ECPR Winter School
University of Bamberg, Bamberg
26 February - 4 March 2016




WD209 - Inferential Network Analysis

Instructor Details

Instructor Photo

Philip Leifeld

Institution:
University of Glasgow

Instructor Bio

Philip Leifeld is a senior lecturer (associate professor) in research methods in the School of Social and Political Sciences, University of Glasgow.

His research focuses on social and political networks, quantitative methods, policy debates, and the study of policy processes.

Philip's work has been published in the American Journal of Political Science, the Journal of Statistical Software, and elsewhere.


Course Dates and Times

Monday 29 February to Friday 4 March 2016
Generally classes are either 09:00-12:30 or 14:00-17:30
15 hours over 5 days

Prerequisite Knowledge

This is an advanced course on network analysis. The course assumes existing knowledge of basic social science research methods at least through generalized linear models (logit, the linear model, etc...) as well as basic knowledge of network analysis. Participants should know the basic anatomy of networks as well as the descriptive tools of network analysis (e.g. measures of centrality, plotting and visualization, etc...). Lastly, all techniques will be demonstrated using the R statistical language. While this is not a course *about* software, basic familiarity with R will be quite helpful for students as I will not go into detail about how to load/manage data or use R's more basic functions. That said, a high level of R (e.g. programming competency) is not necessary. It would be helpful if participants could install R and the packages statnet and xergm before the course starts.

Short Outline

This course will cover topics in the advanced modeling of social and political networks. The primary goal is to move beyond descriptive analysis of networks and develop inferential models; primarily the exponential random graph model (ERGM), but alternative techniques such as the stochastic actor-based model (i.e. SIENA), the quadratic assignment procedure, latent space models, and the temporal network autocorrelation model will be considered as well.

Long Course Outline

This course revolves around the idea of creating probabilistic statistical models of networks. This is a big departure from the descriptive analysis of networks (e.g., measuring the centrality of a node) and also a fairly big departure from the statistical modelling of non-network data with the regression framework. Our goal for the course will be the development of statistical models that can accomplish the same general objectives as regression models (fitting parameters to data with probabilistic models), while accounting for the substantial endogenous complexity that is inherent to network data.

In order to accomplish the above, we will consider two basic approaches to modelling networks. The approach we will spend the most time on involves explicitly modelling the network dependencies present in the data. Starting cross-sectionally, we will introduce the exponential random graph model (ERGM) and consider it in some detail - including specification, estimation, fit checking, diagnosing problems, and limitations. We will then extend our knowledge of this approach to longitudinal networks by considering both the Temporal ERGM and the stochastic actor oriented model (SAOM, more commonly known as SIENA), which are closely related.

We will close by considering alternative approaches to modelling networks, including the latent space network model, in which the network dependencies are projected into the error term rather than explicitly modelled, the quadratic assignment procedure, and the temporal network autocorrelation model, a model for the behaviour of nodes in a network. For each topic we cover, we will also consider how to perform such analyses in R using several example datasets.

Day-to-Day Schedule

Day 
Topic 
Details 
1Introduction and why we need network-specific models (and not regressions).Lecture
2Introduction to the ERGM, form, specification, estimation, and interpretation.Lecture with some lab
3The ERGM, form, specification, estimation, and interpretation (continued).Some lecture with substantial lab
4Longitudinal network models: TERGM and SIENA.Mostly lecture, some lab
5Latent Space Models, QAP, and autocorrelation models.Mostly lecture, some lab
Day-to-Day Reading List

Day 
Readings 
1- Lusher, Dean, Johan Koskinen and Garry Robins. 2013. Exponential Random Graph Models for Social Networks. New York, NY: Cambridge University Press. Chapters 2-5. - Cranmer, Skyler J., Philip Leifeld, Scott McClurg, and Meredith Rolfe (2015): Navigating the Range of Statistical Tools for Inferential Network Analysis. Working Paper.
2- Cranmer, Skyler J. 2011. Inferential Network Analysis with Exponential Random Graph Models. Political Analysis 19: 66-86. - Leifeld, Philip and Volker Schneider 2012. Information Exchange in Policy Networks. American Journal of Political Science 53(3): 731-744.
3- Hunter, David R., Mark S. Handcock, Carter T. Butts, Steven M. Goodreau and Martina Morris. 2008. "ergm: A Package to Fit, Simulate and Diagnose Exponential-Family Models for Networks." Journal of Statistical Software 24(3):1-29. - Goodreau, Steven M., Mark S. Handcock, David R. Hunter, Carter T. Butts and Martina Morris. 2008. "A statnet tutorial." Journal of Statistical Software 24(9): 1-26
4- Leifeld, Philip, Skyler J. Cranmer and Bruce A. Desmarais (2015): Temporal Exponential Random Graph Models with xergm: Estimation and Bootstrap Confidence Intervals. Working Paper. - Leifeld, Philip and Skyler J. Cranmer (2015): A Theoretical and Empirical Comparison of the Temporal Exponential Random Graph Model and the Stochastic Actor-Oriented Model. Working Paper. https://arxiv.org/abs/1506.06696
5- Leifeld, Philip and Skyler J. Cranmer (2015): The Temporal Network Autocorrelation Model. Working Paper.
Software Requirements

All software we use will be free: R and several of its packages (e.g. ergm, statnet, xergm).

