Competition among community colleges in MD

library(XML)

library(RJSONIO)

library(reshape)

library(stringr)

library(dismo)

library(fields)

mydata <- read.csv("E:/MDCommColl.csv", header=T)

gc <- read.table("E:/geocode.txt", header=T)

total <- merge(mydata,gc,by="Address")

dfID<-(expand.grid(total$ID, total$ID))

dfRatio<-(expand.grid(total$ratio, total$ratio))

dfFA2012FT<-(expand.grid(total$FA2012FT, total$FA2012FT))

dfFA2012PT<-(expand.grid(total$FA2012PT, total$FA2012PT))

dfLONG<-(expand.grid(total$longitude, total$longitude))

dfLAT<-(expand.grid(total$latitude, total$latitude))

dfCOLL<-(expand.grid(total$College, total$College))

dfCOLL <- rename(dfCOLL, c(Var1="College", Var2="SameColl"))

dfID <- rename(dfID, c(Var1="ID", Var2="SameID"))

dfCOLL <- rename(dfCOLL, c(Var1="College", Var2="SameColl"))

dfLONG <- rename(dfLONG, c(Var1="longitude", Var2="SameLong"))

dfLAT <- rename(dfLAT, c(Var1="latitude", Var2="SameLat"))

dfFA2012FT <- rename(dfFA2012FT, c(Var1="FA2012FT", Var2="SameFT"))

dfFA2012PT <- rename(dfFA2012PT, c(Var1="FA2012PT", Var2="SamePT"))

dfRatio <- rename(dfRatio, c(Var1="ratio", Var2="SameRatio"))

df <- cbind (dfID, dfRatio, dfLONG, dfLAT, dfCOLL, dfFA2012FT, dfFA2012PT)

df$R <- 6371 # Earth mean radius in km

deg2rad = pi / 180

dlat = deg2rad * (df$SameLat - df$latitude)

dlon = deg2rad * (df$SameLon - df$longitude)

a = sin(dlat / 2) * sin(dlat / 2) + cos(deg2rad * df$latitude) * cos(deg2rad * df$SameLat) * sin(dlon / 2) * sin(dlon / 2)

c = 2 * asin(sqrt(a))

df$miles = df$R * c * 0.621371

df$enrollment.distance.ft = df$FA2012FT*(df$miles**-2)*df$ratio

df$enrollment.distance.pt = df$FA2012PT*(df$miles**-2)*df$ratio

df2 <- df[ which(df$ID!=df$SameID), ]

enrollment.distance.ft.agg <-aggregate(x=df2$enrollment.distance.ft, by=list(ID=df2$SameID), FUN=c("sum"), na.rm=TRUE)

enrollment.distance.pt.agg <-aggregate(x=df2$enrollment.distance.pt, by=list(ID=df2$SameID), FUN=c("sum"), na.rm=TRUE)

enrollment.distance.ft.agg <- rename(enrollment.distance.ft.agg, c(x="d.ft"))

enrollment.distance.pt.agg <- rename(enrollment.distance.pt.agg, c(x="d.pt"))

write.csv(df, "E:/df.csv")

total <- merge(mydata,enrollment.distance.ft.agg,by="ID")

total <- merge(total,enrollment.distance.pt.agg,by="ID")

total$competition.index.ft = (1/total$FA2012FT)*total$ratio*(total$d.ft)

total$competition.index.pt = (1/total$FA2012PT)*total$ratio*(total$d.pt)

write.csv(total, "E:/ci.csv")

(The formula for competition indices is adapted from Misra and Chi (2011) but the ratio of student credit hours to faculty hours taught was added to the equation).

Figure 1. Competition indices - full time enrollment

library(maps)

library(mapproj)

library(RColorBrewer)

library(classInt)

total <- merge(total,gc,by="Address")

map("county", "maryland")

map.axes()

plotvar <- total$competition.index.ft

nclr <- 8

plotclr <- brewer.pal(nclr,"BuPu")

max.symbol.size=10

min.symbol.size=1

class <- classIntervals(plotvar, nclr, style="quantile")

colcode <- findColours(class, plotclr)

symbol.size <- ((plotvar-min(plotvar))/

(max(plotvar)-min(plotvar))*(max.symbol.size-min.symbol.size)

+min.symbol.size)

points(total$longitude, total$latitude, pch=16, col=colcode, cex=symbol.size)

points(total$longitude, total$latitude, cex=symbol.size)

text(-120, 46.5, "Competition indices - full time enrollment")

legend(locator(1), legend=names(attr(colcode, "table")), fill=attr(colcode, "palette"), cex=0.6, bty="n")

Figure 2. Competition indices - part time enrollment