因2019年COVID-19新型冠狀病毒的爆發，加上網際網絡連接速度與行動裝置的成熟，而使網路直播迅速地蓬勃發展，並且網絡直播已應用於各項服務，例如電子商務、娛樂互動、線上教學與會議等情境，以增加企業額外的收益與便利性。 本研究旨在探討使用網絡直播的行為，基於刺激-有機體-反應模型結合科技接受模型中的認知有用性與認知易用性兩個獨立變數，建構了一研究模型。調查在網絡直播串流平台中的環境刺激效應（如直播主吸引力、擬社會人際互動和資訊質量）對觀眾認知狀態（如認知同化、社會臨場感和信任）的正向影響，提出了8個研究假說；在觀眾認知狀態與接受網絡直播的程度中（如認知同化、社會臨場感、信任、認知有用性和認知易用性）對反應行為（如享樂消費、社交分享和衝動性消費）的正向影響，提出了10個研究假說，根據過去參考文獻共提出了18個研究假說。 為了驗證本研究假說的實用性，因此針對台灣為調查對象，其中以台灣X世代（42至57歲）和Y世代（26至41歲）的民眾，調查影響使用網絡直播平台的行為因素。基於515份有效調查數據，進行驗證性因素分析與結構方程模型分析，檢測衡量問卷的信效度與各構面間的因果關係，實證研究模型之假說。實證研究結果顯示：18個研究假說有5個假說不成立，而其餘13個研究假說皆存在著顯著性。其中在刺激面向擬社會互動對社會臨場感的正向影響最為顯著；在有機體面向則信任對社交分享的正向影響最為顯著，然而刺激-有機體-反應模型與科技接受模型中的認知有用性與認知易用性兩個獨立變數是顯示無相關，本研究架構與結果可作為企業在直播策略的規劃參考，以及直播主如何透過環境刺激效應及觀眾認知狀態吸引顧客。 ;Due to the outbreak of COVID-19 in 2019, coupled with the maturity of Internet connection speed and mobile devices, live streaming has flourished rapidly. It has also been applied to various services such as e-commerce, entertainment interaction, online teaching and video conferencing for new business opportunities and customer channels. The purpose of this research is to investigate the behavioral factors affecting the audiences of live streaming services. Based on the stimulus-organism-response model, the research model also includes the factors of perceived usefulness and perceived ease of use from the technology acceptance model. There are 8 research hypotheses related to the influence of environmental stimuli (such as streamer attractiveness, para-social interaction, and information quality) on audience cognitive status (such as cognitive assimilation, social presence and trust) in live streaming platforms. Additionally, 10 research hypotheses related to the influence of audience cognitive status and acceptance of live streaming (such as perceived usefulness and perceived ease of use) on response behaviors (such as hedonic consumption, social sharing and urge to buy impulsively) were proposed as well. To test the practicability of the above mentioned hypotheses, an empirical study was conducted on the Generation X (aged 42 to 57) and Generation Y (aged 26 to 41) in Taiwan. Based on 515 valid survey data, the study applied confirmatory factor analysis and structural equation model analysis to measure the reliability and validity of the questionnaire and to test the research hypotheses. Results show that 5 research hypotheses are not supported, while the other 13 research hypotheses are accepted. Among the relations from stimulus to organism, the result shows that para-social interaction has the most significant influence on social presence. Among the relations from organism to response, the result indicates that trust has the most significant influence on social sharing. The research results can be used as a reference for companies and streamers to develop strategies for live streaming through understanding environmental stimulation effects and audience cognitive status.
