Podcasts aren’t new, but they are definitely trending now. According to Carrie Ryan, author of Serial Box’s Dead Air, podcasts’ popularity can be attributed to the fact that the format uniquely fits our busy lives: “It’s information and entertainment, in bite-size chunks, right at our fingertips. Plus, the format of podcasts allows a deeper dive into many subjects — it creates the possibility of long-form investigations in a very accessible medium. While many of us might not take the time to sit down and read a 20k word length article, we can listen to that article in bite-sized chunks during our commute to work, especially if it’s presented in an entertaining, narrative way.”
The vast variety of podcasts is another catalyst — there’s a podcast for everyone. Want to learn more about celebrities’ secret fears? Check out I’m Afraid That. Can’t get enough of Queer Eye’s Jonathan Van Ness? Get curious with Getting Curious with Jonathan Van Ness. Intrigued by espionage? Spycast. There really is a podcast out there for everyone, and in some cases, many. I have discovered that there are more than a few podcasts written by a for people like me — unashamed data nerds. Here are my favorites.
Data Skeptic is one of the most popular podcasts in the industry and my go-to data-related podcast. Its episodes range from 15 minutes to the occasional hour-long deep dive, but I would guesstimate that the average length falls around 25 minutes — perfect for a quick commute or something to listen to while you make breakfast.
Kyle Polich, the show’s host, earned his BS in computer science and MS in artificial intelligence from the University of Illinois at Chicago. In every episode, Polich approaches — with a healthy dose of scientific skepticism — topics in data science, statistics, machine learning, and artificial intelligence.
Unlike other podcasts, every episode of Data Skeptic does not follow a formula — some episodes feature interviews with industry professionals and innovators (e.g., Feather — a conversation with Hadley Wickham!), others introduce new tools or strategies for handling data, some are structured like mini-lessons (e.g., What is Heteroskedasticity?), and still others are light-hearted explorations of subjects that appeal to data nerds (e.g., Scientific Studies of People’s Relationship to Music and Mapping Dialects with Twitter Data).
While some episodes are pieces of larger series, most episodes of Data Skeptic can be appreciated out of order at your leisure — feel free to jump around and explore the extensive archive of episodes on the Data Skepticsite.Data Skeptic is also available on Apple Podcasts, Google Play, and Stitcher.
DataFramed is the official podcast of DataCamp — a free, educational resource for all things data science, machine learning, and statistics (a site worth checking out if you haven’t already!). Hugo Bowne-Anderson is the podcasts’ host and producer and a data scientist, educator, and amateur stand-up comedian.
Each of the 50+ episodes includes a roughly one hour-long discussion with an industry professional. A little bit long, yes, but the episodical structure means you can pick and choose the topics and guests that interest you, and Bowne-Anderson’s light-hearted personality (and interest in the promotion of data and AI literacy) helps each interview feel more like a friendly conversation.
The podcast attempts to shine a light on the question, “What is Data Science?” by considering what data science can do, and the problems data science professionals are trying to solve through their work. Some of my favorite episodes are Pharmaceuticals and Data Science (with Max Kuhn), Data Security, Data Privacy, and the GDPR (with Katharine Jarmul), and The Credibility Crisis in Data Science (with Skipper Seabold).
Because of its educational nature and variety of topics and perspectives, I recommend this podcast to fledgling data scientists and data science students. Not only is it a good means to learn more about the field, but it’s also a great introduction to the community.
Data Stories is unique in the realm of data-related podcasts because it narrows in on the products of nitty-gritty data analyses: data visualizations. This podcast is hosted by the dynamic duo of Enrico Bertini — a respected researcher in the field of data visualization with a background in computer science — and Moritz Stefaner — a globetrotting “independent ‘Truth & Beauty Operator’ on the crossroads of data visualization, user interface design and information aesthetics.”
Since 2012, they’ve recorded more than 130 episodes, most of which, have featured a conversation between Bertini, Stefaner, and a special guest (this is the show’s standard formula, but there are a few exceptions). Sometimes, they discuss their guest’s projects or work, but occasionally their exchanges are less-specific and address the broader spectrum of data visualizations or issues in the field. One of my favorite episodes is Ep. 104: Visualization Literacy In Elementary School with Basak Alper and Nathalie Riche, which draws the correct conclusion that the promotion of data literacy must include an emphasis on visualization literacy as well.
Not only does this podcast spark ideas about new ways of visualizing data, but it forces one to consider aspects of data visualization — opportunities and potential pitfalls — that one otherwise would not.