analyses/demography/global.R
In joebrew/vilaweb: Reproducible data analysis for www.vilaweb.cat
# Libraries
library(knitr)
library(vilaweb)
library(tidyverse)
library(ggplot2)
library(tidyr)
library(gganimate)
# Functions
mround <- function(x,base){
base*round(x/base)
}
round_percent <- function(x) {
x <- x/sum(x)*100 # Standardize result
res <- floor(x) # Find integer bits
rsum <- sum(res) # Find out how much we are missing
if(rsum<100) {
# Distribute points based on remainders and a random tie breaker
o <- order(x%%1, sample(length(x)), decreasing=TRUE)
res[o[1:(100-rsum)]] <- res[o[1:(100-rsum)]]+1
}
res
}
numberfy <- function(x){
gsub(',', '.', scales::comma(x), fixed = TRUE)
}
# Read in CEO data
new_ceo <- vilaweb::new_ceo
# Read in Catalan demography data
df <- read_delim('data/idescat/t8661.csv', delim = ';', skip = 5)
names(df)[1] <- 'age'
df$age <- gsub(' i més', '', df$age)
df$age <- as.numeric(df$age)
for(j in 2:ncol(df)){
message(j)
vals <- as.character(unlist(df[,j]))
vals <- gsub(',', '.', vals, fixed = TRUE)
vals <- as.numeric(vals)
df[,j] <- vals
}
# Keep only the relevant definitions
df <- df %>% dplyr::select(age, nqx)
# Estimate the last few rows
last_few <- tibble(age = 95:99,
nqx = seq(0.25, 1, length = 5))
# Combine
df <- df %>% filter(age != 95) %>%
bind_rows(last_few)
# Create a table of definitions for the demographic data
df_definitions <-
tibble(variable = c('nPx', 'nDx', 'nMx', 'nqx', 'nax', 'lx',
'ndx', 'nLx', 'Tx', 'ex'),
meaning = c('Mid-year population of those aged x to x+n',
'Number of deaths for those aged x to x+n',
'Age-specific death rates beween ages x to x+n',
'Probability of dying between ages x and x+n',
'Average person-years lived in the interval by those dying in the interval',
'Number of survivors who make it to age x',
'Number of people who will die between age x and x+1',
'Person-years lived between ages x and x+n',
'Person-years lived above age x',
'Expectation of life at age x'))
# Read in place of birth data
lloc <- read_csv('data/idescat/poblacio_per_lloc_de_naixement.csv', skip = 7)
names(lloc) <- c('year', 'Catalunya', 'x', 'y',
"Espanya", "Estranger" , "Total")
lloc <- lloc %>% dplyr::select(year, Catalunya, Estranger, Espanya) %>%
mutate(year = gsub(' (p)', '', year, fixed = TRUE)) %>%
mutate(year = as.numeric(year))
# Make a long version
lloc_long <- lloc %>% tidyr::gather(key, value, Catalunya:Espanya)
# # Basic plot of lloc (exploratory only)
old_lloc <- lloc_long
past_plot <- function(ca = FALSE){
pd <- old_lloc %>%
group_by(year) %>%
mutate(p = value / sum(value) * 100) %>%
ungroup %>%
filter(key == 'Espanya')
if(ca){
the_labs <- labs(title = 'Percentatge de catalans nascuts a Espanya',
caption = 'Dades de l\'Institut d\'Estadística de Catalunya. Gràfic de Joe Brew | @joethebrew | www.vilaweb.cat')
} else {
the_labs <- labs(title = 'Percentage of Catalans born in Spain',
caption = 'Data from the Institut d\'Estadística de Catalunya. Chart by Joe Brew | @joethebrew | www.vilaweb.cat')
}
ggplot(data = pd,
aes(x = year,
y = p)) +
geom_bar(stat = 'identity',
width = 1,
color = 'white',
fill = 'darkorange') +
theme_vilaweb() +
the_labs +
geom_text(aes(label = round(p, digits = 1)),
color = 'white',
nudge_y = -2,
size = 2)
}
# Read in place of birth + year + age data
lloc_dir <- 'data/lloc_i_edat_i_any/'
lloc_files <- dir(lloc_dir)
lloc_list <- list()
read_lloc <- function(x){
data <- suppressMessages(read_delim(x,
delim = ';',
skip = 5)) %>%
dplyr::rename(age = X1) %>%
filter(age != 'Total') %>%
dplyr::select(age,
`Catalunya. Total`,
`Resta d'Espanya. Total`,
`Estranger. Total`)
names(data)[2:4] <- c('Catalunya', 'Espanya', 'Estranger')
year <- unlist(suppressMessages(suppressWarnings(read_delim(x, delim = ';'))[1,1]))
year <- as.numeric(gsub('Catalunya. ', '', as.character(unlist(year)), fixed = TRUE))
data$year <- year
return(data)
}
for(i in 1:length(lloc_files)){
this_file <- paste0(lloc_dir, lloc_files[i])
this_data <- read_lloc(this_file)
message(i)
message(this_data$year[1])
lloc_list[[i]] <- this_data
}
lloc <- bind_rows(lloc_list)
lloc$age <- gsub('anys i més', 'a 99 anys', lloc$age)
split_age <- strsplit(lloc$age, ' ')
lloc$min_age <- as.numeric(unlist(lapply(split_age, function(x){x[2]})))
lloc$max_age <- as.numeric(unlist(lapply(split_age, function(x){x[4]})))
# # Plot (exploratory only)
# pd <- lloc %>% gather(key, value, Catalunya:Estranger)
# ggplot(data = pd %>% filter(year == max(year)),
# aes(x = age,
# y = value,
# fill = key)) +
# geom_bar(stat = 'identity') +
# coord_flip()
# Expand lloc into a one year per row data frame (interpolation)
agey <- tibble(age = 0:99)
age_list <- list()
for(i in 1:nrow(lloc)){
this_row <- lloc[i,]
n <- this_row$max_age - this_row$min_age + 1
new_row <- tibble(age = this_row$min_age:this_row$max_age,
Catalunya = this_row$Catalunya / n,
Espanya = this_row$Espanya / n,
Estranger = this_row$Estranger / n,
year = this_row$year)
# # Get endpoints for smoother interpolation
# end_points <- lloc %>%
# filter(year == this_row$year,
# max_age %in% c(this_row$max_age +5,
# this_row$max_age - 5))
# if(nrow(end_points) == 2){
# the_diffs <- end_points[2,2:4] + end_points[1,2:4]
# gaps <- the_diffs / 7
# vals <- -2:2
# for(i in 1:nrow(new_row)){
# old_vals <- new_row[i,2:4]
# new_vals <- old_vals + (vals[i] * gaps)
# message('Old vals:', sum(old_vals))
# message('New vals: ', round(sum(new_vals)))
# new_row[i,2:4] <- new_vals
# }
# }
age_list[[i]] <- new_row
}
lloc <- bind_rows(age_list)
# Make long
lloc_long <- lloc %>% tidyr::gather(key, value, Catalunya:Estranger)
# # Make plot (exploratory only)
# ggplot(data = lloc_long %>% filter(year == max(year)),
# aes(x = age,
# y = value,
# fill = key)) +
# geom_area() +
# coord_flip()
# # Animation (exploratory only)
# library(gganimate)
# ggplot(data = lloc_long,
# aes(x = age,
# y = value,
# fill = key)) +
# geom_bar(stat = 'identity') +
# coord_flip() +
# labs(title = 'Year: {frame_time}') +
# transition_time(year) +
# ease_aes('linear')
# Estimate what would happen, demographically, with Catalonia's current population
current <- lloc_long %>% filter(year == 2018)
the_list <- list()
years <- 2019:2070
for(i in 1:length(years)){
the_year <- years[i]
# Get last year
if(i == 1){
last_year <- current
} else {
last_year <- the_list[[i-1]]
}
# Join the probability of death
this_year <- last_year %>%
dplyr::select(-year) %>%
left_join(df %>% dplyr::select(age, nqx)) %>%
# replace the value with the mortality-adjusted value
mutate(value = value - (value * nqx)) %>%
mutate(year = the_year) %>%
dplyr::select(-nqx) %>%
mutate(age = age + 1) %>%
filter(age <= 100)
the_list[[i]] <- this_year
}
future <- bind_rows(the_list)
make_animation <- function(ca = FALSE){
if(ca){
legend_title <- 'LLoc de naixement'
pd <- future
the_labs <- labs(title = 'Any: {frame_time}',
x = 'Edat',
y = 'Persones',
subtitle = 'Projecció de la composició de la població catalana\n(assumint cap naixement o migració després de 2019)',
caption = 'Dades de llocs de naixement i mortalitat de l\'Institut d\'Estadística de Catalunya.\nProjeccions i gràfic de Joe Brew | @joethebrew | www.vilaweb.cat')
pd$key <- factor(pd$key,
levels = rev(c('Catalunya',
'Espanya',
'Estranger')))
} else {
legend_title <- 'Place of birth'
pd <- future %>%
mutate(key = ifelse(key == 'Catalunya', 'Catalonia',
ifelse(key == 'Espanya', 'Spain',
ifelse(key == 'Estranger', 'Abroad', key))))
pd$key <- factor(pd$key,
levels = rev(c('Catalonia',
'Spain',
'Abroad')))
the_labs <- labs(title = 'Year: {frame_time}',
x = 'Age',
y = 'People',
subtitle = 'Projection of the composition of Catalonia\'s population\n(assuming no births or migration after 2019)',
caption = 'Data on places of birth and mortality from the Institut d\'Estadística de Catalunya.\nProjections and chart by Joe Brew | @joethebrew | www.vilaweb.cat')
}
cols <- rev(c(colors_vilaweb()[c(5,3)], 'grey'))
label_df <- pd %>%
group_by(key, year) %>%
summarise(value = sum(value, na.rm = TRUE)) %>%
group_by(year) %>%
mutate(p = value / sum(value) * 100) %>%
ungroup %>%
filter(!is.na(p)) %>%
group_by(year) %>%
summarise(value = paste0(legend_title, '\n',paste0(key, ': ', round(p, digits = 1), '%', collapse = '\n'), collapse = ''))
ggplot() +
geom_label(data = label_df,
aes(x = 90,
y = 110000,
label = value)) +
geom_bar(stat = 'identity', width = 1,
data = pd,
aes(x = age,
y = value,
fill = key)) +
coord_flip() +
scale_fill_manual(name = legend_title,
values = cols) +
theme_vilaweb() +
transition_time(year) +
the_labs
}
make_plot <- function(ca = FALSE){
if(ca){
legend_title <- 'LLoc de naixement'
pd <- future
the_labs <- labs(x = 'Any',
y = 'Percentatge',
title = 'Projecció: lloc de naixement dels catalans',
subtitle = '(assumint cap naixement o migració després de 2019)',
caption = 'Dades de llocs de naixement i mortalitat de l\'Institut d\'Estadística de Catalunya.\nProjeccions i gràfic de Joe Brew | @joethebrew | www.vilaweb.cat')
pd$key <- factor(pd$key,
levels = rev(c('Catalunya',
'Espanya',
'Estranger')))
} else {
legend_title <- 'Place of birth'
pd <- future %>%
mutate(key = ifelse(key == 'Catalunya', 'Catalonia',
ifelse(key == 'Espanya', 'Spain',
ifelse(key == 'Estranger', 'Abroad', key))))
pd$key <- factor(pd$key,
levels = rev(c('Catalonia',
'Spain',
'Abroad')))
the_labs <- labs(title = 'Projection: place of birth of Catalans',
x = 'Year',
y = 'Percentage',
subtitle = '(assuming no births or migration after 2019)',
caption = 'Data on places of birth and mortality from the Institut d\'Estadística de Catalunya.\nProjections and chart by Joe Brew | @joethebrew | www.vilaweb.cat')
}
cols <- rev(c(colors_vilaweb()[c(5,3)], 'grey'))
# Group pd
pd <- pd %>%
group_by(key, year) %>%
summarise(value = sum(value, na.rm = TRUE)) %>%
ungroup %>%
group_by(year) %>%
mutate(p = value / sum(value) * 100) %>%
ungroup
ggplot() +
geom_line(data = pd,
aes(x = year,
y = p,
color = key)) +
scale_color_manual(name = legend_title,
values = cols) +
theme_vilaweb() +
the_labs
}
birth_plot <- function(ca = FALSE){
if(ca){
no_know <- 'NS/NC'
catalunya <- 'Catalunya'
ccaa <- 'Espanya'
mon <- 'Estranger'
} else {
no_know <- 'Not sure\nNo answer'
catalunya <- 'Catalonia'
ccaa <- 'Spain'
mon <- 'Abroad'
}
pd <- new_ceo %>%
mutate(neixer = `Em podria dir on va néixer?`,
indepe = `Vol que Catalunya esdevingui un Estat independent?`,
edat = Edat) %>%
mutate(indepe = as.character(indepe),
neixer = as.character(neixer)) %>%
filter(!is.na(neixer),
!is.na(indepe),
!neixer %in% c('No ho sap', 'No contesta')) %>%
mutate(indepe = ifelse(indepe %in% c('No ho sap',
'No contesta'),
no_know,
indepe),
neixer = ifelse(neixer == 'Catalunya', catalunya,
ifelse(neixer %in% c('Altres comunitats autònomes'), ccaa,
mon))) %>%
group_by(neixer, indepe) %>%
tally %>%
ungroup %>%
group_by(neixer) %>%
mutate(p = n / sum(n) * 100) %>%
ungroup
if(ca){
the_labs <- labs(x = 'Lloc de naixement',
y = 'Percentatge',
title = 'Independentisme per lloc de naixement',
caption = paste0('Dades: Baròmetre d\'Opinió Política del Centre d\'Estudios d\'Opinió.\nMostra: ',
numberfy(sum(pd$n)), ' residents de Catalunya amb ciutadania espanyola.\nGràfic: Joe Brew | @joethebrew | www.vilaweb.cat'))
legend_title <- 'A favor\nde la\nindependència?'
