9 May 2018
If you walk into an average Data Science department and ask if anyone has taken a course on Coursera, EdX or DataCamp, you’ll be surprised how many hands go up. Mostly amongst the younger generation, but also more experienced colleagues are MOOCing. MOOCs are a low-cost and flexible way to broaden and deepen knowledge. But there is a mind-boggling amount of choice, so there’s always a risk of having frustrating experiences and ending up disappointed.
We want to share with you how we see it, to help you figure out if and when a MOOC is valuable for you. We can offer three main tips for selecting the best MOOC. We want to share with you how learning actually works and have built our own experiences with MOOCs into it.
First things first, what are MOOCs and where do they come from? MOOC stands for Massive Open Online Course and they have been around since 2011. These early MOOCs were aimed at making academic courses accessible to a wide audience and were attended by hundreds of thousands of students. In 2011, Andrew Ng of Stanford University started setting up the first major MOOC platform: Coursera. Coursera remains one of the biggest providers, although there are now also numerous alternative providers, such as Edx, Udacy, FutureLearn, DataCamp and many more. Class-central.com provides a nice overview of the MOOCs and, as of 2018, now counts more than 30 MOOC platforms and no fewer than 9,400 courses… and counting…
The popularity of MOOCs can be explained by a variety of features. First of all, they are easy to access: the first MOOCs were all free to use. Now, most of them require payment. That payment varies from a mere €20 but can rise to serious sums of money. There are also more and more subscription formats where you buy access to all the content on a MOOC platform a month at a time. Their popularity is a result of other characteristics of MOOCs:
If we zoom in on our own field of work, data analysis and data science, the range of MOOCs on offer is huge now. Do you want to learn Python or R? Before you know it you’ll be choosing among literally hundreds (!) of alternatives. You want to learn how to make a forecasting model or a deep neural net? You’re spoilt for choice. Which one will you choose? The one your friend or colleague did too? Or the one taken by most people? The version you can do at your own pace or rather the one with the strict schedule? The one that gives you a digital certificate for your resume or LinkedIn profile?
Do you recognise some of these hesitations about signing up for a MOOC? Or do you already have a whole series of completed MOOCs but are still left somewhat unsatisfied? Making the decision of whether or not to take a MOOC, and which one is best for you, becomes a lot easier if you first clarify what the most important thing is for you to learn.
To give you a bit more background to how learning actually works, we use Bloom’s Taxonomy. This is a popular methodology to classify different levels of knowledge. The taxonomy is based on the theory that you can think on six different levels. First of all you have the two levels of Knowledge and Comprehending: this is about being able to repeat and summarise information. For data analysis and data science, for example, this means you know there are various different modelling techniques and understand what these techniques do. When you have reached this level, you can start toApply information. At this level you actually start to use those modelling techniques, e.g. to come up with a forecasting model. These three levels, in Bloom’s Taxonomy, are categorised together as lower order thinking.
Then you can start with the next levels, i.e. Analysing and Evaluating. This draws upon higher order thinking: you need to be able to reflect critically, research and be able to identify and resolve problems. For data scientists and data analysts, this means applying learned techniques in a different context. In other fields or on other data, and you can deal with setbacks. If you can do all that, then you can proceed to the last step: to Synthesize. You are able to develop a forecasting model from scratch and adapt it to the context.
Well you might be thinking “OK, I’ll get started with synthesizing and then I’ll get the hang of it all perfectly.” Well, no – to learn properly, you have to really master the lower thinking levels first. You can’t spontaneously start making Deep Neural Networks in Python if you don’t understand what neural networks are and don’t know Python.
Based on our experiences with a number of different MOOCs, MOOC platforms and using Bloom’s Taxonomy, we identify the three biggest tips for fledgling MOOC participants.
1. Clearly identify what is your most important learning objective and what you already know
It sounds obvious but – backed up by Bloom’s Taxonomy – it is important to figure out exactly where you are in terms of your knowledge of the subject before you start on a MOOC. The introduction films are often inspiring, but before you know it you’re being sucked into a MOOC that has little to do with what you want to learn. We have already mentioned the Deep Neural Networks in Python. It’s incredibly interesting, but not wise as a “first MOOC” if both neural networks and Python are both still unexplored territory for you.
Think properly about what you want to learn first: Mastering Python? You could start with an introduction to Python to first master sections such as numpy, pandas and matplotlib. Or are you primarily interested in understanding neural networks? If so, choose a MOOC that focuses on properly explaining the theory behind them and corresponds to your background knowledge. Do you have a clear focus on your learning objective? If so, you can start going through an initial long list of MOOCs on Google or on MOOC sites such as class-central.com.
2. Think about what keeps you on track
A MOOC will keep you busy anything from a few weeks to a couple of months. During your MOOC journey, there will always be unexpected things in life that eat up time. The time you want to be spending on your MOOC. How do you make sure you won’t drop out half way and will get to the finish line? MOOC platforms and MOOCs are very different when it comes to encouraging you to stay on track. Not every strategy will work for you. For some people a big menacing stick works best, for other a carrot, promising something, works better. So, before you start on a MOOC, think about what kind of stimulationworks best for you and what absolutely doesn’t. Go through this checklist when you evaluate your long list of courses:
3. Choose thoroughly and make a plan!
Because MOOCs do not impose much in terms of enty barriers, it can seem – quite incorrectly! – as if it doesn’t really matter which MOOC you choose. And it doesn’t matter when you start on the MOOC either, as surely you can start again at any time anyway? But that’s annoying, as you’ve invested your (free) time! (also read this article about why this can turn out be problematic) So before you start on a carefully selected MOOC check:
Finally, one more important tip for the moment you finally, after weeks of sweating and wading through videos, quizzes and exercises, reach the end of the MOOC. After successfully completing your first MOOC, you are likely to experience a rush aka an acute transcendent state of euphoria and suddenly find yourself in a MOOC flow. You look back with pride on your MOOC journey and that’s a feeling you want to keep! What next MOOC should I take??
Your enthusiasm and sense of pride are of course entirely justified: you did a great job. What’s more, you have gained so much new knowledge that you have understood, mostly remembered and even applied. But if we go back to Bloom for a moment: your learning has only just begun! After all, you are still mainly doing lower order thinking. In order to get to higher order thinking, you will have to apply what you learned in new contexts. You can do that by applying what you learned in your day-to-day activities – at work, in your studies, in the evenings. When you have learned a new analysis technique or programming language, look for opportunities to put what you learned to practice.
So can I not do any new MOOCs for a while? Don’t worry, MOOCs can play a major role in deepening your knowledge, but only if the MOOC properly corresponds to the previous one. Look in detail at the recommended follow-on MOOCs and check whether they go into more depth in the direction that is relevant for you. Some MOOC platforms help you to do this, by putting a series of MOOCs together into one programme, therefore introducing you gradually in more and more depth to a subject area. A great example is the Data Science Specialization by John Hopkins University on Coursera. Various components of the field of data analysis are explained in 10 consecutive MOOCs. You will start to practise them and then continue building on them in the following MOOCs. Therefore, by applying the knowledge from the previous MOOC in the next one, you will become proficient in what you are learning in no time!
Do you want to know more about this subject? Please contact Jurriaan Nagelkerke using the details below
17 April 2023
In every company it’s a struggle to make sure we only keep the documents we want... read more
7 September 2022
We don’t know if you’ve heard already, but there is yet another crisis on our horizon:... read more
15 September 2021
Project-Friday At Cmotions, we love a challenge. Especially those that make us both think and have... read more