• September 21, 2015



    Trey Causey just finished an interview roundabout for data science jobs. He outlines his experiences and describes what interviewers seem to want, what questions to expect, and what to expect from yourself.

    Sooner or later you’re going to find yourself looking for a data science job. Maybe it’s your first one or maybe you’re changing jobs. Even if you’re fully confident in your skills, have no impostor syndrome, and have tons of inside leads at great companies, it’s a tremendously stressful experience. The process of looking for a new job is often one that occurs secretly and confidentially and then is so exhausting that discussing the process is the last thing you want to do. I hope to change that.

    A must-read for those about to get your feet wet.

    See also Causey’s short guide on getting started with data science.

  • September 18, 2015

    In a clean and simple set of slope charts, Alyson Hurt for NPR shows the shifts in power sources — coal, gas, nuclear, hydroelectric, and renewables — from 2004 to 2014. As you might guess, coal power output is down in most states and natural gas is up. On a national scale, the hydroelectric and renewable sources need more time.

    Grab the data from the U.S. Energy Information Administration to look yourself.

  • September 17, 2015


    Maps  /  , ,

    Luis Dilger made a set of fine-looking prints that show city landscapes in 3-D. They look like little cardboard cutouts.
    Read More

  • September 16, 2015

    You know the thing. It’s the triangle of numbers that you learned about in high school. Each number in a row is the sum of the two numbers above it in the previous row. Of course, as explained in the video below, there’s more to it than that. SECRETS REVEALED.

    [via kottke]

  • September 16, 2015

    It is estimated that over 200,000 people have been killed during the Syrian civil war. That’s a lot of lives. Lives. In a striking representation by the New York Times, a dot represents each life lost.

  • Who Earned a Higher Salary Than You

    Work changed over the years. Salaries changed over the years. I was curious: If you compared your personal income from present day, how would it compare to the distribution of salaries in previous decades?

  • September 14, 2015


    Data Sources  / 

    In a very non-government-like release (in a good way), the U.S. Department of Education provides detailed data for college debt, graduation rates, test scores, and more. It’s at the program-level, and there’s even a front-facing College Scorecard that lets you look up information for your university.

    And it doesn’t look and work like an outdated government site. With all of my frustrations with government sites, the education release feels pretty great. It’s as if the department actually wants us to look at the data. Imagine that.

    You can download the data as a single ZIP file, access it via the data.gov API, and most importantly, there’s documentation.

    Seriously, this is good stuff, and if it’s any indicator for where government data is headed, there could be good things to come.

  • Feed Sponsor
    September 14, 2015

    Thanks to Statistics Views for sponsoring the feed this week.


    Statistics Views and John Wiley & Sons Ltd are celebrating World Statistics Day 2015 with a Best Data Visualization competition. Could it be your data visualization that wins you either an Apple watch or $500 worth of Amazon vouchers?!

    Your data visualization needs to relate to international data across 5 continents and be submitted before Friday 2nd October 23:59 (EDT). Entries will be judged by a panel of esteemed members from statistical societies including the American Statistical Association, the Royal Statistical Society and the European Network of Industrial and Business Statistics. The winning data visualisation will be announced on World Statistics Day (20th October 2015).

    You can find everything else you need to know about entering the competition on the Statistics Views website. Good luck!

  • September 14, 2015

    So you have your data neat and tidy in a single spreadsheet, and it’s finally time to explore. There’s a problem though. Maybe you don’t know what to look for or where to start. Maybe you’re not in the mood for a trip to clicksville to make all those charts. With a new exploration tab, Google Sheets might be a good place to start.
    Read More

  • September 11, 2015

    Here’s a fun project to try over the weekend. Hannah Mitt and Andrew Morrison came up with a neat hack using an old Android device and a two-way mirror to make a future-y information display. It shows date, time, and weather, reminders, and the most recent xkcd.

    Just import their project to your device, mount it to the mirror, and mount the whole thing to the wall. Done.

  • September 10, 2015


    Maps  /  , ,

    There are a lot of trees on this planet. But how many trees there actually are is still kind of fuzzy, because the estimates are based on satellite imagery. It’s hard to gauge density. Research by T. W. Crowther et al., recently published in Nature, used on-the-ground sampling to estimate more accurately.

    The global extent and distribution of forest trees is central to our understanding of the terrestrial biosphere. We provide the first spatially continuous map of forest tree density at a global scale. This map reveals that the global number of trees is approximately 3.04 trillion, an order of magnitude higher than the previous estimate. Of these trees, approximately 1.39 trillion exist in tropical and subtropical forests, with 0.74 trillion in boreal regions and 0.61 trillion in temperate regions.

  • September 10, 2015


    Statistics  /  ,

    Apps peak and die on a regular basis. One day everyone is giving an app a go and your feed fills up with links to the service, and the next it’s business as usual. BuzzFeed took a straightforward look at such trends through the eyes of tweets. All they had to do was count tweets that linked to particular service over time.

