Longitudinal social network analysis with RSiena
University of Exeter, August 24-28, 2020
Tom
A.B. Snijders
University of Groningen, University of Oxford
Alessandro
Lomi
University of Italian Switzerland, Lugano
The course will be online. It will consist of an alternation of lectures, Q&A sessions, and practical work in breakout groups of 2-4 participants.
It is assumed that the participants have a good basic understanding of statistical methods, including regression and logistic regression; a good understanding of the basics of social network analysis (e.g., the textbook by Borgatti, Everett, and Johnson); and a good working knowledge of R.
Participants will be requested to read slides on their own; there will be afterwards an opportunity for asking questions and having discussion. Slides are mentioned below (letters A-L). They are available as pairs; the files *_s.pdf are the slides for presentation, the files *_p.pdf are the same presentations but for printing, or for taking notes on your computer (if you have the option to take notes on a pdf file).
For working with R and the RSiena package, scripts are available below (letters a-k). You are requested to have R with the RSiena (or RSienaTest) package installed.
Times are approximate indications;
there will be breaks, these are not all indicated in the schedule.
The program is tentative, especially for the later days, and will be adapted to the interests of the participants.
|
Time |
Topic, RSiena functions |
Slides/scripts |
|
|
||
|
Monday |
||
|
9.00-10.30 |
Introduction of the Stochastic Actor-oriented Model (‘SAOM’) for Network Dynamics; types of research questions, data structures – network panel data. |
A |
|
Assignments 1. Formulate your own research question. 2. For the model of p. 67, check the interpretations on p. 68. 3. For the model of p. 70, with ego as on p. 71, check the probabilities on p. 74 (using the help on p. 73). 4. For the model of p. 76, which of the prescriptions of p. 59-64 are followed?
|
||
|
10.30-11.15 |
Q&A session, discussion of assignments. |
|
|
11.45-12.30 |
First experiences with RSiena, siena07 |
a (lines 1-175) |
|
14.00-14.45 |
Practical work with RSiena |
B / b, d, e |
|
15.00-15.45 |
Model specification |
D / g |
|
16.00-17.00 |
Opportunity for participants to mention Depending on available time: Practical work with RSiena |
|
|
|
||
|
Tuesday |
||
|
9.00-10.45 |
Modeling networks and behaviour; |
C |
|
11.00-12.30 |
Use of RSiena for co-evolution models, siena07 |
f |
|
Assignments (Note that the s50 data set is rather small, and accordingly the sensitivity to discover effects is rather low.) (Rather much time will be taken by the running of siena07. If your machine has sufficiently many processors, use the parameters useCluster and nbrNodes. Else spend your waiting time by reading in the RSiena manual. ) 1. For the s50 data with behavioural variable drink, specify a reasonable model based on similarity, in which social influence of friends on drinking can be tested. 2. Estimate the model and test social influence of drinking. 3. Explore the dependence of friendship on drinking using the 5-parameter model, and come up with a reasonable model including the ego * alter interaction. 4. Based on this, specify a reasonable model based on the ego * alter interaction and average alter. 5. Estimate the model and test social influence of drinking. 6. Make a selection plot. 7. Make an influence plot.
|
||
|
14.00-14.45 |
Practical work with RSiena |
g |
|
15.00-15.45 |
Goodness of fit; sienaTimeTest, sienaGOF |
E |
|
16.00-17.00 |
Practical work with RSiena |
g, h, i, j, k |
|
|
||
|
Wednesday |
||
|
9.00-11.00 |
Other co-evolution models: |
F, E |
|
11.00-12.30 |
Depending on participants’ research interests: |
A,B,C,D,E,F |
|
14.00-15.00 |
Imputation of missing data |
G, H |
|
15.00-17.00 |
Topic depending on participants’ research interest; |
|
|
|
||
|
Thursday |
||
|
9.00-10.00 |
Kai Becker: presentation and discussion |
|
|
10.00-11.00 |
Ana Bravo: presentation and discussion |
|
|
11.00-12.00 |
Steffen Triebel: presentation and discussion |
|
|
14.00-15.30 |
Nondirected Networks (briefly); multilevel longitudinal network analysis; |
A, J, K, L |
|
16.00-17.00 |
Topic depending on participants’ research interest, perhaps arising
from the morning session. |
|
|
|
||
|
Friday |
||
|
9.00-10.00 |
Jinhan Jiao: presentation and discussion |
|
|
10.00-11.00 |
Joobin Ordoobody: presentation and discussion |
|
|
11.00-12.00 |
Andrew Parker: presentation and discussion |
|
|
14.00-16.00 |
Special topics: causality, other effects, effect sizes, influence tables, power, multilevel (multi-mode) networks, comparison with other methods. |
C, I / k |
https://www.stats.ox.ac.uk/~snijders/siena/
There will be a dedicated part of this website with materials and scripts for this course.
