Who Are My Peers? The code I wrote to both learn R and analyze some descriptive qualities of the current Yale undergraduate student body( Classes '13, '14, '15, '16).

gistfile1.r

#Pull in the html
x <- scan("~/Dropbox/Documents/Senior Year/Spring/STAT230/YaleFacebook3.html",
          what="", sep="\n")
#All the student data is within the div class='display_data'
data <- grep('display_data', x, fixed=TRUE)
#The student data is all on the next line of html. The studens variable now contains the 
#line number of the html with all the data
students <- x[data+1]

#each student's info is contained within the div with class='student_text_container'
#so use this to seperate each student into their own index in the array
a <- unlist(strsplit(students, 'student_text_container', fixed = TRUE, 
                     useBytes = TRUE))

#I further break apart the html
b <- unlist(strsplit(a, 'div', fixed = TRUE, useBytes = TRUE))
c <- unlist(strsplit(b, '<br>', fixed = TRUE, useBytes = TRUE))

#Functions
getmonthcount <- function(index) {
  month <- monthslist[index]
  array <- grep(month,c,fixed=TRUE)
  count <- length(array)
  return(count)
}
getcollegecount <- function(index) {
  college <- collegelist[index]
  array <- grep(college,c,fixed=TRUE)
  count <- length(array)
  return(count)
}

getmajorcount <- function(index) {
  major <- majorslist[index]
  array <- grep(major,c,fixed=TRUE)
  count <- length(array)
  return(count)
}

#Find out how many people have each birthday
monthslist <- c("Jan","Feb","Mar","Apr","May","Jun",
                "Jul","Aug","Sep","Oct","Nov","Dec")
months <- data.frame(months= monthslist, count=rep(NA,length(monthslist)))
for (i in 0:(length(monthslist)-1)) months[(i+1),2] <- getmonthcount(i+1)

#Export Graph
png(file="birthdays.png",width=800,height=600)
par(mar=c(5,6,4,2)+0.1)
barplot(months$count, names.arg=months$months, xlim = NULL, cex.names=0.9, 
        ylab="Students", xlab="Birthday", main="Number of Yale Students By Month of Birthday")
dev.off()


#Find out how many people are in each Yale College
collegelist <- c("Berkeley College","Branford College","Calhoun College",
                 "Davenport College","Ezra Stiles College", 
                 "Jonathan Edwards College", "Morse College","Pierson College",
                 "Saybrook College","Silliman College","Timothy Dwight College",
                 "Trumbull College")
colleges <- data.frame(college= collegelist, count=rep(NA,length(collegelist)))
for (i in 0:(length(collegelist)-1)) colleges[(i+1),2] <- getcollegecount(i+1)

#To have a simpler chart we want to simplify the college names
#Need the stringr library in order to do this
library("stringr")
collegelistforprint <- str_replace(collegelist, " College","")
collegelistforprint <- str_replace(collegelistforprint, "Ezra Stiles", "Stiles")
collegelistforprint <- str_replace(collegelistforprint, "Jonathan Edwards","JE")
collegelistforprint <- str_replace(collegelistforprint, "Timothy Dwight","TD")

#Plot the number of students by college
png(file="colleges.png",width=800,height=600)
par(mar=c(5,6,4,2)+0.1)
barplot(colleges$count, names.arg=collegelistforprint, xlim = NULL, cex.names=0.8,
        xlab="College", ylab="Students", main="Number of Students By Residential College")
dev.off()

#An array of undergraduate majors.
majorslist <- c(
  "African American Studies",
  "African Studies",
  "American Studies",
  "Anthropology",
  "Applied Mathematics",
  "Applied Physics",
  "Archaeological Studies",
  "Architecture",
  "Art",
  "Astronomy",
  "Astronomy & Physics",
  "Biology",
  "Chemistry",
  "Chinese",
  "Classical Civilization",
  "Classics",
  "Cognitive Science",
  "Computer Science",
  "Computer Science & Mathematics",
  "Computer Science & Psychology",
  "Computing and the Arts",
  "East Asian Studies",
  "Ecology & Evolutionary Biology",
  "Economics",
  "Economics & Mathematics",
  "Electrical Engineering",
  "Engineering",
  "English",
  "Environmental Studies",
  "Ethics,Politics & Economics",
  "Ethnicity, Race & Migration",
  "Film Studies",
  "French",
  "Geology & Geophysics",
  "German",
  "German Studies",
  "Global Affairs",
  "History",
  "History of Art",
  "Humanities",
  "Italian",
  "Latin American Studies",
  "Linguistics",
  "Literature",
  "Mathematics",
  "Mathematics & Philosophy",
  "Mathematics & Physics",
  "Modern Middle Eastern Studies",
  "Molecular Biophysics & Biochem",
  "Molecular,Cellular,Dev Biology",
  "Music",
  "Near Eastern Languages & Civs",
  "Philosophy",
  "Physics",
  "Physics & Philosophy",
  "Political Science",
  "Psychology",
  "Religious Studies",
  "Russian",
  "Sociology",
  "Spanish",
  "Statistics",
  "Theater Studies",
  "Women'sGender&SexualityStudies",
  "Undeclared")

