SirSamAlot280/The_New_Great_Migration Secret

## The_New_Great_Migration
#Load packages
library(readr)
library(dplyr)
library(ggplot2)
library(RColorBrewer)
library(ggmap)
library(maptools)
library(gtools)
library('gganimate')

#Read in selected IPUMS data
ipums <- read_csv('./data/Great_Migration.csv')

#Filter out 1950 as only MIGPLAC1 and not MIGPLAC5 is available for this data set
ipums2 <- ipums %>% filter(!YEAR %in% c(1950))

#Determine which states are considered Southern
south <- c(1,5,10,11,12,13,21,22,24,28,37,40,45,47,48,51,54)

#Include only individuals who are racially identified as Black
black <- ipums2 %>% filter(RACESING %in% c(2) & STATEFIP %in%
  south & MIGPLAC5>0 & MIGPLAC5<=56 & !(MIGPLAC5 %in% south))

#Create a variable to distinguish between Males and Females
blackgendered <- black %>% mutate(Sex=ifelse(SEX==1, 'Male','Female'))

#Distinguish between education levels for individuals over the age of 18.
educationlvl <-blackgendered %>% select(YEAR,EDUCD,AGE,PERWT) %>% filter(AGE>=18) %>% mutate(Education=factor(ifelse(EDUCD>=0 & EDUCD<=59,'Less than high school',
                                                            ifelse(EDUCD %in% c(60:64),'High school diploma or GED',
                                                            ifelse(EDUCD %in% c(65,70,71,80,81,82,83,90,100,110,111,112,113),'Some College',
                                                            ifelse(EDUCD %in% c(101),'Bachelor Degree',
                                                            ifelse(EDUCD %in% c(114,115,116),'Advanced Degree','Missing'))))))) %>% filter(Education != 'Missing')
educationlvl$Education <- factor(educationlvl$Education,levels=c('Less than high school','High school diploma or GED','Some College','Bachelor Degree', 'Advanced Degree'))
#Group for only the informaton necessary to see the education attainment for each year
#and weigh.
educationlvl2 <- educationlvl %>% group_by(YEAR,Education) %>%
  summarise(People=sum(PERWT))

#Distinguish between family size, focus on only households, and filter out for non-movers
familysize <- blackgendered %>% select(YEAR,SERIAL,AGE,NCHILD,HHWT,GQ,PERWT) %>%
  filter(GQ==1) %>% mutate(n=1)

# (DO I NEED HHWT?) Include the necessary variables to create data frames for number of children and familysize
familysize2 <- familysize %>% group_by(YEAR,SERIAL,HHWT) %>% mutate(FAMSIZE=sum(n)) %>%
  mutate(Family=ifelse(FAMSIZE==1,'Single','Family'))

#Group only the information needed to see the family size for blacks migrants.
familysize3 <- familysize2 %>% group_by(YEAR,Family) %>% summarise(Household=sum(PERWT))

#Distinguish if a family consists of children
familywchild <- familysize %>% filter(AGE>=18) %>%
  mutate(Children=ifelse(NCHILD==0,'No children','One or more children'))

#Group only the information needed to see the number of children black migrants had in their housheold.
familywchild2 <- familywchild %>% group_by(YEAR,Children) %>%
  summarise(Household=sum(PERWT))

#Establish age categories and floor for population pyramid
agecat <- blackgendered %>% select(YEAR,AGE,Sex,PERWT) %>%
  mutate(Age=ifelse(AGE>=80,8,floor(AGE/10)))

#Assign specific ranges for the age categories
rangeagecat <- agecat %>% mutate(Age=factor(Age,labels=c('0-9','10-19','20-29',
                                                         '30-39','40-49','50-59','60-69','70-79','80+')))
#Separate Male and Female for the population Pyramid
agesexsep1 <- rangeagecat %>% group_by(Age,Sex,YEAR) %>% summarise(Number=sum(PERWT))

#Separate Male and Female for the population Pyramid
agesexsep1$Population <- ifelse(agesexsep1$Sex=='Male',-1*agesexsep1$Number,agesexsep1$Number)

#Aggregate to focus on blacks living in a non-southern state five years ago
blackbpl <- black %>% mutate(Birth=ifelse(BPL %in% south, 'South','Non-South')) %>%
  group_by(YEAR,Birth) %>% summarise(Number=sum(PERWT))

#Aggregate for birthplace and year for black population migrating to the SOuth
blackgroup <- black %>% filter(BPL<=56) %>% group_by(YEAR,BPL) %>%
  summarise(Number=sum(PERWT))

