Page 277 - Comparing Political Communication Theories, Cases, and Challenge

P1: JZL/KDD/KAB P2: JZZ
0521828317c11.xml CY425/Esser 0521828317 June 8, 2004 22:9

Political News Journalists

Table 11.1 The Correlation (Pearson’s r) Between Journalists’ Partisan
Beliefs and Their Perception of the Partisan Editorial Position of the
News Organization in Which They Work

United United
States Kingdom Germany Italy Sweden
All Respondents .03 .03 .13 ∗ .20 ∗∗ .10
National Newspaper Journalists .03 .24 ∗ .54 ∗∗ .47 ∗∗ .23 ∗
National Broadcast Journalists .03 −.31 ∗ −.03 .23 ∗ −.15
Local Newspaper Journalists .09 .03 .16 .11 −.01
Local Broadcast Journalists −.11 −.28 ∗ −.03 −.17 .03

p < .05; <.001
∗ ∗∗

German (.54) journalists who worked for the leading national papers
but was also relatively high among their British (.24) and Swedish (.23)
counterparts. When local newspapers in Europe were considered, the
correlations were positive in direction (except for Sweden) but much
weaker than at the national level. In the case of European broadcast-
ing, on the other hand, there was a significant positive correlation only
among Italian national broadcasters. It would appear that, unless jour-
nalists work in an arena where news organizations are overtly partisan,
their partisanship is a small factor in determining the job they hold.
Partisanship in the media, however, is not merely a question of the
news organizations within which journalists work. In the final analysis,
the issue of journalists’ partisanship is a question of whether it affects
their news decisions. If, as news professionals, they make their choices
almost entirely in the context of prescribed journalistic norms and prac-
tices, their partisan beliefs are largely immaterial. Content analysis has
been the primary method for examining this issue. This method, how-
ever, is limited in its ability to isolate and identify bias. It is exceedingly
difficulttodetermine,forexample,whethernegativeorpositivecoverage
of a politician or issue is a result of partisan bias, the nature of events, or
other factors. The problem of inference is magnified when country-to-
country differences are at issue because the watchdog role is emphasized
more heavily in some news systems than it is in others.
Accordingly, we developed a quasi-experimental survey method for
measuring bias that exploits the fact that journalists are accustomed to
making news decisions on the basis of event descriptions. Respondents

257