![Sample our Social Sciences journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/110822/TOC-Subject-Sample-2021-Social-Sciences.jpg"})
ABSTRACT
Today, social network sites (SNS) are a major platform for news consumption. However, their interface differs dramatically from those of news sites. On SNS, news posts lack the traditional news cues that guide audience selection of stories. Instead, they are accompanied by social cues, traces of engagement by other users with news items. This process is known as the socialization of news. This study sought to test the effect of the socialization of news on users’ attention and selection processes. By means of an experimental design employing eye-tracking measurements (N = 86), the study (1) asked whether the existence of various social cues of endorsement (user comments, ‘Likes’, and ‘Reactions’) influenced attention and selection; (2) explored how users divided their attention between news information and social cues, and whether social endorsement affected the attention given by users to news cues; (3) identified differences among users regarding their sensitivity to social endorsement (relying on two psychological traits: self-monitoring and need for cognition). The findings demonstrate that social cues have some effect on attention and selection processes; however, users varied significantly in their response to social endorsement, both in terms of attention and selection. Together, the findings suggest that the socialization of news has changed news consumption processes, and that the consumption of news on SNS might require the development of a new theoretical approach to the consumption of news.
Acknowledgement
The author would like to thank Gal Ravia, Limor Ziv, the editors and anonymous revisers for their helpful comments.
Disclosure statement
No potential conflict of interest was reported by the author.
Notes on contributor
Shira Dvir-Gvirsman is an associate professor at the Department of Communication, Tel Aviv University. Her research focuses on motivational information processing, public opinion, and effects of partisan media [email: [email protected], [email protected]].
Notes
1 Operationalized as attention given to social cues, minus attention given to news information.
2 The decision to use a small number of articles was based on a preliminary observation of the Facebook feeds of five students. We noted that their feeds did not feature a large number of news items. Therefore, in order to mimic a more realistic engagement with news articles, we limited the number of items in the feed.
3 Their headlines were: ‘On modesty, hypocrisy and male domination’; ‘Just for laughs: How to use a rich friend to survive’; ‘Summoning journalists for questioning: crossing the red line’; ‘Trump in a provocative political speech: Obama is no friend to Israel’; ‘April 9, 1988, a ticket to Mashina concert, 15 NIS, June 12, 2016, a ticket to Mashina concert, 174 NIS’; ‘Because of Israel-Today and Ynet, Israel stoops to a new low in its ranking of freedom of the press’; ‘An unconventional projection: the price of apartments in Tel Aviv will to drop soon’; ‘The police refuse to release the documentation of the killing of two brothers in Qalandia checkpoint.’
4 In addition, personal recommendations were counterbalanced for control purposes. for each participant, randomly: half of the posts included a personal recommendation (XX liked this story (two posts) or alternately XX shared this story (two posts) appearing at the top of the post; see ), and half were displayed with the name of the website alone.
5 Given that participants’ heads were not immobilized, significant movements could result in a lost record of eye movement.
6 When testing for slope variability, the model with a random slope was significantly better than the one with a fixed slope for selection and attention, but not for sequence (Selection: ICC = .05; L. ratio = 5.4, p < .05; Fixation: ICC = .09, L. ratio = 142, p < .01; Sequence: ICC = .02, L. ratio = .47, p = .45). However, given that the nesting model is the proper analytical approach for the research design, HLM was still opted as a method of analysis.
7 Interestingly, distinguishing between the three manipulations revealed that while the type of reaction and the presence of user comments had no effect on their own, a high number of ‘likes’ decreased the attention given to news posts (est. = −349.5, SE = 94.2, p < .05).
8 In the case at hand, no significant differences were found between the three manipulations.
9 With regard to H2c and H3c, concerning the effect of the manipulation itself on attention: in the case of sequences, these hypotheses are illogical, since for the manipulation to carry an effect, the participant first needs to observe it. Thus, the manipulation cannot, by definition, affect the sequence of scanning that occurs before it is observed.