社群媒體上假新聞的感知可信度

Abstract

社群媒體上的假新聞傳播已經成為一個嚴重的問題。雖然社群媒體讓任何人都可以輕易地創作並分享新聞，但這樣的過程缺乏了傳統新聞媒體的品質管控機制，因而導致新聞品質下降，使人們更容易相信並傳播假新聞。 為了理解人們為何會相信假新聞，進一步預防假新聞的傳播，我們利用推敲可行性模型（Elaboration likelihood model）來解釋影響社群媒體上假新聞可信度的因素。我們假設中央線索（central cue）和周邊線索（peripheral cue）都會對新聞的感知可信度（news perceived believability）產生正向影響，進而影響用戶在社群媒體上的參與行為。另外我們也探討了社群媒體使用者的動機，如何調節這些線索與新聞可信度。 我們的研究結果顯示，中央線索和周邊線索都顯著影響新聞的感知可信度，且用戶的動機也會對新聞的可信度有直接影響。然而，與我們的假設不同，動機並未調節中央線索和周邊線索對新聞感知可信度的影響。這表明用戶在評估新聞感知可信度時，會依據內容、動機對新聞可信度做出判斷。最後，研究發現感知可信度在中央線索與社群媒體行為之間起到了中介作用，並在周邊線索與部分社群媒體行為(點讚、正面評論)之間起到了中介作用。

The spread of fake news on social media has become a serious problem. Social media allows anyone to create and share news easily without traditional quality control by journalism, leading to a decrease in news quality and making fake news more believable and widely shared. To address this issue, we apply the well-known Elaboration Likelihood Model (ELM) to explain the factors influencing the perceived believability of fake news on social media. We hypothesize that both central and peripheral cues positively impact news believability, which in turn affects user behavior on social media. Additionally, we investigate how user motivation moderates the relationship between these cues and the believability of news. Our findings indicate that both central and peripheral cues significantly influence news perceived believability, and user motivation has an impact on believability. However, motivation does not moderate the effects of central and peripheral cues on believability as hypothesized. This suggests that users rely more on their motivation rather than just the content when evaluating news believability. The study also revealed that perceived believability serves as a mediator in the relationship between central cues and social media behaviors, including liking, positive commenting, sharing, and negative commenting. Additionally, perceived believability mediates the relationship between peripheral cues and social media behaviors such as liking and posting positive comments.