Wednesday 13 February 2013

Michael Littman’s new blog: End-user programming for household devices

I’m excited about the direction that Michael Littman is taking with his new blog. It’s a different argument for “Computing for Everyone.” He’s not making a literacy argument, or a jobs argument. He’s simply saying that our world is filled with computers, and it should be easy to talk to those computers — for everybody. Nobody should be prevented from talking to their own devices.The aspiration of the “ Scratch able Devices” team is to help move us to a future in which end-user programming is commonplace. The short version of the pitch goes like this. We are all surrounded by computers—more and more of the devices we interact with on a daily basis are general purpose CPUs in disguise. The marvelous thing about these machines is that they can carry out activities on our behalf: activities that we are too inaccurate or slow or fragile or inconsistent or frankly important to do for ourselves. Unfortunately, most of us don’t know how to speak to these machines And, even those of us who do are usually barred from doing so by device interfaces that are intended to be friendly but in fact tie our hands.

We seem to be on the verge of an explosion of new opportunities. There are new software systems being created, more ways to teach people about programming, and many many more new devices that we wish we could talk to in a systematic way. The purpose of this blog is to raise awareness of developments, both new and old, that bear on the question of end-user programming.

Deepa Singh
Business Developer
Email Id:-deepa.singh@soarlogic.com
Posted by at No comments:

College Degree, No Class Time Required: Just Religious Faith in Tests

The announcement from U. Wisconsin (that they’ll test students to get a degree, rather than requiring any coursework at all) is showing enormous and unsupported (almost religious) faith in our ability to construct tests, especially online tests. Building reliable and valid assessments is part of my research, and it’s really hard. Can I come up with assessments that are at least as good as having 32 (roughly) teachers assess you over a four year period? I already know that there is a lot that I don’t know how to assess in computing education (because we've tried and failed), e.g., the kinds of process knowledge that one gains in software engineering and senior design classes. I’m sure that there are many assessment experts who are far better than me, so certainly, someone else could do what I could not. Since I’m also a consumer of others’ assessments, I don’t see high-quality assessments (e.g., I trust them, they've been shown to be reliable and valid) that cover everything that we want students to learn. So, no, I do not believe currently that we can build tests to assess an entire computer science undergraduate degree. To create programs like what Wisconsin proposes is having unsupported faith that new assessments will miraculously appear. (“Miraculous” because as far as I can tell, no funding is going into building new assessments, and that’s pretty expensive to do well!)

Now, educators in Wisconsin are offering a possible solution by decoupling the learning part of education from student assessment and degree-granting.Wisconsin officials tout the UW Flexible Option as the first to offer multiple, competency-based bachelor’s degrees from a public university system. Officials encourage students to complete their education independently through online courses, which have grown in popularity through efforts by companies such as Coursera, edX and Udacity.

Deepa Singh
Business Developer
Email Id:-deepa.singh@soarlogic.com
Posted by at No comments:

Why Big Data Mostly Can’t Help Improve Teaching

The Muller research being described in the below post was discussed here previously, and is related to the predict-before-demo work that Eric Mazur presentedat last year’s ICER. The uppermost bit here is that data mining can’t get at this level of abstraction in terms of identifying good teaching. I’m also concerned that data mining can’t help if you lose 80% of your subject pool — you can’t learn about people who aren't there.

But even granting that you can get sufficiently rich information about the students, there’s another hard problem. Let’s say that, thanks to the upgrade in your big data infinite improbability drive made possible by your new Spacey space sprocket, your system is able to flag at least a critical mass of videos taught in the Mueller method as having a bigger educational impact on the students the average educational video by some measure you have identified. Would the machine be able to infer that these videos belong in a common category in terms of the reason for their effectiveness? Would it be able to figure out what Muller did? There are lots of reasons why a video might be more effective than average. And many of those ways are internal to the narrative structure of the video. The machine only knows things like the format of the video, the length, what kind of class it’s in, who the creator is, when it was made, and so on. Other than the external characteristics of the video file, it mostly knows what we tell it about the contents. It has no way for it to inspect the video and deduce that a particular presentation strategy is being used. We are nowhere close to having a machine that is smart enough to do what Muller did and identify a pattern in the narrative of the speaker.

