analyses/filiafobia/prepare_data.R
In joebrew/vilaweb: Reproducible data analysis for www.vilaweb.cat
# Libraries
library(vilaweb)
library(tidyverse)
library(lubridate)
library(databrew)
library(tidylog)
options(scipen = 999)
plot_twitter <- function(ca = TRUE){
pd <- fobia %>% filter(key == 'Hispanofòbia') %>%
group_by(date = as.Date(cut(date, 'month'))) %>%
summarise(n = n(),
rt = sum(retweets_count))
if(ca){
the_labs <- labs(x = 'Mes',
y = 'Piulets (escala logarítmica)',
title = "Piulets amb la paraula de 'hispanofòbia'",
subtitle = 'Dades de Twitter, incloent-hi repiulets',
caption = 'Gràfic de Joe Brew, @joethebrew')
} else {
the_labs <- labs(x = 'Month',
y = 'Piulets (log scale)',
title = "Tweets with the word 'Hispanophobia'",
subtitle = 'Data from Twitter, including retweets',
caption = 'Chart by Joe Brew, @joethebrew.')
}
ggplot(data = pd,
aes(x = date,
y = rt + n)) +
geom_line() +
# geom_path(color = 'darkorange', alpha = 0.6) +
theme_simple() +
scale_y_log10() +
the_labs
}
plot_google <- function(ca = TRUE){
# Read in google trends data (retrieved 23 jan 2020)
google <- read_csv('multiTimeline.csv', skip = 1)
google$date <- as.Date(paste0(google$Month, '-01'))
# google$date <- as.Date(cut(google$date, 'year'))
google$value <- google$`hispanofobia: (Spain)`
if(ca){
the_labs <- labs(x = 'Mes',
y = 'Percentatge (escala logarítmica)',
title = "Cerques de 'hispanofòbia'",
subtitle = 'Dades de Google Trends',
caption = 'Percentatge: cerques mensuals relatives al màxim. Gràfic de Joe Brew, @joethebrew')
} else {
the_labs <- labs(x = 'Month',
y = 'Percentage (log scale)',
title = "Searches for 'Hispanophobia'",
subtitle = 'Data from Google Trends',
caption = 'Percentage: monthly searches relative to the maximum. Chart by Joe Brew, @joethebrew.')
}
pd <- google %>%
group_by(date) %>%
summarise(value = mean(value))
ggplot(data = pd,
aes(x = date,
y = value)) +
geom_line() +
# geom_path(color = 'darkorange', alpha = 0.6) +
theme_simple() +
scale_y_log10(breaks = c(0:3, 5,10,25, 50, 100)) +
the_labs
}
# Read in the territorial survey data
pt <- vilaweb::ceo_percepcio_territorial
# Function for cleaning up the sympathy variable
simpatia_cleaner <- function(x){
x <- as.character(x)
x <- ifelse(x %in% c('No contesta', 'No ho sap'), NA,
ifelse(x == 'Us cauen molt bé 10', '10',
ifelse(x == 'No us cauen ni bé ni malament 5', 5,
ifelse(x == 'Us cauen molt malament 0', '0',
x))))
x <- as.numeric(x)
return(x)
}
pt <- pt %>%
mutate(visits = `I d’aquestes, en concret, em podria dir quantes vegades a viatjat a Catalunya?`,
simpatia = `Grau de simpatia cap als habitants de cadascuna de les diferents comunitats autònomes: Catalans`,
referendum = `Grau d'acord: S’hauria de fer un referèndum a Catalunya perquè els catalans i les catalanes decidissin quina relació volen que hi hagi entre Catalunya i l’Estat espanyol`) %>%
mutate(simpatia = simpatia_cleaner(simpatia)) %>%
mutate(ccaa = `Comunitat autònoma`)
# # Companeros de trabajo
# pt$`Amb independència que treballeu o no, quins són els tres que menys us agradaria tenir com a companys de feina o treball? Catalans`
#
# # Grau de simpatia
# pt$`Grau de simpatia cap als habitants de cadascuna de les diferents comunitats autònomes: Catalans`
# Define the label sets (for factor ordering)
label_df <- tibble(
ccaa = c('Andalusia',
'Aragó',
'Astúries',
'Balears',
'Canàries',
'Cantàbria',
'Castella la Manxa',
'Castella i Lleó',
'Catalunya',
'País Valencià',
'Extremadura',
'Galícia',
'Madrid',
'Múrcia',
'Navarra',
'Euskadi/País Basc',
'La Rioja', 'Ceuta', 'Melilla'),
key = c('Andalusos',
'Aragonesos',
'Asturians',
'Balears',
'Canaris',
'Càntabres',
'Castellanomanxecs',
'Castellanolleonesos',
'Catalans',
'Valencians',
'Extremenys',
'Gallecs',
'Madrilenys',
'Murcians',
'Navarresos',
'Bascos',
'Riojans',
'Ceutencs',
'Melillencs')
)
