This paper aims to explore the effect of participant composition and contribution behavior of the different types of participants on the quality of knowledge generation in online communities.,This study samples all the featured articles in Chinese Wikipedia and performs a Cox regression to reveal how participant composition and contribution behavior affect the quality of articles in different contexts.,The results show that an increase in the number of participants increases the possibility of either enhancing or reducing the article quality. In most cases, the greater the proportion of core members (people who frequently participate in editing), the higher the possibility of enhancing the article quality. Occasional participants’ editorial behavior hinders quality promotion, this negative effect weakens when such editorial behavior becomes more frequent.,The findings help to better leverage the role of online communities in practice and to achieve knowledge collaboration in a more efficient manner. For example, an appropriate centralized organizational form should be established in online communities to improve the efficiency of crowd contributions. And it is worth developing mechanism to encourage participants to frequently participate in editing the article.,This study contributes to the research on the organizational forms of online communities by showing the effect of participant composition and behavior in the new form of organizing on knowledge generation. This study also contributes to the research on wisdom of crowds by revealing who in a group of participants, in what context, and by what means influence knowledge generation.

中文翻译：

---

人群的智慧：参与者组成和贡献行为对维基百科文章质量的影响

本文旨在探讨参与者组成和不同类型参与者的贡献行为对在线社区知识生成质量的影响。结果表明，参与者数量的增加增加了提高或降低文章质量的可能性。在大多数情况下，核心成员（经常参与编辑的人）的比例越大，提高文章质量的可能性就越大。偶尔的参与者的编辑行为阻碍了质量提升，当这种编辑行为变得更加频繁时，这种负面影响就会减弱。研究结果有助于更好地利用在线社区在实践中的作用，并以更有效的方式实现知识协作。例如，应该在在线社区中建立适当的集中式组织形式，以提高人群贡献的效率。值得鼓励的机制是鼓励参与者经常参与文章的编辑。本研究通过显示参与者组成和行为在新的组织形式对知识生成中的作用，为在线社区的组织形式研究做出了贡献。这项研究还揭示了一群参与者在什么情况下，