} else {
the_labs <- labs(x = 'Place of birth',
y = 'Percentage',
title = 'Support for independence by\nplace of birth',
caption = paste0('Data: Baròmetre d\'Opinió Política of the Centre d\'Estudios d\'Opinió.\nMostra: ',
numberfy(sum(pd$n)), ' residents of Catalonia with Spanish citizenship.\nChart: Joe Brew | @joethebrew | www.vilaweb.cat'))
legend_title <- 'In favor\nof\nindependence?'
}
cols <- colors_vilaweb()[c(5,4)]
cols <- c(cols[1], 'grey', cols[2])
ggplot(data = pd,
aes(x = neixer,
y = p,
fill = indepe,
group = indepe)) +
geom_bar(stat = 'identity',
position = position_stack()) +
theme_vilaweb() +
the_labs +
scale_fill_manual(name = legend_title,
values = cols) +
geom_text(aes(label = paste0(round(p, digits = 1), '%')),
position = position_stack(),
vjust = 1,
alpha = 0.6) +
theme(legend.position = 'right') +
theme(plot.caption = element_text(hjust = 0))
}
age_plot <- function(ca = FALSE){
if(ca){
no_know <- 'NS/NC'
catalunya <- 'Catalunya'
ccaa <- 'Espanya'
mon <- 'Estranger'
} else {
no_know <- 'Not sure\nNo answer'
catalunya <- 'Catalonia'
ccaa <- 'Spain'
mon <- 'Abroad'
}
pd <- new_ceo %>%
mutate(neixer = `Em podria dir on va néixer?`,
indepe = `Vol que Catalunya esdevingui un Estat independent?`,
edat = Edat) %>%
mutate(indepe = as.character(indepe),
neixer = as.character(neixer)) %>%
filter(!is.na(neixer),
!is.na(indepe),
!is.na(edat),
!neixer %in% c('No ho sap', 'No contesta')) %>%
mutate(indepe = ifelse(indepe %in% c('No ho sap',
'No contesta'),
no_know,
indepe),
neixer = ifelse(neixer == 'Catalunya', catalunya,
ifelse(neixer %in% c('Altres comunitats autònomes'), ccaa,
mon))) %>%
group_by(edat, indepe) %>%
tally %>%
ungroup %>%
group_by(edat) %>%
mutate(p = n / sum(n) * 100) %>%
ungroup
if(ca){
the_labs <- labs(x = 'Edat',
y = 'Percentatge',
title = 'Independentisme per edat',
caption = paste0('Dades: Baròmetre d\'Opinió Política del Centre d\'Estudios d\'Opinió.\nMostra: ',
numberfy(sum(pd$n)), ' residents de Catalunya amb ciutadania espanyola.\nGràfic: Joe Brew | @joethebrew | www.vilaweb.cat'))
legend_title <- 'A favor\nde la\nindependència?'
point_name <- 'Mida de\nmostra'
} else {
the_labs <- labs(x = 'Age',
y = 'Percentage',
title = 'Support for independence by age',
caption = paste0('Data: Baròmetre d\'Opinió Política of the Centre d\'Estudios d\'Opinió.\nMostra: ',
numberfy(sum(pd$n)), ' residents of Catalonia with Spanish citizenship.\nChart: Joe Brew | @joethebrew | www.vilaweb.cat'))
legend_title <- 'In favor\nof\nindependence?'
point_name <- 'Sample\nsize'
}
cols <- colors_vilaweb()[c(5,4)]
cols <- c(cols[1], 'grey', cols[2])
ggplot(data = pd,
aes(x = edat,
y = p,
color = indepe)) +
geom_point(alpha = 0.6,
aes(size = n)) +
theme_vilaweb() +
the_labs +
scale_color_manual(name = legend_title,
values = cols) +
theme(legend.position = 'right') +
theme(plot.caption = element_text(hjust = 0)) +
geom_smooth(se = FALSE) +
xlim(18, 90) +
scale_size_continuous(name = point_name, breaks = c(25, 100, 200))
}
age_birth_plot <- function(ca = FALSE){
if(ca){
no_know <- 'NS/NC'
catalunya <- 'Catalunya'
ccaa <- 'Espanya'
mon <- 'Estranger'
} else {
no_know <- 'Not sure\nNo answer'
catalunya <- 'Catalonia'
ccaa <- 'Spain'
mon <- 'Abroad'
}
pd <- new_ceo %>%
mutate(neixer = `Em podria dir on va néixer?`,
indepe = `Vol que Catalunya esdevingui un Estat independent?`,
edat = Edat) %>%
mutate(indepe = as.character(indepe),
neixer = as.character(neixer)) %>%
filter(!is.na(neixer),
!is.na(indepe),
!is.na(edat),
!neixer %in% c('No ho sap', 'No contesta')) %>%
mutate(indepe = ifelse(indepe %in% c('No ho sap',
'No contesta'),
no_know,
indepe),
neixer = ifelse(neixer == 'Catalunya', catalunya,
ifelse(neixer %in% c('Altres comunitats autònomes'), ccaa,
mon))) %>%
group_by(edat, indepe, neixer) %>%
tally %>%
ungroup %>%
group_by(edat, neixer) %>%
mutate(p = n / sum(n) * 100) %>%
ungroup
if(ca){
the_labs <- labs(x = 'Edat',
y = 'Percentatge',
title = 'Independentisme per edat',
caption = paste0('Dades: Baròmetre d\'Opinió Política del Centre d\'Estudios d\'Opinió.\nMostra: ',
numberfy(sum(pd$n)), ' residents de Catalunya amb ciutadania espanyola.\nGràfic: Joe Brew | @joethebrew | www.vilaweb.cat'))
legend_title <- 'A favor\nde la\nindependència?'
point_name <- 'Mida de\nmostra'
} else {
the_labs <- labs(x = 'Age',
y = 'Percentage',
title = 'Support for independence by age',
caption = paste0('Data: Baròmetre d\'Opinió Política of the Centre d\'Estudios d\'Opinió.\nMostra: ',
numberfy(sum(pd$n)), ' residents of Catalonia with Spanish citizenship.\nChart: Joe Brew | @joethebrew | www.vilaweb.cat'))
legend_title <- 'In favor\nof\nindependence?'
point_name <- 'Sample\nsize'
}
cols <- colors_vilaweb()[c(5,4)]
cols <- c(cols[1], 'grey', cols[2])
ggplot(data = pd %>% filter(!neixer %in% c('Abroad', 'Estranger')),
aes(x = edat,
y = p,
color = indepe)) +
geom_point(alpha = 0.6,
aes(size = n)) +
theme_vilaweb() +
the_labs +
scale_color_manual(name = legend_title,
values = cols) +
theme(legend.position = 'right') +
theme(plot.caption = element_text(hjust = 0)) +
geom_smooth(se = FALSE) +
xlim(18, 90) +
scale_size_continuous(name = point_name, breaks = c(25, 100, 200)) +
facet_wrap(~neixer, ncol = 2)
}
current_plot <- function(ca = FALSE){
pd <- current %>%
group_by(key) %>%
mutate(p = value / sum(value) * 100)
if(ca){
pd$key <- factor(pd$key,
levels = c('Catalunya',
'Estranger',
'Espanya'))
the_labs <- labs(y = 'Percentatge',
x = 'Edat',
title = 'Catalans: distribució d\'edat per lloc de naixement',
caption = 'Dades de l\'Institut d\'Estadística de Catalunya. Gràfic de Joe Brew | @joethebrew | www.vilaweb.cat')
} else {
pd$key <- factor(pd$key,
levels = c('Catalunya',
'Estranger',
'Espanya'),
labels = c('Catalonia',
'Abroad',
'Spain'))
the_labs <- labs(y = 'Percentage',
x = 'Age',
title = 'Catalans: age distribution by place of birth',
caption = 'Data from the Institut d\'Estadística de Catalunya. Chart by Joe Brew | @joethebrew | www.vilaweb.cat')
}
ggplot(data = pd,
aes(x = age,
y = p)) +
geom_bar(stat = 'identity',
width = 1,
color = 'white') +
facet_wrap(~key) +
coord_flip() +
theme_vilaweb() +
the_labs
}
# Create model of independentism
predictions <-
new_ceo %>% mutate(neixer = `Em podria dir on va néixer?`,
indepe = `Vol que Catalunya esdevingui un Estat independent?`,
edat = as.numeric(Edat)) %>%
filter(`Any de realització del baròmetre` %in% c(2018, 2019)) %>%
# # group together all the very old (not doing for now, just an option to deal with small sample size)
# mutate(edat = ifelse(edat >= 0, 0, edat)) %>%
mutate(indepe = as.character(indepe),
neixer = as.character(neixer)) %>%
filter(!is.na(neixer),
!is.na(indepe),
!neixer %in% c('No ho sap', 'No contesta')) %>%
mutate(indepe = ifelse(indepe %in% c('No ho sap',
'No contesta'),
'NS/NC',
indepe),
neixer = ifelse(neixer == 'Catalunya', 'Catalunya',
ifelse(neixer %in% c('Altres comunitats autònomes'), 'Espanya',
'Estranger'))) %>%
group_by(neixer, indepe, edat) %>%
tally %>%
ungroup %>%
group_by(edat, neixer) %>%
mutate(p = n / sum(n) * 100) %>%
ungroup %>%
group_by(neixer, edat) %>%
summarise(n_si = sum(n[indepe == 'Sí']),
n_no = sum(n[indepe == 'No']),
n_ns = sum(n[indepe == 'NS/NC']),
p_si = sum(n[indepe == 'Sí']) / sum(n),
p_no = sum(n[indepe == 'No']) / sum(n),
p_ns = sum(n[indepe == 'NS/NC']) / sum(n),
p_si_ref = sum(n[indepe == 'Sí']) / sum(n[indepe %in% c('Sí', 'No')]),
p_no_ref = sum(n[indepe == 'No']) / sum(n[indepe %in% c('Sí', 'No')])) %>%
ungroup %>%
dplyr::select(neixer, edat, p_si, p_no, p_ns, p_si_ref, p_no_ref)
# Test to see if current matches up with survey results
test <- current %>%
dplyr::rename(neixer = key,
edat = age) %>%
# mutate(edat = ifelse(edat >= 0, 0, edat)) %>%
left_join(predictions) %>%
# remove those for whom predictions could not be made
filter(!is.na(p_si)) %>%
# calculate number of people
mutate(indepes = value * p_si,
unionistes = value * p_no,
undecided = value * p_ns)
# x <- test %>%
# summarise(indepes = sum(indepes),
# unionistes = sum(unionistes),
# undecided = sum(undecided))
# x
# as.numeric(x) / sum(as.numeric(x))
# Now, using this, we can create projections
projections <- future %>%
dplyr::rename(neixer = key,
edat = age) %>%
# mutate(edat = ifelse(edat >= 0, 0, edat)) %>%
left_join(predictions) %>%
# remove those for whom predictions could not be made
filter(!is.na(p_si)) %>%
# calculate number of people
mutate(indepes = value * p_si,
unionistes = value * p_no,
undecided = value * p_ns) %>%
group_by(year) %>%
summarise(indepes = sum(indepes),
p_indepes = sum(indepes) / (sum(indepes) + sum(unionistes) + sum(undecided)) * 100,
unionistes = sum(unionistes),
p_unionistes = sum(unionistes) / (sum(unionistes) + sum(indepes) + sum(undecided)) * 100,
undecided = sum(undecided),
p_undecided = sum(undecided) / (sum(unionistes) + sum(indepes) + sum(undecided)) * 100,) %>%
mutate(p_si_ref = p_indepes / (100 - p_undecided),
p_no_ref = p_unionistes / (100 - p_undecided))
projection_plot <- function(ca = FALSE){
if(ca){
the_labs <- labs(x = 'Any',
y = '%',
title = 'Independentisme segons la projecció demogràfica',
subtitle = 'Efecte de la mortalitat, població de 2019',
caption = "Model basat en enquestes de 2018-2019:CEO. Dades poblacionals de 2019: IDES.\nEl percentage reflecteix un hipotètic referèndum (traient els 'no sap/no contesta'), edat=18+.\nJoe Brew | @joethebrew | www.vilaweb.cat")
choices <- c('En contra', 'A favor')
legend_title <- 'Opinió sobre la independència'
} else {
the_labs <- labs(x = 'Year',
y = '%',
title = 'Support for independence per demographic projection',
subtitle = 'Effect of mortality only, year 2019',
caption = "Model based on survey data from 2018-2019: CEO. Population data from 2019: IDES.\nThe percentage reflects a hypothetical referendum (without the 'unknown' responses), age=18+.\nJoe Brew | @joethebrew | www.vilaweb.cat")
choices <- c('Opposed', 'In favor')
legend_title <- c('View on independence')
}
ggplot(data = projections %>% mutate(p_si_ref = p_si_ref * 100,
p_no_ref = p_no_ref * 100) %>%
dplyr::select(year, p_si_ref, p_no_ref) %>%
gather(key, value, p_si_ref:p_no_ref) %>%
mutate(key = ifelse(key == 'p_no_ref',
choices[1],
choices[2])),
aes(x = year,
y = value)) +
geom_line(aes(color = key),
size = 2) +
xlim(2018, 2080) +
theme_vilaweb() +
ylim(0, 100) +
geom_hline(yintercept = 50, lty = 2) +
the_labs +
scale_color_manual(name = legend_title,
values = colors_vilaweb()[3:4])
