Network sampling coverage II: The effect of non-random missing data on network measurement

详细信息    查看全文

• 作者：Jeffrey A. Smith^a ; ^{jsmith77@unl.edu} ; James Moody^b ; ^c ; Jonathan H. Morgan^b
• 关键词：Missing data ; Network sampling ; Network bias
• 刊名：Social Networks
• 出版年：2017
• 出版时间：January 2017
• 年：2017
• 卷：48
• 期：Complete
• 页码：78-99
• 全文大小：8388 K
• 卷排序：48

文摘

We examine the effect of missing more/less central nodes on network measurement. We look at measures of centrality, topology, homophily and centralization. Measurement bias is generally worse when central nodes are missing. The effect of missing central nodes varies by measure and network type. Researchers can estimate bias in their own network using our web-based calculator.