fix_levels <- function(var, from_place = FALSE, to_place = TRUE){
if(from_place){
x <- tibble(ccaa = var)
} else {
x <- tibble(key = var)
}
# Join with the labels
joined <- left_join(x, label_df)
# Get the return val
if(to_place){
out <- joined$ccaa
out <- factor(out, levels = label_df$ccaa)
} else {
out <- joined$key
out <- factor(out, levels = label_df$key)
}
return(out)
}
plot_viatges_i_simpatia <- function(ca = TRUE){
# Viatges a Cat i favorabilitat e Cats
pd <- pt %>%
filter(ccaa != 'Catalunya') %>%
filter(!is.na(visits)) %>%
filter(!visits %in% c('No ho sap', 'No contesta')) %>%
group_by(visits) %>%
summarise(simpatia = mean(simpatia, na.rm = TRUE),
n = n())
pd$visits <- factor(pd$visits,
levels = unique(c('No, mai',
levels(pd$visits))))
pd$visits <- factor(pd$visits,
levels = levels(pd$visits),
labels = gsub(' vegad', '\nvegad', c('Mai', levels(pd$visits)[2:length(levels(pd$visits))])))
if(!ca){
pd$visits <- factor(pd$visits,
labels = c('Never', 'Just\nonce',
'2-5 times',
'6-10 times',
'More than\n10 times'))
the_labs <- labs(x = 'How many times have you travelled to Catalonia?',
y = 'Degree of sympathy (0-10) for\nthe residents of Catalonia',
title = "Times travelled to Catalonia and\nfeelings about Catalans",
caption = "Data source: Survey 'Percepció sobre el debat territorial a Espanya' from the Centre d'Estudis d'Opinió.\nChart: @joethebrew")
} else {
the_labs <- labs(x = 'Quantes vegades ha viatjat a Catalunya?',
y = 'Grau de simpatia (0-10) cap\nals habitants de Catalunya',
title = "Vegades que ha viatjat a Catalunya i\ngrau de simpatia cap als catalans",
caption = "Font de dades: Enquesta 'Percepció sobre el debat territorial a Espanya' del Centre d'Estudis d'Opinió.\nGràfic: @joethebrew")
}
library(databrew)
ggplot(data = pd,
aes(x = visits,
y = simpatia)) +
# geom_bar(stat = 'identity') +
# geom_area(alpha = 0.5, color = NA, fill = 'darkorange', group = 1) +
geom_point(size = 10) +
geom_segment(aes(xend = visits,
yend = 5),
alpha = 0.8,
# size = 2,
lty = 3) +
geom_line(alpha = 0.2, group = 1) +
geom_text(aes(label = round(simpatia, digits = 1)),
# nudge_y = 0.2,
# alpha = 0.8,
color = 'white',
size = 3.6) +
the_labs +
theme_simple(black = T, base_size = 16) +
theme(plot.title = element_text(size = 22)) +
ylim(5, 7.2)
}
plot_viatges_i_odi <- function(ca = TRUE){
# Viatges a Cat i favorabilitat e Cats
pd <- pt %>%
filter(ccaa != 'Catalunya') %>%
filter(!is.na(visits)) %>%
filter(!visits %in% c('No ho sap', 'No contesta')) %>%
group_by(visits) %>%
summarise(avg = mean(simpatia, na.rm = TRUE),
n = n(),
n_no_na = length(which(!is.na(simpatia))),
n0 = length(which(simpatia == 0))) %>%
mutate(p0 = n0 / n_no_na * 100)
pd$visits <- factor(pd$visits,
levels = unique(c('No, mai',
levels(pd$visits))))
pd$visits <- factor(pd$visits,
levels = levels(pd$visits),
labels = gsub(' vegad', '\nvegad', c('Mai', levels(pd$visits)[2:length(levels(pd$visits))])))
if(!ca){
pd$visits <- factor(pd$visits,
labels = c('Never', 'Just\nonce',
'2-5 times',
'6-10 times',
'More than\n10 times'))
the_labs <- labs(x = 'How many times have you travelled to Catalonia?',
y = "Percentage who say 'I don't like them at all'",
title = "Extreme dislike of Catalans among Spaniards\nand times travelled to Catalonia",
caption = "Data source: Survey 'Percepció sobre el debat territorial a Espanya' from the Centre d'Estudis d'Opinió.\nChart: @joethebrew")
} else {
the_labs <- labs(x = 'Quantes vegades ha viatjat a Catalunya?',
y = "Percentatge qui diu que els catalans\n'em cauen molt malament'",
title = "Aversió extrema dels catalans i\nvegades que ha viatjat a Catalunya",
caption = "Font de dades: Enquesta 'Percepció sobre el debat territorial a Espanya' del Centre d'Estudis d'Opinió.\nGràfic: @joethebrew")
}
library(databrew)
ggplot(data = pd,
aes(x = visits,
y = p0)) +