}
# Immigracions procedents de la resta d'Espanya. Per sexe, grups d'edat i lloc de naixement.
# https://www.idescat.cat/pub/?id=mm&n=6383
#t6383.csv
ready <- function(number = 6383){
ides_urls <- paste0('https://www.idescat.cat/pub/?id=mm&n=',
number,'&t=', 2005:2017,'00&f=csv')
yearsy <- 2005:2017
ides_list <- list()
for(j in 1:length(ides_urls)){
# Sys.sleep(2)
message(j)
ides <- read_csv(ides_urls[j], skip = 5)
# Keep only the total and below
ides <- ides[(which(ides$X1 == 'Total') +1): nrow(ides),]
ides <- ides %>%
gather(age, value, `De 0 a 14 anys`:`De 60 anys i més`) %>%
dplyr::rename(key = X1) %>%
mutate(key = gsub('nascuts a ', '', key)) %>%
mutate(key = gsub("la resta d'", "", key)) %>%
mutate(key = gsub("l'e", "E", key, fixed = TRUE))
# Fix age
ides$age <- gsub('anys i més', 'a 99 anys', ides$age)
split_age <- strsplit(ides$age, ' ')
ides$min_age <- as.numeric(unlist(lapply(split_age, function(x){x[2]})))
ides$max_age <- as.numeric(unlist(lapply(split_age, function(x){x[4]})))
ides$Total <- NULL
# Expand
migrations_list <- list()
for(i in 1:nrow(ides)){
this_row <- ides[i,]
out <- tibble(age = this_row$min_age:this_row$max_age,
key = this_row$key)
n <- nrow(out)
out$value <- this_row$value / n
migrations_list[[i]] <- out
}
ides <- bind_rows(migrations_list)
ides_list[[j]] <- ides %>% mutate(year = yearsy[j])
}
ides <- bind_rows(ides_list) %>%
group_by(age, key, year) %>%
summarise(value = mean(value))
return(ides)
}
# Get inmig #6383
# exmig 6384
# wexmig 6386
# winbmig 6389
if(!'migdata.RData' %in% dir()){
inmig <- ready(6383)
exmig <- ready(6384)
wexmig <- ready(6386)
winmig <- ready(6389)
save(inmig, exmig, wexmig, winmig,
file = 'migdata.RData')
} else {
load('migdata.RData')
}
transformy <- function(x, since_year = 2005){
x %>% filter(year >= since_year) %>%
group_by(age, key) %>%
summarise(value = mean(value))
}
inmig <- transformy(inmig)
exmig <- transformy(exmig)
wexmig <- transformy(wexmig)
winmig <- transformy(winmig)
# Copmbine all migrations
migrations <-
bind_rows(
inmig %>% mutate(destination = 'Catalunya'),
winmig %>% mutate(destination = 'Catalunya'),
exmig %>% mutate(destination = 'World'),
wexmig %>% mutate(destination = 'World')
)
# # Add births
births <- tibble(age = 0,
key = c('Catalunya', 'Espanya', 'Estranger'),
value = c(66803,
896, # 896 + 241 - 135,
3051))# + 938 - 945))
# changes <- bind_rows(migrations, births)
changes <- migrations
changes <- changes %>%
# dplyr::rename(place_of_birth = key) %>%
mutate(action = ifelse(destination == 'Catalunya',
'Coming',
'Leaving')) %>%
dplyr::select(-destination) %>%
group_by(key, age) %>%
summarise(leaving = sum(value[action == 'Leaving']),
coming = sum(value[action == 'Coming'])) %>%
ungroup
# Make changes a percentage, rather than absolute number
changes <- left_join(changes, current) %>% dplyr::select(-year)
changes <- changes %>%
# mutate(
# leaving = leaving / value,
# coming = coming / value) %>%
dplyr::select(-value)
# Create an adjusted future, to account for changes (migrations and births)
# Estimate what would happen, demographically, with Catalonia's current population
current <- lloc_long %>% filter(year == 2018)
the_list <- list()
years <- 2019:2070
for(i in 1:length(years)){
the_year <- years[i]
# Get last year
if(i == 1){
last_year <- current
} else {
last_year <- the_list[[i-1]]
}
# Join the probability of death
this_year <- last_year %>%
dplyr::select(-year) %>%
left_join(df %>% dplyr::select(age, nqx)) %>%
# replace the value with the mortality-adjusted value
mutate(value = value - (value * nqx)) %>%
mutate(year = the_year) %>%
dplyr::select(-nqx) %>%
mutate(age = age + 1) %>%
filter(age <= 100)
# Add a zero year
this_year <- this_year %>%
bind_rows(births %>% mutate(year = this_year$year[1]))
this_year <- this_year %>% arrange(age)
# Add the changes
this_year <- left_join(this_year, changes)
this_year <- this_year %>%
mutate(value = value + coming - leaving) %>%
# mutate(value = value - (leaving * value) + (coming * value)) %>%
dplyr::select(-coming, -leaving)
the_list[[i]] <- this_year
}
future_adj <- bind_rows(the_list)
# Use the adjusted futures to make adjusted projections
# Now, using this, we can create projections
projections_adj <- future_adj %>%
dplyr::rename(neixer = key,
edat = age) %>%
# mutate(edat = ifelse(edat >= 0, 0, edat)) %>%
left_join(predictions) %>%
# remove those for whom predictions could not be made
filter(!is.na(p_si)) %>%
# calculate number of people
mutate(indepes = value * p_si,
unionistes = value * p_no,
undecided = value * p_ns) %>%
group_by(year) %>%
summarise(indepes = sum(indepes),
p_indepes = sum(indepes) / (sum(indepes) + sum(unionistes) + sum(undecided)) * 100,
unionistes = sum(unionistes),
p_unionistes = sum(unionistes) / (sum(unionistes) + sum(indepes) + sum(undecided)) * 100,
undecided = sum(undecided),
p_undecided = sum(undecided) / (sum(unionistes) + sum(indepes) + sum(undecided)) * 100,) %>%
mutate(p_si_ref = p_indepes / (100 - p_undecided),
p_no_ref = p_unionistes / (100 - p_undecided))
# Make a version with no foreigners
projections_adj_no_estranger <- future_adj %>%
dplyr::rename(neixer = key,
edat = age) %>%
filter(!neixer %in% c('Estranger', 'Abroad')) %>%
# mutate(edat = ifelse(edat >= 0, 0, edat)) %>%
left_join(predictions) %>%
# remove those for whom predictions could not be made
filter(!is.na(p_si)) %>%
# calculate number of people
mutate(indepes = value * p_si,
unionistes = value * p_no,
undecided = value * p_ns) %>%
group_by(year) %>%
summarise(indepes = sum(indepes),
p_indepes = sum(indepes) / (sum(indepes) + sum(unionistes) + sum(undecided)) * 100,
unionistes = sum(unionistes),
p_unionistes = sum(unionistes) / (sum(unionistes) + sum(indepes) + sum(undecided)) * 100,
undecided = sum(undecided),
p_undecided = sum(undecided) / (sum(unionistes) + sum(indepes) + sum(undecided)) * 100,) %>%
mutate(p_si_ref = p_indepes / (100 - p_undecided),
p_no_ref = p_unionistes / (100 - p_undecided))
make_animation_adj <- function(ca = FALSE){
if(ca){
legend_title <- 'LLoc de naixement'
pd <- future_adj
the_labs <- labs(title = 'Any: {frame_time}',
x = 'Edat',
y = 'Persones',
subtitle = 'Projecció de la composició de la població catalana\n(assumint naixement/migració constants)',
caption = 'Dades de llocs de naixement i mortalitat de l\'Institut d\'Estadística de Catalunya.\nProjeccions i gràfic de Joe Brew | @joethebrew | www.vilaweb.cat')
pd$key <- factor(pd$key,
levels = rev(c('Catalunya',
'Espanya',
'Estranger')))
} else {
legend_title <- 'Place of birth'
pd <- future_adj %>%
mutate(key = ifelse(key == 'Catalunya', 'Catalonia',
ifelse(key == 'Espanya', 'Spain',
ifelse(key == 'Estranger', 'Abroad', key))))
pd$key <- factor(pd$key,
levels = rev(c('Catalonia',
'Spain',
'Abroad')))
the_labs <- labs(title = 'Year: {frame_time}',
x = 'Age',
y = 'People',
subtitle = 'Projection of the composition of Catalonia\'s population\n(assuming constant birth and migration rates)',
caption = 'Data on places of birth and mortality from the Institut d\'Estadística de Catalunya.\nProjections and chart by Joe Brew | @joethebrew | www.vilaweb.cat')
}
cols <- rev(c(colors_vilaweb()[c(5,3)], 'grey'))
label_df <- pd %>%
group_by(key, year) %>%
summarise(value = sum(value, na.rm = TRUE)) %>%
group_by(year) %>%
mutate(p = value / sum(value) * 100) %>%
ungroup %>%
filter(!is.na(p)) %>%
group_by(year) %>%
summarise(value = paste0(legend_title, '\n',paste0(key, ': ', round(p, digits = 1), '%', collapse = '\n'), collapse = ''))
ggplot() +
geom_label(data = label_df,
aes(x = 90,
y = 110000,
label = value)) +
geom_bar(stat = 'identity', width = 1,
data = pd,
aes(x = age,
y = value,
fill = key)) +
coord_flip() +
scale_fill_manual(name = legend_title,
values = cols) +
theme_vilaweb() +
transition_time(year) +
the_labs
}
make_plot_adj <- function(ca = FALSE, remove_estranger = F){
if(ca){
legend_title <- 'LLoc de naixement'
pd <- future_adj
the_labs <- labs(x = 'Any',
y = 'Percentatge',
title = 'Projecció: lloc de naixement dels catalans',
subtitle = '(assumint migració/naixement constants)',
caption = 'Dades de llocs de naixement i mortalitat de l\'Institut d\'Estadística de Catalunya.\nProjeccions i gràfic de Joe Brew | @joethebrew | www.vilaweb.cat')
pd$key <- factor(pd$key,
levels = rev(c('Catalunya',
'Espanya',
'Estranger')))
} else {
legend_title <- 'Place of birth'
pd <- future_adj
pd <- pd %>%
mutate(key = ifelse(key == 'Catalunya', 'Catalonia',
ifelse(key == 'Espanya', 'Spain',
ifelse(key == 'Estranger', 'Abroad', key))))
pd$key <- factor(pd$key,
levels = rev(c('Catalonia',
'Spain',
'Abroad')))
the_labs <- labs(title = 'Projection: place of birth of Catalans',
x = 'Year',
y = 'Percentage',
subtitle = '(assuming constant birth/migration rates)',
caption = 'Data on places of birth and mortality from the Institut d\'Estadística de Catalunya.\nProjections and chart by Joe Brew | @joethebrew | www.vilaweb.cat')
}
cols <- rev(c(colors_vilaweb()[c(5,3)], 'grey'))
# Group pd
pd <- pd %>%
group_by(key, year) %>%
summarise(value = sum(value, na.rm = TRUE)) %>%
ungroup %>%
group_by(year) %>%
mutate(p = value / sum(value) * 100) %>%
ungroup
ggplot() +
geom_line(data = pd,
aes(x = year,
y = p,
color = key)) +
scale_color_manual(name = legend_title,
values = cols) +
theme_vilaweb() +
the_labs
}
projection_plot_adj <- function(ca = FALSE, short = FALSE,
remove_estranger = FALSE){
if(ca){
the_labs <- labs(x = 'Any',
y = '%',
title = 'Independentisme segons la projecció demogràfica',
subtitle = 'Efecte de la mortalitat/migració/fertilitat',
caption = "Model basat en enquestes de 2018-2019:CEO. Dades poblacionals de 2019: IDES.\nEl percentage reflecteix un hipotètic referèndum (traient els 'no sap/no contesta'), edat=18+.\nJoe Brew | @joethebrew | www.vilaweb.cat")
choices <- c('En contra', 'A favor')
legend_title <- 'Opinió sobre la independència'
} else {
the_labs <- labs(x = 'Year',
y = '%',
title = 'Support for independence per demographic projection',
subtitle = 'Effect of mortality/migration/fertility',
caption = "Model based on survey data from 2018-2019: CEO. Population data from 2019: IDES.\nThe percentage reflects a hypothetical referendum (without the 'unknown' responses), age=18+.\nJoe Brew | @joethebrew | www.vilaweb.cat")
choices <- c('Opposed', 'In favor')
legend_title <- c('View on independence')
}
if(remove_estranger){
pd <- projections_adj_no_estranger
} else {
pd <- projections_adj
}
g <- ggplot(data = pd %>% mutate(p_si_ref = p_si_ref * 100,
p_no_ref = p_no_ref * 100) %>%
dplyr::select(year, p_si_ref, p_no_ref) %>%
gather(key, value, p_si_ref:p_no_ref) %>%
mutate(key = ifelse(key == 'p_no_ref',
choices[1],
choices[2])),
aes(x = year,
y = value))
if(short){
g <- g + geom_line(aes(color = key),
size = 1) +
xlim(2018, 2035) +
ylim(46, 54)
} else {
g <- g + geom_line(aes(color = key),
size = 2) +
xlim(2018, 2080) +
geom_line(aes(color = key),
size = 2) +
ylim(0, 100)
}
g <- g +
theme_vilaweb() +
geom_hline(yintercept = 50, lty = 2) +
the_labs +
scale_color_manual(name = legend_title,
values = colors_vilaweb()[3:4])
return(g)
}
# # calculations for Ramir
# # independentisme by place of birth by year
# no_know <- 'NS/NC'
# catalunya <- 'Catalunya'
# ccaa <- 'Espanya'
# mon <- 'Estranger'
# pd <- new_ceo %>%
# mutate(neixer = `Em podria dir on va néixer?`,
# indepe = `Vol que Catalunya esdevingui un Estat independent?`,
# edat = Edat,
# any = `Any de realització del baròmetre`) %>%
# mutate(indepe = as.character(indepe),
# neixer = as.character(neixer)) %>%
# filter(!is.na(neixer),
# !is.na(indepe),
# !neixer %in% c('No ho sap', 'No contesta')) %>%
# mutate(indepe = ifelse(indepe %in% c('No ho sap',
# 'No contesta'),
# no_know,
# indepe),
# neixer = ifelse(neixer == 'Catalunya', catalunya,
# ifelse(neixer %in% c('Altres comunitats autònomes'), ccaa,
# mon))) %>%
# filter(indepe != 'NS/NC') %>%
# group_by(neixer, indepe, any) %>%
# tally %>%
# ungroup %>%
# group_by(neixer, any) %>%
# mutate(p = n / sum(n) * 100) %>%
# ungroup %>%
# filter(neixer %in% c('Catalunya', 'Espanya'))
#
# ggplot(data = pd,
# aes(x = any,
# y = p,
# color = indepe)) +
# geom_line() +
# facet_wrap(~neixer)
# write_csv(pd, '~/Desktop/ramir.csv')
avis_plot <- function(ca = FALSE, pares = FALSE){
if(ca){
no_know <- 'NS/NC'
catalunya <- 'Catalunya'
ccaa <- 'Espanya'
mon <- 'Estranger'
} else {
no_know <- 'Not sure\nNo answer'
catalunya <- 'Catalonia'
ccaa <- 'Spain'
mon <- 'Abroad'
}
if(pares){
pd <- new_ceo %>%
# neixer = `Em podria dir on va néixer?`,
mutate(mare = `Em podria dir el lloc de naixement de la seva mare?`,
pare = `Em podria dir el lloc de naixement del seu pare?`) %>%
filter(!mare %in% c('No ho sap', 'No contesta'),
!pare %in% c('No ho sap', 'No contesta')) %>%
mutate(mare = as.character(mare),
pare = as.character(pare)) %>%
mutate(mare = ifelse(mare == 'Catalunya', '1', '0'),
pare = ifelse(pare == 'Catalunya', '1', '0')) %>%
mutate(mare = as.numeric(mare),
pare = as.numeric(pare)) %>%
mutate(avis = mare + pare,
indepe = `Vol que Catalunya esdevingui un Estat independent?`,
edat = Edat)
} else {
pd <- new_ceo %>%
mutate(avis = `Quants dels seus avis/àvies van néixer a Catalunya?`,
# neixer = `Em podria dir on va néixer?`,
indepe = `Vol que Catalunya esdevingui un Estat independent?`,
edat = Edat)
}
pd <- pd %>%
mutate(indepe = as.character(indepe)) %>%#,
# neixer = as.character(neixer)) %>%
filter(#!is.na(neixer),
!is.na(indepe),
!is.na(avis)) %>%
# !neixer %in% c('No ho sap', 'No contesta')) %>%
mutate(indepe = ifelse(indepe %in% c('No ho sap',
'No contesta'),
no_know,
indepe)) %>%#,
# neixer = ifelse(neixer == 'Catalunya', catalunya,
# ifelse(neixer %in% c('Altres comunitats autònomes'), ccaa,
# mon))) %>%
mutate(avis = as.character(avis))
if(!pares){
pd <- pd %>%
mutate(avis = ifelse(avis == 'Cap', '0',
ifelse(avis == 'Un', '1',
ifelse(avis == 'Dos', '2',
ifelse(avis == 'Tres', '3',
ifelse(avis == 'Quatre', '4', NA)))))) %>%
mutate(avis = as.numeric(avis))
}
pd <- pd %>%
filter(!is.na(avis)) %>%
group_by(avis, indepe) %>%
tally %>%
ungroup %>%
group_by(avis) %>%
mutate(p = n / sum(n) * 100) %>%
ungroup
if(ca){
if(pares){
x <- 'Quants pares van néixer a Catalunya'
tt <- 'Independentisme per lloc de naixement dels pares'
} else {
x <- 'Quants avis van néixer a Catalunya'
tt <- 'Independentisme per lloc de naixement dels avis'
}
the_labs <- labs(x = x,
y = 'Percentatge',
title = tt,
caption = paste0('Dades: Baròmetre d\'Opinió Política del Centre d\'Estudios d\'Opinió.\nMostra: ',
numberfy(sum(pd$n)), ' residents de Catalunya amb ciutadania espanyola.\nGràfic: Joe Brew | @joethebrew | www.vilaweb.cat'))
legend_title <- 'A favor\nde la\nindependència?'
