Lab4_Eitel

#Helena Eitel
#Professor Merchant
#QSS 30.05
#Lab Assignment 4: Population Pyramids

library(readr)
library(dplyr)
library(ggplot2)
library(RColorBrewer)

a <- read_csv('./Extract 4.csv')

#recode the sex variable
b <- a %>% mutate(Sex=factor(SEX,labels=c('Male','Female')))
#create age variable
c <- b %>% mutate(Age=ifelse(AGE>=80,8,floor(AGE/10)))
#create vector of labels for age factor variable
agecats <- '0-9'
for (i in 1:7) {
  agecats <- c(agecats,paste(i,'0-',i,9,sep=''))
}
agecats <- c(agecats,'80+')
d <- c %>% mutate(Age=factor(Age,labels=agecats))
#add the generation variable 'Gen'
e <- d %>% mutate(Gen=ifelse(BPL %in% c(430:440),'First Generation',
                             ifelse(MBPL %in% c(430:440) | FBPL %in% c(430:440),'Second Generation','Neither')))
#don't include Alaska or Hawaii until 1960
f <- e %>% filter(YEAR>= 1960 | !(STATEFIP %in% c(2,15)))
#remove anyone who is neither first or second generation
g <- f %>% filter(Gen!='Neither')
h <- g %>% mutate(weight=ifelse(YEAR==1940 & Gen=='Second Generation',SLWT,PERWT))
i <- h %>% group_by(Age,Sex,Gen,YEAR) %>% summarise(Number=sum(weight))
#make all of the male numbers negative so that they move left from the center of the axis
j <- i %>% mutate(Number=ifelse(Sex=='Female',Number*-1,Number))

#create the population pyramid graph
#create a bargraph that you then flip on its side
png('PopulationPyramid1.png',height=500,width=2000)
ggplot(data=j,aes(x=Age,y=Number,fill=Sex)) + 
  geom_bar(data=j[j$Sex=='Male',],stat='identity') +
  geom_bar(data=j[j$Sex=='Female',],stat='identity') +
  coord_flip() +
  facet_grid(Gen~.~YEAR) +
  scale_y_continuous(breaks=c(-600000,-300000,0,300000,600000),labels=c('6','3','0','3','6')) +
  labs(y='Population in Hundreds of Thousands',title='Population Pyramids for Immigrants from Southern Europe and Their Children') +
  scale_fill_brewer(palette='Set1') +
  guides(fill=guide_legend(title='')) +
  theme_bw() + theme(legend.position='bottom')
dev.off()