Publication Type: Conference Paper/Unpublished Manuscript
Abstract: Past research suggests that perceptions of media bias among the public have increased largely because of increased attacks by political elites. Although past research provides strong evidence that charges of bias influence public opinion, it is unclear what these charges entail. My research investigates why political elites accuse the media of being biased, or at least, what evidence is provided to support their accusations. To answer this, I perform a content analysis of newspaper and television coverage of accusations of media bias. For the analysis, I code articles and news transcripts from 2005-2006 for the direction of bias charged against the news media and the type of evidence given (if any) to support these charges. In the end, my research adds insight into why many political elites think the news media is biased.

Publication Type: Conference Paper/Unpublished Manuscript
Review Method: Peer Reviewed
Abstract: Brians and Wattenberg (1996) examined
the ability of the electorate to gain and use information during a
presidential campaign. Brians and Wattenberg studied the 1992 election
using the American National Election Survey. Using two regression
equations predicting either the respondent's evaluation of the
candidate or respondent's knowledge of candidates' issue positions,
Brians and Wattenberg found that early in the campaign people relied
more on candidate evaluations and, later in the campaign, they were
more likely to know something about the candidates' positions on the
issues.
While able to replicate the Brians and Wattenberg study for 1996 and
2000 (in addition to 1992), Benoit, Hansen, and Bough (ongoing) have
moved beyond the arbitrary date delineating early campaign from late
campaign in the Brians and Wattenberg article. Specifically, Benoit,
Hansen, and Bough use the dates of presidential debates to divide the
campaigns into either four periods (for campaigns involving three
debates) or three periods (for the campaign with only two debates),
which yielded significant shifts in the information levels of the
electorate.
This iteration (Bough, Benoit, and Hansen, ongoing) focuses primarily
on information levels in partisans and non-partisans. Specifically, we
divide up the samples by partisanship (and by party) to see whether
partisans are more or less influenced by debate viewing.
**I do not mind presenting this as a posterboard session. In fact, that
may be easier for me to do since this is variations on a
replication.

Publication Type: Conference Paper/Unpublished Manuscript

Publication Type: Paper (prepared oral presentation)
Review Method: Peer Reviewed
Abstract: A replication and extension of Cohen (2003) confirms that for some citizens, partisan cues can overwhelm policy details in decision making and lead to support for ideologically contrary welfare policies. The current work demonstrates conditions under which party cues fail to dominate policy information: partisan ambivalence and ideological salience. An experimental design manipulated policy descriptions across three levels: policy information only, policy and party identification, and policy, party and ideological information. Individuals with ambivalent attitudes toward the parties were not swayed by partisan cues to support ideologically contrary policies compared to univalent individuals. Additionally the inclusion of ideologically salient information removed the party cue effect for univalent individuals. The results coincide with previous work showing ambivalence decreases the impact of party endorsement on decision making. Additionally we have shown when univalent individuals, those most likely to rely on party cues, will disregard them in favor of more relevant ideological information.

Publication Type: Conference Paper/Unpublished Manuscript
Review Method: Peer Reviewed
Abstract: Partisan voters who change their vote choices on the Election Day account for a small portion of the electorate but play a decisive role in a close election. This study examines Paul A. Beck’s (2002) social support theory of partisan defection—a voter who perceives support for a candidate of the opposite party from his/her communication network is likely to defect—with data collected in Taiwan after the 2004 presidential election, the election marked by its two-camp competition, the first time in Taiwan's s history that campaign went like a two-party system. Besides empirical test of the theory, this paper presents that pan-green supporters are more volatile than pan-blue camp supporters, as well as the major factors associated with this pattern.