among for-profit colleges, and one that’s earned them praise, even from some of the industry’s critics. Part of the reason is the university system’s relatively cheap tuition: three-credit undergraduate courses cost $750, an amount that hasn’t changed in a decade.
Affordability is a draw for its 84,000 students, about 85 percent of whom are affiliated with the military. So is flexibility, given service members’ unpredictable schedules and frequent moves. As a result, all courses offered by the university are fully online.
The university assigns e-textbooks as the default for most courses. It also offers an unusual perk to students by not charging for those texts, which are included in tuition.
Since the university itself pays for digital texts, it stands to reap substantial savings by producing course materials internally.
Here’s another example of students-as-customers and ramifications of that perspective. If students dislike a teaching approach (even if it’s demonstrably effective, with empirical support), then the professor is not doing his job, if the job is defined as “satisfy students.”
Some students didn’t take well to Steven Maranville’s teaching style at Utah Valley University. They complained that in the professor’s “capstone” business course, he asked them questions in class even when they didn’t raise their hands. They also didn’t like it when he made them work in teams.
Those complaints against him led the university denying him tenure – a decision amounting to firing, according to a lawsuit Maranville filed against the university this month. Maranville, his lawyer and the university aren’t talking about the case, although the suit details the dispute.
Maranville and his attorney did not return phone calls, but the allegations in the lawsuit raises questions that have been raised and debated about the value of student evaluations and opinions, how negative evaluations play into the career trajectory of affected professors and whether students today will accept teaching approaches such as the Socratic method.
Students doing more homework isn’t as effective for student learning. Homework can influence learning, if it’s quality homework. But if quality of learning is not the outcome variable that we care about, then homework is not an issue. We can just have students watch videos instead.
The quantity of students’ homework is a lot less important than its quality. And evidence suggests that as of now, homework isn’t making the grade. Although surveys show that the amount of time our children spend on homework has risen over the last three decades, American students are mired in the middle of international academic rankings: 17th in reading, 23rd in science and 31st in math, according to results from the Program for International Student Assessment released last December.
In a 2008 survey, one-third of parents polled rated the quality of their children’s homework assignments as fair or poor, and 4 in 10 said they believed that some or a great deal of homework was busywork. A new study, coming in the Economics of Education Review, reports that homework in science, English and history has “little to no impact” on student test scores. (The authors did note a positive effect for math homework.) Enriching children’s classroom learning requires making homework not shorter or longer, but smarter.
Fortunately, research is available to help parents, teachers and school administrators do just that. In recent years, neuroscientists, cognitive scientists and educational psychologists have made a series of remarkable discoveries about how the human brain learns. They have founded a new discipline, known as Mind, Brain and Education, that is devoted to understanding and improving the ways in which children absorb, retain and apply knowledge.
PhD programs, at least here at Georgia Tech, have changed dramatically in the last five years — with the economic downturn and with increased cutbacks in federal and state funding. A PhD student without funding can’t last long in the program now. A student can’t get admitted to the program without a faculty member making a commitment to getting the student funding.
There’s also the case of picking a topic in order to get a job in the future. I recently met with an undergraduate who said that she wants to be a teacher at a four-year liberal arts college. She was talking to me about doing research in CS Education. I told her that I’d be glad to work with her, but it’s not in her best interest to get a PhD in CS Education Research given her job focus. Few teaching-oriented schools hire CS Ed people. She has time now, as an undergraduate, to develop research skills in some area that could lead to a graduate career that would make getting her dream job more likely.
Here in Australia, there is a real boom in CS Ed PhD students. There were 20 at the Doctoral Consortium in Melbourne. University of Adelaide has one now, two more starting in September. One of those starting next year had to re-do her proposal because she’s getting a CS PhD and there wasn’t enough CS in there, but not because she couldn’t do CS Ed. It’s striking that Australia has this boom in CS Ed PhD’s because of student interest — no increase in funding, and I don’t know what the job prospects are like.
Selecting a topic for your PhD dissertation is critical. You need something promising, but do-able within the time available. It has to be something you really care about, but also something that your advisor cares about and is reasonably knowledgable about. It’s already a highly constrained problem. Needing funding from Day One complicates things further. If I need to find the money before the student, then I need to organize PhD admissions around finding the best fit for my particular project. That may not be the best student available–it’s just the one who fits. It feels backwards–buy the shoes first, and then look for someone with the right size feet. It’s also not a time efficient process. If a grant proposal takes six months to be evaluated and I can’t hire someone until the money is in hand, then I may have to wait up to a year to find the person who fits in those shoes. My annual report for year one I fear will end up saying “so far I think we’ve found someone who might want to work on this. We haven’t spent any of the money yet. Can you give us an extra year?” And what if I make an offer of admission to the one person who fits those shoes, and they choose a different graduate program? My fear is that this will lead to 1) lots of unhappy students working on projects that are not their first choice, and 2) lots of projects with no labor available for a medium to long amount of time.