  • Feed Sponsor
    September 9, 2015
    Thanks to Metis for sponsoring the feed this week.


    September 16 – October 28
    Mondays & Wednesdays
    6:30 – 9:30pm
    enroll here


    Enrollments close very soon for the Data Visualization with D3.js course at Metis in New York city and a few seats remain.

    Designed and taught by Kevin Quealy, Graphics Editor for the New York Times, this course is for anyone who wants to be proficient in the use of D3 and seeks expertise visualizing quantitative information. You’ll learn to tell stories and communicate information interactively in ways that are simply not possible outside a web browser.

    The 6-week course is held on Monday and Wednesday evenings from 6:30 – 9:30pm at Metis, 27 East 28th Street, New York City where the Metis Data Science Bootcamp is also held.

    Watch an interview with Kevin Quealy.

    Course Outcomes:

    • Proficiency in using D3 to make static and interactive charts and documents, and in using JavaScript to process and manipulate data.
    • A working conceptual understanding of the field of data visualization, particularly as it relates to the internet and mobile devices.
    • Deep knowledge of the forms and techniques of data visualization and effective display of quantitative information; especially, bar charts, scatterplots, area charts, line charts, choropleth and bubble maps, small multiples, annotation principles; and the strengths and weaknesses of each.

    About Kevin Quealy:

    Kevin Quealy is Graphics Editor at The New York Times and a contributor to The Upshot, the Times’ data-centric vertical about policy, politics and everyday life. He has taught journalism and data visualization courses at N.Y.U., the University of California, Berkeley and the City University of New York.

    Before coming to The New York Times, Kevin served as a Peace Corps volunteer in South Africa. He has a Master’s degree from the Missouri School of Journalism and a B.A. in physics from Gustavus Adolphus College in St. Peter, Minnesota.

  • September 9, 2015

    Stock market spoofers put in orders to buy with the intent to cancel. This can shift prices up with fake interest, or it can shift prices down with a wave of cancellations. The spoofers then take advantage of the shifts by buying and selling accordingly. Bloomberg has an interesting stepper that walks you through the process for how one might catch such spoofers.

    It starts with an overview. A minute of buying, selling, and cancellations whiz across the screen, and all looks hunky-dory. But then it zooms in on the details to show you what to look for, and it doesn’t look like such a flurry anymore.

    The challenge is that regular people cancel orders all the time, and the activity itself is not illegal. More data needed.

  • Venn Diagrams: Read and Use Them the Right Way

    Venn diagrams seem straightforward, but why all the mistakes? Here’s a guide to avoid the snafus.

  • September 8, 2015


    Maps  /  , ,

    Damien Saunder, a cartographer at ESRI, likes to use mapping methods to evaluate tennis player patterns and tendencies.

    When I look at tennis, I see it moving on a grid. I see space and x/y coordinates [position] and I see z values [height], and I see trajectories of the balls, and space opening up. I started GameSetMap to try and educate people of the value of mapping where people are on the court, storing the data in a GIS, and visualizing it.

  • September 8, 2015

    I don’t know the full context of this discussion, but in the interview below, Hans Rosling talks to media person Adam Holm about why we shouldn’t use the media to form our opinions about the world. Media person disputes. Rosling puts foot on table and says Holm is wrong.

    Hard to argue with that.

    See also Rosling’s 2014 TED talk on how to not be ignorant about the world.

  • September 4, 2015


    Statistics  /  ,

    There’s a small site dedicated to Bayesian-informed fantasy football decisions, because of course there is. Here’s the 101 intro.

    Here’s the crux of thinking probabilistically about fantasy football: for any given week, when you start a player you’re picking out one of these little x’s at random. Each x is equally likely to get picked. Each score, however, is not. There are a lot more x’s between 0-10 points than there are between 20 and 30.

  • September 4, 2015


    Data Art  / 

    Weevmee autogenrates a woven-like image, based on your Instagram photos.

    A lot of us are pretty into photographing our lives. We can look back at each individual image and it magically transports us back in time and engenders a memory. That’s great. But at the end of a long, fun year, we had no creative way to evoke the many memories contained in an entire year’s worth of images. We were hungry to create a single, personal manifestation of a year in photos – one that feels artistic enough to grab our attention, but contains enough clarity that it feels uncannily familiar.

    Link your account, set some criteria, and you’re off to the races.

  • September 3, 2015

    The job of a tennis line judge can be though when you have to judge the difference of a few millimeters as a ball speeds by. Sure, it’s easy to complain about bad calls at home, where we get to see replays in slow motion, but it’s more challenging in real life. The Wall Street Journal provides a bit of the experience with an interactive game. Watch video clips from a line judge’s point of view, and try to make the right call.