Lab 4 - LF

Lab 3

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

#Read in the IPUMS data
x <- read_csv('LAB4.csv') %>% filter((BPL>=500 & BPL<=502) | (MBPL>=500 & MBPL<=502) | (FBPL>=500 & FBPL<=502))


#Create a vector for age category labels
agecats <- '0-9'
for (i in 1:7) {
  agecats <- c(agecats, paste(i, '0-', i, 9, sep=''))
}
agecats <- c(agecats,'80+')

#Recode for gender and age
a <- x %>% mutate(Sex=factor(SEX,labels=c('Male','Female')))
b <- a %>% mutate(Age=ifelse(AGE>=80,8, floor(AGE/10)))

#Get it into a factor with labels, as we only have numeric now
c <- b %>% mutate(Age=factor(Age, labels=c(agecats)))

#First or second generation
d <- c %>% mutate(Gen=ifelse(BPL>99, 'First Generation',
                             ifelse(MBPL>99 | FBPL>99, 'Second Generation',
                             'Neither')))

#Sorting out Alaska and Hawaii
e <- d %>% filter(YEAR>=1960 | !(STATEFIP %in% c(2,15)))

#Singling first or second generation
f <- e %>% filter(Gen != 'Neither')

#SLWT for those in 1940 Second Generation 
g1 <- f %>% mutate(Weight= ifelse(YEAR==1940 & Gen=='Second Generation',SLWT,PERWT))

#Aggregating
g <- g1 %>% group_by(Age, Sex, Gen, YEAR) %>% summarise(Number=sum(Weight))
H <- g %>% mutate(Number=ifelse(Sex=='Male',-1 *Number, Number))


#Graphing
ggplot(data=H, aes(x=Age, y=Number, fill=Sex)) +
  geom_bar(data=H[H$Sex=='Male',],stat='identity') + 
  geom_bar(data=H[H$Sex=='Female',],stat='identity') +
  coord_flip() +
  facet_grid(Gen~.~YEAR) +
  scale_y_continuous(breaks = c(-50000,-25000,0,25000,50000),
                     labels = c('5','2.5','0','2.5','5')) +
  labs(y='Population in Tens of Thousands', title= 'Population Pyramids for East Asian Immigrants and their Children') +
  scale_fill_brewer(palette='Set1', guide=guide_legend(reverse=TRUE))+
  theme(legend.position='bottom')