Sensationalism, narrativity and objectivity---modeling ongoing news story practice

Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)


Mass Communications


George A. Comstock

Second Advisor

George A. Comstock


Narrative, Demonization, Valorization, Waco, Ruby Ridge, FBI

Subject Categories

Communication | Journalism Studies | Mass Communication | Social and Behavioral Sciences


The opposing forces of sensationalism and objectivity in the news have not been unambiguously defined. Their concepts are thus in need of explication. By the 1990s, highly competitive market conditions, amid a plethora of channels through which consumers might obtain their news, had caused print journalists to adopt a richer narrative style. Whether such a narrative style veered into sensationalism, and in doing so, abandoned the tenets of objectivity, is the subject of this dissertation.

This dissertation assumes that key dimensions of sensationalism, such as demonizing and valorizing characterization, and its opposite, objective characterization, may be validly and reliably identified, and are discernable statistically via quantitative content analysis.

Four ongoing news events from the 1990s - Ruby Ridge, the Waco incident, the Richard Jewell affair and the Wen Ho Lee matter - were chosen because each generated thousands of newspaper stories, both at the local and national level, and for considerable periods of time. These stories were examined for crests and troughs of narrative characterization, hypothesized to occur via a concept the researcher uses called valence oscillation.

Valence oscillation is conceptualized to occur when news writers radically alter the tenor of their depiction of contending news actors, over the course of an ongoing story, in order to maintain salience, and thus readership. When valence oscillation shows strong excursion from mean characterization, sensationalism is hypothesized to occur.

Hypotheses were derived from this conceptualization to determine if valence oscillation occurred in the news characterization of the main contending actors, and if so, when, and under what conditions. The tested news conditions were: characterizations in local versus the national press, and characterizations in hard news, the op/ed page or the entertainment pages. The general linear model was used to derive t-tests, correlations and analyses of variance to test the presence or absence of valence oscillation.

Results revealed some commonality of characterization among the stories in characterization, particularly in a general tendency of news writers to alter depiction over time to increase salience. Therefore, valence oscillation most probably occurred in the tested samples. However, this did not occur with any greater regularity in the national press versus the local press, nor was it found to be more frequent in the opinion pages, when compared with hard news reportage. An overall negativity of characterization across nearly all news stories and variables was also observed.


Surface provides description only. Full text is available to ProQuest subscribers. Ask your Librarian for assistance.