# geom_bar(stat = 'identity') +
# geom_area(alpha = 0.5, color = NA, fill = 'darkorange', group = 1) +
geom_point(size = 10) +
geom_segment(aes(xend = visits,
yend = 5),
alpha = 0.8,
# size = 2,
lty = 3) +
geom_line(alpha = 0.2, group = 1) +
geom_text(aes(label = round(p0, digits = 1)),
# nudge_y = 0.2,
# alpha = 0.8,
color = 'white',
size = 3.6) +
the_labs +
theme_simple(black = T, base_size = 16) +
theme(plot.title = element_text(size = 22)) +
ylim(3.5, 12)
}
plot_support_referendum <- function(ca = FALSE){
# Viatges a Cat i suport pel referendum
pd <- pt %>%
filter(`Comunitat autònoma` != 'Catalunya') %>%
mutate(visits = `I d’aquestes, en concret, em podria dir quantes vegades a viatjat a Catalunya?`) %>%
filter(!is.na(visits)) %>%
filter(!visits %in% c('No ho sap', 'No contesta')) %>%
mutate(referendum = as.character(referendum)) %>%
mutate(referendum = ifelse(referendum %in% c('Nc', 'Nhs'),
'NS/NC', referendum)) %>%
group_by(visits, referendum) %>%
tally %>%
group_by(visits) %>%
mutate(p = n / sum(n) * 100) %>%
ungroup
pd$visits <- factor(pd$visits,
levels = unique(c('No, mai',
levels(pd$visits))))
pd$visits <- factor(pd$visits,
levels = levels(pd$visits),
labels = gsub(' vegad', '\nvegad', c('Mai', levels(pd$visits)[2:length(levels(pd$visits))])))
pd$referendum <- as.character(pd$referendum)
pd$referendum <- ifelse(pd$referendum %in% c('Nhs', 'Nc'),
'NS/NC', pd$referendum)
pd$referendum <- factor(pd$referendum,
levels = c("Molt d’acord",
"Més aviat d’acord",
"NS/NC",
"Ni d’acord ni en desacord",
"Més aviat en desacord",
"Molt en desacord"))
pd$referendum <- factor(pd$referendum,
levels = rev(levels(pd$referendum)))
if(!ca){
pd$visits <- factor(pd$visits,
labels = c('Never', 'Just\nonce',
'2-5 times',
'6-10 times',
'More than\n10 times'))
pd$referendum <- factor(pd$referendum,
labels = rev(c("Strongly agree",
"Agree",
"Don't know/No answer",
"Neither agree nor disagree",
"Disagree",
"Strongly disagree")))
the_labs <- labs(x = 'How many times have you travelled to Catalonia?',
y = 'Percentage',
title = "Times travelled to Catalonia\nand support for referendum",
subtitle = "Agreement with phrase: 'A referendum should take place so Catalans\ncan decide the relationship they want between Catalonia and the Spanish State'",
caption = "Data source: survey 'Percepció sobre el debat territorial a Espanya' from the Centre d'Estudis d'Opinió.\nChart: @joethebrew")
legend_title <- ''
} else {
the_labs <-
labs(x = 'Quantes vegades ha viatjat a Catalunya?',
y = 'Percentatge',
title = "Vegades que ha viatjat a Catalunya i\nsuport al referèndum",
subtitle = "Grau d'acord amb la frase: 'S'hauria de fer un referèndum a Catalunya perquè els catalans i les\ncatalanes decidissin quina relació volen que hi hagi entre Catalunya i l'Estat espanyol'",
caption = "Font de dades: Enquesta 'Percepció sobre el debat territorial a Espanya'. Centre d'Estudis d'Opinió.\nGràfic: @joethebrew")
legend_title <- ""
}
library(databrew)
library(RColorBrewer)
cols <- brewer.pal(n = 6, 'Spectral')
cols[3] <- grey(c(0.5))
ggplot(data = pd,
aes(x = visits,
y = p)) +
geom_bar(stat = 'identity',
aes(fill = referendum),
alpha = 0.9) +
# geom_text(aes(label = round(simpatia, digits = 1)),
# # nudge_y = 0.2,
# # alpha = 0.8,
# color = 'white',
# size = 3.6) +
the_labs +
theme_simple(black = T, base_size = 16) +
theme(plot.title = element_text(size = 22),
plot.caption = element_text(hjust = 0),
axis.text.x = element_text(size = 8),
plot.subtitle = element_text(size = 10)) +
theme(legend.position = 'right') +
scale_fill_manual(name = legend_title,
values = cols) +
geom_hline(yintercept = 50, lty = 2, alpha = 0.7)
}
# Grau de simpatia i suport pel referendum
plot_simpatia_referendum <- function(ca = TRUE){
pd <- pt %>%
filter(`Comunitat autònoma` != 'Catalunya') %>%
mutate(simpatia = `Grau de simpatia cap als habitants de cadascuna de les diferents comunitats autònomes: Catalans`) %>%