Ruth M. Ripley, Tom A.B. Snijders, Zsófia Boda, Andras
Vörös, and Paulina Preciado (2020). Manual for SIENA version 4.0.
Oxford: University of Oxford, Department of Statistics; Nuffield College.
https://www.stats.ox.ac.uk/~snijders/siena/RSiena_Manual.pdf
Snijders, T.A.B., van de Bunt, G.G., and Steglich, C.E.G. (2010). Introduction to actor-based models for network dynamics. Social Networks, 32, 44-60.
Christian E.G. Steglich, Tom A.B. Snijders, and Michael Pearson (2010). Dynamic Networks and Behavior: Separating Selection from Influence. Sociological Methodology, 40, 329-393.
Robert W. Krause, Mark Huisman, and Tom A.B. Snijders
(2018). Multiple imputation for longitudinal network data. Italian Journal
of Applied Statistics, 30, 33-57.
https://doi.org/10.26398/IJAS.0030-002
Josh A. Lospinoso and Tom A. B. Snijders (2019). Goodness of
fit for Stochastic Actor-Oriented Models. Methodological Innovations,
12, 2059799119884282.
https://doi.org/10.1177/2059799119884282
A.
Statistical Methods for Social Network Dynamics, A: Networks
https://www.stats.ox.ac.uk/~snijders/siena/Net_longi7_A_s.pdf
https://www.stats.ox.ac.uk/~snijders/siena/Net_longi7_A_p.pdf
B.
Statistical Methods for Social Network Dynamics, B: Estimation
https://www.stats.ox.ac.uk/~snijders/siena/Net_longi7_B_s.pdf
https://www.stats.ox.ac.uk/~snijders/siena/Net_longi7_B_p.pdf
C.
Statistical Methods for Social Network Dynamics, C: Networks and
Behavior
https://www.stats.ox.ac.uk/~snijders/siena/Net_longi7_C_s.pdf
https://www.stats.ox.ac.uk/~snijders/siena/Net_longi7_C_p.pdf
D.
Model Specification Recommendations for Siena
https://www.stats.ox.ac.uk/~snijders/siena/Siena_ModelSpec_s.pdf
https://www.stats.ox.ac.uk/~snijders/siena/Siena_ModelSpec_p.pdf
E.
Goodness of fit testing in RSiena
https://www.stats.ox.ac.uk/~snijders/siena/SienaGOF_s.pdf
https://www.stats.ox.ac.uk/~snijders/siena/SienaGOF_p.pdf
F.
Analyzing the Joint Dynamics of Several Networks
https://www.stats.ox.ac.uk/~snijders/siena/Net_one_co_two_s.pdf
https://www.stats.ox.ac.uk/~snijders/siena/Net_one_co_two_p.pdf
G.
Multiple Imputation for RSiena
https://www.stats.ox.ac.uk/~snijders/siena/AdSUMMissingDataMD.html
H.
Multiple Imputation for RSiena - Network and Behavior
https://www.stats.ox.ac.uk/~snijders/siena/MultipleImputationNetworkAndBehavior.html
I.
Statistical Power in Longitudinal Network Studies
https://www.stats.ox.ac.uk/~snijders/siena/Advanced_SIENA_power-studies.pdf
J.
Multilevel analysis of network dynamics
https://www.stats.ox.ac.uk/~snijders/siena/MultiMetaSAOM_s.pdf
https://www.stats.ox.ac.uk/~snijders/siena/MultiMetaSAOM_h.pdf
K.
