Amanda Cox, of The New York Times, made another excellent graphic (and I wouldn’t expect anything less). We see an entire story between Obama and Clinton – positions taken, counties won, and counties lost. Go ahead and take a look. Words bad. Picture good. Ooga. Booga.
Moritz has been working on visualization of a hierarchical glossary carefully named “Glossary Visualization” versions 2-5. Not sure where version 1 is. Being a network graph, I can see this getting chaotic when there are more words (or categories) involved, but then again, maybe that’s all the words. In either case, it beats browsing through words in a dictionary; although, these prototypes don’t include definitions yet.
In the most recent version, words are represented as a DOI tree showing only the categories. Click on a category and view the sub-categories.
A while back, I asked, “What is the best way to learn Actionscript for data visualization?” As I’ve had Actionscript staring me in the face for the past two weeks, I can attest to the idea that the best way to learn is by doing i.e. immersing yourself in a project with a deadline looming in the dark behind you. There have been, however, a few things that have made my life a little easier as I strive for coding nirvana. Keep Reading
Facebook recently released Lexicon which is like a Google Trends or Technorati for wall posts. Type in a word or a group of words, and you can see the buzz for those terms in a time series plot. Daniel sent me this excellent example. Type in party tonight, hangover and you’ll get the above graph. Notice the Saturday spikes for party tonight and the Sunday spikes for hangover? Here’s another one for finals:
It’s interesting to see what people are talking about, and being Facebook walls, there’s this realness to the charts (or maybe that’s just me).
Go ahead. Give Lexicon a try. What interesting queries can you find?
Bernard Kerr, the lead designer for del.icio.us, gave an interesting talk (below) focused on remail (mentioned here) and tagorbitals. At the end, he offers three important lessons.
Reduce Multidimensional Data
After showing many thread arc versions, Kerr says that when you are dealing with multidimensional data, pick two variables; otherwise, you’re going to end up with a big mess. He says this literally, but don’t forget that you can also reduce dimensionality with super special and magical statistical methods.
Use Real Data
You won’t know what you’re really dealing with until you have the real data. You can spend lots of time guessing what the data are going to be, but it’s the real data that will eventually drive your design. This goes for statistics too. Real data leads to real analysis.
Try Adobe Illustrator
What rules of thumb do you follow?
Here’s the talk in full. It’s pretty interesting, if you’ve got about 25 minutes to spare.
This video shows statistics centered around atheism, claiming that atheism is correlated with a healthy society. I don’t want to turn this into a religious debate, but I really don’t like these types of videos, slide shows, etc. It’s not the ideas that bother me, but because some people think it’s a great idea to rattle off a bunch of numbers to “prove” a point. Nevermind the biases, invalid studies, poor analysis, cruddy data, and “results” taken out of context.
We all know about information aesthetics, but what other visualization blogs are out there? While writing for FlowingData I’ve come across some good ones as people send me links (hint) or that I’ve just randomly found. Here are some of the visualization (and mapping) blogs that I enjoy.
Strange Maps – Lots of unique maps from ads, books, papers, etc with very informed commentary.
Well-formed Data – Moritz is interested in interface design, visualization, statistics and data mining and is a freelance visualizer.
Random Etc. – Tom occasionally updates his blog with thoughts, resources, and, well, random etc.
Serial Consign – Greg talks about design and research with some visualization mixed in.
AnyGeo – Covers everything geospatial, although I do wish Glenn would switch to full feeds.
What are some of your favorites that others might not know about?
As one of the organizers of the event, Life After a Statistics Doctoral Program (a conference organized by the doctoral students in Columbia’s Statistics Department), I was excited to be invited to guest post on Nathan’s blog but then realized that my perception of the event would be so different than that of an attendee that perhaps I shouldn’t. Two post-docs from Columbia’s Statistics department, Matt and Kenny, agreed that they would post and they did — once on Andrew Gelman’s blog and once on Nathan’s. Keep Reading
The time may not be very remote when it will be understood that for complete initiation as an efficient citizen of one of the new great complex world wide states that are now developing, it is as necessary to be able to compute, to think in averages and maxima and minima, as it is now to be able to read and write.
For each sin we stretched our imagination to find a workable proxy–murder rates for wrath, per capita billionaires for avarice–then culled the available data sources to rank the cities. Some of the results were surprising: Salt Lake City as America’s Vainest City. Some were not: Detroit as America’s Most Murderous.
