edX: HarvardX Data Science - Visualization Review

Data Science: Visualization is the second part of the new Data Science professional certificate program offered by Harvard University on the edX MOOC platform. The 4 week course builds on the foundation laid by the first course, R Basics, by covering data visualization in R using the ggplot2 library. The course does not require any background besides taking the R Basics, so it is appropriate for learners of all levels. Grading is based on a eight sets of programming exercises administered through DataCamp, an online platform for interactive programming and learning.


Content in Data Science: Visualization is divided into 5 topic sections: an introduction to data visualizations and distributions, an overview of ggplot2, summarizing data with dplyr, common plots and data viz principles. Each topic section is divided into two are more subsections with several short video lectures explaining key ideas and operations followed by programming assessments in DataCamp allowing you to put what you've learned into practice. The lecture videos are high quality and the instruction is clear an concise. The code necessary to produce plots can get a bit verbose and code-filled screens may look intimidating to new R users but the layer by layer design of ggplot makes things easier to digest than they might be with a different plotting library. Using DataCamp to deliver programming assignments is a good decision since an online platform reduces problems that arise when users have different programming environments. The DataCamp exercises are generally well done although as problems become more complex, instructions need to be increasingly thorough and precise to elicit the intended output. If you produce the results using methods different from those the course expects, your answer may be rejected even if it is correct. The exercises do allow you to  get hints or even view the full answer if you get stuck, but doing so will reduce your final grade, so think twice about doing so if you are going for a certificate.


Data Science: Visualization is an accessible introduction to plotting in R using the ggplot2 package with good pacing and a healthy amount of hands-on practice. There are many features in ggplot2 that this course does not cover, but it gives you a solid foundation you can build from if you want to dive into documentation to add extra bells and whistles to your visualizations.


I give Data Science: Visualization 4.5 out of 5 stars: Great.