} else {
if(pares){
x <- 'How many parents born in Catalonia'
tt <- 'Support for independence by\nplace of birth of parents'
} else {
x <- 'How many grandparents born in Catalonia'
tt <- 'Support for independence by\nplace of birth of grandparents'
}
the_labs <- labs(x = x,
y = 'Percentage',
title = tt,
caption = paste0('Data: Baròmetre d\'Opinió Política of the Centre d\'Estudios d\'Opinió.\nMostra: ',
numberfy(sum(pd$n)), ' residents of Catalonia with Spanish citizenship.\nChart: Joe Brew | @joethebrew | www.vilaweb.cat'))
legend_title <- 'In favor\nof\nindependence?'
}
cols <- colors_vilaweb()[c(5,4)]
cols <- c(cols[1], 'grey', cols[2])
ggplot(data = pd,
aes(x = avis,
y = p,
fill = indepe,
group = indepe)) +
geom_bar(stat = 'identity',
position = position_stack()) +
theme_vilaweb() +
the_labs +
scale_fill_manual(name = legend_title,
values = cols) +
geom_text(aes(label = paste0(round(p, digits = 1), '%')),
position = position_stack(),
vjust = 1,
alpha = 0.6) +
theme(legend.position = 'right') +
theme(plot.caption = element_text(hjust = 0))
}
pares_birth_plot <- function(ca = FALSE, pares = FALSE){
if(ca){
no_know <- 'NS/NC'
catalunya <- 'Catalunya'
ccaa <- 'Espanya'
mon <- 'Estranger'
} else {
no_know <- 'Not sure\nNo answer'
catalunya <- 'Catalonia'
ccaa <- 'Spain'
mon <- 'Abroad'
}
if(pares){
pd <- new_ceo %>%
mutate(neixer = `Em podria dir on va néixer?`) %>%
mutate(mare = `Em podria dir el lloc de naixement de la seva mare?`,
pare = `Em podria dir el lloc de naixement del seu pare?`) %>%
filter(!mare %in% c('No ho sap', 'No contesta'),
!pare %in% c('No ho sap', 'No contesta')) %>%
mutate(mare = as.character(mare),
pare = as.character(pare)) %>%
mutate(mare = ifelse(mare == 'Catalunya', '1', '0'),
pare = ifelse(pare == 'Catalunya', '1', '0')) %>%
mutate(mare = as.numeric(mare),
pare = as.numeric(pare)) %>%
mutate(avis = mare + pare,
indepe = `Vol que Catalunya esdevingui un Estat independent?`,
edat = Edat)
} else {
pd <- new_ceo %>%
mutate(avis = `Quants dels seus avis/àvies van néixer a Catalunya?`,
neixer = `Em podria dir on va néixer?`,
indepe = `Vol que Catalunya esdevingui un Estat independent?`,
edat = Edat)
}
pd <- pd %>%
mutate(indepe = as.character(indepe)) %>%
mutate(neixer = as.character(neixer)) %>%
filter(!is.na(neixer),
!is.na(indepe),
!is.na(avis)) %>%
filter(!neixer %in% c('No ho sap', 'No contesta')) %>%
mutate(indepe = ifelse(indepe %in% c('No ho sap',
'No contesta'),
no_know,
indepe),
neixer = ifelse(neixer == 'Catalunya', catalunya,
ifelse(neixer %in% c('Altres comunitats autònomes'), ccaa,
mon))) %>%
mutate(avis = as.character(avis))
if(!pares){
pd <- pd %>%
mutate(avis = ifelse(avis == 'Cap', '0',
ifelse(avis == 'Un', '1',
ifelse(avis == 'Dos', '2',
ifelse(avis == 'Tres', '3',
ifelse(avis == 'Quatre', '4', NA)))))) %>%
mutate(avis = as.numeric(avis))
}
pd <- pd %>%
filter(!is.na(avis)) %>%
group_by(avis, neixer, indepe) %>%
tally %>%
ungroup %>%
group_by(avis, neixer) %>%
mutate(p = n / sum(n) * 100) %>%
ungroup
if(ca){
if(pares){
x <- 'Quants pares van néixer a Catalunya'
tt <- 'Independentisme per lloc de naixement dels pares'
} else {
x <- 'Quants avis van néixer a Catalunya'
tt <- 'Independentisme per lloc de naixement dels avis'
}
the_labs <- labs(x = x,
y = 'Percentatge',
title = tt,
caption = paste0('Dades: Baròmetre d\'Opinió Política del Centre d\'Estudios d\'Opinió.\nMostra: ',
numberfy(sum(pd$n)), ' residents de Catalunya amb ciutadania espanyola.\nGràfic: Joe Brew | @joethebrew | www.vilaweb.cat'))
legend_title <- 'A favor\nde la\nindependència?'
} else {
if(pares){
x <- 'How many parents born in Catalonia'
tt <- 'Support for independence by\nplace of birth of parents'
} else {
x <- 'How many grandparents born in Catalonia'
tt <- 'Support for independence by\nplace of birth of grandparents'
}
the_labs <- labs(x = x,
y = 'Percentage',
title = tt,
caption = paste0('Data: Baròmetre d\'Opinió Política of the Centre d\'Estudios d\'Opinió.\nMostra: ',
numberfy(sum(pd$n)), ' residents of Catalonia with Spanish citizenship.\nChart: Joe Brew | @joethebrew | www.vilaweb.cat'))
legend_title <- 'In favor\nof\nindependence?'
}
cols <- colors_vilaweb()[c(5,4)]
cols <- c(cols[1], 'grey', cols[2])
ggplot(data = pd %>%
mutate(neixer = paste0('Persona nascuda a\n', neixer)),
aes(x = avis,
y = p,
fill = indepe,
group = indepe)) +
geom_bar(stat = 'identity',
position = position_stack()) +
theme_vilaweb() +
the_labs +
scale_fill_manual(name = legend_title,
values = cols) +
geom_text(aes(label = paste0(round(p, digits = 1), '%')),
position = position_stack(),
vjust = 1,
size = 2,
alpha = 0.6) +
theme(legend.position = 'right') +
theme(plot.caption = element_text(hjust = 0),
strip.text = element_text(size = 10)) +
facet_wrap(~neixer)
}
# Read in a old ceo
old_ceo <- vilaweb::old_ceo
ramir_plot <- function(ca = FALSE){
if(ca){
no_know <- 'NS/NC'
catalunya <- 'Catalunya'
ccaa <- 'Espanya'
mon <- 'Estranger'
} else {
no_know <- 'Not sure\nNo answer'
catalunya <- 'Catalonia'
ccaa <- 'Spain'
mon <- 'Abroad'
}
old_pd <- old_ceo %>%
mutate(indepe = `P39. I més concretament, si demà es fes un referèndum per decidir la independència de Catalunya, vostè què faria?`,
neixer = `C100. Lloc de naixement de l'enquestat/ada`) %>%
mutate(indepe = as.character(indepe)) %>%
mutate(indepe = ifelse(grepl('a favor', indepe), 'Sí',
ifelse(grepl('en contra', indepe), 'No',
ifelse(grepl('no aniria|ho sap|contesta', indepe), 'NS/NC', NA)))) %>%
mutate(indepe = ifelse(indepe %in% c('No ho sap',
'No contesta',
'NS/NC'),
no_know,
indepe),
neixer = ifelse(neixer == 'Catalunya', catalunya,
ifelse(neixer %in% c('Altres comunitats autònomes'), ccaa,
mon))) %>%
mutate(any = `Any de realització del baròmetre`) %>%
filter(!is.na(neixer),
!is.na(indepe),
!neixer %in% c('No ho sap', 'No contesta')) %>%
mutate(any = as.numeric(as.character(any)))
new_pd <- new_ceo %>%
mutate(neixer = `Em podria dir on va néixer?`,
indepe = `Vol que Catalunya esdevingui un Estat independent?`,
edat = Edat) %>%
mutate(indepe = as.character(indepe),
neixer = as.character(neixer),
any = `Any de realització del baròmetre`) %>%
filter(!is.na(neixer),
!is.na(indepe),
!neixer %in% c('No ho sap', 'No contesta')) %>%
mutate(indepe = ifelse(indepe %in% c('No ho sap',
'No contesta'),
no_know,
indepe),
neixer = ifelse(neixer == 'Catalunya', catalunya,
ifelse(neixer %in% c('Altres comunitats autònomes'), ccaa,
mon))) %>%
mutate(any = as.numeric(as.character(any))) %>%
dplyr::select(neixer, indepe, any)
pd <- new_pd %>%
bind_rows(
old_pd %>% dplyr::select(neixer, indepe, any)
) %>%
group_by(neixer, indepe, any) %>%
tally %>%
ungroup %>%
group_by(neixer, any) %>%
mutate(p = n / sum(n) * 100) %>%
ungroup
if(ca){
born_in <- 'Nascuts a '
the_labs <- labs(x = 'Lloc de naixement',
y = 'Percentatge',
title = 'Independentisme per lloc de naixement',
caption = paste0('Dades: Baròmetre d\'Opinió Política del Centre d\'Estudios d\'Opinió.\nMostra: ',
numberfy(sum(pd$n)), ' residents de Catalunya amb ciutadania espanyola.\nGràfic: Joe Brew | @joethebrew'))
legend_title <- 'A favor de la independència?'
} else {
born_in <- 'Born in '
the_labs <- labs(x = 'Place of birth',
y = 'Percentage',
title = 'Support for independence by\nplace of birth',
caption = paste0('Data: Baròmetre d\'Opinió Política of the Centre d\'Estudios d\'Opinió.\nMostra: ',
numberfy(sum(pd$n)), ' residents of Catalonia with Spanish citizenship.\nChart: Joe Brew | @joethebrew'))
legend_title <- 'In favor of independence?'
