January 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)
Andy Eggers (Gov)
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

Recent Comments

Recent Entries

Categories

Blogroll

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


« January 26, 2007 | Main | January 30, 2007 »

29 January 2007

Applied Statistics - Holger Lutz Kern

The Applied Statistics Workshop resumes this week with a talk by Holger Lutz Kern, a Ph.D. candidate at Cornell University currently visiting at Harvard. His research focuses on comparative political economy and behavior with a focus on causal inference from observational data. His work has appeared in the Journal of Legislative Studies. He will present a talk entitled "The Effect of Free Foreign Media on the Stability of Authoritarian Regimes: Evidence from a Natural Experiment in Communist East Germany." The talk is based on joint work with Jens Hainmueller. The presentation will be at noon on Wednesday, January 31 in Room N354, CGIS North, 1737 Cambridge St. As always, lunch will be provided. An abstract of the talk follows on the jump:

A common claim in the democratization literature is that free foreign media undermine authoritarian rule. In this case study, we exploit a natural experiment to estimate the causal effect of exposure to West German television on East Germans' political attitudes. While one could receive West German television in most parts of East Germany, residents of Dresden were cut off from West German television due to Dresden's location in the Elbe valley. Using an ordered probit LARF instrumental variable estimator and survey data collected in 1988/9, we find that East Germans who watched West German television were *more* satisfied with life in East Germany and the communist regime. To explain this surprising finding, we demonstrate that East Germans consumed West German television primarily for its entertainment value and not because of its unbiased news reporting. Archival data on the internal debates of the East German regime corroborate our argument.

Posted by Mike Kellermann at 10:11 AM