fig1FP.r

#load libraries
library(readr)
library(dplyr)
library(ggplot2)
library(RColorBrewer)
library(scales)

#load data
a <- read_csv('data/usa_00013(1).csv',col_types=cols(PERWT=col_double()))

#filter out those under 16 and over 65
b <- a %>% filter(AGE>= 16 & AGE<= 65)

#assign sex
c <- b %>%mutate(Sex=factor(SEX, labels=c('Male','Female')))
#Remove Men
cc <- c %>% filter(Sex=='Female')

#Create age groupings*******************************
d <- cc %>% mutate(Agegroup=(factor(ifelse(AGE<=25,1,
                                   ifelse(AGE<=35,2,
                                   ifelse(AGE<=45,3,
                                   ifelse(AGE<=55,4,5)))))))

#label age groupings
e <- d %>% mutate(Agegroup=factor(Agegroup,labels=c('16-25','26-35',
                                          '36-45','46-55','56-65')))

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

#Desrcibe employment:
g <- f %>% mutate(Occ=factor(ifelse(OCC1950>=980,1,2)))

#Label employment:
h <- g %>% mutate(Occ=factor(Occ,labels=c('Not Employed','Employed')))

#Create Marriage groupings
i <- h %>% mutate(Status=factor(ifelse(MARST<=2,1,2)))

#label Marriage groupings
j <- i %>% mutate(Status=factor(Status,labels=c('Married','Single')))

#group to calculate totals of women married/single in each year
k <- j %>% group_by(YEAR,Status,Agegroup) %>% summarize(Total=sum(PERWT))

#group to calculate totals of employed married/single women in each year
l <- j %>% group_by(YEAR,Status,Occ,Agegroup) %>% summarize(Number=sum(PERWT))

#join them so we can divide them
m <- left_join(l,k, by=c('YEAR','Status','Agegroup'))

#Remove the unemployed so its easier to see percentages
n <- m %>% filter(Occ=='Employed')

#Divide to get percents
o <- n %>% mutate(Percent=Number*100/Total)

#Negate single women so that they graph opposite the axis
p <- o %>% mutate(Percent=ifelse(Status=='Single',-1*Percent,Percent))

#Graph
ggplot(data=p,aes(x=Agegroup,y=Percent,fill=Status)) +
  geom_bar(data=p[p$Status=='Single',],stat='identity') +
  geom_bar(data=p[p$Status=='Married',],stat='identity') +
  coord_flip() +
  facet_grid(~YEAR) +
  scale_y_continuous(limits=c(-100,100), breaks=c(-50,0,50),
                     labels=c('50%','0%','50%')) +
  labs(y='Percent Employed',title='Percent Employment among Married and Single Women from 1920-1970',x='Age') +
  scale_fill_brewer(palette='Set1',guide=guide_legend(reverse=TRUE)) +
  theme(legend.position='bottom')+
  theme_bw()
ggsave('Fig1.pdf',width=10, height=7.5)