Tuesday, March 13, 2018

edX: HarvardX Data Science - R Basics Review


Data Science: R Basics is the first course in a new 9-part Data Science professional certificate program offered by HarvardX on the edX MOOC platform. The 4-week course is intended as a first course in data science and programming in R, covering such basics as R data types, vectors, sorting, indexing and functions. The course has no prerequisites, making it a suitable first programming course for students interested in data science. Grading is based on a series of programming exercises offered through DataCamp, a data science learning platform that lets you work through exercises using an interactive web-based programming interface.


Each week of R Basics consists of a series of three to four subsections that cover different aspects of a main section topic for week. Subtopics typically contain a few short lecture videos that span from 2-10 minutes in length followed by a programming assignment. The course instructor Rafael Irizarry does a great job explaining topics in plain language that everyone should be able understand and avoids the pitfall of going off on tangents or alluding to more advanced topics in data science and R that would confuse beginners. The video quality itself is great, with simple crisp visuals, good audio and essentially 100% professor face-time as he is always on screen as a green screen overlay with the code and figures.


As with most programming-focused courses, students are likely to spend most of their time working on the programming assignments in this course. The programming assignments on DataCamp are administered as a series of questions with specific instructions aimed at familiarizing students with R functionality while also performing common data science tasks on a real data set. DataCamp's programming assignment system is great for learning programming for the first time since the specific instructions keep students from feeling like they don't know where to start and you can always get hints or even view the full solutions if you get stuck at the cost of losing points.


Data Science: R Basics is an excellent first course for complete beginners to start their journey with R programming and data science. This course should also prove useful to those who might already know how to program in another language, but would like to learn R. My only real complaint is that the data used throughout the course and its assignments is related to murders in the U.S. which is a bit of a morbid subject for what is meant to be a light and highly accessible introductory course.


I give Data Science: R Basics 5 out of 5 stars: Excellent. 



*Note: The course is relatively short: complete beginners should be able to complete each week with a couple of hours of effort and more advanced learners can probably complete each week in an hour or so. As such, don't expect to learn much outside the very basics of R; topics like visualization and predictive modeling will be covered in greater detail in forthcoming courses in the series.

No comments:

Post a Comment

Note: Only a member of this blog may post a comment.