From the monthly archives:

November 2010

Holographic GapMinding

November 30, 2010

Hans Rosling -- whose lectures are always fascinating -- is hosting a new documentary for the BBC called "The Joy of Stats." A 5 minute clip has been released on YouTube showing a faux-holographic version of Hans' GapMinder visualization package. The graphic overlay is very well done and lets Hans describe the data in an […]

0 comments Read the whole post →

Visualizing politics through time

November 22, 2010

We love choropleths here at TGR, and here's a really great set -- David Sparks has mapped US presidential voting patterns through time to create an excellent visualization of ebbing (and sometimes volatile) political attitudes: Best of all, he did it with R. Please see David's website for more details. Some of his other projects […]

0 comments Read the whole post →

UCF cheating scandal

November 18, 2010

A major cheating scandal at UCF was discovered - and resolved - through a relatively simple statistical analysis of midterm results. The team was able to identify students who cheated on their midterm exam with high confidence. Professor Richard Quinn's announcement of those findings was captured in this video of his lecture: (Via The Daily […]

0 comments Read the whole post →

Tiny, Large, Very, Nice, Dumbest.

November 12, 2010

Here's a great analysis from Ben Blatt of the Harvard Sports Analysis Collective. He looked at three well-known sports writers -- Bill Simmons, Rick Reilly and Jason Whitlock -- and performed a lexical analysis to create a statistical representation of their writing styles. What can you do with that analysis? Well, you can see what […]

0 comments Read the whole post →

Modeling how cats drink

November 11, 2010

I thought this was fascinating -- scientists have modeled how cats drink. Naturally, once you have a model, you want to see how well if fits the data. For example, is there an optimal lapping speed? After calculation of things like the Froude number and the aspect ratio, they were able to figure out how […]

1 comment Read the whole post →

Google Refine

November 11, 2010

Google has launched a new open-source project called Refine (formerly Metaweb's Freebase Gridworks) which allows users to easily clean up and transform large datasets. There is nothing more painful than cleaning data at the command line - I'd even go so far as to say it's impossible to do a good job. Sorry, R. Excel […]

0 comments Read the whole post →

An apology

November 11, 2010

I wrote a post yesterday referencing Forbes' coverage of R, which I was psyched about. But as I was reading, one quote in the post stuck in my head. I couldn't get past how bad I thought it made statisticians look, and ultimately that became the focus of my post. I was pretty off base […]

1 comment Read the whole post →

The turkey statistician

November 10, 2010

C links to a Thanksgiving-appropriate essay by Nassim Taleb in which he presents a a story about a turkey statistician. For 100 days, the turkey is fed and cared for by humans. He arrives at the statistically-significant conclusion that humans genuinely care about his well-being. On the 101st day, the turkey is slaughtered. Most interesting […]

0 comments Read the whole post →

Chicken soup for the global economy

November 8, 2010

Just replace "technology" with "stress": (via Dilbert)

1 comment Read the whole post →

Post-election humor

November 4, 2010

Made me chuckle: (via SMBC)

0 comments Read the whole post →

Misreading misleading charts: rally edition

November 4, 2010

From the "question everything" file, an excellent example of why you should never trust anything without verifying it yourself: a number of conservative mailing lists are forwarding the following image comparing the size of Glen Beck's rally to Jon Stewart's: Note the circles at the bottom, which purport to show the areas involved. The first […]

0 comments Read the whole post →

Breaking up is hard to do (especially on Christmas)

November 2, 2010

David McCandless's TED talk on data visualization is excellent -- you can catch it here -- and Mathias Mikkelsen has highlighted a single analysis that investigates when people are most likely to break up (according to Facebook) (Update: original here): What makes the chart so appealing is how easy it is to understand, despite the […]

0 comments Read the whole post →