I thought it would be useful to take a look at who is taking this class. Its a QMSS class, but how many non-QMSS students are here?
First creating the data frame:
program.df <-as.data.frame.table(table(responses[[2]]),stringsAsFactors=FALSE)
program.df #look at the data frame created
## Var1 Freq
## 1 EdD 1
## 2 Other masters 16
## 3 Other PhD 5
## 4 QMSS First semester 18
## 5 QMSS Second semester + 10
## 6 QMSS third semester 1
## 7 QMSS Third semester 1
## 8 Stat 1
## 9 Statistics 1
## 10 Undergraduate 1
program.df <- program.df[-8,]
program.df[8,"Freq"] <- 2
program.df <- program.df[-6,]
program.df[6,"Freq"] <- 2
names(program.df)[1] <- "Program"
names(program.df)[2] <- "Count"
program.df[4,"Program"] <- "QMSS 1st sem."
program.df[5,"Program"] <- "QMSS 2nd Sem.+"
program.df[6,"Program"] <- "QMSS 3rd Sem."
x.axis <- theme(axis.title.x = element_text(face="bold", colour="#990000", size=20), axis.text.x = element_text(colour="#000000", angle=0, size=14))
y.axis <- theme(axis.title.y = element_text(colour="#000000", size=16), axis.text.y = element_text(angle=0, vjust=0.5, size=14))
title.settings <- ggtitle("Class Make Up")+theme(plot.title=element_text(face="bold", size=28))#why is my tital disappearing????
bar.text <- geom_text(data=program.df,aes(x=Program,y=Count,label=Count),vjust=0)
ggplot(data=program.df, aes(x=Program, y=Count, fill=Program)) + geom_bar(stat="identity")+guides(fill=FALSE)+bar.text+x.axis+y.axis+title.settings
Which program/cohort has the highest proportion of twitter users?
Creating the data frame:
library(plyr)
progtwit <- responses[,c(2,4)]
progtwit[14,"Program"] <- "Stat"
progtwit[28,"Program"] <- "QMSS Third semester"
colnames(progtwit)[2] <- "Twitter"
progtwit$Program[progtwit$Program == "QMSS First semester"] <- "QMSS 1st Sem."
progtwit$Program[progtwit$Program == "QMSS Second semester +"] <- "QMSS 2nd Sem. +"
progtwit$Program[progtwit$Program == "QMSS Third semester"] <- "QMSS 3rd Sem."
twit.title <- ggtitle("Twitter Users by Program")
twit.legend <- theme(legend.text = element_text(size = 20)) + theme(legend.title = element_text(size = 25))
legend.order <- guides(fill = guide_legend(reverse=TRUE))
colors.custom <- scale_fill_manual(values=c("#999999", "#1dcaff"))
#Stacked bar chart
ggplot(progtwit, aes(Program, fill=Twitter))+ geom_bar()+ legend.order + colors.custom + twit.title + x.axis + y.axis + twit.legend
#Dot plot
dotplot <- geom_dotplot(stackgroups=TRUE, binpositions="all")
ggplot(progtwit, aes(Program, fill=Twitter))+ dotplot + legend.order + colors.custom + twit.title + x.axis + y.axis+twit.legend
Suggestions: look into Rcharts, have a button to switch between yes/no’s in each category. Also think about adding proportions and/or frequency to the top of the bars. USE TWITTER BLUE(for users)/GREY(for not)! Consider: Dot plot, not bar chart —