OnlineFail

100% Online Class – 300% Fail

It was sexy, so enticing. Take the Stanford Advanced Program Management course (SAPM) online and save 50% over the on-campus, in-person cost. The online course featured the same professors, same lectures, the same class materials – all online, all accessible at any time. Take the classes at your leisure, complete the test at the end of each class, six classes, and voila a SAPM Certificate.

Is it possible to digitize college content, drive out costs, deliver terrific content and create prodigious prodigy? 

College Classes as Digital Content

I consider myself a smart guy. Not brilliant, but certainly capable of completing an online program. I did hesitate when my company refused to pay for the program – which means that I’m on the hook for the $9000 tuition. Academic capacity (check), economic capacity (conditional check), execution (big red ink "Fail"). 

The first copy of Digital Content is expensive to make. The second copy is not. Consider the obvious examples of music, books, and software. How many hours, weeks, months, years, does it take to create a book, or produce an album, or write a program? How long does it take to produce a movie? Once it is in digital format, what is the cost of the second copy?

Achilles only had one weak spot.

Massive Open Online Courses – The “MOOC”

The internet is great at distributing content. The MOOC is “a course of study made available over the Internet without charge to a very large number of people” (Oxford Online Dictionary).

The first course in the SAPM program includes a section about Jeffrey Moore’s work “Crossing the Chasm” which leverages the Rogers Adoption of Innovation curve. The MOOC model needs to address some significant product failures: enrollment and completion statistics are poor, course design and course delivery do not guarantee education outcomes, and a poorly defined customer is causing poor product alignment and is blocking adoption. 

Other Articles on the Subject: 

• Chronicle of Higher Education (April 2015): “What is Being Learned from MOOCs”
• Preparing for the Digital University: (Study)  http://linkresearchlab.org/PreparingDigitalUniversity.pdf

Issue 1 - Enrollment vs Completion Stats (Adoption)

Justin Riech (@bjfr) from EdWeek.org provides evidence of sub 5% completion rates in the article “TheTruthiness of MOOC Completion Rates”. He cites The Atlantic “The Tricky Task of Figuring out What Makes a MOOCSuccessful.

“A 5 percent completion rate is low for a conventional college course but high for other forms of media with which MOOCs share much in common.  One of the first HarvardX courses, JusticeX, was originally produced as a PBS series by WGBH in Boston. Professor Michael Sandel has 12 video lectures from that series that were posted on YouTube in September of 2009. The first video has nearly 5 million views. The second has 1.2 million views. By the fifth video, views have declined to about 200,000 views for each video.

Rather than decry this “5 percent completion rate” as a crisis in public broadcasting, we find it remarkable that Michael Sandel can post a 45 minute video lecture on moral reasoning and get 5 million views. Anyone who watched one video or all twelve is a little wiser for their efforts.”  

The Atlantic article seems to prefer a liberal definition of success – if 5 million people see something it makes the world a better place. This seems to answer a slightly different question: “is the world a better place if knowledge is easy to access?” It seems evident that access is not the same as application. Where does "viral" have more meaning that "useful"?

The Horizons Tracker blog culls a quick profile of MOOC attendees from a study based on the California Community College system. Sadly, The Horizons Tracker does not link to the study, but I am including their interesting summary of student profiles for discussion:

• Bystandars are students who are probably at the lowest end of the engagement ladder.  They’ll sign up for a course, but then not really engage with it.  Some might not even log-in to the course once it begins.
• Collectors on the other hand are slightly further up the ladder.  They’ll consume the video content provided by the course, but they won’t do a great deal of interaction with fellow students.
• Viewers are similar to the collectors in that their primary means of engagement with the course is via the lectures.  Despite watching the content however, the viewer is unlikely to complete many of the assignments.
• Solvers tend to be polar opposite to the viewers.  They’ll do a lot of the assignments, without necessarily having watched the lectures beforehand.
• All rounders are undoubtedly the Holy Grail however, as these are the people that do it all.  They’ll watch the lectures and do the assignments. 

My undergraduate classes at the University of San Francisco had a similar “funnel” shape. Early lectures were in classes up to 150, and my senior year classes were under 20. As a student progresses in a major, the audience shrinks to match the specificity of the curriculum. 

