Sociology Social Network Analysis for Social Scientists

Transcription

1 Institute for Social Sciences Proseminar Sociology Social Network Analysis for Social Scientists Spring Quarter 2017 Proseminar Information Classroom Andrews Room, SS&H 2203 Time Wednesdays 3:40 PM 5:40 PM CRN Website ISS Fellows Bob Faris, Associate Professor of Sociology rwfaris@ucdavis.edu Chris Smith, Assistant Professor of Sociology chmsmith@ucdavis.edu Description Social network analysis (SNA) is a method for investigating social structures through the use of network and graph theories. It is used across a wide range of disciplines, from biology to sociology. The techniques covered in this proseminar will be applicable to any number of data types and disciplines. In addition to the primer on social network data, visualization, and analysis, this proseminar will rely heavily on the use of the free statistical and graphical platform R. Overview Weeks 1 6 cover the very basic nuts and bolts of social network analysis. These modules assume that participants have little to no background in social network analysis or in the statistical package R, but these modules do assume some familiarity with quantitative data and analysis. Weeks 7 10 are advanced topics on modeling and computation. When appropriate we will break up the workshop with 10-minute research spotlights with UC Davis social scientists very briefly discussing their social network analysis research projects. The goal of these research spotlights is for students to become familiar with some of the social network research taking place across the UC Davis Social Sciences and to stimulate ideas on how to apply the social network analysis tools presented in the proseminar.

2 Evaluation There is no graded work or formal evaluation for this proseminar. We have designed the proseminar to introduce various SNA topics to participants, and we have designed the materials as a resource for participants to continue their SNA learning and practice outside of the classroom. Enrollment in Soc 298 does require a pass/no pass evaluation at the end of the quarter for transcripts. Calendar Topic Description Files Week 1 - April 5 Introduction to Social Networks and R -introduce some basic concepts in SNA -overview of research using SNA across the social sciences -6 degrees of separation -install R and RStudio software -introduction to R and R syntax -install the statnet package in R Introduction Slides Introduction RStudio introduction.r (script) Recommended: Acton, Ryan M., and Lorien Jasney "An Introduction to Networks Analysis with R and statnet." Sunbelt XXXII Workshop Series. -defining social networks -directed and undirected networks -sociomatrices and edgelists data structures -where to find relational data Week 2 - April 12 Relational Data Matthew Thompson, Sociology, on institutional nomination data -reverse code a network using two data structures -examine the data structures in spreadsheet software -load data into RStudio -convert different datasets into networks with statnet and igraph -access data built into statnet Data Slides Data RStudio Data RStudio Files data.r (script)

3 Week 3 - April 19 Two-Mode Networks -affiliation data -analysis of two-mode networks -one-mode projections of two-mode networks Chris Smith, Sociology, on arrest data -load affiliation data into RStudio -analyze two-mode networks in igraph -generate one mode projections of two mode networks using the igraph package Two Mode Slides twomode.r (script) Week 4 - April 26 Basic Visualization -basic network visualization and customization of one-mode and two-mode networks -visualizing and saving networks in R Visualization Slides visualization.r (script) Week 5 - May 3 Graph-Level Indices -descriptive statistics available at the graph-level -components -density -centralization Cuihua (Cindy) Shen, Communication -hand calculations of graph-level indices -graph-level indices with statnet commands Week 6 - May 10 Node-Level Indices -descriptive statistics available at the node-level -degree -k-core -distance -betweenness Chris Smith, Sociology, on gender in early 1900s Chicago organized crime -hand calculations of node-level -node-level indices with statnet

4 Week 7 - May 17 Exponential Random Graph Models (ERGM) -logic of ERG modeling -parameters -convergence -goodness of fit Michael Levy, Environmental Science and Policy -ERGM tutorial in statnet -statnet online Week 8 - May 24 statnet with Prof. Carter Butts Professor Carter Butts from UC Irvine will be joining us as a guest speaker. Professor Carter Butts is one of the original designers of the statnet package in R. Week 9 - May 31 Big Data and Advanced Visualization with UC Davis Center for Visualization -visualization of complex data -multiplex networks -longitudinal networks -large dense graphs Week 10 - June 7 Stochastic Actor- Oriented Models (SAOM) -logic of stochastic actor-oriented models -model assumptions -SAOM applications Bob Faris, Sociology, adolescent substance use -SAOM in RSiena

5 We did not design this proseminar with a textbook or heavy reading list in mind because we believe that learning methods requires trying the methods. However, if you want to continue your SNA journey into the social sciences we recommend the following resources. Print Resources Kadushin, Charles Understanding Social Networks: Theories, Concepts, and Findings. New York: Oxford University Press. Kadushin s book is a good overview of SNA measurements and applications in social science. It is a beginner level book. Lusher, Dean, Johan Koskinen, and Garry Robins Exponential Random Graph Models for Social Networks: Theory, Methods, and Applications. New York: Cambridge University Press. An excellent textbook for the formulas behind ERGMs. The book does have a preference for PNet software over the statnet package in R, but discusses both packages when appropriate. Wasserman, Stanley, and Katherine Faust Social Network Analysis: Methods and Applications. New York: Cambridge University Press. Wasserman and Faust wrote the bible for social network analysis. This book has been cited over 26,000 times and is often a go-to citation for defining measures in SNA. Electronic Resources R Project Keep your R software up to date and access frequently asked questions. RStudio Download RStudio and its updates here. statnet, statnet wiki, and statnet_help archives Online resources for statnet, which loads several R packages for social network analysis. Anyone can subscribe to the statnet listserve hosted through the University of Washington and access and search through almost a decade of archived threads.