It’s always good to remember to take these with a grain of salt, since you don’t really know much about the metrics used and how useful these metrics really are. Usually, rankings like these involve a lot of assumptions about the data.
They are of course still interesting and fun to look at though. Apparently, I moved from one America’s most gluttonous cities to one of the most violent and lustful.
This past Friday, Columbia University stat graduate students hosted a symposium on careers for students in statistics. Kenneth Shirley, a stat post doc, was nice enough to write this guest post about the conference so that we can all learn from it. There were two panels – academic and industry – including representation from Google, AT & T, and Pfizer.
Yesterday’s conference at Columbia about career opportunities for Statistics Ph.D. graduates was a great success. It was organized by the graduate students in Columbiaâ€™s Stats department and advertised on the web here:
Andrew Gelman made some opening remarks, and then there were two panel discussions, each with five professional statisticians. The first panel consisted of academic statisticians, and the second panel consisted of industry statisticians. Here are some comments I found interesting. Keep Reading
Transactions Graph, by Burak Arikan, is a piece placing personal transactions in network graph. Each node represents a transaction while connections (or edges) shows a relationship between transactions based on time and spending category. The thicker the edge the greater the total of the two connected transactions. Viewers are also able to scroll through time to watch how transactions evolve. Keep Reading
Stefanie Posavec, maps literary works at the Sheffield Galleries On the Map exhibit. There are several parts to Stefanie’s piece mapping sentence length, writing style, and structure. From the looks of things, it looks like the parsing process was manual and involved a lot of highlighting and circling of things. I could be wrong though. For some reason, long and manual labor makes me appreciate things more. Keep Reading
Check out this lovely use of Chernoff Faces by Steve Wang of Swarthmore College. This method of visualization was developed by none other than mathematician-statistician-physicist Herman Chernoff in 1973. These faces were designed on the premise that people could easily understand facial expressions. With that in mind, Chernoff used facial characteristics to represent multivariate data.
If you like, you can make your own Chernoff faces with this R library.
Energy consumption grows more and more concern, and with the popularity of Mr. Gore’s An Inconvenient Truth, just about everyone is at the very least, semi-aware of energy consumption. These 21 visualizations and designs were created to increase that awareness, so that maybe, a few more people will turn off the light when they leave a room. I think Peter Crabb said it best (which I borrowed from Tiffany Holmes’ ecoviz paper):
[P]eople do not use energy; they use devices and products. How devices and products are designed determines how we use them, which in turn determines rates of energy depletion.
Here they are – 21 dashboards, ambient devices, games, and calculators. Keep Reading
Chris Harrison put together a series of Internet maps that show how cities are interconnected by router configuration. Similar to Aaron Koblin’s Flight Patterns, Chris chose to map only the data, which makes an image that looks a lot like strands of silk stretched from city to city. With these maps, viewers gain a sense of connectivity in the world – and as expected the U.S. and Europe are a lot brighter than the rest. Keep Reading
Let me introduce you to the greatest data visualization of all time. FlowingData readers, greatest data visualization of all time. Greatest data visualization of all time, FlowingData readers. It will blow your mind and affect you to your very core. I haven’t felt this way since 1987 when I first started to walk.
…and OF COURSE the YouTube embed isn’t working, so I guess the link will have to suffice. Ladies and gentleman, be prepared to get up and dance. Here is the greatest visualization that you will ever see. You can thank me in the comments.
Thank you to everyone who left comments and participated in this celebratory contest over the past ten days and for all of the congratulatory wishes. I read every single comment and it only confirms my belief that FlowingData readers are awesome. My favorite discussions were those around the Google API and the redesign of Dolores Labs color cloud. I was also amused by the introduction of the term statcore by Dibyo.
I had a lot of fun running this contest, and really felt like there was this excitement revolving around data. That makes me happy. I hope that now, even though there’s no prize up for grabs, that all of you will continue to leave comments and add to the conversation. Interacting with all of you is one of my favorite parts about FlowingData.
Also, thanks a lot to Andrew, Kaiser, and Tony for helping me promote the contest.
More Contests Ahead
On that note, seeing how this contest was so successful, you should look forward to more contests ahead. I’m thinking end of April. Maybe Tufte’s second book? Or maybe a movie. I don’t know, what do you guys think should be the prize for the next FlowingData contest?
Thanks again, everyone. Here’s to the start of a good week.
P.S. Don’t forget to tell your friends! We’re still working towards 5,000.