October 2007
Sun Mon Tue Wed Thu Fri Sat
  1 2 3 4 5 6
7 8 9 10 11 12 13
14 15 16 17 18 19 20
21 22 23 24 25 26 27
28 29 30 31      

Authors' Committee

Chair:

Matt Blackwell (Gov)

Members:

Martin Andersen (HealthPol)
Kevin Bartz (Stats)
Deirdre Bloome (Social Policy)
John Graves (HealthPol)
Rich Nielsen (Gov)
Maya Sen (Gov)
Gary King (Gov)

Weekly Research Workshop Sponsors

Alberto Abadie, Lee Fleming, Adam Glynn, Guido Imbens, Gary King, Arthur Spirling, Jamie Robins, Don Rubin, Chris Winship

Weekly Workshop Schedule

Recent Comments

Recent Entries

Categories

Blogroll

SMR Blog
Brad DeLong
Cognitive Daily
Complexity & Social Networks
Developing Intelligence
EconLog
The Education Wonks
Empirical Legal Studies
Free Exchange
Freakonomics
Health Care Economist
Junk Charts
Language Log
Law & Econ Prof Blog
Machine Learning (Theory)
Marginal Revolution
Mixing Memory
Mystery Pollster
New Economist
Political Arithmetik
Political Science Methods
Pure Pedantry
Science & Law Blog
Simon Jackman
Social Science++
Statistical modeling, causal inference, and social science

Archives

Notification

Powered by
Movable Type 4.24-en


« October 22, 2007 | Main | October 26, 2007 »

25 October 2007

Visualizing UK Politicians

Since I saw Fernanda Viegas and Martin Wattenberg's presentation on Many Eyes a few weeks ago in our Applied Stats workshop, I've been itching to use their visualization tools on some of my own data. Tonight I made a treemap of the dataset of UK politicians that Jens Hainmueller and I have been developing. (The data consists of over 6000 candidates who ran for the House of Commons between 1950 and 1970.) I set up the visualization such that each box in the treemap is sized to indicate the number of campaigns for each combination of party and occupation (eg Conservative barristers) and the color reflects "proportion attending Oxbridge." But you can play with it via the menus at the bottom of the visualization and cut the data the way you want: you can make the size reflect "proportion female" and the color reflect "proportion elected," and you can make it show party by occupation instead of occupation by party. I've embedded the visualization here; you can make comments tied to a particular view of the data on the many-eyes site. I'm eager to hear reactions, whether on the visualization or on the brand-new data.

Posted by Andy Eggers at 1:50 AM