mutate(simpatia = simpatia_cleaner(simpatia)) %>%
mutate(referendum = as.character(referendum)) %>%
mutate(referendum = ifelse(referendum %in% c('Nc', 'Nhs'),
'NS/NC', referendum)) %>%
group_by(referendum, simpatia) %>%
tally %>%
group_by(simpatia) %>%
mutate(p = n / sum(n) * 100) %>%
ungroup
pd$referendum <- as.character(pd$referendum)
pd$referendum <- ifelse(pd$referendum %in% c('Nhs', 'Nc'),
'NS/NC', pd$referendum)
pd$referendum <- factor(pd$referendum,
levels = c("Molt d’acord",
"Més aviat d’acord",
"NS/NC",
"Ni d’acord ni en desacord",
"Més aviat en desacord",
"Molt en desacord"))
pd$referendum <- factor(pd$referendum,
levels = rev(levels(pd$referendum)))
if(!ca){
pd$referendum <- factor(pd$referendum,
labels = c("Strongly agree",
"Agree",
"Don't know/No answer",
"Neither agree nor disagree",
"Disagree",
"Strongly disagree"))
the_labs <- labs(x = 'How much do you like Catalans (0-10)?',
y = 'Percentage',
title = "Sympathy/antipathy for Catalans\nand support for referendum",
subtitle = "Agreement with phrase: 'A referendum should take place so Catalans\ncan decide the relationship they want between Catalonia and the Spanish State'",
caption = "Data source: survey 'Percepció sobre el debat territorial a Espanya' from the Centre d'Estudis d'Opinió.\nChart: @joethebrew")
legend_title <- ''
} else {
the_labs <- labs(x = 'Grau de simpatia (escala 0 a 10)',
y = 'Percentatge',
title = "Grau de simpatia pels catalans i\nsuport al referèndum",
subtitle = "Grau d'acord amb la frase: 'S'hauria de fer un referèndum a Catalunya perquè els catalans i les\ncatalanes decidissin quina relació volen que hi hagi entre Catalunya i l'Estat espanyol'",
caption = "Font de dades: Enquesta 'Percepció sobre el debat territorial a Espanya. Centre d'Estudis d'Opinió.\nGràfic: @joethebrew")
}
cols <- brewer.pal(n = 6, 'Spectral')
cols[3] <- grey(c(0.5))
ggplot(data = pd,
aes(x = simpatia,
y = p)) +
geom_bar(stat = 'identity',
aes(fill = referendum),
alpha = 0.9) +
# geom_text(aes(label = round(simpatia, digits = 1)),
# # nudge_y = 0.2,
# # alpha = 0.8,
# color = 'white',
# size = 3.6) +
the_labs +
theme_simple(black = T, base_size = 16) +
theme(plot.title = element_text(size = 22),
plot.caption = element_text(hjust = 0),
axis.text.x = element_text(size = 8),
plot.subtitle = element_text(size = 10)) +
theme(legend.position = 'right') +
scale_fill_manual(name = '',
values = cols) +
geom_hline(yintercept = 50, lty = 2, alpha = 0.7) +
scale_x_continuous(breaks = 0:10)
}
plot_que_es_catalunya <- function(){
# Que és Catalunya
pd <- pt %>%
filter(`Comunitat autònoma` != 'Catalunya') %>%
mutate(visits = `I d’aquestes, en concret, em podria dir quantes vegades a viatjat a Catalunya?`) %>%
filter(!is.na(visits)) %>%
filter(!visits %in% c('No ho sap', 'No contesta')) %>%
mutate(what_is_cat = as.character(`I parlant de Catalunya, quin terme preferiu per a referir-vos a Catalunya?`)) %>%
mutate(what_is_cat = ifelse(what_is_cat %in% c('No ho sap', 'No contesta', 'Altres (especificar)'), 'NS/NC/Altres', what_is_cat)) %>%
group_by(visits, what_is_cat) %>%
tally %>%
group_by(visits) %>%
mutate(p = n / sum(n) * 100) %>%
ungroup
pd$visits <- factor(pd$visits,
levels = unique(c('No, mai',
levels(pd$visits))))
pd$visits <- factor(pd$visits,
levels = levels(pd$visits),
labels = gsub(' vegad', '\nvegad', c('Mai', levels(pd$visits)[2:length(levels(pd$visits))])))
pd$what_is_cat <- factor(pd$what_is_cat,
levels = c("Un país",
"Una nació",
"Una nacionalitat",
"NS/NC/Altres",
"Una comunitat autònoma",
"Una regió"))
pd$what_is_cat <- factor(pd$what_is_cat, levels = rev(levels(pd$what_is_cat)))
library(databrew)
library(RColorBrewer)
cols <- brewer.pal(n = 6, 'Spectral')
# cols[2:3] <- 'white'
cols[2:3] <- grey(c(0.4, 0.7))
ggplot(data = pd,
aes(x = visits,
y = p)) +
geom_bar(stat = 'identity',
