First, Second, and Third deliverables: Data types and data transformations (SSNA 2023)
msgc
June 13, 2023
Useful commands or R Markdown Cheat Sheet
@rowling2019harry said this
@canche2014localized
1 Create and describe your own two-mode edgelist with at least 15 human actors following a two-mode network describe the elements
The following edgelist represents a set of relationships among participants of the Statistical Network Analysis class ascribed to their academic departments. These relationships are stored in the first and second columns. The third column is an attribute of the human actors and the fourth column is an attribute of the relationships recoredd in columns 1 and 2. The following code serves to create this dataset.
@canche2017heterogeneous said this blah blah
dtm2<- read.csv("https://raw.githubusercontent.com/msgc/mapping/master/dtm2.csv")
str(dtm2)## 'data.frame': 22 obs. of 4 variables:
## $ Actor : chr "Manuel" "Manuel" "Abigail" "Marissa" ...
## $ Department: chr "HEd" "IEDP" "EdPol" "IEDP" ...
## $ RUser : chr "Med" "Med" "New" "New" ...
## $ Year : int 2 2 3 1 2 1 1 1 2 1 ...#Loading one of our main packages
# install.packages("igraph")
library(igraph)## Warning: package 'igraph' was built under R version 4.1.3##
## Attaching package: 'igraph'## The following objects are masked from 'package:stats':
##
## decompose, spectrum## The following object is masked from 'package:base':
##
## union# dtm2[,1:2] reads our relationships
g<-graph.data.frame(dtm2[ ,c(1,2,4)]) #This will read the first two columns as relationships and the third an edge attribute
names(dtm2)## [1] "Actor" "Department" "RUser" "Year"g<-graph.data.frame(dtm2[,c("Actor","Department","Year")]) #Same as line 23 but with column names
plot(g)
#Creating our dataset within R
#dtm2<- read.csv("G:/My Drive/SSNA 2017 Penn/SSNA 2018/Deliverables/dtm2.csv")
dtm2<-data.frame(Actor=c("Manuel","Manuel","Abigail","Marissa","Alexander","Ian", "Magali","Giovi","Matt","Emily","Michael","SoYeong","Alexis","Nikita","Fengyin","Liu", "Jimin", "Zhiyuan","Siyu" ,"Qing", "Kate", "Manuel"),#write it in quotation marks separated by commas
Department=c("HEd","IEDP", "EdPol","IEDP","Sociology","IEDP","IEDP","IEDP","IEDP","QM","QM","IEDP", "TLL","BDS","QM","QM","QM","QM","QM" ,"QM", "TLTE", "QM"),#
RUser=c("Med","Med","New","New","Med","New","New","New","New", "Med","Med","New","New","New","New","NEW","Med","New","New","New", "Med", "Med"), #Med, New, Adv
Year=c(2,2,3,1,2,1,1,1,2,1,2,1,1,1,2,2,3,2,2,2,2,2)) #Actual numbers no quotations
# Looking at the first six elements of our dataset
head(dtm2)## Actor Department RUser Year
## 1 Manuel HEd Med 2
## 2 Manuel IEDP Med 2
## 3 Abigail EdPol New 3
## 4 Marissa IEDP New 1
## 5 Alexander Sociology Med 2
## 6 Ian IEDP New 11.1 Read your edgelist as a network and create a sociogram
#install.packages("igraph")
library(igraph)
# dtm2[,1:2] reads our relationships
g<-graph.data.frame(dtm2[,c(1,2,4)]) #This will read the first two columns as relationships, you should probably expand more on this given our class discussion.
plot(g)
2 II. UPDATE transformation using code retaining human actors (column 1 of original edgelist)
As a preamble I tabulate the unique number of human and non-human actors
table(duplicated(dtm2$Actor))#I am interested in the first element##
## FALSE TRUE
## 20 2table(duplicated(dtm2$Department)) #The first element plus the first element of the previous line add the total number of actors##
## FALSE TRUE
## 8 14g#prints the graph information that corroborates the number of actors. What is the meaning of 22 here? ## IGRAPH 272f308 DN-- 28 22 --
## + attr: name (v/c), Year (e/n)
## + edges from 272f308 (vertex names):
## [1] Manuel ->HEd Manuel ->IEDP Abigail ->EdPol
## [4] Marissa ->IEDP Alexander->Sociology Ian ->IEDP
## [7] Magali ->IEDP Giovi ->IEDP Matt ->IEDP
## [10] Emily ->QM Michael ->QM SoYeong ->IEDP
## [13] Alexis ->TLL Nikita ->BDS Fengyin ->QM
## [16] Liu ->QM Jimin ->QM Zhiyuan ->QM
## [19] Siyu ->QM Qing ->QM Kate ->TLTE
## [22] Manuel ->QMv<-as.matrix(get.adjacency(g))
dim(v)## [1] 28 28v<-v[1:table(duplicated(dtm2$Actor))[1],(table(duplicated(dtm2$Actor))[1]+1):ncol(v)]#length(V(g)$name)
dim(v)#what does this mean? You should probably expand on this.## [1] 20 8Now we are ready to dothe II point
v1<-v%*%t(v) #Actor relationships given their common ascription, probably expand more on this
dim(v1)## [1] 20 20To do the III point3.
v2<-t(v)%*%v #Community relationships, probably expand more on this
dim(v2)## [1] 8 82.1 Questions
After executing the following code address the following questions:
nofcomm<-diag(v1)
nofcomm#keep or save that information## Manuel Abigail Marissa Alexander Ian Magali Giovi Matt
## 3 1 1 1 1 1 1 1
## Emily Michael SoYeong Alexis Nikita Fengyin Liu Jimin
## 1 1 1 1 1 1 1 1
## Zhiyuan Siyu Qing Kate
## 1 1 1 1nofactors<-diag(v2)
nofactors## HEd IEDP EdPol Sociology QM TLL BDS TLTE
## 1 7 1 1 9 1 1 1diag(v1)<-0
g1<-graph.adjacency(v1, mode = "undirected")
plot(g1)
g1edgelist<-get.edgelist(g1)
head(g1edgelist)## [,1] [,2]
## [1,] "Manuel" "Marissa"
## [2,] "Manuel" "Ian"
## [3,] "Manuel" "Magali"
## [4,] "Manuel" "Giovi"
## [5,] "Manuel" "Matt"
## [6,] "Manuel" "Emily"diag(v2)<-0
g2<-graph.adjacency(v2, mode = "undirected")
plot(g2)
g2edgelist<-get.edgelist(g2)
head(g2edgelist)## [,1] [,2]
## [1,] "HEd" "IEDP"
## [2,] "HEd" "QM"
## [3,] "IEDP" "QM"2.2 What is the meaning of the diagonal for point II?
Your answer
2.3 What is the meaning of the diagonal for point III?
Your answer
2.4 What would have happened if you did not added zeroes to the diagonals in points II and III?
Your answer
2.5 Explain the meaning of the sociograms resulting from points II and III.
Your answer
2.6 Why did you have to add the option undirected in the code?
Your answer