}
cols <- colors_vilaweb()[c(5,4)]
cols <- c(cols[1], 'grey', cols[2])
ggplot(data = pd %>%
mutate(neixer = paste0(born_in, neixer)) %>%
mutate(neixer = gsub('in Abroad', 'Abroad', neixer)) %>%
mutate(neixer = gsub('Estranger', "l'Estranger", neixer)),
aes(x = any,
y = p,
color = indepe,
group = indepe)) +
geom_line() +
geom_point() +
theme_vilaweb() +
scale_color_manual(name = legend_title,
values = cols) +
# geom_text(aes(label = paste0(round(p, digits = 1), '%')),
# vjust = 1,
# alpha = 0.6) +
theme(legend.position = 'bottom') +
theme(plot.caption = element_text(hjust = 0)) +
facet_wrap(~neixer) +
scale_x_continuous(name = '',breaks = min(pd$any):max(pd$any)) +
theme(axis.text.x = element_text(angle = 90, hjust = 1, vjust = 0.5)) +
the_labs
}
# ramir_plot(ca = T)
joebrew/vilaweb documentation built on Sept. 11, 2020, 3:42 a.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
# Libraries
library(knitr)
library(vilaweb)
library(tidyverse)
library(ggplot2)
library(tidyr)
library(gganimate)
# Functions
mround <- function(x,base){
base*round(x/base)
}
round_percent <- function(x) {
x <- x/sum(x)*100 # Standardize result
res <- floor(x) # Find integer bits
rsum <- sum(res) # Find out how much we are missing
if(rsum<100) {
# Distribute points based on remainders and a random tie breaker
o <- order(x%%1, sample(length(x)), decreasing=TRUE)
res[o[1:(100-rsum)]] <- res[o[1:(100-rsum)]]+1
}
res
}
numberfy <- function(x){
gsub(',', '.', scales::comma(x), fixed = TRUE)
}
# Read in CEO data
new_ceo <- vilaweb::new_ceo
# Read in Catalan demography data
df <- read_delim('data/idescat/t8661.csv', delim = ';', skip = 5)
names(df)[1] <- 'age'
df$age <- gsub(' i més', '', df$age)
df$age <- as.numeric(df$age)
for(j in 2:ncol(df)){
message(j)
vals <- as.character(unlist(df[,j]))
vals <- gsub(',', '.', vals, fixed = TRUE)
vals <- as.numeric(vals)
df[,j] <- vals
}
# Keep only the relevant definitions
df <- df %>% dplyr::select(age, nqx)
# Estimate the last few rows
last_few <- tibble(age = 95:99,
nqx = seq(0.25, 1, length = 5))
# Combine
df <- df %>% filter(age != 95) %>%
bind_rows(last_few)
# Create a table of definitions for the demographic data
df_definitions <-
tibble(variable = c('nPx', 'nDx', 'nMx', 'nqx', 'nax', 'lx',
'ndx', 'nLx', 'Tx', 'ex'),
meaning = c('Mid-year population of those aged x to x+n',
'Number of deaths for those aged x to x+n',
'Age-specific death rates beween ages x to x+n',
'Probability of dying between ages x and x+n',
'Average person-years lived in the interval by those dying in the interval',
'Number of survivors who make it to age x',
'Number of people who will die between age x and x+1',
'Person-years lived between ages x and x+n',
'Person-years lived above age x',
'Expectation of life at age x'))
# Read in place of birth data
lloc <- read_csv('data/idescat/poblacio_per_lloc_de_naixement.csv', skip = 7)
names(lloc) <- c('year', 'Catalunya', 'x', 'y',
"Espanya", "Estranger" , "Total")
lloc <- lloc %>% dplyr::select(year, Catalunya, Estranger, Espanya) %>%
mutate(year = gsub(' (p)', '', year, fixed = TRUE)) %>%
mutate(year = as.numeric(year))
# Make a long version
lloc_long <- lloc %>% tidyr::gather(key, value, Catalunya:Espanya)
# # Basic plot of lloc (exploratory only)
old_lloc <- lloc_long
past_plot <- function(ca = FALSE){
pd <- old_lloc %>%
group_by(year) %>%
mutate(p = value / sum(value) * 100) %>%
ungroup %>%
filter(key == 'Espanya')
if(ca){
the_labs <- labs(title = 'Percentatge de catalans nascuts a Espanya',
caption = 'Dades de l\'Institut d\'Estadística de Catalunya. Gràfic de Joe Brew | @joethebrew | www.vilaweb.cat')
} else {
the_labs <- labs(title = 'Percentage of Catalans born in Spain',
caption = 'Data from the Institut d\'Estadística de Catalunya. Chart by Joe Brew | @joethebrew | www.vilaweb.cat')
}
ggplot(data = pd,
aes(x = year,
y = p)) +
geom_bar(stat = 'identity',
width = 1,
color = 'white',
fill = 'darkorange') +
theme_vilaweb() +
the_labs +
geom_text(aes(label = round(p, digits = 1)),
color = 'white',
nudge_y = -2,
size = 2)
}
# Read in place of birth + year + age data
lloc_dir <- 'data/lloc_i_edat_i_any/'
lloc_files <- dir(lloc_dir)
lloc_list <- list()
read_lloc <- function(x){
data <- suppressMessages(read_delim(x,
delim = ';',
skip = 5)) %>%
dplyr::rename(age = X1) %>%
filter(age != 'Total') %>%
dplyr::select(age,
`Catalunya. Total`,
`Resta d'Espanya. Total`,
`Estranger. Total`)
names(data)[2:4] <- c('Catalunya', 'Espanya', 'Estranger')
year <- unlist(suppressMessages(suppressWarnings(read_delim(x, delim = ';'))[1,1]))
year <- as.numeric(gsub('Catalunya. ', '', as.character(unlist(year)), fixed = TRUE))
data$year <- year
return(data)
}
for(i in 1:length(lloc_files)){
this_file <- paste0(lloc_dir, lloc_files[i])
this_data <- read_lloc(this_file)
message(i)
message(this_data$year[1])
lloc_list[[i]] <- this_data
}
lloc <- bind_rows(lloc_list)
lloc$age <- gsub('anys i més', 'a 99 anys', lloc$age)
split_age <- strsplit(lloc$age, ' ')
lloc$min_age <- as.numeric(unlist(lapply(split_age, function(x){x[2]})))
lloc$max_age <- as.numeric(unlist(lapply(split_age, function(x){x[4]})))
# # Plot (exploratory only)
# pd <- lloc %>% gather(key, value, Catalunya:Estranger)
# ggplot(data = pd %>% filter(year == max(year)),
# aes(x = age,
# y = value,
# fill = key)) +
# geom_bar(stat = 'identity') +
# coord_flip()
# Expand lloc into a one year per row data frame (interpolation)
agey <- tibble(age = 0:99)
age_list <- list()
for(i in 1:nrow(lloc)){
this_row <- lloc[i,]
n <- this_row$max_age - this_row$min_age + 1
new_row <- tibble(age = this_row$min_age:this_row$max_age,
Catalunya = this_row$Catalunya / n,
Espanya = this_row$Espanya / n,
Estranger = this_row$Estranger / n,
year = this_row$year)
# # Get endpoints for smoother interpolation
# end_points <- lloc %>%
# filter(year == this_row$year,
# max_age %in% c(this_row$max_age +5,
# this_row$max_age - 5))
# if(nrow(end_points) == 2){
# the_diffs <- end_points[2,2:4] + end_points[1,2:4]
# gaps <- the_diffs / 7
# vals <- -2:2
# for(i in 1:nrow(new_row)){
# old_vals <- new_row[i,2:4]
# new_vals <- old_vals + (vals[i] * gaps)
# message('Old vals:', sum(old_vals))
# message('New vals: ', round(sum(new_vals)))
# new_row[i,2:4] <- new_vals
# }
# }
age_list[[i]] <- new_row
}
lloc <- bind_rows(age_list)
# Make long
lloc_long <- lloc %>% tidyr::gather(key, value, Catalunya:Estranger)
# # Make plot (exploratory only)
# ggplot(data = lloc_long %>% filter(year == max(year)),
# aes(x = age,
# y = value,
# fill = key)) +
# geom_area() +
# coord_flip()
# # Animation (exploratory only)
# library(gganimate)
# ggplot(data = lloc_long,
# aes(x = age,
# y = value,
# fill = key)) +
# geom_bar(stat = 'identity') +
# coord_flip() +
# labs(title = 'Year: {frame_time}') +
# transition_time(year) +
# ease_aes('linear')
# Estimate what would happen, demographically, with Catalonia's current population
current <- lloc_long %>% filter(year == 2018)
the_list <- list()
years <- 2019:2070
for(i in 1:length(years)){
the_year <- years[i]
# Get last year
if(i == 1){
last_year <- current
} else {
last_year <- the_list[[i-1]]
}
# Join the probability of death
this_year <- last_year %>%
dplyr::select(-year) %>%
left_join(df %>% dplyr::select(age, nqx)) %>%
# replace the value with the mortality-adjusted value
mutate(value = value - (value * nqx)) %>%
mutate(year = the_year) %>%
dplyr::select(-nqx) %>%
mutate(age = age + 1) %>%
filter(age <= 100)
the_list[[i]] <- this_year
}
future <- bind_rows(the_list)
make_animation <- function(ca = FALSE){
if(ca){
legend_title <- 'LLoc de naixement'
pd <- future
the_labs <- labs(title = 'Any: {frame_time}',
x = 'Edat',
y = 'Persones',
subtitle = 'Projecció de la composició de la població catalana\n(assumint cap naixement o migració després de 2019)',
caption = 'Dades de llocs de naixement i mortalitat de l\'Institut d\'Estadística de Catalunya.\nProjeccions i gràfic de Joe Brew | @joethebrew | www.vilaweb.cat')
pd$key <- factor(pd$key,
levels = rev(c('Catalunya',
'Espanya',
'Estranger')))
} else {
legend_title <- 'Place of birth'
pd <- future %>%
mutate(key = ifelse(key == 'Catalunya', 'Catalonia',
ifelse(key == 'Espanya', 'Spain',
ifelse(key == 'Estranger', 'Abroad', key))))
pd$key <- factor(pd$key,
levels = rev(c('Catalonia',
'Spain',
'Abroad')))
the_labs <- labs(title = 'Year: {frame_time}',
x = 'Age',
y = 'People',
subtitle = 'Projection of the composition of Catalonia\'s population\n(assuming no births or migration after 2019)',
caption = 'Data on places of birth and mortality from the Institut d\'Estadística de Catalunya.\nProjections and chart by Joe Brew | @joethebrew | www.vilaweb.cat')
}
cols <- rev(c(colors_vilaweb()[c(5,3)], 'grey'))
label_df <- pd %>%
group_by(key, year) %>%
summarise(value = sum(value, na.rm = TRUE)) %>%
group_by(year) %>%
mutate(p = value / sum(value) * 100) %>%
ungroup %>%
filter(!is.na(p)) %>%
group_by(year) %>%
summarise(value = paste0(legend_title, '\n',paste0(key, ': ', round(p, digits = 1), '%', collapse = '\n'), collapse = ''))
ggplot() +
geom_label(data = label_df,
aes(x = 90,
y = 110000,
label = value)) +
geom_bar(stat = 'identity', width = 1,
data = pd,
aes(x = age,
y = value,
fill = key)) +
coord_flip() +
scale_fill_manual(name = legend_title,
values = cols) +
theme_vilaweb() +
transition_time(year) +
the_labs
}
make_plot <- function(ca = FALSE){
if(ca){
legend_title <- 'LLoc de naixement'
pd <- future
the_labs <- labs(x = 'Any',
y = 'Percentatge',
title = 'Projecció: lloc de naixement dels catalans',
subtitle = '(assumint cap naixement o migració després de 2019)',
caption = 'Dades de llocs de naixement i mortalitat de l\'Institut d\'Estadística de Catalunya.\nProjeccions i gràfic de Joe Brew | @joethebrew | www.vilaweb.cat')
pd$key <- factor(pd$key,
levels = rev(c('Catalunya',
'Espanya',
'Estranger')))
} else {
legend_title <- 'Place of birth'
pd <- future %>%
mutate(key = ifelse(key == 'Catalunya', 'Catalonia',
ifelse(key == 'Espanya', 'Spain',
ifelse(key == 'Estranger', 'Abroad', key))))
pd$key <- factor(pd$key,
levels = rev(c('Catalonia',
'Spain',
'Abroad')))
the_labs <- labs(title = 'Projection: place of birth of Catalans',
x = 'Year',
y = 'Percentage',
subtitle = '(assuming no births or migration after 2019)',
caption = 'Data on places of birth and mortality from the Institut d\'Estadística de Catalunya.\nProjections and chart by Joe Brew | @joethebrew | www.vilaweb.cat')
}
cols <- rev(c(colors_vilaweb()[c(5,3)], 'grey'))
# Group pd
pd <- pd %>%
group_by(key, year) %>%
summarise(value = sum(value, na.rm = TRUE)) %>%
ungroup %>%
group_by(year) %>%
mutate(p = value / sum(value) * 100) %>%
ungroup
ggplot() +
geom_line(data = pd,
aes(x = year,
y = p,
color = key)) +
scale_color_manual(name = legend_title,
values = cols) +
theme_vilaweb() +
the_labs
}
birth_plot <- function(ca = FALSE){
if(ca){
no_know <- 'NS/NC'
catalunya <- 'Catalunya'
ccaa <- 'Espanya'
mon <- 'Estranger'
} else {
no_know <- 'Not sure\nNo answer'
catalunya <- 'Catalonia'
ccaa <- 'Spain'
mon <- 'Abroad'
}
pd <- new_ceo %>%
mutate(neixer = `Em podria dir on va néixer?`,
indepe = `Vol que Catalunya esdevingui un Estat independent?`,
edat = Edat) %>%
mutate(indepe = as.character(indepe),
neixer = as.character(neixer)) %>%
filter(!is.na(neixer),
!is.na(indepe),
!neixer %in% c('No ho sap', 'No contesta')) %>%
mutate(indepe = ifelse(indepe %in% c('No ho sap',
'No contesta'),
no_know,
indepe),
neixer = ifelse(neixer == 'Catalunya', catalunya,
ifelse(neixer %in% c('Altres comunitats autònomes'), ccaa,
mon))) %>%
group_by(neixer, indepe) %>%
tally %>%
ungroup %>%
group_by(neixer) %>%
mutate(p = n / sum(n) * 100) %>%
ungroup
if(ca){
the_labs <- labs(x = 'Lloc de naixement',
y = 'Percentatge',
title = 'Independentisme per lloc de naixement',
caption = paste0('Dades: Baròmetre d\'Opinió Política del Centre d\'Estudios d\'Opinió.\nMostra: ',
numberfy(sum(pd$n)), ' residents de Catalunya amb ciutadania espanyola.\nGràfic: Joe Brew | @joethebrew | www.vilaweb.cat'))
legend_title <- 'A favor\nde la\nindependència?'