aes(fill = what_is_cat),
alpha = 0.9) +
# geom_text(aes(label = round(simpatia, digits = 1)),
# # nudge_y = 0.2,
# # alpha = 0.8,
# color = 'white',
# size = 3.6) +
labs(x = 'Quantes vegades ha viatjat a Catalunya?',
y = 'Percentatge',
title = "Vegades que ha viatjat a Catalunya i\nopinió sobre què és Catalunya",
subtitle = "xxxyyyzzz",
caption = "Font de dades: Enquesta 'Percepció sobre el debat territorial a Espanya. Centre d'Estudis d'Opinió.\n. Gràfic: @joethebrew") +
theme_simple(black = T, base_size = 16) +
theme(plot.title = element_text(size = 22),
plot.caption = element_text(hjust = 0),
axis.text.x = element_text(size = 8),
plot.subtitle = element_text(size = 10)) +
theme(legend.position = 'right') +
scale_fill_manual(name = "Grau d'acord",
values = cols) +
geom_hline(yintercept = 50, lty = 2, alpha = 0.7)
}
# Grau de simpatia cap als habitants
plot_simpatia_matrix <- function(ca = FALSE, var = 'avg', roundy = 1, reverse_color = FALSE, text_size = 4){
library(RColorBrewer)
pd <- pt %>%
mutate(weight = `Coeficients de ponderació`) %>%
mutate(ccaa = `Comunitat autònoma`) %>%
dplyr::select(contains('Grau de simpatia cap als habitants'),
contains('ccaa'),
weight)
names(pd) <- gsub("Grau de simpatia cap als habitants de cadascuna de les diferents comunitats autònomes: ", '', names(pd))
# Make long
pd <- pd %>% gather(key, value, Andalusos:Ceutencs)
pd$value <- simpatia_cleaner(pd$value)
pda <- pd
# Now agg and get averages
pd <- pd %>%
filter(!ccaa %in% c('Ceuta', 'Melilla'),
!key %in% c('Ceutencs')) %>%
group_by(ccaa, key) %>%
summarise(n = n(),
n_no_na = length(which(!is.na(value))),
avg = mean(value, na.rm = TRUE),
avgx = weighted.mean(x = value, w = weight, na.rm = TRUE),
p75 = quantile(value, 0.75, na.rm = TRUE),
p25 = quantile(value, 0.25, na.rm = TRUE),
n_zero = length(which(value == 0)),
n1 = length(which(value <= 1)),
n2 = length(which(value <= 2)),
n3 = length(which(value <= 3)),
n4 = length(which(value <=4)),
n5 = length(which(value <= 5)),
n6 = length(which(value <= 6)),
n10 = length(which(value == 10))) %>%
mutate(p0 = n_zero / n_no_na * 100,
p1 = n1 / n_no_na * 100,
p2 = n2 / n_no_na * 100,
p3 = n3 / n_no_na * 100,
p4 = n4 / n_no_na * 100,
p5 = n5 / n_no_na * 100,
p10 = n10 / n_no_na * 100)
pd <- pd %>% ungroup %>%
mutate(gent_de = fix_levels(ccaa, from_place = TRUE, to_place = TRUE),
cap_a = fix_levels(key, from_place = FALSE, to_place = TRUE))
if(ca){
the_labs <- labs(x = 'Grau de simpatia de les persones d\'aqui',
y = '...cap a les persones d\'aqui',
title = 'Grau de simpatia cap als habitants dels\nhabitants segons comunitat autònoma',
caption = "Font de dades: Enquesta 'Percepció sobre el debat territorial a Espanya' del Centre d'Estudis d'Opinió.\nGràfic: @joethebrew")
} else {
the_labs <- labs(x = 'How people from here feel...',
y = '...about people from here',
title = "How people from one area feel\nabout people from another area",
caption = "Data source: Survey 'Percepció sobre el debat territorial a Espanya' from the Centre d'Estudis d'Opinió.\nChart: @joethebrew")
}
# Define the variable
pd$y <- as.numeric(unlist(pd[,var] ))
# Colors
cols <- colorRampPalette(brewer.pal(n = 9, name = 'Spectral'))(10)
if(reverse_color){
cols <- rev(cols)
}
ggplot(data = pd,
aes(x = gent_de,
y = cap_a)) +
geom_point(size = 7.5,
aes(color = y)) +
# geom_point(size = 5,
# aes(#size = avg,
# # color = p2
# # color = avg
# )) +
# ggthemes::theme_fivethirtyeight() +
scale_color_gradientn(colours = cols) +
the_labs +
theme_simple() +
theme(axis.text.x = element_text(angle = 90,
hjust = 1)) +
geom_text(aes(label = round(y, digits = roundy)), size = text_size) +
theme(legend.position = 'none')
}
# Get average
get_avg <- function(where_from = 'Catalunya',
var = 'avg',
where_to = label_df$ccaa[label_df$ccaa != 'Catalunya']){
pd <- pt %>%
dplyr::select(contains('Grau de simpatia cap als habitants'),
