Monday, 18 February 2013

Keep on MOOCing

Hello all, apologies for the wait, been a bit busy with MOOCs.

So, brace yourself, this post is going to be a bit wordy, but hopefully worth it if you are curious about Massive  Open Online Courses. You can find out a lot about MOOCs  by Googling, but what you can find out here are my experiences as a student of three MOOCs and how they compare.

My 1st MOOC was +Alberto Cairo's Introduction to Infographics and Data Visualisations in association with the Knight Centre, or #iidviz. It was the direct cause of this blog, which was a way of posting my assignments.

My current MOOC is Jeff Leek's +Coursera Data Analytics - it doesn't have its own hashtag, but a few (a very few) are using #cdatan, and is the reason I've been too busy to post.

My brand new MOOC is canvas network's  Learning Analytics Knowledge #LAK13, which actively recommends students to start a blog as part of the course requirements. (Got that covered)

What they all have in common is that they concern representing data, which I have decided I'm interested in, not least because as a philosopher up until 2002 I was concerned with theories of representation, and as a health informatics data analyst since 2004, my days are spent wrestling meaning out of lists of numbers and variables. 

So, let's start with the present: Jeff Leek's #cdatan

This course is HARD. I'm no dummy, I have a PhD (though that's no guarantee) and I work with data every day, but this course contains vast amounts of new information, new techniques and new concepts. This is not meant to dissuade anyone, it is reassuringly difficult. On the basis that anything worth having should come at a cost, I feel that I am genuinely pushing myself. Jeff has to teach a lot of people a lot of things in a short time, and he is doing that by conscientiously providing as much information & instruction as he can. 

It is an indication of both the value and the challenge of this course that (what seems to me) a disproportionate number of those who introduce themselves on the fora possess higher degrees, and all, regardless of prior education, genuinely appear to be  seeking to add to their skillset and enhance their professional capability.

My criticisms of this course (for I have them) are not to do with content, nor workload. They concern structure and communication. 

The structural problem is that the weekly lectures and quizs are strictly drip-fed to the students. There is good reason to control the timing of the peer assessed assignments, but there is no reason to withhold the course materials until their allotted release date. By making everything you can available at the outset, you allow people to work both at their own pace, and to manage their time far more efficiently. This is something that could easily be changed, and some on the fora have requested it, but Jeff Leek does not seem to be reading the messages.

This is my second criticism. The level of communication between tutor and students is virtually non-existent, to the extent that recently students had to mount a semi-rebellion to gain his attention by voting up a single critical post en masse - even if it wasn't a precise expression of their concerns. This, combined with the failure to encourage engagement on social media by setting up at least a hashtag (necessary because he would have to inform everyone of it) if not a Google+ community, lends quite a sterile feel to what should be a far more collaborative experience. 


+Alberto Cairo's #iidviz was a wholly enjoyable and collaborative experience. Cairo himself was everywhere, offering praise or constructive criticism in the fora, highlighting particularly interesting or important posts, even engaging in the off topic fora (we had a very interesting conversation about dataviz and comic books). He was active on twitter, posting both personally and through #iidviz, retweeting student's posts and posts of those he was following. His obvious enthusiasm and engagement defined the course and.

Also - and remember this was my first MOOC so I wasn't entirely sure that this was out of the ordinary - he was maximally responsive to peoples requests for change where they had problems. Deadlines were extended, the entire structure of the forum system was revised to make it easier for people to engage with each other, lectures were posted early to give people more time, and I think that one quiz was  waived altogether to allow students to concentrate on their final assignment.

If I hadn't undertaken my current MOOC, I never would have recognised how remarkable #iidviz was, and how much Jeff Leek, and I guess @coursera, could learn from it. I'd tweet Jeff to come and read this blog, but I don't know his handle..or if he has one. 

I should add that the course materials for #iidviz were just as valuable as those on #cdatan, although with a much broader scope. This I believe is just about a different focus, #cdatan is narrowly focussed on data analysis, whereas #iidviz takes in a wide sweep of historical, political and real-life examples.

I heartily recommend both MOOCs, despite my wish that #cdatan could have a more social media dimension. They complement each other perfectly, and its a fact that Jeff has the less sexy end of the data visualisation stick. I have every intention of completing #cdatan, and making the tough skills that Jeff Leek is passing on my own (even if I have to take it twice - not joking)

#LAK13 is a very interesting MOOC and I have still to see more than an introductory video. I enrolled because I was advised that you could engage with it asynchronously, despite originally being reluctant because of the #cdatan workload. This is indeed an advantage, as it speaks to a criticism I made above about @coursera's method of drip feeding. 

As it turns out, it also speaks to my desire for a greater exploitation of social media in delivering the learning experience. This seems to be a major driver of a course which is primarily run by educational specialists. I am just as curious about how the course 'feels' as to learning more about the tools involved such as Gephi and Tableau (full) which I have played with before... should also let me turn this back into a #viziblog instead of a #chattyblog 

(this "joke" added for the brave who read all the way down to the bottom of the page - statistically, you are few)