} else {
the_labs <- labs(x = 'Place of birth',
y = 'Percentage',
title = 'Support for independence by\nplace of birth',
caption = paste0('Data: Baròmetre d\'Opinió Política of the Centre d\'Estudios d\'Opinió.\nMostra: ',
numberfy(sum(pd$n)), ' residents of Catalonia with Spanish citizenship.\nChart: Joe Brew | @joethebrew | www.vilaweb.cat'))
legend_title <- 'In favor\nof\nindependence?'
}
cols <- colors_vilaweb()[c(5,4)]
cols <- c(cols[1], 'grey', cols[2])
ggplot(data = pd,
aes(x = neixer,
y = p,
fill = indepe,
group = indepe)) +
geom_bar(stat = 'identity',
position = position_stack()) +
theme_vilaweb() +
the_labs +
scale_fill_manual(name = legend_title,
values = cols) +
geom_text(aes(label = paste0(round(p, digits = 1), '%')),
position = position_stack(),
vjust = 1,
alpha = 0.6) +
theme(legend.position = 'right') +
theme(plot.caption = element_text(hjust = 0))
}
age_plot <- function(ca = FALSE){
if(ca){
no_know <- 'NS/NC'
catalunya <- 'Catalunya'
ccaa <- 'Espanya'
mon <- 'Estranger'
} else {
no_know <- 'Not sure\nNo answer'
catalunya <- 'Catalonia'
ccaa <- 'Spain'
mon <- 'Abroad'
}
pd <- new_ceo %>%
mutate(neixer = `Em podria dir on va néixer?`,
indepe = `Vol que Catalunya esdevingui un Estat independent?`,
edat = Edat) %>%
mutate(indepe = as.character(indepe),
neixer = as.character(neixer)) %>%
filter(!is.na(neixer),
!is.na(indepe),
!is.na(edat),
!neixer %in% c('No ho sap', 'No contesta')) %>%
mutate(indepe = ifelse(indepe %in% c('No ho sap',
'No contesta'),
no_know,
indepe),
neixer = ifelse(neixer == 'Catalunya', catalunya,
ifelse(neixer %in% c('Altres comunitats autònomes'), ccaa,
mon))) %>%
group_by(edat, indepe) %>%
tally %>%
ungroup %>%
group_by(edat) %>%
mutate(p = n / sum(n) * 100) %>%
ungroup
if(ca){
the_labs <- labs(x = 'Edat',
y = 'Percentatge',
title = 'Independentisme per edat',
caption = paste0('Dades: Baròmetre d\'Opinió Política del Centre d\'Estudios d\'Opinió.\nMostra: ',
numberfy(sum(pd$n)), ' residents de Catalunya amb ciutadania espanyola.\nGràfic: Joe Brew | @joethebrew | www.vilaweb.cat'))
legend_title <- 'A favor\nde la\nindependència?'
point_name <- 'Mida de\nmostra'
} else {
the_labs <- labs(x = 'Age',
y = 'Percentage',
title = 'Support for independence by age',
caption = paste0('Data: Baròmetre d\'Opinió Política of the Centre d\'Estudios d\'Opinió.\nMostra: ',
numberfy(sum(pd$n)), ' residents of Catalonia with Spanish citizenship.\nChart: Joe Brew | @joethebrew | www.vilaweb.cat'))
legend_title <- 'In favor\nof\nindependence?'
point_name <- 'Sample\nsize'
}
cols <- colors_vilaweb()[c(5,4)]
cols <- c(cols[1], 'grey', cols[2])
ggplot(data = pd,
aes(x = edat,
y = p,
color = indepe)) +
geom_point(alpha = 0.6,
aes(size = n)) +
theme_vilaweb() +
the_labs +
scale_color_manual(name = legend_title,
values = cols) +
theme(legend.position = 'right') +
theme(plot.caption = element_text(hjust = 0)) +
geom_smooth(se = FALSE) +
xlim(18, 90) +
scale_size_continuous(name = point_name, breaks = c(25, 100, 200))
}
age_birth_plot <- function(ca = FALSE){
if(ca){
no_know <- 'NS/NC'
catalunya <- 'Catalunya'
ccaa <- 'Espanya'
mon <- 'Estranger'
} else {
no_know <- 'Not sure\nNo answer'
catalunya <- 'Catalonia'
ccaa <- 'Spain'
mon <- 'Abroad'
}
pd <- new_ceo %>%
mutate(neixer = `Em podria dir on va néixer?`,
indepe = `Vol que Catalunya esdevingui un Estat independent?`,
edat = Edat) %>%
mutate(indepe = as.character(indepe),
neixer = as.character(neixer)) %>%
filter(!is.na(neixer),
!is.na(indepe),
!is.na(edat),
!neixer %in% c('No ho sap', 'No contesta')) %>%
mutate(indepe = ifelse(indepe %in% c('No ho sap',
'No contesta'),
no_know,
indepe),
neixer = ifelse(neixer == 'Catalunya', catalunya,
ifelse(neixer %in% c('Altres comunitats autònomes'), ccaa,
mon))) %>%
group_by(edat, indepe, neixer) %>%
tally %>%
ungroup %>%
group_by(edat, neixer) %>%
mutate(p = n / sum(n) * 100) %>%
ungroup
if(ca){
the_labs <- labs(x = 'Edat',
y = 'Percentatge',
title = 'Independentisme per edat',
caption = paste0('Dades: Baròmetre d\'Opinió Política del Centre d\'Estudios d\'Opinió.\nMostra: ',
numberfy(sum(pd$n)), ' residents de Catalunya amb ciutadania espanyola.\nGràfic: Joe Brew | @joethebrew | www.vilaweb.cat'))
legend_title <- 'A favor\nde la\nindependència?'
point_name <- 'Mida de\nmostra'
} else {
the_labs <- labs(x = 'Age',
y = 'Percentage',
title = 'Support for independence by age',
caption = paste0('Data: Baròmetre d\'Opinió Política of the Centre d\'Estudios d\'Opinió.\nMostra: ',
numberfy(sum(pd$n)), ' residents of Catalonia with Spanish citizenship.\nChart: Joe Brew | @joethebrew | www.vilaweb.cat'))
legend_title <- 'In favor\nof\nindependence?'
point_name <- 'Sample\nsize'
}
cols <- colors_vilaweb()[c(5,4)]
cols <- c(cols[1], 'grey', cols[2])
ggplot(data = pd %>% filter(!neixer %in% c('Abroad', 'Estranger')),
aes(x = edat,
y = p,
color = indepe)) +
geom_point(alpha = 0.6,
aes(size = n)) +
theme_vilaweb() +
the_labs +
scale_color_manual(name = legend_title,
values = cols) +
theme(legend.position = 'right') +
theme(plot.caption = element_text(hjust = 0)) +
geom_smooth(se = FALSE) +
xlim(18, 90) +
scale_size_continuous(name = point_name, breaks = c(25, 100, 200)) +
facet_wrap(~neixer, ncol = 2)
}
current_plot <- function(ca = FALSE){
pd <- current %>%
group_by(key) %>%
mutate(p = value / sum(value) * 100)
if(ca){
pd$key <- factor(pd$key,
levels = c('Catalunya',
'Estranger',
'Espanya'))
the_labs <- labs(y = 'Percentatge',
x = 'Edat',
title = 'Catalans: distribució d\'edat per lloc de naixement',
caption = 'Dades de l\'Institut d\'Estadística de Catalunya. Gràfic de Joe Brew | @joethebrew | www.vilaweb.cat')
} else {
pd$key <- factor(pd$key,
levels = c('Catalunya',
'Estranger',
'Espanya'),
labels = c('Catalonia',
'Abroad',
'Spain'))
the_labs <- labs(y = 'Percentage',
x = 'Age',
title = 'Catalans: age distribution by place of birth',
caption = 'Data from the Institut d\'Estadística de Catalunya. Chart by Joe Brew | @joethebrew | www.vilaweb.cat')
}
ggplot(data = pd,
aes(x = age,
y = p)) +
geom_bar(stat = 'identity',
width = 1,
color = 'white') +
facet_wrap(~key) +
coord_flip() +
theme_vilaweb() +
the_labs
}
# Create model of independentism
predictions <-
new_ceo %>% mutate(neixer = `Em podria dir on va néixer?`,
indepe = `Vol que Catalunya esdevingui un Estat independent?`,
edat = as.numeric(Edat)) %>%
filter(`Any de realització del baròmetre` %in% c(2018, 2019)) %>%
# # group together all the very old (not doing for now, just an option to deal with small sample size)
# mutate(edat = ifelse(edat >= 0, 0, edat)) %>%
mutate(indepe = as.character(indepe),
neixer = as.character(neixer)) %>%
filter(!is.na(neixer),
!is.na(indepe),
!neixer %in% c('No ho sap', 'No contesta')) %>%
mutate(indepe = ifelse(indepe %in% c('No ho sap',
'No contesta'),
'NS/NC',
indepe),
neixer = ifelse(neixer == 'Catalunya', 'Catalunya',
ifelse(neixer %in% c('Altres comunitats autònomes'), 'Espanya',
'Estranger'))) %>%
group_by(neixer, indepe, edat) %>%
tally %>%
ungroup %>%
group_by(edat, neixer) %>%
mutate(p = n / sum(n) * 100) %>%
ungroup %>%
group_by(neixer, edat) %>%
summarise(n_si = sum(n[indepe == 'Sí']),
n_no = sum(n[indepe == 'No']),
n_ns = sum(n[indepe == 'NS/NC']),
p_si = sum(n[indepe == 'Sí']) / sum(n),
p_no = sum(n[indepe == 'No']) / sum(n),
p_ns = sum(n[indepe == 'NS/NC']) / sum(n),
p_si_ref = sum(n[indepe == 'Sí']) / sum(n[indepe %in% c('Sí', 'No')]),
p_no_ref = sum(n[indepe == 'No']) / sum(n[indepe %in% c('Sí', 'No')])) %>%
ungroup %>%
dplyr::select(neixer, edat, p_si, p_no, p_ns, p_si_ref, p_no_ref)
# Test to see if current matches up with survey results
test <- current %>%
dplyr::rename(neixer = key,
edat = age) %>%
# mutate(edat = ifelse(edat >= 0, 0, edat)) %>%
left_join(predictions) %>%
# remove those for whom predictions could not be made
filter(!is.na(p_si)) %>%
# calculate number of people
mutate(indepes = value * p_si,
unionistes = value * p_no,
undecided = value * p_ns)
# x <- test %>%
# summarise(indepes = sum(indepes),
# unionistes = sum(unionistes),
# undecided = sum(undecided))
# x
# as.numeric(x) / sum(as.numeric(x))
# Now, using this, we can create projections
projections <- future %>%
dplyr::rename(neixer = key,
edat = age) %>%
# mutate(edat = ifelse(edat >= 0, 0, edat)) %>%
left_join(predictions) %>%
# remove those for whom predictions could not be made
filter(!is.na(p_si)) %>%
# calculate number of people
mutate(indepes = value * p_si,
unionistes = value * p_no,
undecided = value * p_ns) %>%
group_by(year) %>%
summarise(indepes = sum(indepes),
p_indepes = sum(indepes) / (sum(indepes) + sum(unionistes) + sum(undecided)) * 100,
unionistes = sum(unionistes),
p_unionistes = sum(unionistes) / (sum(unionistes) + sum(indepes) + sum(undecided)) * 100,
undecided = sum(undecided),
p_undecided = sum(undecided) / (sum(unionistes) + sum(indepes) + sum(undecided)) * 100,) %>%
mutate(p_si_ref = p_indepes / (100 - p_undecided),
p_no_ref = p_unionistes / (100 - p_undecided))
projection_plot <- function(ca = FALSE){
if(ca){
the_labs <- labs(x = 'Any',
y = '%',
title = 'Independentisme segons la projecció demogràfica',
subtitle = 'Efecte de la mortalitat, població de 2019',
caption = "Model basat en enquestes de 2018-2019:CEO. Dades poblacionals de 2019: IDES.\nEl percentage reflecteix un hipotètic referèndum (traient els 'no sap/no contesta'), edat=18+.\nJoe Brew | @joethebrew | www.vilaweb.cat")
choices <- c('En contra', 'A favor')
legend_title <- 'Opinió sobre la independència'
} else {
the_labs <- labs(x = 'Year',
y = '%',
title = 'Support for independence per demographic projection',
subtitle = 'Effect of mortality only, year 2019',
caption = "Model based on survey data from 2018-2019: CEO. Population data from 2019: IDES.\nThe percentage reflects a hypothetical referendum (without the 'unknown' responses), age=18+.\nJoe Brew | @joethebrew | www.vilaweb.cat")
choices <- c('Opposed', 'In favor')
legend_title <- c('View on independence')
}
ggplot(data = projections %>% mutate(p_si_ref = p_si_ref * 100,
p_no_ref = p_no_ref * 100) %>%
dplyr::select(year, p_si_ref, p_no_ref) %>%
gather(key, value, p_si_ref:p_no_ref) %>%
mutate(key = ifelse(key == 'p_no_ref',
choices[1],
choices[2])),
aes(x = year,
y = value)) +
geom_line(aes(color = key),
size = 2) +
xlim(2018, 2080) +
theme_vilaweb() +
ylim(0, 100) +
geom_hline(yintercept = 50, lty = 2) +
the_labs +
scale_color_manual(name = legend_title,
values = colors_vilaweb()[3:4])