Deepa Singh
Business Developer
Email Id:-deepa.singh@soarlogic.com
Posted by at No comments:

Reaching an intellectual peak: Should Everyone Go to College?

Ann Sobel has an article in IEEE Computer asking “Should everyone go to College?” as part of a special issue on education. Her answer is, “No.” She might be right, but I disagree with her argument. For example, below she suggests that students should avoid college if they “have already reached their intellectual peak.” Modern cognitive science suggests that fluid intelligence “peaks” in students’ 20′s, but other forms of intelligence develop and grow throughout one’s life.I’m particularly concerned about this article appearing in IEEE Computer. Thinking that high school is enough for a computing job is (a) wrong and (b) counter-productive at the high school level, since it encourages the instruction to be more vocational and less about developing computing concepts that could be used in post-secondary instruction. I’m particularly worried about what an emphasis on high school computing education means for under-represented minorities. A high-school only IT job will earn, on average, far less than a college degree IT job. Emphasizing high school IT jobs may mean trapping more under-represented minorities “in the shallow end.”

Ann identifies several important issues that prevent students from succeeding in college, like lack of adequate preparation and cost. I see those as challenges to be addressed, not roadblocks. If the context of the piece is taken seriously (i.e., high school degrees as preparation for jobs like those of IEEE Computer readers), then we have to consider the far more considerable issues of inadequate numbers and preparation for teachers. We are challenged to produce enough high school teachers to cover Exploring CS or CS:Principles, both of which de-emphasize programming compared to a traditional CS1. If we wanted students to be ready to get an IT job right out of high school, they better learn some serious vocational computing skills, from network management, to database administration, to low-level coding. How are we going to develop enough high school teachers to teach all of that?!? Here’s my bottom-line: “Should everyone go to college?” If you want a job in computing, yes. Students can attend a community college to help improve these test scores, but this route doesn’t always work, particularly when students have already reached their intellectual peak. While students have the potential for intellectual growth, if they can’t grow sufficiently, they should be supported in considering myriad rewarding career paths that don’t require a college degree.

Deepa Singh
Business Developer
Email Id:-deepa.singh@soarlogic.com
Posted by at No comments:

Thursday 7 February 2013

Berners-Lee calls for computer science education for children

The first part of what Berners-Lee says below makes sense to me. Students don’t really see computing until there’s breakdown. Interfaces carefully hide away the computing underneath. But it does not really make sense for children to check log files and re-write scripts when something breaks with Twitter or Facebook. I think he may be confounding two reasons for knowing about computing education: (a) to understand your world (“I think I have a clue why this broke, or even why it was working in the first place”) versus (b) to be able to construct in that world — to be a producer, as well as a consumer. Both are good reasons for learning about computing, but it’s not always the case that you can construct around every bug that comes up, especially with large social networks.

“A quarter of the planet uses the web,” he cautioned, “then within this quarter of people who may tweet and use social networks and so on, there’s a fairly small set of people who code. But when you look at those people, they have the ability to make a computer do whatever they can imagine.“I think a lot of folks growing up today, when they open a computer, it’s like opening a refrigerator. It’s an appliance, it’s white goods, there’s some stuff in it, if it needs more in it you stock it, you put more music in it, you play it. And If it breaks it’s: ‘Mom, can I have a new one’.“It’s not actually ‘what went wrong there? Let me go in there, lets look at the log files, what crashed, why didn’t it have the right permissions, lets see if we can re-write that script so that it works in the new version of the operating system.’”


Deepa Singh
Business Developer
Email Id:-deepa.singh@soarlogic.com


Posted by at No comments:

A Dream Deferred: How access to STEM is denied to many students before they get in the door

A realistic description of the barriers into STEM for students who are not from the schools that are expected to succeed. I can believe that these kinds of problems exist. Figuring out a way around them is the hard part.
This type of pre-judging of students happens all too often. Students from poor and poorly performing school districts, students who wear sagging pants or speak slang or with accents, students that may not make good grades, students from single-parent/multi-generation homes – these kids are denied an opportunity to participate at the gate. I cannot count the number of students I have encountered who have promise but absolutely no idea where to start or how to get started.

