Brand managers use celebrity microbloggers to endorse their products on microblogs. Previous studies on celebrity endorsement mechanisms concentrated on source factors such as celebrity's characteristics and celebrity—product congruence. This study introduces a new audience factor: the fan–celebrity parasocial interaction (PSI) to explore the celebrity endorsement mechanism within a microblog context. The study hypothesizes that PSI and source factors (credibility, attractiveness, and congruence) significantly influence endorsement effectiveness. The results of an online survey (N = 862) indicate that PSI and celebrity–product congruence are salient antecedents of endorsement effectiveness. PSI serves as a mediator of the effect of source attractiveness on endorsement effectiveness. Source credibility and celebrity–product congruence are mediators between PSI and endorsement effectiveness. The study develops and tests a conceptual model to illustrate the influential mechanism of celebrity endorsement on microblog platforms.

A content analysis of three U.S. Internet newspapers has found that Internet newspapers gave more priority to providing textual information than graphic information, and large graphics were more likely to appear on homepages than on frontpages and news article pages. The news links and the multiple communication channels adopted by Internet newspapers in web page design created a new environment of communication, involving more than host newspaper and initial audience. With interconnected links, the traditional one-to-many newspaper publishing process turned into many-to-many communication centered with and facilitated by the host Internet newspapers. The interconnected news links brought in audience participation in producing newspaper content and providing information beyond the original newspaper content, which demonstrates a shift of balance of communicative power from sender to receiver.

This study examined the third-person effect and the optimistic bias in Internet communication and to what degree sufficiency resource affected the third-person effect and the optimistic bias. Findings demonstrated the third-person effect and the optimistic bias prevalent in traditional media use were also apparent in Internet communication but vary in their relationships with sufficiency resources and other predicting factors. There was a positive relationship between the third-person effect and the optimistic bias involving others. Two indicators of sufficiency resource, computer skill and computer knowledge, and perceived protection ability were significant predictors of the optimistic bias but not that of the third-person effect. Activeness in Internet use was positively predicted by computer knowledge and perceived benefit.

No abstract

This content analysis of the coverage of five television networks and eight newspapers found that both broadcast and print media focused coverage of the Sept. 11 terrorists attacks on facts, but differences do appear in coverage frames and source use.