The effects of three types of newspaper inaccuracies on reader perceptions of reporter and editor bias

The purpose of this study was to determine the effect of news story inaccuracies on readers' perceptions of reporter and editor bias. Three types of inaccuracies were stipulated for research purposes: belief-discrepant, belief-supportive, and belief-unrelated. It was hypothesized that belief-discrepant inaccuracies detected in news stories, unlike belief-supportive or belief-unrelated inaccuracies, would cause such stories to be judged intentionally biased. It was also hypothesized that there would be no difference in perceived reporter-editor bias between a news story with a belief-unrelated inaccuracy and an accurate version of the same story.

An after-only laboratory experiment was conducted with 93 women elementary, secondary, and retired teachers. Equal numbers of subjects were randomly assigned to four groups. One group read a news story with a belief-discrepant inaccuracy; a second group, a belief-supportive inaccuracy; and a third group, a belief-unrelated inaccuracy. The fourth group read an accurate version of the story. The three types of inaccuracies affected the same sentence in the story. Otherwise, stories for all groups were identical.

Prior to reading the stories, subjects were given a brier transcript and told that it represented an "official record" of a Congressional subcommittee hearing. In that record, an unidentified educator expressed the opinion that teachers were not responsible for declining SAT scores. The testimony was taken from subjects when they had completed reading it. They were then given the news story, and told that it represented an actual newspaper account of the educator's testimony. Subjects were also told that the reporter of the fictitious story was present at the subcommittee hearing and heard the testimony which they had already read in the "record." The story with the belief-discrepant inaccuracy had the educator say that teachers were responsible for declining SAT scores. . In the story with the belief-supportive inaccuracy, the educator said teachers "cannot be blamed" for the scores. The belief-unrelated inaccuracy consisted of innocuous synonyms for two of the educator's words.

A one-way analysis of variance coupled with Scheffé's multiple comparison test showed that in accordance with the hypothesis significantly more subjects in the belief-discrepant group perceived reporter bias in the news story than did those in the accurate group. There was also no perceived difference in reporter bias between those in the belief-unrelated and accurate groups, a finding the hypothesis also predicted. No significant difference in reporter bias was perceived by the belief-supportive group compared to the belief-discrepant group; but this was not predicted.

For perceived editor bias, only the predicted lack of significant difference between the belief-unrelated and accurate groups was supported by the statistical tests. The other aspects of the hypothesis were not supported. Explanations were offered for the findings, as well as suggesticns for future research.

The results were offered as new evidence for mass media power, in contrast to the once prevalent view of limited media effects. . It

vi was contended that editors and publishers should find the research of interest as a means of identifying, and correcting for, a heretofore unsuspected cause of reader bias perceptions.