}
# Immigracions procedents de la resta d'Espanya. Per sexe, grups d'edat i lloc de naixement.
# https://www.idescat.cat/pub/?id=mm&n=6383
#t6383.csv
ready <- function(number = 6383){
ides_urls <- paste0('https://www.idescat.cat/pub/?id=mm&n=',
number,'&t=', 2005:2017,'00&f=csv')
yearsy <- 2005:2017
ides_list <- list()
for(j in 1:length(ides_urls)){
# Sys.sleep(2)
message(j)
ides <- read_csv(ides_urls[j], skip = 5)
# Keep only the total and below
ides <- ides[(which(ides$X1 == 'Total') +1): nrow(ides),]
ides <- ides %>%
gather(age, value, `De 0 a 14 anys`:`De 60 anys i més`) %>%
dplyr::rename(key = X1) %>%
mutate(key = gsub('nascuts a ', '', key)) %>%
mutate(key = gsub("la resta d'", "", key)) %>%
mutate(key = gsub("l'e", "E", key, fixed = TRUE))
# Fix age
ides$age <- gsub('anys i més', 'a 99 anys', ides$age)
split_age <- strsplit(ides$age, ' ')
ides$min_age <- as.numeric(unlist(lapply(split_age, function(x){x[2]})))
ides$max_age <- as.numeric(unlist(lapply(split_age, function(x){x[4]})))
ides$Total <- NULL
# Expand
migrations_list <- list()
for(i in 1:nrow(ides)){
this_row <- ides[i,]
out <- tibble(age = this_row$min_age:this_row$max_age,
key = this_row$key)
n <- nrow(out)
out$value <- this_row$value / n
migrations_list[[i]] <- out
}
ides <- bind_rows(migrations_list)
ides_list[[j]] <- ides %>% mutate(year = yearsy[j])
}
ides <- bind_rows(ides_list) %>%
group_by(age, key, year) %>%
summarise(value = mean(value))
return(ides)
}
# Get inmig #6383
# exmig 6384
# wexmig 6386
# winbmig 6389
if(!'migdata.RData' %in% dir()){
inmig <- ready(6383)
exmig <- ready(6384)
wexmig <- ready(6386)
winmig <- ready(6389)
save(inmig, exmig, wexmig, winmig,
file = 'migdata.RData')
} else {
load('migdata.RData')
}
transformy <- function(x, since_year = 2005){
x %>% filter(year >= since_year) %>%
group_by(age, key) %>%
summarise(value = mean(value))
}
inmig <- transformy(inmig)
exmig <- transformy(exmig)
wexmig <- transformy(wexmig)
winmig <- transformy(winmig)
# Copmbine all migrations
migrations <-
bind_rows(
inmig %>% mutate(destination = 'Catalunya'),
winmig %>% mutate(destination = 'Catalunya'),
exmig %>% mutate(destination = 'World'),
wexmig %>% mutate(destination = 'World')
)
# # Add births
births <- tibble(age = 0,
key = c('Catalunya', 'Espanya', 'Estranger'),
value = c(66803,
896, # 896 + 241 - 135,
3051))# + 938 - 945))
# changes <- bind_rows(migrations, births)
changes <- migrations
changes <- changes %>%
# dplyr::rename(place_of_birth = key) %>%
mutate(action = ifelse(destination == 'Catalunya',
'Coming',
'Leaving')) %>%
dplyr::select(-destination) %>%
group_by(key, age) %>%
summarise(leaving = sum(value[action == 'Leaving']),
coming = sum(value[action == 'Coming'])) %>%
ungroup
# Make changes a percentage, rather than absolute number
changes <- left_join(changes, current) %>% dplyr::select(-year)
changes <- changes %>%
# mutate(
# leaving = leaving / value,
# coming = coming / value) %>%
dplyr::select(-value)
# Create an adjusted future, to account for changes (migrations and births)
# Estimate what would happen, demographically, with Catalonia's current population
current <- lloc_long %>% filter(year == 2018)
the_list <- list()
years <- 2019:2070
for(i in 1:length(years)){
the_year <- years[i]
# Get last year
if(i == 1){
last_year <- current
} else {
last_year <- the_list[[i-1]]
}
# Join the probability of death
this_year <- last_year %>%
dplyr::select(-year) %>%
left_join(df %>% dplyr::select(age, nqx)) %>%
# replace the value with the mortality-adjusted value
mutate(value = value - (value * nqx)) %>%
mutate(year = the_year) %>%
dplyr::select(-nqx) %>%
mutate(age = age + 1) %>%
filter(age <= 100)
# Add a zero year
this_year <- this_year %>%
bind_rows(births %>% mutate(year = this_year$year[1]))
this_year <- this_year %>% arrange(age)
# Add the changes
this_year <- left_join(this_year, changes)
this_year <- this_year %>%
mutate(value = value + coming - leaving) %>%
# mutate(value = value - (leaving * value) + (coming * value)) %>%
dplyr::select(-coming, -leaving)
the_list[[i]] <- this_year
}
future_adj <- bind_rows(the_list)
# Use the adjusted futures to make adjusted projections
# Now, using this, we can create projections
projections_adj <- future_adj %>%
dplyr::rename(neixer = key,
edat = age) %>%
# mutate(edat = ifelse(edat >= 0, 0, edat)) %>%
left_join(predictions) %>%
# remove those for whom predictions could not be made
filter(!is.na(p_si)) %>%
# calculate number of people
mutate(indepes = value * p_si,
unionistes = value * p_no,
undecided = value * p_ns) %>%
group_by(year) %>%
summarise(indepes = sum(indepes),
p_indepes = sum(indepes) / (sum(indepes) + sum(unionistes) + sum(undecided)) * 100,
unionistes = sum(unionistes),
p_unionistes = sum(unionistes) / (sum(unionistes) + sum(indepes) + sum(undecided)) * 100,
undecided = sum(undecided),
p_undecided = sum(undecided) / (sum(unionistes) + sum(indepes) + sum(undecided)) * 100,) %>%
mutate(p_si_ref = p_indepes / (100 - p_undecided),
p_no_ref = p_unionistes / (100 - p_undecided))
# Make a version with no foreigners
projections_adj_no_estranger <- future_adj %>%
dplyr::rename(neixer = key,
edat = age) %>%
filter(!neixer %in% c('Estranger', 'Abroad')) %>%
# mutate(edat = ifelse(edat >= 0, 0, edat)) %>%
left_join(predictions) %>%
# remove those for whom predictions could not be made
filter(!is.na(p_si)) %>%
# calculate number of people
mutate(indepes = value * p_si,
unionistes = value * p_no,
undecided = value * p_ns) %>%
group_by(year) %>%
summarise(indepes = sum(indepes),
p_indepes = sum(indepes) / (sum(indepes) + sum(unionistes) + sum(undecided)) * 100,
unionistes = sum(unionistes),
p_unionistes = sum(unionistes) / (sum(unionistes) + sum(indepes) + sum(undecided)) * 100,
undecided = sum(undecided),
p_undecided = sum(undecided) / (sum(unionistes) + sum(indepes) + sum(undecided)) * 100,) %>%
mutate(p_si_ref = p_indepes / (100 - p_undecided),
p_no_ref = p_unionistes / (100 - p_undecided))
make_animation_adj <- function(ca = FALSE){
if(ca){
legend_title <- 'LLoc de naixement'
pd <- future_adj
the_labs <- labs(title = 'Any: {frame_time}',
x = 'Edat',
y = 'Persones',
subtitle = 'Projecció de la composició de la població catalana\n(assumint naixement/migració constants)',
caption = 'Dades de llocs de naixement i mortalitat de l\'Institut d\'Estadística de Catalunya.\nProjeccions i gràfic de Joe Brew | @joethebrew | www.vilaweb.cat')
pd$key <- factor(pd$key,
levels = rev(c('Catalunya',
'Espanya',
'Estranger')))
} else {
legend_title <- 'Place of birth'
pd <- future_adj %>%
mutate(key = ifelse(key == 'Catalunya', 'Catalonia',
ifelse(key == 'Espanya', 'Spain',
ifelse(key == 'Estranger', 'Abroad', key))))
pd$key <- factor(pd$key,
levels = rev(c('Catalonia',
'Spain',
'Abroad')))
the_labs <- labs(title = 'Year: {frame_time}',
x = 'Age',
y = 'People',
subtitle = 'Projection of the composition of Catalonia\'s population\n(assuming constant birth and migration rates)',
caption = 'Data on places of birth and mortality from the Institut d\'Estadística de Catalunya.\nProjections and chart by Joe Brew | @joethebrew | www.vilaweb.cat')
}
cols <- rev(c(colors_vilaweb()[c(5,3)], 'grey'))
label_df <- pd %>%
group_by(key, year) %>%
summarise(value = sum(value, na.rm = TRUE)) %>%
group_by(year) %>%
mutate(p = value / sum(value) * 100) %>%
ungroup %>%
filter(!is.na(p)) %>%
group_by(year) %>%
summarise(value = paste0(legend_title, '\n',paste0(key, ': ', round(p, digits = 1), '%', collapse = '\n'), collapse = ''))
ggplot() +
geom_label(data = label_df,
aes(x = 90,
y = 110000,
label = value)) +
geom_bar(stat = 'identity', width = 1,
data = pd,
aes(x = age,
y = value,
fill = key)) +
coord_flip() +
scale_fill_manual(name = legend_title,
values = cols) +
theme_vilaweb() +
transition_time(year) +
the_labs
}
make_plot_adj <- function(ca = FALSE, remove_estranger = F){
if(ca){
legend_title <- 'LLoc de naixement'
pd <- future_adj
the_labs <- labs(x = 'Any',
y = 'Percentatge',
title = 'Projecció: lloc de naixement dels catalans',
subtitle = '(assumint migració/naixement constants)',
caption = 'Dades de llocs de naixement i mortalitat de l\'Institut d\'Estadística de Catalunya.\nProjeccions i gràfic de Joe Brew | @joethebrew | www.vilaweb.cat')
pd$key <- factor(pd$key,
levels = rev(c('Catalunya',
'Espanya',
'Estranger')))
} else {
legend_title <- 'Place of birth'
pd <- future_adj
pd <- pd %>%
mutate(key = ifelse(key == 'Catalunya', 'Catalonia',
ifelse(key == 'Espanya', 'Spain',
ifelse(key == 'Estranger', 'Abroad', key))))
pd$key <- factor(pd$key,
levels = rev(c('Catalonia',
'Spain',
'Abroad')))
the_labs <- labs(title = 'Projection: place of birth of Catalans',
x = 'Year',
y = 'Percentage',
subtitle = '(assuming constant birth/migration rates)',
caption = 'Data on places of birth and mortality from the Institut d\'Estadística de Catalunya.\nProjections and chart by Joe Brew | @joethebrew | www.vilaweb.cat')
}
cols <- rev(c(colors_vilaweb()[c(5,3)], 'grey'))
# Group pd
pd <- pd %>%
group_by(key, year) %>%
summarise(value = sum(value, na.rm = TRUE)) %>%
ungroup %>%
group_by(year) %>%
mutate(p = value / sum(value) * 100) %>%
ungroup
ggplot() +
geom_line(data = pd,
aes(x = year,
y = p,
color = key)) +
scale_color_manual(name = legend_title,
values = cols) +
theme_vilaweb() +
the_labs
}
projection_plot_adj <- function(ca = FALSE, short = FALSE,
remove_estranger = FALSE){
if(ca){
the_labs <- labs(x = 'Any',
y = '%',
title = 'Independentisme segons la projecció demogràfica',
subtitle = 'Efecte de la mortalitat/migració/fertilitat',
caption = "Model basat en enquestes de 2018-2019:CEO. Dades poblacionals de 2019: IDES.\nEl percentage reflecteix un hipotètic referèndum (traient els 'no sap/no contesta'), edat=18+.\nJoe Brew | @joethebrew | www.vilaweb.cat")
choices <- c('En contra', 'A favor')
legend_title <- 'Opinió sobre la independència'
} else {
the_labs <- labs(x = 'Year',
y = '%',
title = 'Support for independence per demographic projection',
subtitle = 'Effect of mortality/migration/fertility',
caption = "Model based on survey data from 2018-2019: CEO. Population data from 2019: IDES.\nThe percentage reflects a hypothetical referendum (without the 'unknown' responses), age=18+.\nJoe Brew | @joethebrew | www.vilaweb.cat")
choices <- c('Opposed', 'In favor')
legend_title <- c('View on independence')
}
if(remove_estranger){
pd <- projections_adj_no_estranger
} else {
pd <- projections_adj
}
g <- ggplot(data = pd %>% mutate(p_si_ref = p_si_ref * 100,
p_no_ref = p_no_ref * 100) %>%
dplyr::select(year, p_si_ref, p_no_ref) %>%
gather(key, value, p_si_ref:p_no_ref) %>%
mutate(key = ifelse(key == 'p_no_ref',
choices[1],
choices[2])),
aes(x = year,
y = value))
if(short){
g <- g + geom_line(aes(color = key),
size = 1) +
xlim(2018, 2035) +
ylim(46, 54)
} else {
g <- g + geom_line(aes(color = key),
size = 2) +
xlim(2018, 2080) +
geom_line(aes(color = key),
size = 2) +
ylim(0, 100)
}
g <- g +
theme_vilaweb() +
geom_hline(yintercept = 50, lty = 2) +
the_labs +
scale_color_manual(name = legend_title,
values = colors_vilaweb()[3:4])
return(g)
}
# # calculations for Ramir
# # independentisme by place of birth by year
# no_know <- 'NS/NC'
# catalunya <- 'Catalunya'
# ccaa <- 'Espanya'
# mon <- 'Estranger'
# pd <- new_ceo %>%
# mutate(neixer = `Em podria dir on va néixer?`,
# indepe = `Vol que Catalunya esdevingui un Estat independent?`,
# edat = Edat,
# any = `Any de realització del baròmetre`) %>%
# mutate(indepe = as.character(indepe),
# neixer = as.character(neixer)) %>%
# filter(!is.na(neixer),
# !is.na(indepe),
# !neixer %in% c('No ho sap', 'No contesta')) %>%
# mutate(indepe = ifelse(indepe %in% c('No ho sap',
# 'No contesta'),
# no_know,
# indepe),
# neixer = ifelse(neixer == 'Catalunya', catalunya,
# ifelse(neixer %in% c('Altres comunitats autònomes'), ccaa,
# mon))) %>%
# filter(indepe != 'NS/NC') %>%
# group_by(neixer, indepe, any) %>%
# tally %>%
# ungroup %>%
# group_by(neixer, any) %>%
# mutate(p = n / sum(n) * 100) %>%
# ungroup %>%
# filter(neixer %in% c('Catalunya', 'Espanya'))
#
# ggplot(data = pd,
# aes(x = any,
# y = p,
# color = indepe)) +
# geom_line() +
# facet_wrap(~neixer)
# write_csv(pd, '~/Desktop/ramir.csv')
avis_plot <- function(ca = FALSE, pares = FALSE){
if(ca){
no_know <- 'NS/NC'
catalunya <- 'Catalunya'
ccaa <- 'Espanya'
mon <- 'Estranger'
} else {
no_know <- 'Not sure\nNo answer'
catalunya <- 'Catalonia'
ccaa <- 'Spain'
mon <- 'Abroad'
}
if(pares){
pd <- new_ceo %>%
# neixer = `Em podria dir on va néixer?`,
mutate(mare = `Em podria dir el lloc de naixement de la seva mare?`,
pare = `Em podria dir el lloc de naixement del seu pare?`) %>%
filter(!mare %in% c('No ho sap', 'No contesta'),
!pare %in% c('No ho sap', 'No contesta')) %>%
mutate(mare = as.character(mare),
pare = as.character(pare)) %>%
mutate(mare = ifelse(mare == 'Catalunya', '1', '0'),
pare = ifelse(pare == 'Catalunya', '1', '0')) %>%
mutate(mare = as.numeric(mare),
pare = as.numeric(pare)) %>%
mutate(avis = mare + pare,
indepe = `Vol que Catalunya esdevingui un Estat independent?`,
edat = Edat)
} else {
pd <- new_ceo %>%
mutate(avis = `Quants dels seus avis/àvies van néixer a Catalunya?`,
# neixer = `Em podria dir on va néixer?`,
indepe = `Vol que Catalunya esdevingui un Estat independent?`,
edat = Edat)
}
pd <- pd %>%
mutate(indepe = as.character(indepe)) %>%#,
# neixer = as.character(neixer)) %>%
filter(#!is.na(neixer),
!is.na(indepe),
!is.na(avis)) %>%
# !neixer %in% c('No ho sap', 'No contesta')) %>%
mutate(indepe = ifelse(indepe %in% c('No ho sap',
'No contesta'),
no_know,
indepe)) %>%#,
# neixer = ifelse(neixer == 'Catalunya', catalunya,
# ifelse(neixer %in% c('Altres comunitats autònomes'), ccaa,
# mon))) %>%
mutate(avis = as.character(avis))
if(!pares){
pd <- pd %>%
mutate(avis = ifelse(avis == 'Cap', '0',
ifelse(avis == 'Un', '1',
ifelse(avis == 'Dos', '2',
ifelse(avis == 'Tres', '3',
ifelse(avis == 'Quatre', '4', NA)))))) %>%
mutate(avis = as.numeric(avis))
}
pd <- pd %>%
filter(!is.na(avis)) %>%
group_by(avis, indepe) %>%
tally %>%
ungroup %>%
group_by(avis) %>%
mutate(p = n / sum(n) * 100) %>%
ungroup
if(ca){
if(pares){
x <- 'Quants pares van néixer a Catalunya'
tt <- 'Independentisme per lloc de naixement dels pares'
} else {
x <- 'Quants avis van néixer a Catalunya'
tt <- 'Independentisme per lloc de naixement dels avis'
}
the_labs <- labs(x = x,
y = 'Percentatge',
title = tt,
caption = paste0('Dades: Baròmetre d\'Opinió Política del Centre d\'Estudios d\'Opinió.\nMostra: ',
numberfy(sum(pd$n)), ' residents de Catalunya amb ciutadania espanyola.\nGràfic: Joe Brew | @joethebrew | www.vilaweb.cat'))
legend_title <- 'A favor\nde la\nindependència?'
