Comedy as a Gateway
Page 9
Hypothesis 2: Over time, the rate of increase in attention to the campaign in television
news will be greater for viewers of late-night comedy than those who do not watch late-
night comedy.
Previous research (Young & Tisinger, 2006) has identified viewers of The Daily Show to be
more interested in and knowledgeable about politics than other late-night audiences. If this is the
case, the news attention of Daily Show viewers may be relatively less responsive to late-night
coverage of the primaries and instead be maintained at high levels throughout the campaign. To
help better understand the differences between various late-night programs in their effects on
audiences, we therefore propose the following research question:
Research Question 1: Will the trend in hard news attention over the course of the
campaign differ between audiences of The Daily Show and audiences of other late-night
programs like the Tonight Show with Jay Leno or the Late Show with David Letterman?
Data and Methods
Data used in this study come from the 2004 National Annenberg Election Survey
(NAES), conducted by the Annenberg Public Policy Center at the University of Pennsylvania.
The 2004 NAES employed a rolling cross-sectional design, which involves sampling a series of
random cross-sections of the target population over a period of time (Johnston & Brady, 2002).
With the NAES design, cross-sections were taken on a daily basis. A random sample of
telephone numbers was first selected using random digit dialing (RDD). These numbers were
then divided into random subsamples, or replicates, which were released for telephone
interviewing each night. Every evening, interviews were not only attempted with the numbers
released on that day but also with all other numbers that had not been contacted from previous
days. In this way, each night’s sample includes individuals who are easy to reach (i.e., those
Convention