Computing for Data Analysis Review (Coursera/John Hopkins)

Cousera's month-long computing for data analysis class offered by John Hopkins University is just wrapping up this week, so it’s time for another course review!


Computing for Data Analysis is an introduction to the R programming language, but not an introduction to programming. This course is designed for people who already know how to program. The course description makes it seem like the class is intended for everyone, even those who do not know how to program at all, which caused a lot of pain in the class forums. The class description estimates the effort required to complete the class at 3-5 hours per week. Although this range is probably accurate for weeks 1 and 4, week 2 and especially week 3 are likely to take significantly longer than that. I wouldn't say the course is impossible for someone who hasn't programmed before, but I imagine it would take dozens of hours to understand what's going on, because the lectures move through material at a rate and level of sophistication that assumes prior programming experience. For example, in week two the lectures cover control flow, loops, defining functions, various built in functions, scoping, higher order functions and debugging. That is a ton of ground to cover in one week of an introductory class that claims 3-5 hours of effort per week. Thus if you have no prior experience I'd avoid this course in favor of a true introduction to computer science course like Udacity 101, Rice's intro to CS on Courera, MIT 6.00, or Harvard CS50. All of these are great courses that either have sessions starting up soon or let you learn at your own pace at any time.


Computing for Data Analysis provides a decent overview of R, but the format is not ideal for learning a new language. The lectures are generally 10-25 minutes with no interactive programming exercises to do as you go along. It’s a good idea to follow along and do the commands he talks about on your own so that you at least get some practice. There’s some good material in the lectures, but they leave a lot out as well. You’ll probably end up spending a substantial amount of time Googling about basic R functions to complete the programming assignments. Thankfully, the course was only a month long, which isn't too much of a commitment for learning the basic nuts and bolts of a new language. This isn't a bad course for someone who already knows programming and wants to learn a bit of R, but it isn't fun or exciting in any way.


I give it 3 out of 5 stars: Satisfactory.