WHICH SOURCE MATTERS FOR NEWS ON THE WEB? AN INVESTIGATION OF THE RELATIONSHIP BETWEEN SOURCE ATTRIBUTION AND THE PERCEPTION OF NEWS CREDIBILITY.
 
Student researchers:

Keunmin Bae, Hyunjin Kang, Bo-Youn Lee, Shaoke Zhang (PhD Students)
This paper is based on a project from the "Psychological Aspects of Communication Technology" graduate course.

Faculty Supervisor:

Dr. S. Shyam Sundar

Introduction

A recent report by the Pew Research Center (2006) illustrated the importance of the Internet as a news source. This report said that nearly one in three Americans regularly get news from the Internet, a huge leap from a decade ago when only one in fifty Americans acquired news online. Unlike traditional or offline media, there are often multiple layers of sources in the Internet, including the name of the portal site and that of the news agency for a single online news story, or multiple news cues on a single page, such as posted times and the number of viewers. Despite the rising research interest in the Internet as news media, however, it has not been thoroughly explored how multiple source layers such as names of news agencies and portal sites appearing on the interface influence readers’ perception of news. This study attempts to answer which source layer people attribute news to in the environment of multiple source layers and how the credibility of each source cue influences the credibility of news story.

Research Questions

RQ: What is the relationship between the level of involvement, the credibility of multi-layered source cues, and source attribution, AND news credibility?
Credibility is perceived as an attribute of the source, and readers’ response to the message is considered to be under influence of this attribution.
H1: The credibility of a news agency will be positively related to the credibility of the news story.
H2: The credibility of a portal site will be positively related to the credibility of the news story.

According to dual-process models, source credibility is considered heuristic cues and can influence attitudes toward arguments by biasing thoughts, even when people processed the message systematically. However, the models also posit that central route vs. peripheral route (or systematic processing vs. heuristic processing) activated in accordance with involvement.
H3: There will be an interaction between the level of involvement and the level of news agency credibility for the credibility of news.
H4: There will be an interaction between the level of involvement and the level of portal credibility for the credibility of news.

Previous literature found evidence for a cue-cumulation effect in the Internet environment when source credibility is low.
H5: There will be an interaction between the level of news agency credibility and the level of portal site credibility for the credibility of news.

Based on previous literature of credibility and dual-process models, two hypotheses were formulated.
H6a: People with low involvement in the news will be more sensitive to the credibility of a portal site when evaluating news credibility
H6b: People with high involvement in the news will be more sensitive to the credibility of a news agency when evaluating news credibility.

Method

The hypotheses were explored via an online between-subjects experiment that manipulated two levels of involvement in the news, two levels of portal site cue credibility, and two levels of news agency cue credibility for a 2 x 2 x 2 full-factorial design. All study participants (N = 218) were randomly assigned to one of the eight stimuli conditions. Participants were asked to read a piece of news on a simulated online news page and then to answer both close-ended and open-ended questions about the perceived credibility of the news and of its sources and of their involvement level.

Results

There was no significant main effect of the manipulated IVs on perceived news credibility. However there were indirect effects of the manipulated portal and news agency credibility on news credibility. Therefore, the indirect effect of manipulated source credibility on news credibility (i.e., manipulated source credibility ? perceived source credibility ? perceived news credibility) was explored to test the hypotheses.

H1: Not supported. There was no significant main effect of perceived source credibility on news credibility.

H2: Supported. The perceived portal credibility had main effects on the perceived news credibility, F (1, 199) = 33.77, p < .001.

H3: Not supported. There was no interaction between the level of involvement and the level of news agency credibility for the credibility of news.

H4: Not supported. There was no interaction between the level of involvement and the level of portal credibility for the credibility of news.

H5: Supported. The interaction between portal credibility and news agency credibility on news credibility was significant. F (1, 199) = 6.08, p < .05; H5 was supported. In the low involvement condition, the perceived portal credibility had main effects on the perceived news credibility F (1, 95) = 18.78, p < .0001.

H6a: Partially supported. In the low involvement condition, the perceived portal credibility had main effects on the perceived news credibility F (1, 95) = 18.78, p < .0001. There were no main effects of perceived news agency credibility and no interaction between perceived portal credibility and news agency credibility.

H6b: Supported. In the high involvement condition, there was a significant interaction between perceived portal credibility and perceived news agency credibility on news credibility F (1, 104) = 15.28, p < .001.

Conclusions

The result indicated that when people are highly involved in a news story, they care about the credibility of both proximal and distal sources in evaluating news credibility. Therefore, for highly-involved people, it was suggested that a source cue with low perceived credibility in multiple layers of sources cues would negatively influence the perceived news credibility, whereas low-involved people only care about the perceived credibility of the most proximate source cue.
This study found evidence that the relative heuristic values of source cues varied according to perceived psychological distances; if a source cue of a news story appears psychologically proximate to readers, readers tend to attribute the story to the source.
In addition, the current study supported HSM rather than ELM. The results suggested that even people who have high involvement in the news were influenced by source credibility which is commonly considered as heuristic cue, when they evaluate a news story’s credibility.
The results suggested that portals of low or unknown credibility can gain in credibility by publishing feeds from credible news agencies.

For more details regarding the study contact

Dr. S. Shyam Sundar by e-mail at sss12@psu.edu or by telephone at (814) 865-2173

>> Return to abstract listings page
>> Return to the main research page

Media Effects Research Lab at College of Communications, Penn State University