Here is an article in the San Jose Mercury News that shows how ‘flipped learning’ is taking off. It is encouraging to see the possibilities of revolutionizing education with this model.  The only thing that surprises me is that they wrote an article on flipped learning and didn’t  mention the videos done by places like the Khan Academy.   From the article:

“Teachers note that making the videos and coming up with project activities to fill class time is a lot of extra work up front”

Don’t they understand that there are people out there creating great videos and are making them available for free?  Don’t they understand that all it takes is one great teacher to create a video series and then thousands of teachers can use it?

Technology Review has an interesting article about the new Massive Open Online Courses (MOOCS) that are becoming very popular.  The Nicholas Carr makes an interesting comparison between the hype surrounding MOOCs and a description of the revolutionary new idea from a century ago – correspondence courses.  Apparently they were so popular that in the 1920s that there were 4 times as many “distance learning” students as there were traditional students.  It was thought that these courses would transform education.  Yet today, not many people point to correspondence courses a watershed moment in education.  So will MOOCs follow the same trajectory?

I think there is one huge factor that will give MOOCS long term viability – cost.  Providing instruction to a remote student via snail mail is moderately less expensive than providing the same instruction in person.   So colleges in the 1920s still had to charge for enrollment.  But with a MOOC the cost of delivering a course to one more student approaches nothing, allowing the course to be  free to anyone.  The most costly parts of the class – lectures, grading, tutoring, and moderated discussions – have largely been automated.

Mr. Carr also points to the higher drop-out rate seen in MOOCS.   His example is that out of the 155,000 people that signed up for the an MIT course on electronic circuits, 85% had dropped out before completing the first set of problems, and only 7000 (5%) ended up passing the class.  But this is one of those ‘the glass is 95% empty’ kind of things. In a normal year only 175 students complete this course.  With the online version 40 times as many students have taken and passed the course, with a marginal cost of almost nothing.

And what of the 95% who dropped out?  I think that is the wrong term. When a course is free it encourages students to ‘drop IN‘ to a class and see what it is like. Most may have found that the course was too hard or just not for them.  Others may have been surprised to discover an aptitude for the material or an interest in something new. They may have continued on to become one of the 5%.  The traditional college structure would make this kind of experimentation prohibitively expensive. But with the MOOCs this becomes free.

I just found another site for online courses called Udacity. Actually these are not just online courses but these qualify as MOOCS (Massive Open Online Courses). The course is taken by anywhere from 10,000 to 100,000 people at once – that is the ‘massive’ part.  Then these students then take part in an equally massive discussion board. They even have the option of taking a proctored exam and getting a certification for the course (for a small fee).

According to this article in Slashdot, a recent AI course was taught to a group that included both Stanford traditional students and online students.  The highest scoring Stanford student in the class was only the 411th highest scoring student when compared to the online students. That sounds like the definition of a disruptive technology to me.

Currently there are less than 20 courses in the catalog, but new courses are being added and the the course that are there are very high qualify.  Who better to teach the Artificial Intelligence course then Sebastian Thrun and Shayan Doroudi, the team from Google and Stanford that developed Google’s autonomous driving cars.

I just got done reading an article on spanking. as well as reading the study mentioned in the article.   I like to read studies like these to see if I can find evidence of bias.  I often find bias introduced by the researchers themselves and then magnified by the people who quote the study.  Here is the beginning of the article in the LA Times:

A child who is spanked, slapped, grabbed or shoved as a form of punishment runs a higher risk of becoming an adult who suffers from a wide range of mental and personality disorders…

and later in the article:

They concluded that the nation’s physicians should explicitly tell parents that physical punishment, including spanking, smacking and slapping, “should not be used on children of any age.

Now the average person reading this article would get the clear impression that it is the spanking that causes the mental disorders. Nowhere is it mentioned that it may actually be the other way around – that people who have a mental disorders might exhibit specific behaviors as children.  And that these behaviors might lead to more severe punishments.  But if you dig through the study you will found exactly that idea:

Findings from this research should be considered in light of several important limitations.  First, the cross-sectional design precludes determining any causal inferences in the  relationship between harsh physical punishment and mental disorders.

In other words, there is no evidence to determine which is the cause and which is the effect.  And yet the study recommends that pediatricians tell parents not to use physical punishment in any circumstances because it is “associated” with some mental disorders:

Harsh physical punishment in the absence of child maltreatment is associated with mood disorders, anxiety disorders, substance abuse/dependence, and personality disorders in a general population sample.

