ggplot2 using census data

ggplot2-census.R

require("maps")
require("ggplot2")
require("sp")

census_data = "ACS_10_1YR_S1901/ACS_10_1YR_S1901_with_ann.csv"
t = read.csv(census_data, skip=6)

sinfo = data.frame(region=t$Geography, 
                   mean=t$Estimate.48, 
                   median=t$Estimate.44,
                   discrep=(t$Estimate.48 / t$Estimate.44) - 1)

sinfo$k10 = ctable$Estimate.4
sinfo$k15 = ctable$Estimate.8
sinfo$k25 = ctable$Estimate.12
sinfo$k35 = ctable$Estimate.16 
sinfo$k50 = ctable$Estimate.20
sinfo$k75 = ctable$Estimate.24
sinfo$k100 = ctable$Estimate.28
sinfo$k150 = ctable$Estimate.32
sinfo$k200 = ctable$Estimate.36
sinfo$k1000 = ctable$Estimate.40

sinfo$region = casefold(sinfo$region)
sinfo$union = 0

ustates = c(
  "california", "colorado", "connecticut", "delaware", "hawaii", "illinois", "iowa",
  "maine", "maryland", "massachusetts", "nevada", "new hampshire", "new york",
  "oregon", "rhode island", "vermont", "washington", "wisconsin")

sinfo$union[sinfo$region %in% ustates] = 1

state_centers = data.frame(
  region=casefold(state.name),
  center=state.center)

m3 <- merge(map_data('state'), sinfo, by='region', all=T)
m3 <- merge(m3, state_centers, by='region', all=T)
m3 <- m3[order(m3$order),]

m3$union = as.factor(m3$union)

p <-ggplot(m3, aes(x=long, y=lat, group=group, fill=k100 + k150)) + 
  geom_polygon() + 
  geom_point(aes(x=center.x, y=center.y, color=union)) +
  scale_fill_gradient(low="red", high="darkgreen") +
  labs(x="", y="", fill="Income");

d=data.frame(region=sinfo$region,income=sinfo$net)
d$r2 = factor(d$region, levels = d$region[order(d$income)])
d$union = sinfo$union
p <- qplot(income, r2, data=d, color=union)
show(p)