Draw a Venn diagram of the student profiles in the Horizons and the EdWeek. How would the overlap compare to a traditional college Venn? My guess: Colleges do not suffer from significant numbers of bystanders, collectors or viewers. Sure, some students hang-out in college for a long time, and student loans allow some to attempt college. Current tuition for a year at my Alma Mater, USF, tops $42,000. How many classes can you act as bystander, collector, viewer at those rates? The low cost model of online classes and MOOCs does not seem to drive the proper behavior which is the completion of a course of study.

Returning to the SAPM discussion about product adoption: Can MOOC students that take a single course even be considered “adopters”? 

In Rogers’ example – the farmers adopted (used) new techniques to improve their crop yield. They had to take the risk to change. In Moore’s example, from a product point of view, the marginal students are being represented as the tornado (leading MOOC to "Main Street". Can education be sampled like the newest hit song and be declared a success? The Web 2.0 world of “build it free and they will come” posits the argument: get people into your funnel – fast, worry about monetizing them later. But many economist can provide a lecture about the perils of “risk-less” and “costless” decisions.  

Why are MOOCs successful overseas, compared to the US? My very simple hypothesis: being a marginal student overseas is a guarantee of a harsh life. Ignoring a path to a better life is horribly expensive.

Issue 2 – The Frankenstein Syllabus (Design)

The SAPM class is designed as a collection of videos from classroom lectures, online case studies, supporting videos, and a couple of books (Gerstner and Moore). Exactly the curriculum assortment that should make online classes so attractive. Each section had a quiz, and some had basic exercises to recap class concepts. But the class final exam was sheer ugliness.

Imagine a class taught by five professors, with dozens of links to external supporting material including videos of several prominent company leaders. Everyone is describing the elephant from their point of view. Now, imagine that the final exam is created by a fourth party -- that cannot see the elephant either. If criticism of the use of Adjunct Professors is any indicator, the use of teaching assistants or, (worse) third and fourth-party administrators is a terrible weakness.

Education is now offered on a continuum from pure in-class to pure online. The middle road is described as the blended classroom or the hybrid classroom – which has varying commitments to in-person or virtual-class participation. Online classes deprecate the contribution of the role played by lecturers and teachers. Teachers play the adhesive role of coaches and mentors. Pure online (MOOC or otherwise) deprecates both student-teacher interaction and student-student interaction.

My undergraduate degree was 100% in class (in the early 1980s). My MBA was a hybrid offering with formal class times and set times for online participation. The SAPM program is 100% online. Over thirty years the classroom dynamics changed. The interplay of ideas and the cross-pollination of cultural perspective have been totally lost. Presenting any observation for conversation, whether it is straw man or fable or parable, or business history, or simple opinion would elicit various perspectives in a class setting. Discuss: Did Steve Jobs’ paranoid need for secrecy help or hinder Apple?

If class material changes, substantially or otherwise, can the course be edited, or must a new course be designed? I may own several versions of the same song -- the LP version, the EP version, the 7”, the Live version, the 10-year reunion re-release. Really, It’s the Same Old Song. It likely changed over time and was played with minor variations every single time. Consumption of digital music is not a good comparison to the consumption of knowledge.

Worse, what happens when students complete the coursework and then fail the final exam? (Yes, it took me more than a few attempts to pass the SAPM final test). Did the material change, and not the test? Did the test reflect preference for the opinion of Professor A over textbook A, or video expert C? The SAPM is decidedly Pass/Fail – but does not provide the instructor with feedback on efficacy. Heck, after re-reading the last statement I'm not convinced there is an instructor, there is not even a proctor, only a test scoring algorithm.

What I Really Needed: The Uber Tutor

Does the MOOC model lead to the need for on-demand Private Tutors? Tyler Cowen (@tylercowen)  of the Marginal Revolution Blog makes a compelling argument: “You do know that private tutors are the missing secret element in MOOCs,right?  All sorts of internet learning will go better when private tutors are available on demand.  Yet for the non-wealthy it does not always make sense to hire a private tutor on an ongoing basis.  Still, you might have a few questions which can be cleared up in fifteen minutes or so, if only the person were available on relatively short notice.” Again, the wealthy and the motivated can pass the test, check the box, and start signaling. But that's not really the goal, is it?