Multilevel analysis of network dynamics using sienaBayes
https://www.stats.ox.ac.uk/~snijders/siena/sienaBayes_s.pdf
L.
Missing data in RSiena
https://www.stats.ox.ac.uk/~snijders/siena/RSiena_MissingData_s.pdf
https://www.stats.ox.ac.uk/~snijders/siena/RSiena_MissingData_p.pdf
a. https://www.stats.ox.ac.uk/~snijders/siena/basicRSiena.r
b. https://www.stats.ox.ac.uk/~snijders/siena/Rscript01DataFormat.R
with some basic information about R, networks, data formats etc;
with an example data file arclistdata.dat.
c. https://www.stats.ox.ac.uk/~snijders/siena/RSienaSNADescriptives.R with some descriptives, using package sna.
d. https://www.stats.ox.ac.uk/~snijders/siena/Rscript02SienaVariableFormat.R for how to specify data as variables in RSiena, and specify the model; using the s50 data set.
e. https://www.stats.ox.ac.uk/~snijders/siena/Rscript03SienaRunModel.R for how to carry out the estimation and look at the results.
f. https://www.stats.ox.ac.uk/~snijders/siena/Rscript04SienaBehaviour.R for how to specify models for dynamics of networks and behaviour.
g. https://www.stats.ox.ac.uk/~snijders/siena/g129_net.r the script used for Section 4 of slides D (model specification, five-parameter model)
i. https://www.stats.ox.ac.uk/~snijders/siena/sienaGOF_vdB.R an illustration for goodness of fit.
j. https://www.stats.ox.ac.uk/~snijders/siena/SelectionTables.r for making selection tables and plots.
k. https://www.stats.ox.ac.uk/~snijders/siena/InfluenceTables.r for making influence tables and plots.
l. https://www.stats.ox.ac.uk/~snijders/siena/RscriptMultipleGroups_meta.R an example of meta-analysis with the metafor package and siena08.
Kayo Fujimoto, Tom A.B. Snijders, and Thomas W. Valente (2018). Multivariate dynamics of one-mode and two-mode networks: Explaining similarity in sports participation among friends. Network Science, 6, 370-395.
Yuval Kalish (2019). Stochastic Actor-Oriented Models for
the Co-Evolution of Networks and Behavior: An Introduction and Tutorial. Organizational
Research Methods, in press.
DOI: http://dx.doi.org/10.1177/1094428118825300
Nynke M. D. Niezink, Tom A. B. Snijders, and Marijtje van Duijn (2019). No longer discrete: Modeling the dynamics of social networks and continuous behavior. Sociological Methodology, 49, 295-340.
Tom A.B. Snijders. (2011). Statistical Models for Social Networks. Annual Review of Sociology, 37, 129-151.
Tom A.B. Snijders (2016). The Multiple Flavours of Multilevel Issues for Networks. Chapter 2 in Emmanuel Lazega and Tom A.B. Snijders (eds.), Multilevel Network Analysis for the Social Sciences, Cham: Springer, 2016.
Tom A.B. Snijders and Chris Baerveldt (2003). A Multilevel Network Study of the Effects of Delinquent Behavior on Friendship Evolution. Journal of Mathematical Sociology 27, 123-151.
Tom A.B. Snijders and Alessandro Lomi (2019). Beyond Homophily: Incorporating Actor Variables in Statistical Network Models. Network Science, 7, 1-19.
Tom A.B. Snijders, Alessandro Lomi, and Vanina Jasmine Torló (2013). A model for the multiplex dynamics of two-mode and one-mode networks, with an application to employment preference, friendship, and advice. Social Networks, 35, 265-276.
Christoph Stadtfeld, Tom A. B. Snijders, Christian Steglich,
and Marijtje van Duijn (2018). Statistical Power in Longitudinal Network
Studies, Sociological Methods and Research, 2018.
DOI:
https://doi.org/10.1177/0049124118769113
René Veenstra, Jan Kornelis Dijkstra, Christian Steglich and Maarten H. W. Van Zalk (2013). Network-Behavior Dynamics. Journal of Research on Adolescence, 23, 399-412.
Further see https://www.stats.ox.ac.uk/~snijders/siena/siena_articles.htm