随着社会化媒体的发展，在线人际交往在人们的社交活动中占据了越发重要的位 置，甚至一定程度上改变了传统的线下人际交往方式。本文以特纳的“人际交往整合模 型”为框架，探讨个体在线网络关系构建的影响因素。本文抽取了某大型社交网站60万 对好友关系，采用事件史分析方法，考察了个体特征、对偶特征以及网络结构特征对在 线关系构建的影响。本研究发现，用户的网络结构特征相比于其他特征对个体在线关系 构建更为重要，人类的在线社会交往行为并非“一厢情愿”，而是很大程度上受到网络 外部结构的影响；对在线社交网络的量化分析发现，“择优连接”原则在在线社交网络 中并不适用。相反，人们倾向于选择与自己连接度相似的用户，这一定程度上修正了巴 拉巴斯 “择优连接”原则对于网络演化的普适性。

With the prosperity of the social media, online social interaction has played an increasingly important and evolutionary role in personal interactions. Borrowing Turner’s composite model of social interaction as the theoretical framework, this study examines the effects of three sets of factors, including characteristics of individual users, characteristics of dyadic friendship, and characteristics of network structure, on the likelihood of the friendship formation. This study finds that the network structure effects (i.e., the triangular closure process, connectedness of personal network, and degree similarity) dominate tie formation, in comparison with the dyadic characteristics and the actors’ characteristics. Moreover, this study found that “preferential attachment” is not applicable in explaining online friendship formation.