All posts from FlowingData

A couple of months back, WNYC's The Brian Lehrer Show asked listeners who have moved to or away from New York some questions. They asked current zipcode, previous zipcode, year of move, and some other questions. BLS then posted the data and let information and data folk have a go at it. Here are the results.
My favorite is Moritz Stefaner's interactive (above). You've seen his stuff around here before. Red indicates more moving out and blue indicates more people moving in. The New York area is enlarged in the white circle since most moves happened within the state, and the rest are placed using a damped distance function. Updating bar graphs on the right provide more context. Check it out, even if just for the cool factor.
A static look of the data by designer Andrea Stranger is pretty interesting too. The map below shows moves from 2000 to 2010, and it's accompanied by smaller maps for each year.

Check out the rest here. Some good stuff in there.

Where Americans are moving
New York Times on how they design their graphics
My Last Day as a New York Times Graphics Editor Intern

I trust we've all seen the OldSpice YouTube campaign by now? This graphic from Know Your Meme categorizes videos by who they were directed to and how many views they received. For example, a video to Joe Blow would be in the low-profile category, while responses to Alyssa Milano go to the high-profile category.
It's fun subject matter and interesting to look at, but it does miss some chances to be more informative. I mean, the story here isn't just number of views. It's more about the distribution of views and videos. We don't really get that from these visuals. Instead the main thing we get out of the graphic is the five most viewed videos.
That's not to say it's bad. But there is a good amount of room for improvement. How can we make this graphic more meaningful and tell a more complete story? Here's the full-sized version. Leave your constructive comments below.
[via]

Discuss: Driving is why you’re fat?
Discuss: Flowchart on drinkable water in the world
BP tries to mislead you with graphs

The tennis US Open is in full swing, and since you're at work, you probably need a way to keep up with all of the matches. In a collaboration between the US Open and IBM, this real-time display shows you what's going on during any given match.
New for the 2010 US Open, the USTA and IBM are introducing a "beta" release of US Open PointStream -- a new way to enhance your US Open experience. PointStream visualizes match data and stats in real-time to give you insight into the way your favorite players are performing.
It's sort of confusing at first, especially if you don't know anything about tennis. There are just so many more encodings than you're used to interpreting. That said, if you're a tennis nut and have it running in the background, it's easy to see how this could be useful as an ambient display, as opposed to an analysis tool.
Here are the main encodings. Each player has a color: green or blue. The inner ring is a calculated measure of who currently is at an advantage during the match. The outer ring shows progress of the current match. Bar height indicates serving speed and white lines indicate aces. Small squares on top of bars show winners. Finally, the rings in the middle show previous sets.
Still confused and have suggestions on how to improve it? Let them know. It's still in beta, and they're taking feedback. I was unsure what I was looking at at first, but it's grown on me after staring at it for a little.
[Thanks, Jeff]

Many Eyes on The Times U.S. Open Blog
How to Create a Real-Time Web Traffic Map for Your Site
Researchers Map Chaos Inside Cancer Cell

FlowingData reader Chris asks:
I was wondering, have you ever considered doing a Chernoff faces tutorial for R? I think Chernoff faces are pretty interesting and I haven't seen much about them on the web.
This wasn't the first time someone's asked how to make Chernoff faces, so I did a quick search. Guess what. There's an R library for that. This tutorial describes how to apply Chernoff faces to your own data.
Chernoff Faces
The point of Chernoff faces is to display multiple variables at once by positioning parts of the human face, such as ears, hair, eyes, and nose, based on numbers in a dataset. The assumption is that we can read people's faces easily in real life, so we should be able to recognize small differences when they represent data. Now that's a pretty big assumption, but debate aside, they're fun to make.
We've seen them applied to baseball players and judge ratings. In this tutorial, we'll look at US crime rate by state.
Download R
Like in previous tutorials, we'll be using R (surprise, surprise), the software environment for statistical computing and graphics, to make our Chernoff faces, so if you haven't already, download and install R first before moving on. It's free, open-source, and a one-click install. Go on, I'll wait for you.
Step 1. Install package
Once you've opened up R, the first thing we need to do is install the aplpack (Another Plot Package) package by Peter Wolf. Go to the the "Packages & Data" menu in R, and select the "Package Installer." Select "CRAN (binaries)" in the dropdown menu if it's not already on that, and then click on "Get List." Scroll down to "aplpack" and click on the "Install Selected" button and installation should begin.

