Page 234 - Information and American Democracy Technology in the Evolution of Political Power

P1: IBE/IRP/IQR/IRR
12:12
0 521 80067 6
CY101-Bimber
August 13, 2002
CY101-05
Information Technology and Political Engagement
47 percent of those with access to only 56 percent between late 1996 and
early 2001.

INFORMATION TECHNOLOGY AND
POLITICAL ENGAGEMENT
The diffusion of Internet technology during the fourth information rev-
olution has reached a point where more people have access to the Internet
than vote in presidential elections, and where roughly half of these use
thetechnologyinpoliticalways.Accordingtothepsychologicalapproach
to explaining how people absorb political information, the wide acces-
sibility of this new technology should not be altering overall levels of
engagement. Data on the factors that influence attention to political in-
formation across different media begin to build the empirical case for
this broad hypothesis. To examine media choice, I drew on data from the
2000 NES. To begin, I examined the correspondence between use of the
Internet for political information and attention to the campaigns on tele-
vision and in newspapers. The theory predicts that overall associations
between different media should be high. In both cases, this expectation
is supported by the data. The more people read about the campaigns in
newspapers or watch shows on television, the more likely they are also
to use the Internet for political information. There is scant evidence for
a substitution effect of new media for old in which respondents use the
Internet exclusively. Instead, Internet use is additive, and in a general way,
the media-rich do seem to get richer.
To see what common underlying factors might influence people’s use
of the Internet, television, and newspapers for political information, I
built a statistical model using logistic regression predicting people’s at-
tention to the campaigns that year. The model includes independent
variables for education, age, sex, race, level of trust in the news media,
leveloftrustinotherpeople,beliefinthehelpfulnessofothers,andoverall
interest in the campaigns. I ran this model using dichotomous dependent
variables for whether respondents obtained campaign information from
the Internet, whether they watched at least one television show about
the campaign, and whether they paid at least a little attention to the
campaigns in newspapers. 44 I hypothesized that, in general, education,

44
Unfortunately, the NES measures use of the Internet for campaign information only
as a dichotomous variable, so a good deal of useful information about frequency or
intensity is not available.

217