} else {
if(pares){
x <- 'How many parents born in Catalonia'
tt <- 'Support for independence by\nplace of birth of parents'
} else {
x <- 'How many grandparents born in Catalonia'
tt <- 'Support for independence by\nplace of birth of grandparents'
}
the_labs <- labs(x = x,
y = 'Percentage',
title = tt,
caption = paste0('Data: Baròmetre d\'Opinió Política of the Centre d\'Estudios d\'Opinió.\nMostra: ',
numberfy(sum(pd$n)), ' residents of Catalonia with Spanish citizenship.\nChart: Joe Brew | @joethebrew | www.vilaweb.cat'))
legend_title <- 'In favor\nof\nindependence?'
}
cols <- colors_vilaweb()[c(5,4)]
cols <- c(cols[1], 'grey', cols[2])
ggplot(data = pd,
aes(x = avis,
y = p,
fill = indepe,
group = indepe)) +
geom_bar(stat = 'identity',
position = position_stack()) +
theme_vilaweb() +
the_labs +
scale_fill_manual(name = legend_title,
values = cols) +
geom_text(aes(label = paste0(round(p, digits = 1), '%')),
position = position_stack(),
vjust = 1,
alpha = 0.6) +
theme(legend.position = 'right') +
theme(plot.caption = element_text(hjust = 0))
}
pares_birth_plot <- function(ca = FALSE, pares = FALSE){
if(ca){
no_know <- 'NS/NC'
catalunya <- 'Catalunya'
ccaa <- 'Espanya'
mon <- 'Estranger'
} else {
no_know <- 'Not sure\nNo answer'
catalunya <- 'Catalonia'
ccaa <- 'Spain'
mon <- 'Abroad'
}
if(pares){
pd <- new_ceo %>%
mutate(neixer = `Em podria dir on va néixer?`) %>%
mutate(mare = `Em podria dir el lloc de naixement de la seva mare?`,
pare = `Em podria dir el lloc de naixement del seu pare?`) %>%
filter(!mare %in% c('No ho sap', 'No contesta'),
!pare %in% c('No ho sap', 'No contesta')) %>%
mutate(mare = as.character(mare),
pare = as.character(pare)) %>%
mutate(mare = ifelse(mare == 'Catalunya', '1', '0'),
pare = ifelse(pare == 'Catalunya', '1', '0')) %>%
mutate(mare = as.numeric(mare),
pare = as.numeric(pare)) %>%
mutate(avis = mare + pare,
indepe = `Vol que Catalunya esdevingui un Estat independent?`,
edat = Edat)
} else {
pd <- new_ceo %>%
mutate(avis = `Quants dels seus avis/àvies van néixer a Catalunya?`,
neixer = `Em podria dir on va néixer?`,
indepe = `Vol que Catalunya esdevingui un Estat independent?`,
edat = Edat)
}
pd <- pd %>%
mutate(indepe = as.character(indepe)) %>%
mutate(neixer = as.character(neixer)) %>%
filter(!is.na(neixer),
!is.na(indepe),
!is.na(avis)) %>%
filter(!neixer %in% c('No ho sap', 'No contesta')) %>%
mutate(indepe = ifelse(indepe %in% c('No ho sap',
'No contesta'),
no_know,
indepe),
neixer = ifelse(neixer == 'Catalunya', catalunya,
ifelse(neixer %in% c('Altres comunitats autònomes'), ccaa,
mon))) %>%
mutate(avis = as.character(avis))
if(!pares){
pd <- pd %>%
mutate(avis = ifelse(avis == 'Cap', '0',
ifelse(avis == 'Un', '1',
ifelse(avis == 'Dos', '2',
ifelse(avis == 'Tres', '3',
ifelse(avis == 'Quatre', '4', NA)))))) %>%
mutate(avis = as.numeric(avis))
}
pd <- pd %>%
filter(!is.na(avis)) %>%
group_by(avis, neixer, indepe) %>%
tally %>%
ungroup %>%
group_by(avis, neixer) %>%
mutate(p = n / sum(n) * 100) %>%
ungroup
if(ca){
if(pares){
x <- 'Quants pares van néixer a Catalunya'
tt <- 'Independentisme per lloc de naixement dels pares'
} else {
x <- 'Quants avis van néixer a Catalunya'
tt <- 'Independentisme per lloc de naixement dels avis'
}
the_labs <- labs(x = x,
y = 'Percentatge',
title = tt,
caption = paste0('Dades: Baròmetre d\'Opinió Política del Centre d\'Estudios d\'Opinió.\nMostra: ',
numberfy(sum(pd$n)), ' residents de Catalunya amb ciutadania espanyola.\nGràfic: Joe Brew | @joethebrew | www.vilaweb.cat'))
legend_title <- 'A favor\nde la\nindependència?'
} else {
if(pares){
x <- 'How many parents born in Catalonia'
tt <- 'Support for independence by\nplace of birth of parents'
} else {
x <- 'How many grandparents born in Catalonia'
tt <- 'Support for independence by\nplace of birth of grandparents'
}
the_labs <- labs(x = x,
y = 'Percentage',
title = tt,
caption = paste0('Data: Baròmetre d\'Opinió Política of the Centre d\'Estudios d\'Opinió.\nMostra: ',
numberfy(sum(pd$n)), ' residents of Catalonia with Spanish citizenship.\nChart: Joe Brew | @joethebrew | www.vilaweb.cat'))
legend_title <- 'In favor\nof\nindependence?'
}
cols <- colors_vilaweb()[c(5,4)]
cols <- c(cols[1], 'grey', cols[2])
ggplot(data = pd %>%
mutate(neixer = paste0('Persona nascuda a\n', neixer)),
aes(x = avis,
y = p,
fill = indepe,
group = indepe)) +
geom_bar(stat = 'identity',
position = position_stack()) +
theme_vilaweb() +
the_labs +
scale_fill_manual(name = legend_title,
values = cols) +
geom_text(aes(label = paste0(round(p, digits = 1), '%')),
position = position_stack(),
vjust = 1,
size = 2,
alpha = 0.6) +
theme(legend.position = 'right') +
theme(plot.caption = element_text(hjust = 0),
strip.text = element_text(size = 10)) +
facet_wrap(~neixer)
}
# Read in a old ceo
old_ceo <- vilaweb::old_ceo
ramir_plot <- function(ca = FALSE){
if(ca){
no_know <- 'NS/NC'
catalunya <- 'Catalunya'
ccaa <- 'Espanya'
mon <- 'Estranger'
} else {
no_know <- 'Not sure\nNo answer'
catalunya <- 'Catalonia'
ccaa <- 'Spain'
mon <- 'Abroad'
}
old_pd <- old_ceo %>%
mutate(indepe = `P39. I més concretament, si demà es fes un referèndum per decidir la independència de Catalunya, vostè què faria?`,
neixer = `C100. Lloc de naixement de l'enquestat/ada`) %>%
mutate(indepe = as.character(indepe)) %>%
mutate(indepe = ifelse(grepl('a favor', indepe), 'Sí',
ifelse(grepl('en contra', indepe), 'No',
ifelse(grepl('no aniria|ho sap|contesta', indepe), 'NS/NC', NA)))) %>%
mutate(indepe = ifelse(indepe %in% c('No ho sap',
'No contesta',
'NS/NC'),
no_know,
indepe),
neixer = ifelse(neixer == 'Catalunya', catalunya,
ifelse(neixer %in% c('Altres comunitats autònomes'), ccaa,
mon))) %>%
mutate(any = `Any de realització del baròmetre`) %>%
filter(!is.na(neixer),
!is.na(indepe),
!neixer %in% c('No ho sap', 'No contesta')) %>%
mutate(any = as.numeric(as.character(any)))
new_pd <- new_ceo %>%
mutate(neixer = `Em podria dir on va néixer?`,
indepe = `Vol que Catalunya esdevingui un Estat independent?`,
edat = Edat) %>%
mutate(indepe = as.character(indepe),
neixer = as.character(neixer),
any = `Any de realització del baròmetre`) %>%
filter(!is.na(neixer),
!is.na(indepe),
!neixer %in% c('No ho sap', 'No contesta')) %>%
mutate(indepe = ifelse(indepe %in% c('No ho sap',
'No contesta'),
no_know,
indepe),
neixer = ifelse(neixer == 'Catalunya', catalunya,
ifelse(neixer %in% c('Altres comunitats autònomes'), ccaa,
mon))) %>%
mutate(any = as.numeric(as.character(any))) %>%
dplyr::select(neixer, indepe, any)
pd <- new_pd %>%
bind_rows(
old_pd %>% dplyr::select(neixer, indepe, any)
) %>%
group_by(neixer, indepe, any) %>%
tally %>%
ungroup %>%
group_by(neixer, any) %>%
mutate(p = n / sum(n) * 100) %>%
ungroup
if(ca){
born_in <- 'Nascuts a '
the_labs <- labs(x = 'Lloc de naixement',
y = 'Percentatge',
title = 'Independentisme per lloc de naixement',
caption = paste0('Dades: Baròmetre d\'Opinió Política del Centre d\'Estudios d\'Opinió.\nMostra: ',
numberfy(sum(pd$n)), ' residents de Catalunya amb ciutadania espanyola.\nGràfic: Joe Brew | @joethebrew'))
legend_title <- 'A favor de la independència?'
} else {
born_in <- 'Born in '
the_labs <- labs(x = 'Place of birth',
y = 'Percentage',
title = 'Support for independence by\nplace of birth',
caption = paste0('Data: Baròmetre d\'Opinió Política of the Centre d\'Estudios d\'Opinió.\nMostra: ',
numberfy(sum(pd$n)), ' residents of Catalonia with Spanish citizenship.\nChart: Joe Brew | @joethebrew'))
legend_title <- 'In favor of independence?'
}
cols <- colors_vilaweb()[c(5,4)]
cols <- c(cols[1], 'grey', cols[2])
ggplot(data = pd %>%
mutate(neixer = paste0(born_in, neixer)) %>%
mutate(neixer = gsub('in Abroad', 'Abroad', neixer)) %>%
mutate(neixer = gsub('Estranger', "l'Estranger", neixer)),
aes(x = any,
y = p,
color = indepe,
group = indepe)) +
geom_line() +
geom_point() +
theme_vilaweb() +
scale_color_manual(name = legend_title,
values = cols) +
# geom_text(aes(label = paste0(round(p, digits = 1), '%')),
# vjust = 1,
# alpha = 0.6) +
theme(legend.position = 'bottom') +
theme(plot.caption = element_text(hjust = 0)) +
facet_wrap(~neixer) +
scale_x_continuous(name = '',breaks = min(pd$any):max(pd$any)) +
theme(axis.text.x = element_text(angle = 90, hjust = 1, vjust = 0.5)) +
the_labs
}
# ramir_plot(ca = T)
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.