Media Effects Research Lab - Research Archive
Engagement with news content in online social networks
Student Researcher(s)
Anne Oeldorf-Hirsch (Ph.D Candidate);
Faculty Supervisor
INTRODUCTION
In the last decade, the newspaper industry has a seen a major decline in sales due to a number of factors, but undoubtedly in large part due to the. However, despite the increase in the number of news sources, as many as 34% of individuals younger than 25 say that they get no news on a typical day, up from 25% just ten years before. Fortunately, the Internet provides new forms of interaction that may prove essential for engaging users in the news. For instance, Facebook’s status update feature has become the key ingredient of the site’s news feed and now allows users to quickly share external content in addition to posting text updates. This access to a variety of external content may well make Facebook a primary information source for its users.
This paper explores the potential benefits of both sharing news content and seeing shared news content on a social networking site, in terms of engagement in that news content. First, sharing news content on Facebook may have important empowering effects for those who post links to external content. Further, those who learn of news stories through their friends may also become more involved in the story. Discussing this content could further engage them, and seeing public discussion by friends could also have such an effect.
RESEARCH QUESTION(S) AND HYPOTHESES:
H1: Sharing a news story on Facebook through status updates will lead Messengers to feel greater involvementin the content of the news story compared to only reading the news story.
H2a: Receiving comments on shared content will lead Messengers to feel a greater sense of influence than not receiving comments on shared content, by way of feeling a sense of community.
H2b: Receiving recommendations (in the form of “likes”) on shared content will lead Messengers to feel a greater sense of influence than not receiving recommendations on shared content, by way of feeling a sense of agency.
H3a: Sharing a news story on Facebook with an opinion will lead to greater involvement in the content of the news story compared to posting the news story without an opinion.
H3b: Sharing a news story on Facebook with a question will lead to greater involvement in the content of the news story compared to posting the news story without a question.
H3c: Sharing a news story on Facebook with a comment or question will lead Messengers to feel a greater sense of influence than sharing the news story without a comment or question, and this will be mediated by a sense of agency.
H4a: Tagging friends in the post when sharing a news story on Facebook will lead Messengers to feel greater involvement in the content of the news story compared to posting the news story without tagging friends.
H4b: Tagging friends in the post when sharing a news story on Facebook will lead Messengers to feel a greater sense of influence than not tagging friends, by way of feeling a sense of community.
H5a: Commenting on a news story shared on Facebook by a Messenger will lead the Discussant to feel more informed about the news content, compared to only reading the news story.
H5b: Commenting on a news story shared on Facebook by a Messenger will lead the Discussant to feel more involved in the news content, compared to only reading the news story.
H6a: Viewing others’ comments along with the posted news story will lead the Receiver to feel more informed about the news content, compared to only reading the story.
H6b: Viewing others’ comments along with the posted news story will lead the Receiver to feel more involved in the news content, compared to only reading the story.
H6c: Viewing others’ comments along with the posted news story will make the Receiver more likely to comment on the story, compared to only reading the story.
H6d: Viewing others’ comments along with the posted news story will make the Receiver more likely to repost the story, compared to only reading the story.
H7a: Receivers will like a news story posted by Messengers on Facebook more than those reading the story on a news site.
H7b: Receivers will find a news story posted by Messengers on Facebook to be more representative than those reading the story on a news site.
H7c: Receivers will find a news story posted by Messengers on Facebook to be of higher quality than those reading the story on a news site.
H7d: Receivers will perceive news stories posted by Messengers on Facebook as less credible than those reading the story on a news site.
H8a: The higher the number of recommendations posted in response to a news story link shared by a Messenger, the greater the likelihood that Receivers will follow the link and read more of the news story.
H8b: The higher the number of recommendations posted in response to a news story link shared by a Messenger, the greater the Receiver’s involvement in the story.
RQ1: What are the effects of various levels of broadcasting (direct message, wall post, and news feed) on the predicted outcomes of sharing news stories on Facebook?
METHOD
An online experiment was conducted using a 3 (Broadcast Level: News Feed vs. Wall Post vs. Direct Message) x 3 (Elaboration: Opinion vs. Question vs. No Comment) x 2 (Involving Friends: Tag vs. No Tag) between-subjects experimental design. The conditions are not fully crossed because tagging is not available in all conditions, leaving 13 “Messenger” conditions. In addition, some participants were randomly assigned to one of three “Receiver” conditions: Discussant, Recommender, or Reader. The remaining participants were randomly assigned to a control condition, for a total of 17 study conditions.