So how would I improve this study?  I would limit the study to children who had siblings, and then I would change the focus of the question.  Rather than asking, ‘did you (personally) experience physical punishment as a child?’ I would have asked ‘did your parents use physical punishment on you or your siblings’.  This way you are focusing on the relationship between the parenting style and the offspring outcome – separate from the behavior of individual children.

Earlier this year two students from Stanford launched Coursera, which offers online courses from several ivy league schools. These courses include graded quizzes and certificates of completion.  This article in Forbes is great.  My favorite quote is this one:

One major benefit of the online classroom over the lecture hall is the ease with which data can be collected and used to improve courses. Data on which test questions are most frequently missed, to the amount of time spent on each question is collected. Through the data points, they can better understand what works and what doesn’t work – and use that information to improve the courses.

The graded quizzes are a huge step forward for student feedback.  Coursera also allows the students to organize into study groups, which gives some of the peer interaction at the heart of traditional colleges. Coursera’s website includes a description of their teaching methodology, and the research on which it is based.  The mention studies that show the ideal method is a combination of online and traditional, which may be why some of Coursera’s partner institutions are considering encouraging their own students to use Coursera resources while taking classes.  I think we are approaching a tipping point where the online model becomes as acceptable as the traditional model.

And here is the exciting part. Even if it can’t produce students at the level of a Harvard graduate the effect of educating millions of students – students that can’t currently afford a traditional college education – is huge.  The only current barrier preventing the full impact in the developing world is language, since all of the courses are in English.  But through translation, or through cloning of the model by foreign universities, the language barrier will be not be permanent.  So I can envision the student potential that is currently locked up in developing countries finally starting to be realized, which will raise living standards across the board.

Another huge money sink in college (and grade school), is in the cost of textbooks. I am sure that in some subjects it makes sense to replace texts every year, but in many subjects this is not necessary.  I remember as a music major being expected to buy the new edition of the music history book, when the history covered in that book hadn’t changed in many years.

There is a new movement, copied from the open-source software movement, to create open source textbooks.  A professor or a group of professors get together and write a text book, then make it available for free on the internet. Other professors are free to take the base materials and customize them to create exactly the materials that they want. Students can use digital versions for free or they can make or buy print copies for minimal costs.

The main challenge with these open courses is that they are often hard to find and even harder to evaluate.  But a recently announced project at the University of Minnesota could change that.  They have begun an Open Academics Textbook Catalog to list and evaluate the available open source textbooks.  All of the textbooks listed must be:

• Openly Licensed (e.g. Creative Commons)
• Complete (similar in scope to traditional textbooks)
• Suitable for General Use
• Available in Print

Then faculty volunteer to review textbooks in their area of expertise.  The University’s College of Education and Human Development (CEHD) plans to provide stipends of $500 to $1,000 to professors who review and adopt open textbooks. The CEHD is also giving iPads to all incoming freshmen so they can use the less expensive open textbook formats.

College is very expensive and the biggest expense of any college is professor salaries.  So it makes sense to make the most efficient use of the professors’ time.  So why do we have  thousands of professors all over the country delivering the same exact lecture?  Some will do a poor job and some will do a great job.  So why not find the 2 or 3 best lectures on each topic and have the students all watch that lecture on video? Does anyone seriously think that sitting in a live (but boring) lecture is more effective than watching a great lecture on video? What about questions and discussion? You can have the local professor available to answer questions, and you can even schedule discussion time as a class.  My point is that the lecture itself doesn’t need to be live.  In many core classes it doesn’t make sense to pay thousands of different professors to all prepare and deliver the same lecture year after year.

I get catalogs from a company called  The Teaching Company and that is exactly what they do.  They take lectures from the best professors in each field and make DVDs of them for anyone to watch.  These lectures are relatively expensive and they are designed for consumers rather than college students, but both of those things could be changed so that college lectures all over are done by the best lecturers available.

This is already happening in high schools.  A group called TED (Technology, Entertainment and Design) is launching a web service were short video lessons are made available, for free, on the internet.  They take the best lectures, produce them with quality video and professional animation and put it where anyone can watch it.  These are already being used by high school teachers all over the world.

There is also the Kahn Academy, produced by Salman Kahn.  It started out as video math tutorials for his  family and now has a collection of 3000  videos covering a dozen or more subjects.

Even Yale and MIT have started to make some courses available on the web:

Open Yale Courses provides free and open access to a selection of introductory courses taught by their professors, including videos of lectures. The site says,  “The aim of the project is to expand access to educational materials for all who wish to learn.”

MIT has an Open Courseware Project with the course materials for over 2000 courses on line, and about 150 of those include video lectures.