摘要
对狭义的微博影响力进行了研究,通过界定影响微博信息的多层结构,即用户特征和微博属性特征,将微博转发量视为衡量微博影响力的标尺,建立多层线性模型对微博转发量的影响因素进行研究,并应用非参数拔靴法改进了模型的统计推断结果。研究表明,宏观层面的用户特征不仅显著影响微博转发量,并通过跨层交互影响微博属性特征变量进而影响微博转发量。在此基础上,进一步从理论上探讨了用户特征变量与微博属性变量影响微博转发量的原因以及提高微博影响力的可行措施。
This study focused on the narrow sense of micro blog＇s influence, namely, every single micro blog^s influence. By defining the multilevel structure of micro blog information, including user＇ s attributes and micro blog attributes, this research took repost quantity as the objective measurement of micro blog＇s influence and built a Hierarchical Linear Model to find out key factors that affect repost quantity. Nonparametric bootstrap method was applied to improve the statistics inference of outcomes of model fitting. The results showed that the user＇s attributes not only significantly affected the repost quantity of micro blog, but also exerted cross-level effect on micro blog attributes and thus affected repost quantity. Finally, the reasons why these macro-level and micro-level variables could affect the repost quantity were discussed and some possible measures to improve the micro blog＇s influence were provided.
出处
《管理学报》
CSSCI
北大核心
2014年第7期1062-1068,共7页
Chinese Journal of Management
基金
国家自然科学基金资助项目(70872103)
关键词
微博影响力
转发量
信息多层结构
多层线性模型
Micro Blog＇s influence
repost quantity
Micro-blog＇s Multilayer Structure
HLM
