R Code for HIST 90.01 Race Assignment

R Code for RACED Assignment.R

library(readr)
library(dplyr)
library(ggplot2)
library(scales)
library(grid)
library(RColorBrewer)

setwd('/Users/joshpearl/Documents/Dartmouth/Junior Year 16W-16X/16W/HIST 90.01- US Hist Through Census/Race Assignment')

a<-read_csv('usa_00004.csv')
# Remove Alaska, Hawaii and overseas military
b<-filter(a, STATEFIP < 60 & (YEAR >= 1960 | !(STATEFIP %in% c(2,15))))
c<-select(b, YEAR, PERWT, RACED)
unique(c$RACED)

# Labels must go in numeric order not the order given by the code
# Get proper labels to count NA from 1850 onwards
d<-mutate(c, RACED=factor(RACED, labels=c("White", 
      "White (Blank)", "Portuguese", "Mexican", "Puerto Rican", "Black", 
      "Mulatto", "Native American", "Chinese", "Japanese", "Filipino", 
      "Asian Indian", "Korean", "Hawaiian", "Other")))

# To get NA compared to everyone else (all the other races), create 
# new variable where the variable is NA and another variable for !=NA 
# and then sum those variables
e<-summarise(group_by(d, YEAR, RACED), NUMBER=sum(PERWT))

# Separate NA from all other races
f<-mutate(e, NATIVE=ifelse(RACED=="Native American", "Native American", 
                                    "Rest of Population (Not Native)"))
g<-summarise(group_by(f, YEAR, NATIVE), POPULATION=sum(NUMBER))
h<-ggplot(g, aes(x=YEAR, y=POPULATION, fill=NATIVE)) + geom_bar(stat="identity") + 
            labs(title="Native American Population Compared to U.S. Population", x="Year", y="Population") + 
            guides(fill=guide_legend(reverse=TRUE))
print(h)

#Create graph of just Native American population over time, must isolate Native American variable, delete FALSE rows
i<-summarise(group_by(g, YEAR, NATIVE=='Native American', POPULATION))
j<-i[i$'NATIVE == "Native American"',]
k<-j[-2]

#Graph just Native American population
l<-ggplot(k, aes(x=YEAR, y=POPULATION)) + 
  geom_bar(stat='identity') + 
  labs(title="Native American Population 1850 - 1950", x='Year', y='Population') +
  scale_y_continuous(labels=scales::comma) + scale_fill_brewer(palette='Greens')
print(l)

#Graph percentages and population of Native Americans and other races, using RACED
m<-select(b, YEAR, RACED, PERWT)
unique(c$RACED)
n<-mutate(m, RACED=factor(RACED, labels=c("White", 
                                          "White (Blank)", "Portuguese", "Mexican", "Puerto Rican", "Black", 
                                          "Mulatto", "Native American", "Chinese", "Japanese", "Filipino", 
                                          "Asian Indian", "Korean", "Hawaiian", "Other")))
o<-summarise(group_by(n, YEAR, RACED), NUMBER=sum(PERWT))
p<-summarise(group_by(n, YEAR), TOTALPOP=sum(PERWT))
q<- merge(o, p, by='YEAR')
r<-ggplot(q, aes(x=YEAR, y=NUMBER, fill=RACED))+geom_bar(stat='identity')+geom_text(aes(label=ifelse(RACED=='Native American', paste(round(NUMBER/TOTALPOP*100, digits=0),
    '%', sep=''), ''),y=NUMBER/2), size=4)+guides(fill=guide_legend(reverse=TRUE))+ggtitle("Percentage of Native Americans and Population Compared to all Races")
print(r)