When Barb and I went to the NCWIT summit in May last year, we were asked to be interviewed with our daughters for the NCWIT Sit With Me campaign. The campaign hasn’t launched yet, but there is a video up on the website now which is really compelling. And it’s fun for me to see all four of us in it!
Sometimes you have to sit down to stand up. Starting later this year, we will be sitting to inspire more women to participate in technical careers. When the “Sit With Me” campaign launches, we will sit to raise awareness of the crisis facing our field. We need more women and men to sit with us. Our collective support will help promote a world of technology that is as broad and creative as the people it serves.
via Sit With Me.
I love the direction, to make research careers more family-friendly, but am wondering if these steps will do it. Deferring awards for one year makes sense. Paying for technicians and staff to maintain labs for a year? Not sure that really works for computer science.
Barb and I were talking about the positive impacts of stopping tenure in the United States. It might make academic careers more family-friendly. No more intense tenure race during peak years for having children.
The 10-year initiative will include provisions for grant recipients to defer their awards for up to one year for the care of a newborn or newly adopted child, allow recipients to suspend their grants for parental leave, and pay for technicians and staff to maintain labs for researchers on leave.
The plan also includes efforts to publicize these family friendly opportunities, encourage more research into workplace flexibility, and design outreach that introduces girls to women at the top of research fields.
I’m in Melbourne, Australia, where I spoke yesterday at the Melbourne Computing Educational Conventicle. It was an exciting, high-energy event. Almost 90 attendees from around Melbourne came to talk about issues, from cheating and Android, to the definition of Multimedia. Even more exciting was the news that their doctoral consortium the day before had 20 PhD students, exploring research issues related to CS education. That’s amazing! Below is Catherine Lang, the chair and organizer of the event, opening the Conventicle. (Thanks to ACM Distinguished Speakers program for providing me support for this.)
Heard an interesting talk yesterday from ACDICT, Australian Council of Deans in ICT. They’re funding work in “Grand Challenges in CS education.”. Will be interesting to see what they identify.
Today, Barbara ran a workshop for teachers (middle and high school) on building apps for Android. She ended up touching on lots of other issues, from Lightbot to CS Unplugged. She had about 20 teachers.
Monday, we head to Adelaide for two talks there on Tuesday.
It’s a short piece, but it makes the point that Senator Casey gets it — CS education is shrinking, just as the demand is growing.
The availability of introductory high school computer science courses has decreased by 17 percent since 2005, he said, and the number of Advanced Placement computer science courses has dropped by 33 percent.
“Just when we need more students to focus on this course of study it’s going in the wrong direction and rapidly in the wrong direction,” Mr. Casey said.
Moreover, women and minorities are under-represented among those taking existing courses, he said.
Using Wikis for undergraduate courses was invented at Georgia Tech. We started in 1997, long before Wikipedia. Ward Cunningham talks about our work in his book “The Wiki Way.” Our paper on how we designed the Swiki (or CoWeb) at CSCW 2000 is, I believe, the earliest reference to wikis in the ACM Digital Library. Jochen “Jeff” Rick built the Swiki software that we use today, and he did his dissertation on his extensions to Swiki.
We published a technical report in 2000 about the varied uses of Swikis that we saw around Georgia Tech’s campus. Some classes were having students create a public case library. Others were have cross-semester discussions between current and past students. Others had public galleries of student work.
All of that ended yesterday.
Georgia Tech’s interpretation of FERPA is that protected information includes the fact that a student is enrolled at all. The folks at GT responsible for oversight of FERPA realized that a student’s name in a website that references a course is evidence of enrollment. Yesterday, in one stroke, every Swiki ever used for a course was removed. None of those uses I described can continue. For example, you can’t have cross-semester discussions or public galleries, because students in one semester of a course can’t know the identities of other students who had taken the course previously.
Seymour Papert coined the term constructionism to describe a setting for constructivism to occur.
Constructionism–the N word as opposed to the V word–shares constructivism’s connotation of learning as “building knowledge structures” irrespective of the circumstances of the learning. It then adds the idea that this happens especially felicitously in a context where the learner is consciously engaged in constructing a public entity, whether it’s a sand castle on the beach or a theory of the universe.
Constructionism relies on the fact that the entity being constructed is public. The public nature influences the student’s motivation for doing it and doing it well. If it’s not public, it’s not constructionism. We can no longer have students construct public entities on the Web anymore for education at Georgia Tech. It may be that FERPA demands that no school can use the Web to post student work publicly.
An interesting take, from the NYTimes, on how Stanford’s president sees the on-line Stanford AI class experiment. The virtual campus is for “specialized programs,” for going beyond the undergraduate experience. The evidence that technology leads to better learning isn’t there, and the undergraduate experience is better face-to-face. Further, Universities need the money — someone has to pay for the content.
