Helping a dean to resolve his dilemma.
setwd("C:/Users/CJ With HP/Desktop/IIM Lucknow/Datasets")
dean.df <- read.csv(paste("Data - Deans Dilemma.csv",sep=""))
median(dean.df$Salary)
## [1] 240000
mytable <- with(dean.df,table(Placement))
format(round(prop.table(mytable)*100,2),nsmall=2)
## Placement
## Not Placed Placed
## "20.20" "79.80"
placed.df <- dean.df[which(dean.df$Placement=="Placed"),]
median(placed.df$Salary)
## [1] 260000
aggregate(placed.df$Salary,by=list(Gend = placed.df$Gender),mean)
## Gend x
## 1 F 253068.0
## 2 M 284241.9
library(lattice)
hist(placed.df$Percent_MBA,main="MBA Performance of placed students",xlab = "Percent_MBA",ylab="Count",xlim=c(50,80))
notplaced.df <- dean.df[which(dean.df$Placement=="Not Placed"),]
par(mfrow=c(1,2))
hist(placed.df$Percent_MBA,main="MBA Performance of\nplaced students",xlab = "Percent_MBA",ylab="Count",xlim=c(50,80),breaks=3)
hist(notplaced.df$Percent_MBA,main="MBA Performance of\nnot placed students",xlab = "Percent_MBA",ylab="Count",xlim=c(50,80),breaks=3)
par(mfrow=c(1,1))
boxplot(Salary ~ Gender,data=placed.df,horizontal = TRUE,xlab = "Salary",ylab="Gender")
placedET.df <- dean.df[which(dean.df$Placement=="Placed" & dean.df$Entrance_Test!="None"),]
library(car)
scatterplotMatrix(formula=~Salary+Percent_MBA+Percentile_ET,cex=0.5,data=placedET.df,diagonal="density")