Education as a Packaged Product

Universities continue to miss the mark when choosing online material. Phil Hill (@PhilOnEdTech) makes the point. “Rather than asking “which students need UF Online and what support do they need”, the institution is asking “what do we need and how can we use UF Online to fill any gaps?"

A university or a company that designs a product to fill its perceived gaps – rather than listening to their customers would fail the session about Moore. Consider the example of IBM pre- Louis Gerstner).  Does the removal of research and facilitating a social community – which leaves only teaching) improve outcomes? Feldstein doesn’t think so.

Paul Gordon Brown – argues that knowledge acquisition is not the hard part, access to opportunities is a bit more difficult, teaching cognitive and employable skills are much harder, and personal transformation requires human intervention. This video covers all of those topics and describes the lag that accompanies transition to the new learning model.

Secret Ingredient

The World Economic Forum addresses the question “ Why OnlineLearning Will Fail” in their article by Jason Pitts (Feb 2015)  “Danish study quantifies the extent to which educational outcomes are not the only valuable service that universities deliver – namely that a further and large component of the implicit demand for education is assortative mating – and MOOCs might not be so good at that.”. Brown cites Gustav Bruze or proof that college is a place to meet prospective husbands and wives (a valuable use of signaling).

In addition to signaling for perspective mates, college graduation is a signal to potential employers.  Bryan Caplan (@bryan_caplan) of George Mason University argues that this signal is suffering from diminished returns in the post “Bryan Caplan on College Signaling an Human Capital”. “ So you have to do something that only smart, hardworking people will not find prohibitively costly. Interestingly, lowering the costs of a degree degrades the value of the signal. This will drive credential inflation, forcing the truly smart and hardworking to pursue graduate studies, or some other costly pointless thing, in order to differentiate themselves.”

Realigning the Product Online Options:

How do MOOC and the pure online learning model overcome the three issues of enrollment and completion course design and delivery and poor product alignment?

Redefine the Target Audience

Option One: Remedial and Paired Track offerings and the corequisite model. This option requires qualification testing – to determine the collection of course materials, and for strict assessment of student learning. The arguments have already started, and neither side seems to have the data to back their position (opinions are in abundance). 

“There's no question there is a movement afoot across the country to implement corequisite remediation and to do it to scale,” said Bruce Vandal, vice president of Complete College America, which receives most of its funding from the Bill and Melinda Gates Foundation.

Inside Higher Ed is not convinced. “…we have concerns is that a corequisite [approach] might not serve students who are the most vulnerable or the most needy… “It doesn't allow us to address the diversity of students that come to us with different levels of academic preparedness… what might work for one student might not work for others."

First World Offering (Texas Example)

Complete College Texas makes their case in this document. “Fewer than one-third of Texas students entering four-year universities graduate within four years… fewer than one-sixth of students complete community college in two-years. Over the course of five years, more than $440 million in state appropriations and grant aid goes to university dropouts, as well as more than $60 million in federal grants. Each year, Texans earn $57 million worth of college credits that are not needed for their degrees.”

Their Recommendation (from the Texas Tribune):

• Financial Support tied to College Performance
• Corequisite (rather than prerequisite courses)
• Promotion of "On-Time" graduation timeline w/15 units per semester, raising from the minimum 9 units)

Developing World Offering

If the underlying assumption of developing world offerings removes student motivation as an obstacle, then the problem focused squarely on assessment and credentials. Will student investment, specifically in time spent, lead to a credential with any signal strength? Is the student willing to be learned, but unattractive to employers?

Bottom Line

Product adoption has implied cost. If online classes and MOOC want to unseat the traditional education model it will need to re-align its’ product offering. At this point, Online Learning and MOOC appear to be a product searching for a customer. Colleges and universities will continue to dominate the educational product offerings providing a strong signal for employers (and probably for marriageable partners), while completing an online course or MOOC signals something else.