Project 1.R

#Code for Project 1

#install packages - this only needs to run once per machine
install.packages('readr')
install.packages('dplyr')
install.packages('ggplot2')
install.packages('scales')
install.packages('grid')

#load packages
library(readr)
library(dplyr)
library(ggplot2)
library(scales)
library(grid)

#set working directory
setwd('/Users/robertmoffitt/Desktop/R Code')

#read in data
a <- read.csv('usa_00002.csv')
head(a)


#select year, perwt, sex, race and region
b <- select(a,YEAR,PERWT,SEX,RACE,REGION)
head(c)

#factor sex variable
c <- mutate(b,SEXF=factor(SEX,labels=c('male','female')))
head(c)

#factor RACE variable to just white and black
d <- mutate(c,Race=ifelse(RACE==1,'white',ifelse(RACE==2,'black','other')))

#create new variable for region
e <- mutate(d,REGIONA=ifelse(REGION<=13,'Northeast',ifelse(REGION<=23,'Midwest',ifelse(REGION<=34,'South',ifelse(REGION<=43,'West','other')))))

#take out regions I am not looking at
f <- filter(e,REGIONA!='other')

#select new variables
g <- select(f,YEAR,PERWT,SEXF,Race,REGIONA)
head(e)

#sum across unique combinations of year, sex, race and region
h <- summarise(group_by(g,YEAR,SEXF,Race,REGIONA),NUMBER=sum(PERWT))
head(g)

#Graph with race as the filled variable, and by gender and region
l <- ggplot(h,aes(x=YEAR,y=NUMBER,fill=Race)) + 
  geom_bar(stat='identity') + 
  facet_grid(SEXF~.~REGIONA)
print(l)

n <- ggplot(h,aes(x=YEAR,y=NUMBER,fill=Race)) + 
  geom_bar(stat='identity',position="fill") + 
  facet_grid(SEXF~.~REGIONA) +
  scale_y_continuous(labels = scales::percent)
print(n)

#Add labels to the non-percent Graph
m <- l + labs(title='Population by Race, Gender and Region',x='Year',y='Number')
print(m)

#Add labels to percent Graph
o <- n + labs(title='Population by Race, Gender and Region',x='Year',y='Number')
print(o)