On Thu, 4 May 2006, jkatkin(a)fas.harvard.edu wrote:
Hi everyone,
Below is a revised version of our abstract. Thanks in advance for any comments!
Jen and Jacqueline
The Media Followed the Parties, but Not Equally on All Issues: A Closer Look
at Agenda Formation in the 1997 British General Election
the main part of the title is not a clear message; nothing you can do
about that if those are your results of course.
A small body of literature has attempted to determine the flow of agenda-setting
during political campaigns: do the media lead the parties or vice versa? Using
time-series cross-sectional data from the 1997 British general election,
Brandenburg (2002) employs ordinary least squares regression to conclude that
agenda formation was unidirectional, from parties to media. We replicate these
results but find evidence in first differences that parties conceivably
i'd drop 'in first differences' to keep this focused on the substance.
followed the media as well. We therefore re-estimate
Brandenburg's models using
negative binomial regression, given the over-dispersed count-data. Changes in
media coverage that chronologically follow changes in party coverage *are*
greater than party changes that follow media changes. Additionally, because the
negative binomial model permits variation in party influence across policy
dimensions, we demonstrate that the parties particularly affected the media's
tendency to cover the economy, welfare, education and foreign policy. OLS
treats jumps between policy dimensions identically and thus improperly suggests
a uniform impact across all issues.
doesn't the negbin model also assume fixed jumps, even if different sized?
Gary
(154 words)
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l