The sample contained 333 individuals, ranging in age from 18 to 63 years (M = 27.34, SD = 9.72), with 67% being female. Seventy-nine percent of the sample was White, 11% was Asian, 5% Black, 2% Hispanic, .6% Native American, and 1% identified as “other,” mainly South Asian. Approximately two-thirds of the participants were students (64% full-time, 4% part-time) and the non-students were employed mostly in a university setting with occupations ranging from professors to administrative staff.
Individuals were randomly assigned to conditions using a script online that randomly selects one of the 17 questionnaire URLs to load. All study instructions and questionnaire items were hosted on SurveyMonkey. Messengers were given detailed instructions on how to select and post a news story, add an opinion or question, and tag friends, depending on condition. Receivers were given detailed instruction on how to find a friend’s posted story and read it, comment on it, or “like” it. Participants completed a questionnaire immediately following the task, and again one week later, about the news story, their posting activity, dependent variables of if interest, and their news reading and Facebook use habits.
RESULTS
Involvement in the news story was not significantly higher for Messengers (vs. Control): F(1, 171) = 1.08, p = .30. However, testing the effect on Involvement at Time 2 produced a significant result, F(1, 130) = 3.96, p < .05, ω2 = .02, with those in the Messenger condition feeling significantly more involved in the story one week later (M = 3.46) than those in the control group (M = 2.92).For Messengers, Elaboration had a significant effect on Involvement, F(2, 142) = 4.50, p < .05, η2 = .02, such that those who asked a question about the news story in their post felt significantly greater involvement (M = 4.33) than those who posted their opinion about the story in their post (M = 3.55). Involving Friends did not have a significant impact on Involvement in the news story: F(1, 74) = .01, p = .91. However, Involving Friends was tested its effects on Community, and the results show a trend toward significance: F(1, 74) = 2.74, p = .10, with those who tagged their friends feeling a higher sense of community (M = 7.00) than those who did not tag their friends (M = 6.87).
For Receivers, the effect of Role on Liking was significant, F(1, 46) = 7.93, p < .05, η2 = .09, with those in the Receiver condition liking the story significantly more (M = 4.11) than those in the control condition (M = 3.53). The effect of Role on Representativeness also approached significance: F(1, 46) = 3.91, p = .05, but with those in the control condition finding the news story more Representative (M = 5.19) than those in the Receiver condition (M = 4.55).
According to the Messengers’ ratings of the comments received, they were generally favorable, though somewhat superficial. To test for mediating effects of Comment Favorability between the predictors and outcome variables, a bootstrapping method for testing indirect effects of mediators developed by Preacher and Hayes (2008) was run with 5000 bootstrap samples. As shown in Figure 1, the results indicate that Posting on a Facebook wall led indirectly, through favorable comments, to feeling more informed (point estimate of -.4422, and 95% BCa CI of -.9716, -.1635), wanting to know more (point estimate of -.3744, and 95% BCa CI of -.9233, -.0372), greater interest in the topic (point estimate of -.4001, and 95% BCa CI of -.9164, -.0906), greater involvement (point estimate of -.3512, and 95% BCa CI of -.8903, -.0145), and a greater sense of influence (Empowerment; point estimate of -.1619, and 95% BCa CI of -.4131, -.0056).
Figure 1. Indirect effects of where a news story is posted on outcomes through comment favorability as a mediator.
CONCLUSIONS
The promise of social networking sites to engage our youth in news and public affairs appears to be just that—a promise. Unlike blogs, these sites do not seem to carry the cachet of “publishing.” Study participants did not feel agentic when posting or forwarding a news story to their networks. However, their engagement in news appeared to be stimulated by support from their respective networks. It is clear from our findings that the psychological appeal of Facebook is in the social features that it offers (and continues to invent) rather than in its personal affordances. Four key elements of Facebook seem to be critical for breeding user engagement with news content—(1) Wall; (2) Likes; (3) Tags; and (4) Comments. The Wall appears to be the new “public sphere” that is quite crucial for shaping discourse about news and public affairs. The publicness of the Wall is especially critical; users do not feel the same sense of empowerment when they simply post or message others via private means. In addition to its publicness, the interface features that enable others to respond to one’s posts are particularly important. This is clearly illustrated by our mediation findings showing that others’ comments on one’s post, especially their perceived favorability and the number of Likes, have a significant impact on one’s interest and involvement in the news topic, how informed one feels, and the desire to find out more information on the news topic. Tagging friends seems to stimulate their endorsement of one’s posts, thereby creating a heightened sense of community.
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