Social Network Analysis

Worksheet 1: Introduction

Author

Termeh Shafie

R packages for network analysis

Throughout the course we will use a variety of different packages of doing network analysis, modeling and visualization. Make sure to install them all and have them ready to load when needed:

install.packages("igraph")   
install.packages("statnet")  #installs ergm, network, and sna
install.packages("snahelper")
install.packages("netUtils")
install.packages("ggraph")
install.packages("backbone")
install.packages("netrankr")
install.packages("signnet")
install.packages("intergraph")
install.packages("graphlayouts")
install.packages("visNetwork")
install.packages("patchwork")
install.packages("edgebundle")
install.packages("ggplot2")
install.packages("gganimate")
install.packages("ggforce")
install.packages("rsiena")

install.packages("remotes")
remotes::install_github("schochastics/networkdata")

Load Packages

We will strat with loading the following packages:

library(igraph)
library(networkdata)
library(netUtils)

Exploring packages and functions

(Interactive Session)

Handling package conflicts

never load sna and igraph at the same time

library(sna)
data("flo_marriage")
degree(flo_marriage)
Error in FUN(X[[i]], ...): as.edgelist.sna input must be an adjacency matrix/array, edgelist matrix, network, or sparse matrix, or list thereof.

If for any reason you have to load both, you can circumvent the error by explicitly stating package first

igraph::degree(flo_marriage)
  Acciaiuoli      Albizzi    Barbadori     Bischeri   Castellani       Ginori 
           1            3            2            3            3            1 
    Guadagni Lamberteschi       Medici        Pazzi      Peruzzi        Pucci 
           4            1            6            1            3            0 
     Ridolfi     Salviati      Strozzi   Tornabuoni 
           3            2            4            3

The package intergraph can be used to transform an igraph object into a network object and vice versa.

#install.packages("intergraph")
library(intergraph)
asNetwork(flo_marriage)
 Network attributes:
  vertices = 16 
  directed = FALSE 
  hyper = FALSE 
  loops = FALSE 
  multiple = FALSE 
  bipartite = FALSE 
  total edges= 20 
    missing edges= 0 
    non-missing edges= 20 

 Vertex attribute names: 
    vertex.names wealth X.priors X.ties 

No edge attributes
degree(asNetwork(flo_marriage))
 [1]  2  6  4  6  6  2  8  2 12  2  6  0  6  4  8  6

You can unload a package without restarting R/RStudio.

detach("package:sna", unload = TRUE)

Handling network data

Inbuilt network data

The networkdata package includes around 1000 datsets and more than 2000 networks. Throughout the course will use several examples using data from this package. Spend some time exploring datasets in this package (you will be asked to choose and work on one of them for you empirical study).

data(package = "networkdata")