[R] Social Network Analysis and R

Uwe Ligges ligges at statistik.uni-dortmund.de
Wed Apr 23 08:54:55 CEST 2003


Emmanuel F. Koku wrote:
> I'm interested in computing densities of a number of small ego-centred or local 
> networks. I've installed both the R program and the sna package.  My next task 
> is importing the data into R. 
> 
> Our data was entered into Spss in an edgelist format in which each case 
> specifies the link between an alter and an ego. For example: 
> 
> 10101  10102  1 
> 10101  10103  3 
> 10102  10104  2 
> 10201  10203  4 
> 10203  10204  2 
> 
> The following data fragment for example shows an ego-centred network of 2 
> persons: 101 and 102. Person 101 for example reports of a closeness of ties 
> between alters 01 and 02 = value of 1; alters 01 and 03 - value of 3, etc. 
> Similarly Person 102 reports on the closeness of ties between his alters 01 
> and 03 = value of 4; and alters 03 and 04=value of 2. 
> 
> We have about 350 of these small ego networks represented as 560 cases in 
> edgelist format in spss (each line or ego-alter relational link represents a 
> case in spss). 
> 
> First, how do I write out the commands in R to read the edgelists ? 
> 
> Secondly, is there any way to tell R that the first 3 characters of the ego 
> and alter values are actually identifiers for the ego networks -- so that 
> densities are computed for each these? 
> 
> I've begun reading the reference manual on data reading and import facilities 
> and hope that would help me get further along. 

[The question arises why you did not finish it before asking.]

See ?read.fwf for importing the data.

Uwe Ligges



More information about the R-help mailing list