Alternatively, you can also just type this in the R console:
install.packages("aplpack")
Step 2. Load the data
Next we need to load the data into the R environment. Like I said, we'll be looking at crime rates by state. I got the data from Infochimps, which is actually from Table 301 of the 2008 US Statistical Abstract, but it's typically a headache going through dot gov navigation, so I avoid it when I can.
I cleaned the datafile I got from Infochimps a little bit more so it only includes the numbers we're interested in. You can find it here, but you don't need to download it. We'll load it directly into R via the URL using the read.csv() command.
crime
To view the data, type the following:
crime[1:6,]
This shows you the first six lines of our dataset. Note that there are eight columns. The first column is state name, with the exception of the row for US average and District of Columbia later on. The rest of the columns are seven categories of crime.
Step 3. Make some faces
Once the data is in, it's actually really easy to make some faces using the faces() function from the aplpack package. So far we've only installed the package, so now we'll load it:
library(aplpack)
If you get errors when you try to load, you might want to check to see if you installed the library correctly.
Okay, let's make some faces:
faces(crime[,2:8])
Here we're telling R to use the faces() function, using columns 2 through 8 of our crime data. Remember, the first column is state name. You get something that looks like this:

Step 4. Change Features
This is pretty much what we want except for two things. The first is that the faces are labeled with numbers. That isn't of much use without a key. The second is that some of the faces are smiling. For more positive datasets like quality of life or baseball stats, that would make sense. The higher the value, the better. This is crime data though. The higher the value, the worse. Smiles for rate of larceny theft doesn't seem quite right.
Unfortunately, the faces() function doesn't let us choose what face parts to associate with each metric, so we need to find a workaround. According to the documentation (view by typing ?faces), the curve of the smile is applied to the sixth column in the input matrix, which is crime in this case.
Ah. Here's what we'll do. We make the sixth column in our data all the same value. That way all smile curves will be neutral. Here's how we can do that:
crime_filled
The cbind() function combines multiple columns to form a matrix. In the above, we combine the first six columns of crime, stick a column of zeros whose length matches the number of rows in our crime data, and then we end with the last two columns in crime. We save the new matrix into a variable called crime_filled. Similar to in Step 2, you can type the following to see the first rows of crime_filled.
crime_filled[1:6,]
Notice the new column of zeros?
Now use faces() with crime_filled:
faces(crime_filled[,2:8])
We get similar faces, but with no more smiles:

Step 5. Add labels
Instead of numbers, it'd be much more useful to include state names. Easy.
faces(crime_filled[,2:8], labels=crime_filled$state)
It's the same as previous, but we use the labels argument to use the state column in crime_filled to label with state names.

Much more useful now. We can easily associate the faces with a state. It's a little cluttered, but we can fix that up easy in Illustrator.
Step 6. Fix up in Illustrator (optional)
You can pretty much stop here if you like, but as most of you know, I like to save the image as a PDF, bring it into Adobe Illustrator (aff), and clean things up to make it more readable. You can also try Inkscape, the open-source alternative, although I've never tried it.
After some label cleanup and some annotation, here's our final result. What's going on there Washington, D.C.?

Not too bad, right?
Read the R documentation on faces() for more details on what else you can do with the function. Remember, documentation is your friend when it comes to making full use of R.
Now go on. Have some fun with your new Chernoff toy.
Got a visualization question? Post it in the forums.