contains('ccaa'),
contains('weight'))
names(pd) <- gsub("Grau de simpatia cap als habitants de cadascuna de les diferents comunitats autònomes: ", '', names(pd))
# Make long
pd <- pd %>% gather(key, value, Andalusos:Ceutencs)
pd$value <- simpatia_cleaner(pd$value)
# Now agg and get averages
pd <- pd %>%
# filter(!ccaa %in% c('Ceuta', 'Melilla'),
# !key %in% c('Ceutencs')) %>%
group_by(ccaa, key) %>%
summarise(n = n(),
n_no_na = length(which(!is.na(value))),
avg = mean(value, na.rm = TRUE),
p75 = quantile(value, 0.75, na.rm = TRUE),
p25 = quantile(value, 0.25, na.rm = TRUE),
n_zero = length(which(value == 0)),
n1 = length(which(value <= 1)),
n2 = length(which(value <= 2)),
n3 = length(which(value <= 3)),
n4 = length(which(value <=4)),
n5 = length(which(value <= 5)),
n6 = length(which(value <= 6)),
n10 = length(which(value == 10))) %>%
mutate(p0 = n_zero / n_no_na * 100,
p1 = n1 / n_no_na * 100,
p2 = n2 / n_no_na * 100,
p3 = n3 / n_no_na * 100,
p4 = n4 / n_no_na * 100,
p5 = n5 / n_no_na * 100,
p10 = n10 / n_no_na * 100)
pd <- pd %>%
mutate(gent_de = fix_levels(ccaa, from_place = TRUE, to_place = TRUE),
cap_a = fix_levels(key, from_place = FALSE, to_place = TRUE))
pd$y <- as.numeric(unlist(pd[,var]))
pd %>% ungroup %>%
filter(gent_de %in% where_from,
cap_a %in% where_to) %>%
summarise(avg = mean(y, na.rm = T)) %>%
.$avg
}
# For each ccaa, what do the people there think of catalans compared to what catalans think of them
plot_compare <- function(ca = TRUE){
pd <- pt %>%
dplyr::select(contains('Grau de simpatia cap als habitants'),
contains('ccaa'),
contains('weight'))
names(pd) <- gsub("Grau de simpatia cap als habitants de cadascuna de les diferents comunitats autònomes: ", '', names(pd))
# Make long
pd <- pd %>% gather(key, value, Andalusos:Ceutencs)
pd$value <- simpatia_cleaner(pd$value)
# Rename
pd <- pd %>%
mutate(gent_de = fix_levels(ccaa, from_place = TRUE, to_place = TRUE),
cap_a = fix_levels(key, from_place = FALSE, to_place = TRUE)) %>%
dplyr::select(gent_de,
cap_a, value) %>%
group_by(gent_de, cap_a) %>%
summarise(value = mean(value, na.rm = TRUE))
# Keep only what people think about catalans or vice-versa
pd <- pd %>%
filter(gent_de == 'Catalunya' | cap_a == 'Catalunya')
# Define the ccaa
pd$cap_a <- as.character(pd$cap_a); pd$gent_de <- as.character(pd$gent_de)
pd$ccaa <- ifelse(pd$cap_a == 'Catalunya', pd$gent_de,
ifelse(pd$gent_de == 'Catalunya', pd$cap_a,
NA))
pd <- pd %>%
filter(!(cap_a == 'Catalunya' & gent_de == 'Catalunya')) %>%
filter(!ccaa %in% c('Ceuta', 'Melilla'))
# Make a new var
if(ca){
levs <- c('Valoració mitjana dels habitants d\'aquesta\ncomunitat autònoma cap als catalans',
'Valoració mitjana dels catalans cap als\nhabitants d\'aquesta comunitat autònoma')
the_labs <- labs(x = '',
y = "Valoració mitjana (0-10)",
title = 'Grau de simpatia/antipatia entre catalans\ni espanyols de diferents comunitats autònomes')
} else {
levs <- c('How people from this area feel\nabout Catalans',
'How Catalans feel about people\nfrom this area')
the_labs <- labs(x = '',
y = "Degree to which they 'like' the other (0-10)",
title = 'Degree to which Catalans and Spaniards\nof different regions like/dislike one another')
}
pd$var <- ifelse(pd$cap_a == 'Catalunya', levs[1], levs[2])
pd$var <- factor(pd$var, levels = levs)
fl <- sort(unique(pd$ccaa))
pd$ccaa <- factor(pd$ccaa, levels = ,
labels = gsub('Castella', 'Cas.', gsub('/País Basc','', fl)))
ggplot(data = pd,
aes(x = var,
y = value)) +
geom_line(aes(group = ccaa), alpha = 0.7) +
geom_point(aes(color = var), size = 9) +
facet_wrap(~ccaa) +
theme_simple() +
theme(axis.text.x = element_blank()) +
scale_color_manual(name = '',
values = c('darkorange', 'lightblue')) +
geom_text(aes(label = round(value, digits = 1))) +
the_labs +
ylim(min(pd$value) - 1,
max(pd$value) + 1) +
theme(legend.position = 'top')