Hardware Requirements

Students to bring their own laptops.

Literature


Berardo, Ramiro and John T. Scholz. 2010. "Self-Organizing Policy Networks: Risk, Partner Selection, and Cooperation in Estuaries." American Journal of Political Science 54(3):632-649.

Butts, C. T. (2008a). A relational event framework for social action. Sociological Methodology, 38(1):155–200.

Butts, C. T. (2008b). network: A package for managing relational data in R. Journal of Statistical Software, 24(2):1–36.

Cranmer, Skyler J., Bruce A. Desmarais and Elizabeth Menninga. 2012. "Complex Dependencies in the Alliance Network." Conflict Management and Peace Science 29(3): 279-313.

Dekker, D., Krackhardt, D., and Snijders, T. A. B. (2007). Sensitivity of MRQAP tests to collinearity and autocorrelation conditions. Psychometrika, 72(4):563– 581.

Desmarais, Bruce A. and Skyler J. Cranmer. 2012. "Micro-Level Interpretation of Exponential Random Graph Models with Application to Estuary Networks." Policy Studies Journal 40(3): 402-434.

Desmarais, Bruce A. and Skyler J. Cranmer. 2012. "Statistical Mechanics of Networks: Estimation and Uncertainty." Physica A 391(4):1865-1876.

Goodreau, Steven .M., James A. Kitts and Martina Morris. 2009. "Birds of a feather, or friend of a friend? Using exponential random graph models to investigate adolescent social networks." Demography 46(1):103-125.

Handcock, M. S., Hunter, D. R., Butts, C. T., Goodreau, S. M., and Morris, M. (2008). statnet: Software tools for the representation, visualization, analysis and simulation of network data. Journal of Statistical Software, 24(1):1–11.

Hanneke, Steve, Wenjie Fu and Eric P. Xing. 2010. "Discrete Temporal Models of Social Networks." Electronic Journal of Statistics 4:585-605.

Hoff, Peter D., Adrian E. Raftery and Mark S. Handcock. 2002. "Latent Space Approaches to Social Network Analysis." Journal of the American Statistical Association 97(460):pp. 1090-1098.

Ingold, Karin and Philip Leifeld (forthcoming): Structural and Institutional Determinants of Influence Reputation: A Comparison of Collaborative and Adversarial Policy Networks in Decision Making and Implementation. Journal of Public Administration Research and Theory.

Krackhardt, David. 1988. Predicting with networks: Nonparametric multiple regression analysis of dyadic data. Social Networks 10(4):359-381.

Krivitsky, Pavel N. and Mark S. Handcock. 2008. "Fitting Latent Cluster Models for Networks with latentnet." Journal of Statistical Software 24(5):1-23.

Leenders, R. T. A. J. (2002). Modeling social influence through network autocorrelation: Constructing the weight matrix. Social Networks, 24(1):21–47.

Lerner, J., Bussmann, M., Snijders, T. A. B., and Brandes, U. (2013). Modeling frequency and type of interaction in event networks. Corvinus Journal of Sociology and Social Policy, 4:3–32.

Lusher, Dean, Johan Koskinen and Garry Robins. 2013. Exponential Random Graph Models for Social Networks. New York, NY: Cambridge University Press.

Morris, M., Handcock, M. S., and Hunter, D. R. (2008). Specification of exponential- family random graph models: Terms and computational aspects. Journal of Statistical Software, 24(4):1–24.

Robins, G., Pattison, P., Kalish, Y., and Lusher, D. (2007). An introduction to exponential random graph (p*) models for social networks. Social Networks, 29(2):173–191.

Snijders, Tom A.B., Gerhard G. van de Bunt and Christian E.G. Steglich. 2010. "Introduction to stochastic actor-based models for network dynamics." Social Networks 32(1):44 – 60.

The following other ECPR Methods School courses could be useful in combination with this one in a ‘training track .
Recommended Courses Before

Winter School WD102 Introduction to Applied Social Network Analysis (NOT STRICTLY REQUIRED)

Additional Information

Disclaimer

The information contained in this course description form may be subject to subsequent adaptations (e.g. taking into account new developments in the field, specific participant demands, group size etc.). Registered participants will be informed in due time in case of adaptations.

Note from the Academic Convenors

By registering to this course, you certify that you possess the prerequisite knowledge that is requested to be able to follow this course. The instructor will not teach these prerequisite items. If you are not sure if you possess this knowledge to a sufficient level, we suggest you contact the instructor before you proceed with your registration.


Share this page
 

"History is past politics, and politics is past history" - E.A. Freeman


Back to top