#Aggregate to create new categories
blackgroup2 <- blackgroup %>% mutate(Population=factor(ifelse(Number<250,1,
                                                       ifelse(Number<1000,2,
                                                                            ifelse(Number<4500,3,
                                                                                   ifelse(Number<20000,4,5))))))
levels(blackgroup2$Population) <- c('1-250','251-999','1,000-4,499','4,500-19,999','20,000+')

#Create Population Pyramid for Black Population
png('Black_Population_Pyramids.png',height=750,width=2000)
ggplot(data=agesexsep1,aes(x=Age,y=Population,fill=Sex)) +
  geom_bar(data=agesexsep1[agesexsep1$Sex=='Male',],stat='identity') +
  geom_bar(data=agesexsep1[agesexsep1$Sex=='Female',],stat='identity') +
  coord_flip() +
  facet_wrap(~YEAR) +
  scale_y_continuous(breaks=c(-50000,-25000,0,25000,50000),
                     labels=c('0.05','0.025','0','0.025','0.05')) +
  labs(y='Population in Thousands',title='Population Pyramids for Southern-born African-Americans Migrating Back to the South, 1940,1960-2000') +
  scale_fill_brewer(palette ='Set1',guide=guide_legend(reverse=TRUE))+
  theme_bw() + theme(legend.position='bottom')
dev.off()

#Create bar graph for blacks moving to the South who where born in Southern
#and non-Southern states
png('Black_Population_Birthplace.png',height=500,width=1000)
graph1 <- ggplot(blackbpl, aes(x=YEAR,y=Number,fill=Birth))+
  geom_bar(stat='identity') +
  labs(fill='Where People are Born',title='Number of African-Americans Born in a Southern or Non-Southern State Migrating to the South, 1940,1960-2000',x='YEAR',y='Number of African-Americans') +
  scale_fill_brewer(palette ='Set1')+
  scale_y_continuous(labels=scales::comma)
print(graph1)
dev.off()

#Create bar graph for black migrating by themselves or with family
png('Black_Population_Family.png',height=500,width=1000)
graph2 <- ggplot(familysize3, aes(x=YEAR,y=Household,fill=Family)) +
  geom_bar(stat='identity') +
  labs(fill='',title='The Number of African-American Households Migrating to the South as Single or a Family, 1940,1960-2000',x='YEAR',y='Number of African-Americans Households') +
  scale_fill_brewer(palette ='Set1')+
  scale_y_continuous(labels=scales::comma)
print(graph2)
dev.off()

#Create bar graph for households with and without children
png('Black_Population_Children.png',height=500,width=1000)
graph3 <- ggplot(familywchild2, aes(x=YEAR,y=Household,fill=Children))+
  geom_bar(stat='identity') +
  labs(fill='',title='Number of African-Americans Families Moving to the South With and Without Children, 1940,1960-2000',x='YEAR',y='Number of African-American Households') +
  scale_fill_brewer(palette ='Set1')+
  scale_y_continuous(labels=scales::comma)
print(graph3)
dev.off()

#Create bar graph for educational attainment for African-Americans
png('Black_Population_Education.png',height=500,width=1000)
educationlvl2$Education <- factor(educationlvl2$Education, levels = c('Advanced Degree','Bachelor Degree','Some College','High school diploma or GED','Less than high school'))
graph4 <- ggplot(educationlvl2, aes(x=YEAR,y=People,fill=Education))+
  geom_bar(stat='identity') +
  labs(fill='Education Level',title='Educational Attainment of African-Americans Moving to the South, 1940,1960-2000',x='YEAR',y='Number of African-American Households') +
  scale_fill_brewer(palette ='Set1')+
  scale_y_continuous(labels=scales::comma)
print(graph4)
dev.off()

#Create bar graph for the number of children in migrating black households
#Read in maps
mapdata <- read_csv('data/map.csv',col_types=cols(STATEFIP=col_integer()))

#Create a map of the United States
map1 <- ggplot() + theme_nothing(legend=TRUE) +
  geom_polygon(data=mapdata, aes(x=long,y=lat,group=group),fill='white',color='black')
png('map.png',width=1500,height=1000)
print(map1)
dev.off()

#Join data to the map
blackgroupjoin <- left_join(blackgroup2,mapdata, by=c('BPL'='STATEFIP'))

#Combine every map year produce and animate
map2 <- map1 + scale_fill_brewer(palette = 'Blues') +
  geom_polygon(data=blackgroupjoin,aes(x=long,y=lat,group=group,fill=Population,frame=YEAR),color='black') +
  labs(title='The Birthplace of African-Americans Moving to the South, ')
gg_animate(map2,ani.width=1500,ani.height=1000,'blackmigmap1940,1960-2000.gif')
	#Load packages
	library(readr)
	library(dplyr)
	library(ggplot2)
	library(RColorBrewer)
	library(ggmap)
	library(maptools)
	library(gtools)
	library('gganimate')

