Policy or Politics?
1
Introduction
In October, 2003, the Korean President who had been elected just 8 months before
proposed an unprecedented National referendum the result of which might lead him to step down.
His shocking announcement, although the referendum was not called due to national opposition,
followed the nation’s rapidly dropping approval rating of his performance. In a poll conducted
by a newspaper company, for example, only 16.5% of the Koreans supported him.
What can explain such a poor presidential approval rating in his early incumbency?
Considering that social judgment, including evaluation of a president, involves selective
activation of prior knowledge (Wyer & Srull, 1989), and that political information is almost
exclusively provided by mass media, it is necessary to investigate how mass media have
portrayed the president.
Regarding media effect on public opinion of presidential performance, the study of
priming has combined quantitative analysis of media texts and poll data (Pan & Kosicki, 1997;
Krosnick & Kinder, 1990; Iyengar & Simon, 1993). One may note that the idea of priming is
based on accessibility of knowledge, a psychological concept, determined by frequency or
recency of the previous activation of the knowledge. Since the concept of frequency and recency
is quantitative, media analysis in previous priming studies has mainly delved into quantifying the
news coverage of specific issues (e.g. economic recession, Gulf War, and Iran-Contra disclosure)
rather than going through the content of the news coverage of the issues. (Pan & Kosicki, 1997;
Krosnick & Kinder, 1990; Iyengar & Simon, 1993).
However, frequency or recency of positive/negative issues covered by news media is just
one way that media influence public opinion. Framing research, which initially had focused on
media content assuming media-produced frames will influence pubic opinion, also started
Page 1 of 27 
Convention