I am sharing my thoughts on making data analytics accessible - and when I say a accessible, I mean accessible to those who may not be even considering it or those who consider it out of bounds for them. This was also the guiding philosophy for my projects in the “Programming for Everyone” series of guided projects. And I will narrate my experience here, to illustrate things we need to consider.
There are two challenges here: 1) Reaching out to those folks and 2) Ensuring that once they sign up, they don’t switch off. Let’s assume that 1) is taken care of, say, thanks to Coursera’s reach. How do we make sure that learners stay on, once they sign up. To do so, we need to make sure that a) they feel confident that they can complete the project and b) we also need to sustain their interest.
a) To sustain their interest, we need to show them that the project will help them build something useful and/or interesting.
b) And to help them feel confident that they can complete the course, the instruction has to be simple. In addition, they should not feel that they need to go through a lot of concepts/foundational material, before they can get to building something useful.
When developing my guided project, b) was a challenge for me, since I was also teaching Programming Foundations. I consciously covered the bare minimum of R Foundations and I had to make it interesting with examples from real life, examples that the learners could relate to. And I wanted to combine the foundational material with the material that could help them create the Web App in one project - splitting them into two ran the risk of learners giving up knowing that they have to go through two projects to create the Web App. And that is how I landed up with a initial project with 25 tasks. . I eventually ended up splitting it into multiple projects.
The challenges with teaching Data Analytics to everyone are slightly bigger. When showing learners how to create a Web App, one can take a guess at what would be an interesting app for a large set of learners. That may not be true with data, since data can be very specific to a person or a group.
What would be of interest would be the methods to work with data. And here, we run the risk of overloading the learners with methods, given the plethora of methods and tools.
A good balance would be to show them methods interspersed with useful examples; rather interesting examples, that they can relate to. So this adds another dimension that we need to factor in: finding interesting datasets.
The idea here is to show learners that data analytics is accessible and show them that it is easy to get started: we arm them with concepts and confidence, tell them that there are enough open source resources available, that they can take the journey on their own. That is what I am aiming to achieve and I look forward to thoughts and suggestions.
PS: Based on some of the titles I have seen in the Peer Review feedback requests, there are quite a few data analytics guided projects already created. I haven’t had a chance to review those. In case any of those projects have taken an approach that I have described, please let me know and I will be happy to review them.