	#Read in selected IPUMS data
	ipums <- read_csv('./data/Great_Migration.csv')

	#Filter out 1950 as only MIGPLAC1 and not MIGPLAC5 is available for this data set
	ipums2 <- ipums %>% filter(!YEAR %in% c(1950))

	#Determine which states are considered Southern
	south <- c(1,5,10,11,12,13,21,22,24,28,37,40,45,47,48,51,54)

	#Include only individuals who are racially identified as Black
	black <- ipums2 %>% filter(RACESING %in% c(2) & STATEFIP %in%
	south & MIGPLAC5>0 & MIGPLAC5<=56 & !(MIGPLAC5 %in% south))

	#Create a variable to distinguish between Males and Females
	blackgendered <- black %>% mutate(Sex=ifelse(SEX==1, 'Male','Female'))

	#Distinguish between education levels for individuals over the age of 18.
	educationlvl <-blackgendered %>% select(YEAR,EDUCD,AGE,PERWT) %>% filter(AGE>=18) %>% mutate(Education=factor(ifelse(EDUCD>=0 & EDUCD<=59,'Less than high school',
	ifelse(EDUCD %in% c(60:64),'High school diploma or GED',
	ifelse(EDUCD %in% c(65,70,71,80,81,82,83,90,100,110,111,112,113),'Some College',
	ifelse(EDUCD %in% c(101),'Bachelor Degree',
	ifelse(EDUCD %in% c(114,115,116),'Advanced Degree','Missing'))))))) %>% filter(Education != 'Missing')
	educationlvl$Education <- factor(educationlvl$Education,levels=c('Less than high school','High school diploma or GED','Some College','Bachelor Degree', 'Advanced Degree'))
	#Group for only the informaton necessary to see the education attainment for each year
	#and weigh.
	educationlvl2 <- educationlvl %>% group_by(YEAR,Education) %>%
	summarise(People=sum(PERWT))

	#Distinguish between family size, focus on only households, and filter out for non-movers
	familysize <- blackgendered %>% select(YEAR,SERIAL,AGE,NCHILD,HHWT,GQ,PERWT) %>%
	filter(GQ==1) %>% mutate(n=1)

	# (DO I NEED HHWT?) Include the necessary variables to create data frames for number of children and familysize
	familysize2 <- familysize %>% group_by(YEAR,SERIAL,HHWT) %>% mutate(FAMSIZE=sum(n)) %>%
	mutate(Family=ifelse(FAMSIZE==1,'Single','Family'))

	#Group only the information needed to see the family size for blacks migrants.
	familysize3 <- familysize2 %>% group_by(YEAR,Family) %>% summarise(Household=sum(PERWT))

	#Distinguish if a family consists of children
	familywchild <- familysize %>% filter(AGE>=18) %>%
	mutate(Children=ifelse(NCHILD==0,'No children','One or more children'))

	#Group only the information needed to see the number of children black migrants had in their housheold.
	familywchild2 <- familywchild %>% group_by(YEAR,Children) %>%
	summarise(Household=sum(PERWT))

	#Establish age categories and floor for population pyramid
	agecat <- blackgendered %>% select(YEAR,AGE,Sex,PERWT) %>%
	mutate(Age=ifelse(AGE>=80,8,floor(AGE/10)))

	#Assign specific ranges for the age categories
	rangeagecat <- agecat %>% mutate(Age=factor(Age,labels=c('0-9','10-19','20-29',
	'30-39','40-49','50-59','60-69','70-79','80+')))
	#Separate Male and Female for the population Pyramid
	agesexsep1 <- rangeagecat %>% group_by(Age,Sex,YEAR) %>% summarise(Number=sum(PERWT))

	#Separate Male and Female for the population Pyramid
	agesexsep1$Population <- ifelse(agesexsep1$Sex=='Male',-1*agesexsep1$Number,agesexsep1$Number)

	#Aggregate to focus on blacks living in a non-southern state five years ago
	blackbpl <- black %>% mutate(Birth=ifelse(BPL %in% south, 'South','Non-South')) %>%
	group_by(YEAR,Birth) %>% summarise(Number=sum(PERWT))

	#Aggregate for birthplace and year for black population migrating to the SOuth
	blackgroup <- black %>% filter(BPL<=56) %>% group_by(YEAR,BPL) %>%
	summarise(Number=sum(PERWT))