Chernoff Faces to Display Baseball Managers From 2007 MLB Season
An Easy Way to Make a Treemap
How to Make a Heatmap – a Quick and Easy Solution

Auger Loizeau, in collaboration with Reyer Zwiggelaar and Bashar Al-Rjoub, describe their smart-home project Happylife. It monitors facial expressions and movements to estimate a family's mood, displayed via four glowing orbs on the wall, one for each member:
We built a visual display linked to the thermal image camera. This employs facial recognition to differentiate between members of the family. Each member has one rotary dial and one RGB LED display effectively acting like emotional barometers. These show current state and predicted state, the predicted state being based on years of accumulated statistical data.
They also include a few quite beautiful vignettes from a family that has Happylife in their home. While there are no concrete metrics or instructions on how to read the displays, the family does draw some kind of emotional insights and sometimes finds comfort in the glow:
It was that time of the year. All of the Happylife prediction dials had spun anti-clockwise, like barometers reacting to an incoming storm. we lost David 4 years ago and the system was anticipating our coming sadness. We found this strangely comforting.

The irony here is that despite being called Happylife, the vignettes are actually kind of depressing, as are the cold, blue glowing orbs. And in The Veldt by Ray Bradbury, a story from the anthology that this project draws inspiration from, a new technology called the "Happylife home" is introduced. The parents are killed by virtual deadly lions. Coincidentally, that's last on my list of ways I want to die.
[via]

Detailed View of the Kennedy Family Tree
Displaying Data as Efficiently as Possible
Big Happy Republican Family

Connoisseur of scaled rounded rectangles, bubbles, and triangles, David McCandless of Information is Beautiful talks data visualization in recently posted TED talk (below). He explains how information design can help us get through information glut on the Web and how simple charts can show patterns that we never would have seen otherwise. He uses his own works and collaborations as evidence.
Right around the 8-minute mark, David makes an interesting point about learning design. He explains that he started as a programmer, and then was a writer for about twenty years, but only recently started designing. He picked up the skills on his own. David argues that he's not unique, and that actually, everyone (like you) can do stuff like this too, because we've all been exposed to so much information that we have a sense of what makes it understandable.
That sounds about right. Obviously, a formal design education is going to help, but a lot of the skills you learn come from practice and experience.
With the proper resources to help you with foundations such as software, code, and basic principles, you open up a whole bunch of different paths that you can explore on your own.
Watch David's full talk below.

[Thanks, Lara]

Why Isn’t Data Visualization More Popular?
Daily Design Workout – DONE by Jonas Buntenbruch
Data Visualization Blogs You Might Not Know About

This animation by the National Astronomy and Ionosphere Center shows a beautiful view of the past 30 years of asteroid discoveries, using data culled by Ted Bowell et al.
As time passes, asteroid are highlighted white and then colored by how closely it comes it comes to our inner solar system. Earth crossers are red, Earth approachers are yellow, and all others are colored green.
What you get is a view of the solar system's planets and asteroids orbiting the sun and these sparkles in sky. As automated sky scanning systems come online in the 1990s, we see waves of discoveries. Then starting at the beginning of 2010, we see a discovery pattern as a result of the Wide-field Infrared Survey Explorer, which has been tasked with mapping all infrared light in the sky.
Watch the full video below.

[Thanks, Cay]

Do you know who owns the Arctic? As it turns out, it's a pretty messy subject:
In August 2007 Russian scientists sent a submarine to the Arctic Ocean seabed at 90° North to gather data in support of Russia's claim that the North Pole is part of the Russian continental shelf. The expedition provoked a hostile reaction from other Arctic littoral states and prompted media speculation that Russia's action might trigger a "new Cold War" over the resources of the Arctic.
Luckily things are at least a little more in control now though. Well, sort of. Canada, Denmark and the US still need to define their continental shelf limits. Keep in mind that the shelf can be more than 200 nautical miles from these countries' coastal baselines.
The International Boundaries Research Unit provides this map [pdf] of claimed boundaries and areas that will potentially be claimed in the future.
[via]

Nmap visualizes site popularity as scaled icons. Favicons, that is. They're that little icon that shows in your address bar or when you bookmark a site in your browser. If you're reading this on FlowingData, you should see a little red icon next to the URL. The larger the icon, the more popular the site is, based on Alexa traffic data. In whole, the image is a giant 37,440 by 37,440 pixels image. Google is 11,936 x 11,936 pixels. Facebook is 6,736 × 6,736 pixels. Yahoo is 6,544 × 6,544 pixels.
And I'm happy to report FlowingData is a whopping 80 × 80 pixels. Aw yeah. Zoom in about a billion times, and you'll find the icon in there:

Is your site in there? Enter a URL in the search field to find out.
The Web is a big and diverse place. You know that. I know that. But it's always nice to see just how diverse it is.
[via | Thanks, Elise]

This is part two in a seven-part series on basic rules for graph design. Rule #1 was to check your data. Today we cover rule #2: explain your encodings.
The design of every graph follows a familiar flow. You get the data, you encode the data with circles, bars, or colors, and then you let others read it. The readers have to decode your encodings at this point. What do those circles, bars, or colors represent?
William S. Cleveland and Robert McGill have written about encodings in detail. Some encodings work better than others. But it won't matter what you choose if readers don't know what the encodings represent. If they can't decode, the time you spent designing your graphic goes to waste.
We sometimes see this lack of context with graphics that are somewhere in between data art and infographic. We definitely see it a lot with data art. I understand why. I mean, a label or legend could totally mess up the vibe of a piece of work, but at the very least, include some information in a short description paragraph. It helps others appreciate your efforts.
Other times we see this in actual data graphics, which can be very frustrating for readers. Sometimes we forget, because we're actually working with the data. Remember that readers usually come to a graphic blind and lack the context that we, as the creators, already know.
In the end, whatever the case is, make sure you double check or have someone look things over to see if it makes sense to an outsider. Add some annotation if there's anything that causes confusion.
How to explain your encodings
So how do we make sure readers can decode our encodings? We explain what they mean with labels, legends, and keys. Which one you choose will vary depending on the situation. For example, if you have bubbles on a map, you could have some scaled bubbles with number labels to provide scale. If you use a color scale, provide a key for what each color means, or you can directly label things if there is enough space and not too many categories.
As you can see, these are all very easy things to do, but they can make a huge difference on how your graphic reads, so explain your encodings wisely.
Stay tuned for rule #3: label your axes.

Wilson Andrews and Alicia Parlapiano report for The Washington Post on how the fight over tax breaks affects your bottom line:
Tax cuts enacted under former president George W. Bush are set to expire at year's end, and lawmakers are battling over whether to extend them before the November elections. Most Republicans want to extend all of the cuts, saying that any increase in taxes will hold back the economic recovery. President Obama and Democratic leaders would extend many of the cuts but say tax breaks for top earners should expire to pare down deficits. Each plan would affect average tax rates for income groups differently.
Each row represents an income group, and you can flip between letting Bush's tax cuts expire, shifting to Obama's plan, and extending the current cuts. Bubbles on the right show the average tax change per taxpayer for each income group. Switch from the first option (letting all cuts expire) to the second (Obama's plan), and you'll notice some changes for top earners.
Moving from left to right, you start with a base rate, and then provisions increase that rate with the final new rate at the end of the row. This part slightly confused me — and maybe this is because I've been a graduate student for the past five years and taxes have never been a major concern for me — but after some fiddling it seems to make sense.
Overall, and correct me if I'm wrong here, under Obama's plan, taxes go down compared to if we were to just let the cuts expire and would go up, compared to Bush's cuts.
However, if we extend cuts or implement new ones, it "costs" the government a few trillion dollars over ten years.
[via]

What country has the best education? Health? Quality of life? Thomas Klepl and Adam Clarkson of Newsweek take a look at important metrics for the world's best countries. It's basically a parallel coordinates plot turn on its side. Each show represents a metric, and each circle in a row is a country.
Select a country from the list on the left or by directly interacting with the plot. If a country is top in all categories, like Finland, then all of the scores are going to be on the right. Burkina Faso, on the other hand, is all the way to the left. Of course, this is only the "top" 100 countries.
You can also filter by geographic regions, income, and population groups.
While I'm not totally sure about the ranking system and methodology, it's an interesting look.
[Thanks, Adam]