The purpose of this on-line textbook is to provide readers with an introduction to statistical applications of social network data. The text outlines the differences between variable-oriented statistical analyses and relation-oriented analyses, and using real data provided by the authors, takes the reader through examples that demonstrate how to test for association between attributes embedded in networks, between multiple networks themselves, and between the attributes and the networks they are embedded within. The text also goes on to examine models of network selection.
It is assumed that the reader has basic knowledge of statistical approaches used to describe distributions, estimate parameters of those distributions, and test hypotheses about those parameters. It is also assumed that readers have a basic understanding of social network data, and recommended that readers refer to the following text by Hanneman and Riddle (2005) when necessary: http://faculty.ucr.edu/~hanneman/nettext/
You are invited to use and redistribute this text freely -- but please acknowledge the source.
Apkarian, Jacob and Robert A. Hanneman. 2016. Statistical Analysis of Social Networks. Jamaica, NY: City University of New York, York College (http://web.york.cuny.edu/~japkarian/).
the full text as a single .pdf file here.***
This textbook is filled with examples that present statistical analyses of social network data using the following software packages: UCINET, Stata, PNET, StOCHNET. The data that is used for the examples was collected from an upper division social networks class taught in 2011. The data are a population census taken over four waves and contain information about student attributes and acquaintanceship. The data are available for download in multiple formats below. Readers are encouraged to work through the examples themselves while they progress through the textbook. The codebook for the data can be downloaded here in .pdf format.
The graphic serving as a backdrop for the title at the top of this page was created using the classroom data set. Each node represents a student, with men represented as circles, and women as squares. The colors indicate the work groups assigned to students at the beginning of the semester. The black lines connecting the nodes display acquaintanceship ties at wave 2 (after completion of one third of the semester).