Sports news on Twitter edition 1: popularity

For the last several months I have followed 11 different sports news sources on Twitter. Other than my love of sports, my objective in doing this has been to characterize these sports news sources in a variety of ways using R and the twitteR package. Below is a graphic of the 11 sports news sources I chose to analyze, along with their popularity as measured by number of followers.

The two ESPN sources “espn” and “SportsCenter” clearly dominate the field, with Sports Illustrated’s “SInow” coming in a distant third. The Twitter popularity of the ESPN news sources is no doubt a consequence of the leading role of ESPN in the televised delivery of sports, but I wonder if the ESPN Twitter feeds are actually superior to the other Twitter feeds as news sources.

What do you think?

#This code ONLY works once you have authenticated with Twitter
#See twitteR documentation for more details on authenticating

#Specify Twitter screen names
users<-lookupUsers(c('espn', 'Sinow', 'BleacherReport',
'FOXSports', 'YahooSports', 'SportsCenter', 'USATODAYsports', 
'CBSSports', 'sportingnews', 'Deadspin', 'NBCSports'))

#Function to put user data into a data.frame for plotting
users.to.df<-function(x){
  u.df<-data.frame(matrix(nrow=length(x), ncol=16))
  for(i in 1:length(x)){
    u.df[i,]<-x[[i]]$toDataFrame()
  }
  names(u.df)<-c("description", "statusesCount", "followersCount", 
  "favoritesCount", "friendsCount", "url", "name", "created", 
  "protected", "verified", "screenName", "location", "id", 
  "listedCount", "followRequestSent", "profileImageUrl")
  u.df$screenName<-names(x)
  return(u.df)
}
users.df<-users.to.df(users)

#Reorder factor levels for plotting followers in decreasing order
users.df<-users.df[order(-users.df[,3]),]
users.df$screenName<-factor(users.df$screenName,levels=users.df$screenName)

#Plot counts of Twitter followers
library(ggplot2)
ggplot(users.df,aes(screenName,followersCount)) + 
  geom_bar(stat="identity",fill="#253494") +
  xlab("") + ylab("Followers") + theme_bw() + 
  theme(axis.text.x=element_text(angle=45,hjust=1,size=13)) +
  theme(axis.title.y=element_text(vjust=.25,size=14))