	#Aggregate to create new categories
	blackgroup2 <- blackgroup %>% mutate(Population=factor(ifelse(Number<250,1,
	ifelse(Number<1000,2,
	ifelse(Number<4500,3,
	ifelse(Number<20000,4,5))))))
	levels(blackgroup2$Population) <- c('1-250','251-999','1,000-4,499','4,500-19,999','20,000+')

	#Create Population Pyramid for Black Population
	png('Black_Population_Pyramids.png',height=750,width=2000)
	ggplot(data=agesexsep1,aes(x=Age,y=Population,fill=Sex)) +
	geom_bar(data=agesexsep1[agesexsep1$Sex=='Male',],stat='identity') +
	geom_bar(data=agesexsep1[agesexsep1$Sex=='Female',],stat='identity') +
	coord_flip() +
	facet_wrap(~YEAR) +
	scale_y_continuous(breaks=c(-50000,-25000,0,25000,50000),
	labels=c('0.05','0.025','0','0.025','0.05')) +
	labs(y='Population in Thousands',title='Population Pyramids for Southern-born African-Americans Migrating Back to the South, 1940,1960-2000') +
	scale_fill_brewer(palette ='Set1',guide=guide_legend(reverse=TRUE))+
	theme_bw() + theme(legend.position='bottom')
	dev.off()

	#Create bar graph for blacks moving to the South who where born in Southern
	#and non-Southern states
	png('Black_Population_Birthplace.png',height=500,width=1000)
	graph1 <- ggplot(blackbpl, aes(x=YEAR,y=Number,fill=Birth))+
	geom_bar(stat='identity') +
	labs(fill='Where People are Born',title='Number of African-Americans Born in a Southern or Non-Southern State Migrating to the South, 1940,1960-2000',x='YEAR',y='Number of African-Americans') +
	scale_fill_brewer(palette ='Set1')+
	scale_y_continuous(labels=scales::comma)
	print(graph1)
	dev.off()

	#Create bar graph for black migrating by themselves or with family
	png('Black_Population_Family.png',height=500,width=1000)
	graph2 <- ggplot(familysize3, aes(x=YEAR,y=Household,fill=Family)) +
	geom_bar(stat='identity') +
	labs(fill='',title='The Number of African-American Households Migrating to the South as Single or a Family, 1940,1960-2000',x='YEAR',y='Number of African-Americans Households') +
	scale_fill_brewer(palette ='Set1')+
	scale_y_continuous(labels=scales::comma)
	print(graph2)
	dev.off()

	#Create bar graph for households with and without children
	png('Black_Population_Children.png',height=500,width=1000)
	graph3 <- ggplot(familywchild2, aes(x=YEAR,y=Household,fill=Children))+
	geom_bar(stat='identity') +
	labs(fill='',title='Number of African-Americans Families Moving to the South With and Without Children, 1940,1960-2000',x='YEAR',y='Number of African-American Households') +
	scale_fill_brewer(palette ='Set1')+
	scale_y_continuous(labels=scales::comma)
	print(graph3)
	dev.off()

	#Create bar graph for educational attainment for African-Americans
	png('Black_Population_Education.png',height=500,width=1000)
	educationlvl2$Education <- factor(educationlvl2$Education, levels = c('Advanced Degree','Bachelor Degree','Some College','High school diploma or GED','Less than high school'))
	graph4 <- ggplot(educationlvl2, aes(x=YEAR,y=People,fill=Education))+
	geom_bar(stat='identity') +
	labs(fill='Education Level',title='Educational Attainment of African-Americans Moving to the South, 1940,1960-2000',x='YEAR',y='Number of African-American Households') +
	scale_fill_brewer(palette ='Set1')+
	scale_y_continuous(labels=scales::comma)
	print(graph4)
	dev.off()

	#Create bar graph for the number of children in migrating black households
	#Read in maps
	mapdata <- read_csv('data/map.csv',col_types=cols(STATEFIP=col_integer()))

	#Create a map of the United States
	map1 <- ggplot() + theme_nothing(legend=TRUE) +
	geom_polygon(data=mapdata, aes(x=long,y=lat,group=group),fill='white',color='black')
	png('map.png',width=1500,height=1000)
	print(map1)
	dev.off()

	#Join data to the map
	blackgroupjoin <- left_join(blackgroup2,mapdata, by=c('BPL'='STATEFIP'))

	#Combine every map year produce and animate
	map2 <- map1 + scale_fill_brewer(palette = 'Blues') +
	geom_polygon(data=blackgroupjoin,aes(x=long,y=lat,group=group,fill=Population,frame=YEAR),color='black') +
	labs(title='The Birthplace of African-Americans Moving to the South, ')
	gg_animate(map2,ani.width=1500,ani.height=1000,'blackmigmap1940,1960-2000.gif')