Monday, July 30, 2018

Educational buffet


CC BY-NC-ND Some rights reserved by iamphl on Flickr
Subscription based platforms like Spotify and Netflix have been a massive success allowing you access to a massive library of music and films for a flat rate monthly subscription. A similar platform for magazines, Readly, offers you access to a wide range of monthly magazines on a similar flat-rate subscription basis. Although it is debatable how much the artists benefit much from this model but it's an improvement on the free downloading of the Napster days. So how this can be applied to education?

While most of the media focus has been on MOOCs over the last few years there's another side to online education that is galloping along almost unnoticed. There are many platforms that offer a vast range of short training courses provided by individual educators, colleges or companies where the learner pays a fee and some of that money goes to the course creators. The most prominent platforms in this niche are UdemySkillshareTeachable and Lynda.com, but there are many more. Udemy has been around for many years now and is as a market place where educators can create and offer an online course and earn money on the registrations. Other platforms can have more in-house course production or various forms of quality control on the courses published. the simplest form of quality control is by learners reviews and ratings. There are two basic types of business model: one that charges learners a price per course and the course creator gets a share of that or the subscription model where the learner pays a monthly fee to access the whole range of courses and some of the income is distributed among the contributors.

You could call this field just-in-time learning for anyone who wants to get a quick overview of a new concept or how to use a particular application. It's more the modern equivalent of all the books with titles like A beginner's guide to X or Teach yourself Y. They don't pretend to be like a university course or to provide interaction with teachers and other learners. They guide you through the process, allow you to test your knowledge and maybe some kind of practical task. If you want to learn some more advanced functions in PhotoShop or the principles of lean management then this is a good place to start but it's not the place for deeper learning and collaboration, nor does it even pretend to be so.

One of these companies, Skillshare, is highlighted in an article in EdSurge, Can a Subscription Model Work for Online Learners and Teachers? Skillshare Just Raised $28 Million to Find Out. For a very affordable (at least for learners in developed countries) you get access to a self-service buffet of courses and so far they have amassed around 5 million users. Learn as much as you like for a monthly fee.

There are roughly 1,000 courses available on Skillshare for free. For full access to the more than 22,000 classes currently on its platform, there’s a subscription fee (either $15 per month or $99 a year). About 30 to 50 percent of this subscription revenue goes to a royalty pool that pays Skillshare teachers based on their share of all the minutes of video watched in a month. The company claims that the average Skillshare teacher makes about $3,000 a year, with top earners raking in as much as $40,000.

There is a useful overview of different online course platforms that use some kind of subscription model on the site Medium, The Economics of Teaching in an Online Learning Marketplace. They all offer attractive packaging and presentation for your course and a marketplace to attract participants but as ever you need to weigh up the costs of using the platform with what you get out of it.

Are MOOCs heading in this direction? There are already special prices for course packages and the differences between the MOOC providers and the course platforms are narrowing. Online education is becoming increasingly commercial and I think these platforms fill an important niche in terms of professional development and lifelong learning. However I still think that universities should also offer truly open education to those who are unable to access more traditional forms and cannot afford the commercial variety. As the majority of the MOOC platforms become more commercial we mustn't forget all the less hyped open education that is being conducted by committed universities and partnerships all over the world. That's where the really interesting development is taking place.

Thursday, July 26, 2018

Learning analytics - is there an off switch?

CC0 Public domain on pxhere
Returning to the theme of learning analytics, I wonder if there will be any way of opting out from being tracked and coached. The potential of learning analytics is in tracking a student's progress, suggesting resources, offering extra practice and constant feedback on assignments as well as monitoring performance. The vision is to have a personal tutor by your side round the clock and provide your human tutor with alerts if your are struggling in some way. The trouble with this is that sometimes you need to be able to switch off the surveillance and just practice in peace. You seldom perform to the best of your ability if you know someone (or something) is watching you.

I recommend you read an article by John Warner in Inside Higher EdThe Problems of Real-Time Feedback in Teaching Writing. He objects to real-time AI feedback on students' writing on the grounds that it does more harm than good. Too much feedback too often can destroy the creative process and simply leading the students to write in order to satisfy the system. There is a time for providing feedback and there must also be time for trial and error and experimentation without the feeling that you are being observed and assessed.

For example, when learning to play the guitar, it’s useful to have some periods of real-time feedback where a teacher may be able to correct a flaw like a bad hand position, but you also need to go lock yourself in your room and practice, likely making a bunch of unpleasant noises in the process. Imagine trying to do this while being constantly reminded that your noises really are unpleasant.