}
# Autonomia
# Pel que fa a les relacions entre Catalunya i Espanya, creieu que Catalunya ha assolit...
# pt$`Pel que fa a les relacions entre Catalunya i la resta de l’Estat espanyol, què diríeu que és més necessari?`
joebrew/vilaweb documentation built on Sept. 11, 2020, 3:42 a.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
# Libraries
library(vilaweb)
library(tidyverse)
library(lubridate)
library(databrew)
library(tidylog)
options(scipen = 999)
plot_twitter <- function(ca = TRUE){
pd <- fobia %>% filter(key == 'Hispanofòbia') %>%
group_by(date = as.Date(cut(date, 'month'))) %>%
summarise(n = n(),
rt = sum(retweets_count))
if(ca){
the_labs <- labs(x = 'Mes',
y = 'Piulets (escala logarítmica)',
title = "Piulets amb la paraula de 'hispanofòbia'",
subtitle = 'Dades de Twitter, incloent-hi repiulets',
caption = 'Gràfic de Joe Brew, @joethebrew')
} else {
the_labs <- labs(x = 'Month',
y = 'Piulets (log scale)',
title = "Tweets with the word 'Hispanophobia'",
subtitle = 'Data from Twitter, including retweets',
caption = 'Chart by Joe Brew, @joethebrew.')
}
ggplot(data = pd,
aes(x = date,
y = rt + n)) +
geom_line() +
# geom_path(color = 'darkorange', alpha = 0.6) +
theme_simple() +
scale_y_log10() +
the_labs
}
plot_google <- function(ca = TRUE){
# Read in google trends data (retrieved 23 jan 2020)
google <- read_csv('multiTimeline.csv', skip = 1)
google$date <- as.Date(paste0(google$Month, '-01'))
# google$date <- as.Date(cut(google$date, 'year'))
google$value <- google$`hispanofobia: (Spain)`
if(ca){
the_labs <- labs(x = 'Mes',
y = 'Percentatge (escala logarítmica)',
title = "Cerques de 'hispanofòbia'",
subtitle = 'Dades de Google Trends',
caption = 'Percentatge: cerques mensuals relatives al màxim. Gràfic de Joe Brew, @joethebrew')
} else {
the_labs <- labs(x = 'Month',
y = 'Percentage (log scale)',
title = "Searches for 'Hispanophobia'",
subtitle = 'Data from Google Trends',
caption = 'Percentage: monthly searches relative to the maximum. Chart by Joe Brew, @joethebrew.')
