Given that you’re currently reading a data science blog, you’re probably well aware that online resources for an informal education in data science abound. Blogs are a great place to start (here, here, here, here, here), but topics and pedagogical quality are –— let’s be honest –— scattershot at best. No comment on the usefulness of this particular blog…

As I’ve mentioned before, MOOCs are rapidly maturing and expanding into a viable educational resource (with some caveats and warnings, of course). Personally, I’ve learned a great deal through Coursera courses in the past, and am currently enrolled in two: Data Analysis with Jeff Leek and Natural Language Processing with Michael Collins. (Both are now in progress, but maybe it’s not too late to sign up!) Even though I have a great deal of experiential education in these topics —– and formal education in scientific data analysis –— following along with an instructor helps me consolidate prior knowledge and incorporate new information into a more structured conceptual framework. Yes, this is an obvious statement on how one learns.

Attending a conference (or viewing videos/slides after-the-fact) is a great way to network and learn about recent developments in the field. I’ve been meaning to do this for ages but am continually surprised when a major data conference comes and/or goes before I’m made aware of it. Sigh. On that note, Strata 2013 just came to a close, and some videos and presentations are available here and here. For example, check out how one video game publisher is updating its data pipeline and philosophy:

I’ve discussed these and other informal/online resources for learning data science before, but what about people interested in a formal (and perhaps accredited) data science education? Fortunately, data science has been expanding more and more into a traditional forum: Academia.*

A couple weeks ago NYU announced the launch of a new Center for Data Science and the first offering of an M.S. in Data Science this coming fall (now accepting applications). It looks as if they’ll also be incorporating data science into several programs over a range of disciplines — very exciting news! I’m also happy to see that one of their “associated faculty” members is Kyle Cranmer, a high-energy physicist working on the ATLAS experiment, as I used to be.

Last fall, Columbia University’s Department of Statistics offered Introduction to Data Science and published a really excellent blog to go along with it. Incidentally, two of the instructors for the course, Rachel Schutt and Cathy O’Neill aka mathbabe, now share an office with me! :) They’re greatly expanding their efforts with a new Institute for Data Sciences and Engineering. According to their website, the IDSE “is in the process of developing interdisciplinary graduate certification programs, certificates and Master’s degrees to support IDSE’s educational mission. Part-time, full-time and online study opportunities will be available beginning Fall 2013.” It looks like the instructor for my Coursera NLP course is associated with the IDSE –— small world!

And this is just the beginning! Ryan Swanstrom of Data Science 101 (one of my favorite blogs) has compiled a list of data science (or related) programs in higher education. Most are graduate programs, but a handful are for undergrads. I’m a big proponent of self-directed learning and experiential education, but there’s certainly something to be said for a structured, formal education in such a diverse, multi-disciplinary field. And it’s always good to have options! :) Plus, an oft-cited 2011 study by the McKinsey Global Institute projects a shortage of about 150,000 individuals with data science skills by 2018 —– so get learning, people!

* A footnote for the sticklers and cynics: Yes, data science (broadly defined) has long had a place in Academia under the scope of Computer Science and Mathematics/Statistics departments, but no, it’s not just sexy rebranding of an existing discipline. Considering that data science is fundamentally multi-disciplinary, I believe that the recent development of dedicated data science programs represents something new and belongs to a broader trend in education away from excessive specialization.