The market for “continuing education” is potentially much larger than undergraduate. After those four years of college, there are a lot more years in a rapidly-changing workplace. Maybe that’s where the real money lies? Could providing the on-going, lifelong learning be the place where some of the costs for the face-to-face undergraduate education are carried? Maybe the content gets paid for by the lifelong learners, and the undergrads get it at reduced cost? Recall what John Daniel said about the US Open University — it failed because it went after undergraduate first, not graduate.
John Hennessy, Stanford’s president, gave the university’s blessing to Thrun’s experiment, which he calls “an initial demonstration,” but he is cautious about the grander dream of a digitized university. He can imagine a virtual campus for some specialized programs and continuing education, and thinks the power of distributed learning can be incorporated in undergraduate education — for example, supplanting the large lecture that is often filled with students paying more attention to their laptops. He endorses online teaching as a way to educate students, in the developing world or our own, who cannot hope for the full campus experience.
But Hennessy is a passionate advocate for an actual campus, especially in undergraduate education. There is nothing quite like the give and take of a live community to hone critical thinking, writing and public speaking skills, he says. And it’s not at all clear that online students learn the most important lesson of all: how to keep learning.
As The Times’s Matt Richtel recently reported, there is remarkably little data showing that technology-centric schooling improves basic learning. It is quite possible that the infatuation with technology has diverted money from things known to work — training better teachers, giving kids more time in school.
THE Stanford president is hardly a technophobe. Hennessy came up through computer engineering, used his sabbatical to start a successful microprocessor company, and sits on the boards of Google and Cisco Systems.
“In the same way that a lot of things go into the cost of a newspaper that have nothing to do with the quality of the reporting — the cost of newsprint and delivery — we should ask the same thing about universities,” Hennessy told me. “When is the infrastructure of the university particularly valuable — as it is, I believe, for an undergraduate residential experience — and when is it secondary to the learning process?”
But, he notes, “One has to think about the sustainability of all these things. In the end, the content providers have to get paid.”
- Visit a Local High School: Send computer science clubs or groups of student advocates to area middle/high schools to advise them about computing, career opportunities, classes needed to prepare for college, etc. See http://www.ncwit.org/roadshow for guidelines and templates for taking your show “on the road”.
- Invite Pre-college Students to Your Campus: Host a hands-on workshop/open house for parents, counselors and high school students to: explore the world of computer science; learn about the career opportunities and salaries; and discover what’s special about your program. You might consider using CS Unplugged to teach lessons that explain how computers work without using computers. Visit http://www.ncwit.org/unplugged for activities and instructions.
- Host an Open House for Non-Majors and Community College Students: Host a computer science open house/social hour for non-majors and local community colleges to share with them the opportunities available in computer science and develop interest in computer science classes. This should be a hands-on opportunity for students to try computing first hand.
- Once you take the CSEdWeek pledge, click the ‘share’ button to share your commitment throughout your networks.
- ‘Like’ CSEdWeek on Facebook at http://www.facebook.com/CSEdWeek and join the conversation.
- Blog, tweet, and post to spread the word and raise awareness. Use the hashtag: #CSEdWeek.
CNN Money is trying to answer a simple question: How diverse is Silicon Valley? The answer: Tech companies won’t tell them! Google, Apple, and Yahoo claim that to do so would be releasing a “trade secret.” Really? The Netflix argument below is particularly aggravating. “We just hire the best people” has long been exposed as a lie, since technology hiring is not really a meritocracy. Instead, “just hire the best” seems to be code for “we hire the mostly White and Asian males who are friends with our current White and Asian male workers.”
This is a real problem for computing — we can’t make it better if we don’t see it. It’s a real problem for efforts to broaden participation in computing education — if we generated a more diverse computer workforce, could they get jobs? (Thanks to Amber Settle for pointing out this article.)
Are these programs working? That’s impossible to measure. Microsoft refused to release its workforce demographic data. Sixteen other companies contacted repeatedly by CNNMoney also declined or ignored our request: Apple, Amazon, Cisco, eBay, Facebook, Google, Groupon, Hewlett-Packard, Hulu, IBM, LinkedIn, LivingSocial, Netflix, Twitter, Yelp and Zynga.
“Every company talks about their lovely diversity programs … but they won’t give us their data,” says Aditi Mohapatra, senior sustainability analyst at Calvert Investments, which invests in socially responsible companies and conducts its own diversity research. “What gets measured, gets managed. We need something tangible and public.”
Some companies shrug off those criticisms. Netflix (NFLX) is a “veritable UN,” according to spokesman Steve Swasey: “We don’t do anything to get diversity; we just get it. We don’t focus on it, and we don’t talk about it. It just is what it is. And we get the best people.”