We all need our own quiet spaces to concentrate, experiment, test, reflect and discuss: a play room, sandbox, hideaway, tree house. You don't want anyone to see or hear your embarrassing mistakes and it's best that many early attempts are discarded and lost without trace. This is one reason why students seldom post in the forums of our learning management systems. They only post when required and soon learn that everything they write there is in the university's system and may be used in evidence against them. As a result they create their own closed communities to discuss coursework, away from all risk of assessment. 

We need to use AI/learning analytics wisely and make sure we allow students the right to escape when they need to. The risk is that we destroy creativity by offering too much support and personalisation.

Writing is thinking, writing is thinking, writing is thinking, and sometimes when we’re working on our thinking, we have to be left alone.

Sunday, July 8, 2018

Is edtech simply automating tradition?

CC BY-SA Some rights reserved by ISAPUT on Wikimedia Commons
For a few years the internet was an exciting new space where we thought we could build a better world through innovation, borderless collaboration and a culture of sharing. Today big business reigns supreme and technology is being used for control, surveillance and commercial gain. After several years of speculation and experimentation we are now beginning to see how analytics and artificial intelligence are going to be used in education. As governments become increasingly concerned with accountability, profitability and competition, technology is feeding this drive with statistics.

A post by Mike CrowleyGoogle, ISTE, and the Death of EdTech, reflects on the recent ISTE conference and sees worrying tendencies that today's edtech industry is increasingly about control rather than creativity. Technology is being used to make traditional practice more efficient rather than fostering real learning.

All week, the ISTE Expo Hall gushed with predictions that Artificial Intelligence will automate the traditional classroom. Auto-grading is now a thing. Marketeers touted emerging tools to help students study smarter, by enabling them to memorise more information faster. Ultimately, though, the Big News, like Google Classroom, is the proliferation of software that allows teachers to monitor and keep students on-task. The lofty ambitions for education were summed up thus. Make tradition more efficient. Stop bored students from cheating on mindless, low-level assessments. Deliver content like Windows 98 is the next big thing. New tools. Old thinking. Systems, not learners.

The tyranny of tradition is extremely hard to escape from. The traditional educational paradigm of learning facts that can be easily tested and categorised is a perfect fit for the smart systems of today. We can use analytics and AI to provide the statistical illusion that students have achieved the learning objectives we set them. But technology can and should be able to do so much more and that's why Crowley states:

Make no mistake about it. EdTech as we currently know it is dead, it’s over. We should retire the phrase right now. If education is to be the target of an industry that has grown increasingly obsessed with standardization, control, automation, and delivery efficiencies, then we must opt out.
Related to this is the use of automated grading of assignments that has been under development for several years now. This involves computer analysis of a written assignment to check for coherence, argumentation, linguistic style, grammar and fluency. An article on NPR, More States Opting To 'Robo-Grade' Student Essays By Computer, examines recent experience of automated grading at several US universities. Many educators are impressed by the accuracy of these tools, often giving similar grades to human examiners and if the computer is unsure it flags for teacher assessment. The potential savings are obvious and very attractive to cash-strapped institutions.

Several states including Utah and Ohio already use automated grading on their standardized tests. Cyndee Carter, assessment development coordinator for the Utah State Board of Education, says the state began very cautiously, at first making sure every machine-graded essay was also read by a real person. But she says the computer scoring has proven "spot-on" and Utah now lets machines be the sole judge of the vast majority of essays. In about 20 percent of cases, she says, when the computer detects something unusual, or is on the fence between two scores, it flags an essay for human review. But all in all, she says the automated scoring system has been a boon for the state, not only for the cost savings, but also because it enables teachers to get test results back in minutes rather than months.

Impressive indeed, but there will always be ways to hack the system and already there are examples of text generators that produce nonsensical texts that satisfy the grading system's preferences. This type of cat-and-mouse game has nothing to do with learning. When the only purpose of writing an assignment is to get a grade then people will simply do whatever it takes to get that grade as easily as possible. If the assignment has real meaning to more people than simply an examiner (or a robot) then the game element disappears. We need to develop methods of assessing student ability based on the impact of their projects on real people. Maybe traditional for-teacher's-eyes-only assignments are the problem.

One clear use for automated grading systems is to help the students improve their writing skills. The system can provide repeated formative assessment opportunities, helping students to improve the coherence and style of their assignments. They can submit several times during the writing process, something that no teacher could ever have time to deal with, and then submit the final version to the teacher for grading. However a more collaborative approach already used by many teachers is to encourage peer assessment with fellow students providing feedback during the writing process. This method benefits everyone and the students can learn a lot from being actively involved in the assessment process.

The real purpose of technology should be to facilitate human collaboration and learning. I love using digital tools to facilitate creative and collaborative activities and that's where educational technology should be focused. The use of big data in education, however is worrying and we need to be very careful not to be tempted into handing over control of our students to big business.