社交媒體(SNS)的黑暗面指的是使用社交媒體所帶來的負面影響效果。其中，孤獨感是最被受到關注的議題。孤獨感牴觸了社群媒體服務最初想提供人們增強社交的功能，並對於個體產生嚴重的後果。有鑑於SNS領域的文獻過於零散，且對於孤獨感的解釋仍未具一致性，本研究進行系統性的文獻回顧與整合，更全面的釐清影響的脈絡。研究透過兩階段進行模型驗證。第一階段以系統性文獻回顧，運用Meta Analysis - Structural Equation Modeling (MASEM)建構出研究模型。第二階段基於第一階段模型，進行實證調查與分析。研究共計彙整了35篇文獻，並調查了1,144份有效問卷。研究結果顯示，社會比較是影響使用者產生孤獨感與社交焦慮的負向外在行為。且社會比較的加入，有助於提升整體模型解釋力，這將提供社交媒體領域在其生態系建構更有利的文獻。另一方面，本研究也發現錯失恐懼是影響使用者孤獨感與社交媒體疲勞的內在主要的負面情緒反應。社交媒體疲勞對於孤獨感並未有顯著的影響。另外社交媒體的特性所帶來資訊超載，則是影響使用者內在負面情緒的重要前因。本研究結果將提供未來網路心理學領域的學者作為參考方向及建議。;The dark side of Social network service (SNS) refers to the negative effects of using SNS. One of the effects that have received the most attention is loneliness. Loneliness contradicts what SNS services are supposed to do, that is, to enhance social interaction, causing serious consequences for individuals. Since literature in the SNS field is too fragmented and there are still a lot of gaps in the understanding of loneliness, systematic literature review and integration can help clarify the context of loneliness more comprehensively. Therefore, this study carried out a model verification to construct the research model through two stages: systematic literature review and meta-analysis structural equation modeling (MASEM). Based on the first-stage model, empirical investigation and analysis were carried out again. A total of 35 articles were reviewed and 1,144 valid questionnaires were included. The results showed that social comparison is a negative external behavior that affects users′ loneliness and social anxiety. Moreover, the addition of social comparison helped improve the explanatory power of the overall model, which could provide more favorable literature for the construction of the SNS ecosystem. On the other hand, this study found that fear of missing out was the main internal negative emotional response that affects users′ loneliness and SNS fatigue. SNS fatigue did not significantly affect loneliness. In addition, information overload caused by the characteristics of SNS is an important antecedent that affects users′ internal negative emotions. The results and suggestions stipulated in this study could serve as a reference for future scholars in the field of network psychology.