I have seen in at the high school and college level - professors that turn away students with GEDs or those who struggle academically, but who show up anyway. So many students who have been dismissed or passed over by teachers, guidance counselors, and professors because s/he may not be polished enough for top-level science. (Whatever that means.)

Deepa Singh
Business Developer
Email Id:-deepa.singh@soarlogic.com
Posted by at No comments:

Special issue of ACM TOCE on Computing in Schools

The ACM journal Transactions on Computing Education is going to have a special issue devoted to Computer Science Education in K-12 Schools. Well worth exploring.

Recent activities in several countries, for example in the USA, the United Kingdom, New Zealand and South Korea, show a growing awareness of the importance of rigorous computer science education (CSE) for a successful, self-responsive and self-deciding life in the modern world. Consequently, serious efforts are made to introduce or to improve CSE in schools that will be followed by other countries, as we hope. Yet, for any country that wants to improve CSE in schools, it would be advisable to learn from the experiences that were made somewhere else. Nevertheless, those experiences were gathered under preconditions and circumstances that usually differ strongly from country to country. Unfortunately, the short format of conventional scientific papers prevents most reports about such experiences from covering all relevant aspects of the respective context. To produce relief, this Special Issue of TOCE aims to collect extensive, detailed case studies that discuss as many relevant aspects as possible, for example regarding the category system that was proposed in 2011 by the ITiCSE Working Group about Informatics in Secondary Education [1].

Deepa Singh
Business Developer
Email Id:-deepa.singh@soarlogic.com
Posted by at No comments:

Wednesday 6 February 2013

Demographics on GT’s first Coursera MOOC: Computational Investing by Tucker Balch

My colleague Tucker Balch posted on his blog the detailed demographics of his Coursera MOOC (the first at Georgia Tech), “Computational Investing.” He got 41% of the completers to respond to his survey, but only 2.6% of those who enrolled but did not complete. That’s a remarkable response rate, so it’s a great snapshot into who completes a course like this.

A big caveat up-front: This is “Computational Investing.” It’s clearly an elective subject, so we would expect demographics to shift from what we might hope to see in a required course (like CS1 or data structures) or a common upper-level course (like AI).

Some of the results that I found intriguing:
  • I predicted that CS course MOOC completers would be 80% white or Asian and 90% male. I underestimated. Tucker’s course was 88.6% white or Asian and 91% male.
  • 73.3% of completers came from OECD countries (as a measure of “developed”), and half of those were from the US. So, were the completers people who couldn’t get access to higher education otherwise? Nope. Over 10% had their PhD’s, and over 40% had their Master’s degree. Less than 10% of the completers only had a high school degree.
  • The discussion forums were not how most students asked questions. Everyone reads (over 95%), but only 33% post — which is pretty similar to the lack of participation that we documented years ago in engineering courses using Wikis. That doesn’t mean that the collaboration forums aren’t contributing to learning, but it does mean that it’s not substituting for discussion in the classroom.
Deepa Singh
Business Developer
Web Site:-http://www.gyapti.com
Blog:- http://gyapti.blogspot.com
Email Id:-deepa.singh@soarlogic.com
Posted by at No comments:

Measuring the Success of Online Education: Duolingo and MOOCs

Was anyone else bothered by the argument in this NYtimes blog post? ”MOOCS aren’t effective in terms of completion rates; Duolingo is not a MOOC; Duolingo is effective.” So…what does that tell us about MOOCs?The paper on Duolingo effectiveness is pretty cool. I think it’s particularly noteworthy that more prior knowledge of Spanish led to less of an effect of Duolingo. I wonder if that’s because Duolingo is essentially using a worked example model, and worked examples do suffer from the expertise reversal effect.

Moreover, there are early indications that the high interactivity and personalized feedback of online education might ultimately offer a learning structure that can’t be matched by the traditional classroom.Duolingo, a free Web-based language learning system that grew out of a Carnegie Mellon University research project, is not an example of a traditional MOOC. However, the system, which now teaches German, French, Portuguese, Italian, Spanish and English, has roughly one million users and about 100,000 people spend time on the site daily.
 
 Deepa Singh
Business Developer
Web Site:-http://www.gyapti.com
Blog:- http://gyapti.blogspot.com
Email Id:-deepa.singh@soarlogic.com
Posted by at No comments:
Subscribe to: Posts (Atom)