Network Sampling

Network sampling designs select subsets of the nodes and relationships in a network for observation and study. Some network samples are drawn in order to estimate properties of a network—like density, centralization, or segmentation—or parameters of stochastic [Page 1046]network models. Other network samples identify pools of subjects for survey studies about characteristics of a population of nodes. This entry describes some major network sampling designs employed in these ways and concludes with considerations for researchers when making inferences from network samples.

All network sampling designs select subsets of both the nodes and the relationships that comprise a network. Some designs explicitly sample nodes from a list or other sampling frame, and then observe them and the relationships in which they are involved. Others start by sampling ...