#Number of students for each major
majors <- data.frame(majors = majorslist, count=rep(NA,length(majorslist)))
for (i in 0:(length(majorslist)-1)) majors[(i+1),2] <- getmajorcount(i+1)

#Plot Students by Major
png(file="allmajors.png",width=800,height=1200)
par(las=2)
par(mar=c(5,15,4,1)+0.1)
barplot(majors$count, names.arg=majorslist, space=0.8, 
        horiz=TRUE, xlim = NULL, cex.names=1,
        xlab="Number of Students", main="Numer Of Students With Each Declared Major")
dev.off()

##Look at the most popular majors
mostmajors <- subset(majors, count > 50)
mostmajors <- subset(mostmajors, majors != "Undeclared")

#Graph Most Popular Majors
png(file="mostpopularmajors.png",width=800,height=600)
par(las=2)
par(mar=c(5,13,4,2)+0.1)
barplot(mostmajors$count, names.arg=mostmajors$majors, space=0.5, horiz=TRUE, xlim = NULL, cex.names=0.9,
        xlab="Number of Students", main="The Most Popular Majors at Yale")
dev.off()

##Look at the least common majors
fewmajors <- subset(majors,count <30)

#Graph Least Popular Majors
png(file="leastpopularmajors.png",width=800,height=900)
par(las=2)
par(mar=c(5,13,4,2)+0.1)
barplot(fewmajors$count, names.arg=fewmajors$majors, space=0.7, horiz=TRUE, xlim = NULL, cex.names=0.9,
        xlab="Number of Students", main="The Least Popular Majors at Yale")
dev.off()

#Goal: the number of majors by college
#create data frame with majors in left column
majorsbycollege <- data.frame(major=mostmajors$majors)
#Run through each college to get the number of studnets with the most popular
#majors per college
for (i in 1:(length(collegelist)))
{
  collegetocheck <- collegelist[i]
  collegeheader <- collegelistforprint[i]
  #Add a Column & Header for that college to the matrix
  majorsbycollege[collegeheader] <- NA
  #go through each major to check the count for the college currently assessing
  #j is the row (the major)
  for (j in 1:(length(mostmajors$majors)))
  {
   majortocheck <- majorsbycollege[j,1]
   #array of locations for students with the given major
   studentswithmajor <- grep(majortocheck, a, fixed=TRUE)
   #array of locations for students with the given college
   studentswithcollege <- grep(collegetocheck,a,fixed=TRUE)
   occurances = as.integer(0)
   #check for the number of matches: 
   occurancelocations <- grep(TRUE, studentswithmajor %in% studentswithcollege)
   occurances <- occurances + as.integer(length(occurancelocations))
   majorsbycollege[j,(i+1)] <- occurances
  }
}

#Engineering by College
engineering <- grep("Engineering",majorsbycollege[,1])
png(file="engineering.png",width=800,height=600)
par(mar=c(5,6,4,2)+0.1)
barplot(as.integer(majorsbycollege[engineering,2:(length(collegelist)+1)]), names.arg=collegelistforprint,
        cex.names=0.8, xlab="College", main="Number of Engineering Majors by College")
dev.off()

#Computer Science by College
computer <- grep("Computer Science", majorsbycollege[,1])
png(file="compsci.png",width=800,height=600)
par(mar=c(5,6,4,2)+0.1)
barplot(as.integer(majorsbycollege[computer,2:(length(collegelist)+1)]), names.arg=collegelistforprint,
        cex.names=0.8, xlab="College", main="Number of Computer Science Majors by College")
dev.off()

#Economics by College
economics <- grep("Economics", majorsbycollege[,1])
png(file="econ.png",width=800,height=600)
par(mar=c(5,6,4,2)+0.1)
#I have to use economics[1] because economics[2] is the row for Ethics, Politics & Econ
barplot(as.integer(majorsbycollege[economics[1],2:(length(collegelist)+1)]), names.arg=collegelistforprint,
        cex.names=0.8, xlab="College", main="Number of Economics Majors by College")
dev.off()

#English by College
english <- grep("English", majorsbycollege[,1])
png(file="english.png",width=800,height=600)
par(mar=c(5,6,4,2)+0.1)
barplot(as.integer(majorsbycollege[english,2:(length(collegelist)+1)]), names.arg=collegelistforprint,
        cex.names=0.8, xlab="College", main="Number of English Majors by College")
dev.off()

#Physics by College
physics <- grep("Physics", majorsbycollege[,1])
png(file="physics.png",width=800,height=600)
par(mar=c(5,6,4,2)+0.1)
barplot(as.integer(majorsbycollege[physics,2:(length(collegelist)+1)]), names.arg=collegelistforprint,
        cex.names=0.8, xlab="College", main="Number of Physics Majors by College")
dev.off()

#Mathematics by College
mathematics <- grep("Mathematics", majorsbycollege[,1])
png(file="mathematics.png",width=800,height=600)
par(mar=c(5,6,4,2)+0.1)
barplot(as.integer(majorsbycollege[mathematics,2:(length(collegelist)+1)]), names.arg=collegelistforprint,
        cex.names=0.8, xlab="College", main="Number of Mathematics Majors by College")
dev.off()