Problems with self-learning and MOOCs
Why are MOOCs not taking off, why do we still need classrooms and colleges?
I think MOOCs are a revolutionary idea but they still have a long way to go, in this post I try to discuss issues with MOOCs which I have first hand experienced.
Philip Guo in his blog post Why take classes (when you can just learn everything from the internet or books)?, talks about the benefits of classes over self-learning and MOOCs. He says
self-teaching works well for some people some of the time, there are benefits that classes provide over learning on your own:
Along with that, He lists the following benefits of classroom based learning over self-learning and MOOCs
Accountability, Deadlines, Practice, Community, Curation, Context, Enthusiasm, Mentoring and Equity
I completely agree with Philip Guo on the benefits of classrooms but classrooms have big downsides. They are not accessible to masses, not everyone can afford, or has a schedule which allows them to go to classes(e.g: working professionals).
Today there are tons of platforms out there, ready to teach you something in exchange for your money. They will also give you a certificate to verify that you learned from them.
Side-note: is that certificate worth anything, does that certificate hold any value? Popular opinion on the internet says that those certifications offer zero to very little value. employers value projects much more than just plain certificates
I am sure you bought some online courses but stopped midway when you got bored, or stuck during the assignment and dropped out, the same happened with me, I registered for online courses on edx and then decided to do something else. Reminder emails about assignment due dates kept showing up in inbox but either we didn’t end up starting that course, or even if we started we didn’t finish it.
Along with these issues, a lot of times I am lost in the sea of resources. I don’t know where to start, what to read up first or what should be the path in an unexplored area.
The ML Experience
When I started learning ML, I was deep down in the big list of free and paid resources but didn’t had time to try out all of them. I also had hard time figuring out which one to pick and where to start. I decided to start with basics but then I got bored because I already knew most of the things that were discussed in basic courses, so I decided to jump to advanced courses but turned out it was too advanced for me. I was doing a binary search for the right course in the sea of content and I starting suffering from content overload. I was looking for structure and guidance on how to approach this foreign field, so I did what anyone would do, read up on how to break in this new area but almost all of those blog post and Reddit threads were filled with links to resources which I already had in my list, I already tried most of them so I was lost.
This was my experience trying to break in a new field with self-learning, and failing miserably
my curated sea of resources, notice the scroll bars
If we talk about ML as filed which is on top of the hype cycle right now. I think this problem is happening with lots of folks, you hear lots of ML courses are showing up left and right, some of them are great (i.e fast.ai), but some of them are scamming students and ripping people off and plagiarizing.
I think people fall for these scams is because they are overwhelmed by the sea of content, they might be suffering from content overload too. They are looking for community, guidance, and peers who can hold them accountable and keep them motivated. These scams are taking advantage of students by claiming to offer the above.
I am not saying all online courses are bad, I had a very good experience with Udacity and edx but I also feel that they are lacking in a lot of things that are offered by classrooms.
The Dropout Problem
Other then content overload, another problem with self-learning is that dropout rates are very high for online courses, even when the course is paid, learners will not complete fully or they will just do enough to get the certificate.
A recent study by academics at the Massachusetts Institute of Technology found that online courses had an astronomical dropout rate of about 96 per cent on average over five years. – Moocs struggle to lift rock-bottom completion rates, Financial Times
I have dropped out from multiple online courses, have skipped assignments that are not part of the certificate, so I think I can answer why I dropped out?
For me, the answer is that it stopped being fun, even though I was learning, I wasn’t enjoying. It might be because, I was just sitting in front of the computer screen watching hours of videos. The whole learning experience felt too dry, because of which my motivation significantly dropped by the end of the course. I didn’t had any accountability, or network which pushed me, encouraged me to keep going.
Philip Guo also mentions it in his blog post that the self-learning requires very high interstice motivation. Along with that, I would also like to add that even with high interstice motivation you may find yourself demotivated at the end of an online course(i.e motivation depletion), and due to lack of motivation you might drop out.
You need feedback loops that keep your motivation level high throughout the whole journey. These feedback loops which help you recharge your motivation are essential.
We need to address these issues of motivation depletion and content overload to make self-learning a viable option for the masses. If we don’t address these issues then I am afraid that self-learning will only appeal to a small, specialized section of the population.
So the obvious question here is how do we address these problems, and the answer is I don’t know.
I want to know how we can solve these problems like motivation depletion and content overload. Have you had a similar experience or maybe a different experience with MOOCs?. Let me know your thoughts.
I don’t want to say that there is an issue with self-learning and MOOCs as an idea. They are amazing, but they need more work to be more practical for the masses. Some MOOCs are trying out various ways to build community by using Facebook Groups, Reddit and other community mediums where learners interact and help each other, so I think overall things are moving in the good direction.