For those who work with R and d3.js, the differences between the two are obvious. But for those who are brand new to this world, the names might as well be gibberish. This quick primer is for the latter group.
Read More

Stay safe, Florida.

The Simpsons is in its 27th season. That’s a lot of d’ohs. Todd W. Schneider had a look at the dialogue over the years, prominent characters, and declining ratings.

Did I mention 27 seasons? How is that even possible?

Artist Nicholas Rouguex visualized some musical scores in his project Off the Staff.

I can’t read music but I can parse it. The talent of reading music has always escaped me which is a little ironic considering I grew up in a musical family. However, I’ve always enjoyed how sheet music looks so I took a shot at visualizing the notes from musical scores and the result is this series of posters.

Notes were placed in a circular layout, evenly spaced. Each colored circle represents a note, and the size of the circle represents the length of the note. Colors represent instruments. Put it all together, and you get these diagrams that look like tie dye.

In an update to his two-party map, political scientist David Sparks shows the evolution of the two-party system across the country, since 1920. Red and blue. Keeps on moving.

[arve url=”https://youtu.be/I9aLRsMTk_o”]

Picking basketball players for a professional team is no easy task. College and high school players picked high in the draft might end up average in the NBA, and those picked low might actually be fine professional players. Russell Goldenberg for Polygraph dives into 20 years of NBA draft picks to find the biggest busts, the biggest steals, and convergence to the mean.

After the first presidential debate, #TrumpWon was a trending topic on Twitter, which led many to believe that there were bots involved — maybe from Russia. It didn’t help that a fake map of Saint Petersburg with a bunch of bubbles on it went viral too. The real reasons for the trending hashtag are much more mundane.

The general public kind of knows about data privacy issues. But not really. Or they know, and they’re willing to trade for the ability to share things easily on social media.

I liken it to people who eat hot dogs but avoid animal parts that aren’t in the shape of a steak. As long as it’s packaged right and you can’t see the bits, it must be okay. It’s similar to why people burn the American Community Survey but flip through their phone no problem.

ProPublica aims to dig into the black box a bit in their ongoing coverage on machine bias. Most recently they released the first of a four-episode video series: Breaking the Black Box. They started with Facebook. The video is accompanied by a Chrome extension that “lets you see what Facebook says it knows about you” and more interestingly, lets you rate the items and send accuracy back to ProPublica for analysis.

Check it out on ProPublica.

Each year, the National Highway Traffic Safety Administration releases data for traffic accidents that resulted in deaths. I briefly looked at the data a while back. Others did too. For years, there have been fewer deaths than the year before, but the 2015 data shows a 7.2 percent increase.

So, the NHTSA released the data earlier this year and, along with the White House, formally reached out to the data community to analyze the data.

DOT is aggressively seeking ways to improve safety on the roads. From our work with the auto industry to improve vehicle safety, to new solutions to behavioral challenges like drunk, drugged, distracted and drowsy driving, we know we need to find novel solutions to old challenges.

We’re also looking to accelerate technologies that may make driving safer, including connected and highly automated vehicles.

But we need your help, too! Data Science is a team sport.

Yes.

We know that income increased by a lot in 2015, but for who? If it’s just the rich getting richer, then the rise doesn’t mean much for a lot of people, but as Quoctrung Bui for the Upshot points out, this was not the case.

Real median incomes rose 5.2 percent in 2015 — phenomenal growth by economic standards. And 3.5 million people moved out of poverty. But more important, 2015 was encouraging to economists because of where income growth was concentrated: the poor and middle class.

I’m not 100% sure what to call this chart type, but the going term seems to be barcode chart. Each vertical line represents a year, and each row represents an income bracket. Mouse over a bar and you see the change in income for all brackets during the corresponding year.

Using data from their Washington Post-ABC News polling, the Washington Post compares shifting support for Clinton and Trump among several demographic groups.

The presidential contest is often compared to a horse race, with the candidates fighting to finish in first place on Election Day. We offer a bit of a different metaphor here. The campaign is also a series of simultaneously fought tug-of-war matches for different demographic groups — based on gender, age, and race/ethnicity, among others.

By rotating the time series to put time on the vertical and candidate lead values on the horizontal, the graphic provides a different metaphor: more left, more right, and mostly middle. I like it.

Thermal cameras, which use infrared to detect heat, provide images of temperature. Firefighters can use them to find people in smokey rooms, law enforcement can use them for surveillance, and technicians can use them to detect power faults. Brea Souders used one to take pictures of everyday foods. [via kottke]

Missing data is everywhere. Or, I guess technically it’s nowhere. You know what I mean. Missing data is common, especially with temporal data over long periods of time. Just look at country-level estimates for anything over ten years, and you’ll understand. Peter Beshai for Bocoup provides five solutions for line charts with missing data, using d3.js.

This is what happens when there is a disconnect between data and what it represents. So much wrong.
Read More

A challenge of using geographic maps to show data is that larger regions inevitably get more visual real estate. Cartograms try to solve this problem by sizing regions by the data instead of land mass. Tilegrams by Pitch Interactive makes it easy to construct tiled versions at the US state level.

Upload your data, adjust the resolution to your liking, and export for your own purposes. Nice.

In case you didn’t know, there’s a drought here in California so there are rules for when you can and can’t water your grass and plants. Not everyone adheres to those rules though. And some households really don’t follow the rules. In Los Angeles, or more specifically, Bel Air, there are a handful of households using millions of gallons per year. Michael Corey and Lance Williams for Reveal used satellite data to guess which ones.

I mainly share this though for the title of their post that explains how they did it: Now this is a story all about how we found the Wet Princes of Bel Air. Genius.

Nate Cohn for the Upshot provides a statistics lesson in the context of election forecasts and why they differ so much.

[P]ollsters make a series of decisions when designing their survey, from determining likely voters to adjusting their respondents to match the demographics of the electorate. These decisions are hard. They usually take place behind the scenes, and they can make a huge difference.

To make the point, the Upshot gave four pollers the same raw data to interpret. Results varied.