> getwd()
> setwd("Desktop/college-scorecard-raw-data-030216/")
> list.files()
> mydata = read.csv("merged_2013_PP.csv")
> df = data.frame(mydata)
> head(df)
> nrow(df)
[1] 7793
> ncol(df)
[1] 1729
> 

df1 = df[,c(rep(TRUE, 900),rep(FALSE, 829))]
df2 = df[,c(TRUE, rep(FALSE, 899),rep(TRUE, 829))]
> write.csv(df1, file = "df1.csv")
> write.csv(df2, file = "df2.csv")
 write.csv(names(df), file = "col_names.csv")

> df[2692,21]
[1] NULL
Levels: NULL
> df[2692,20]
[1] NULL
Levels: NULL
> names(df)[20]
[1] "LOCALE"
> names(df)[20:21]
[1] "LOCALE"  "locale2"
> df[2692,20:21]
     LOCALE locale2
2692   NULL    NULL
> 



> reading = df[df[,2]==19610,]
> reading['normDist']=exp(-reading['mi_to_zipcode'])

> zips = df[,2]
> length(zips)
[1] 1506704
> zips = unique(zips)
> length(zips)
[1] 29246
> write.csv(zips,"all_zip_codes.csv")


dat = readLines("columns.csv")
smalldf = df[,dat]





Column descriptions
1 = id
4 = Name of college
5 = City
5 = State
7 = zip (or zip+4)