}
pd <- google %>%
group_by(date) %>%
summarise(value = mean(value))
ggplot(data = pd,
aes(x = date,
y = value)) +
geom_line() +
# geom_path(color = 'darkorange', alpha = 0.6) +
theme_simple() +
scale_y_log10(breaks = c(0:3, 5,10,25, 50, 100)) +
the_labs
}
# Read in the territorial survey data
pt <- vilaweb::ceo_percepcio_territorial
# Function for cleaning up the sympathy variable
simpatia_cleaner <- function(x){
x <- as.character(x)
x <- ifelse(x %in% c('No contesta', 'No ho sap'), NA,
ifelse(x == 'Us cauen molt bé 10', '10',
ifelse(x == 'No us cauen ni bé ni malament 5', 5,
ifelse(x == 'Us cauen molt malament 0', '0',
x))))
x <- as.numeric(x)
return(x)
}
pt <- pt %>%
mutate(visits = `I d’aquestes, en concret, em podria dir quantes vegades a viatjat a Catalunya?`,
simpatia = `Grau de simpatia cap als habitants de cadascuna de les diferents comunitats autònomes: Catalans`,
referendum = `Grau d'acord: S’hauria de fer un referèndum a Catalunya perquè els catalans i les catalanes decidissin quina relació volen que hi hagi entre Catalunya i l’Estat espanyol`) %>%
mutate(simpatia = simpatia_cleaner(simpatia)) %>%
mutate(ccaa = `Comunitat autònoma`)
# # Companeros de trabajo
# pt$`Amb independència que treballeu o no, quins són els tres que menys us agradaria tenir com a companys de feina o treball? Catalans`
#
# # Grau de simpatia
# pt$`Grau de simpatia cap als habitants de cadascuna de les diferents comunitats autònomes: Catalans`
# Define the label sets (for factor ordering)
label_df <- tibble(
ccaa = c('Andalusia',
'Aragó',
'Astúries',
'Balears',
'Canàries',
'Cantàbria',
'Castella la Manxa',
'Castella i Lleó',
'Catalunya',
'País Valencià',
'Extremadura',
'Galícia',
'Madrid',
'Múrcia',
'Navarra',
'Euskadi/País Basc',
'La Rioja', 'Ceuta', 'Melilla'),
key = c('Andalusos',
'Aragonesos',
'Asturians',
'Balears',
'Canaris',
'Càntabres',
'Castellanomanxecs',
'Castellanolleonesos',
'Catalans',
'Valencians',
'Extremenys',
'Gallecs',
'Madrilenys',
'Murcians',
'Navarresos',
'Bascos',
'Riojans',
'Ceutencs',
'Melillencs')
)
fix_levels <- function(var, from_place = FALSE, to_place = TRUE){
if(from_place){
x <- tibble(ccaa = var)
} else {
x <- tibble(key = var)
}
# Join with the labels
joined <- left_join(x, label_df)
# Get the return val
if(to_place){
out <- joined$ccaa
out <- factor(out, levels = label_df$ccaa)
} else {
out <- joined$key
out <- factor(out, levels = label_df$key)
}
return(out)
}
plot_viatges_i_simpatia <- function(ca = TRUE){
# Viatges a Cat i favorabilitat e Cats
pd <- pt %>%
filter(ccaa != 'Catalunya') %>%
filter(!is.na(visits)) %>%
filter(!visits %in% c('No ho sap', 'No contesta')) %>%
group_by(visits) %>%
summarise(simpatia = mean(simpatia, na.rm = TRUE),
n = n())
pd$visits <- factor(pd$visits,
levels = unique(c('No, mai',
levels(pd$visits))))
pd$visits <- factor(pd$visits,
levels = levels(pd$visits),
labels = gsub(' vegad', '\nvegad', c('Mai', levels(pd$visits)[2:length(levels(pd$visits))])))
if(!ca){
pd$visits <- factor(pd$visits,
labels = c('Never', 'Just\nonce',
'2-5 times',
'6-10 times',
'More than\n10 times'))
the_labs <- labs(x = 'How many times have you travelled to Catalonia?',
y = 'Degree of sympathy (0-10) for\nthe residents of Catalonia',
title = "Times travelled to Catalonia and\nfeelings about Catalans",
caption = "Data source: Survey 'Percepció sobre el debat territorial a Espanya' from the Centre d'Estudis d'Opinió.\nChart: @joethebrew")
} else {
the_labs <- labs(x = 'Quantes vegades ha viatjat a Catalunya?',
y = 'Grau de simpatia (0-10) cap\nals habitants de Catalunya',
title = "Vegades que ha viatjat a Catalunya i\ngrau de simpatia cap als catalans",
caption = "Font de dades: Enquesta 'Percepció sobre el debat territorial a Espanya' del Centre d'Estudis d'Opinió.\nGràfic: @joethebrew")
}
library(databrew)
ggplot(data = pd,
aes(x = visits,
y = simpatia)) +
# geom_bar(stat = 'identity') +
# geom_area(alpha = 0.5, color = NA, fill = 'darkorange', group = 1) +
geom_point(size = 10) +
geom_segment(aes(xend = visits,
yend = 5),
alpha = 0.8,
# size = 2,
lty = 3) +
geom_line(alpha = 0.2, group = 1) +
geom_text(aes(label = round(simpatia, digits = 1)),
# nudge_y = 0.2,
# alpha = 0.8,
color = 'white',
size = 3.6) +
the_labs +
theme_simple(black = T, base_size = 16) +
theme(plot.title = element_text(size = 22)) +
ylim(5, 7.2)
}
plot_viatges_i_odi <- function(ca = TRUE){
# Viatges a Cat i favorabilitat e Cats
pd <- pt %>%
filter(ccaa != 'Catalunya') %>%
filter(!is.na(visits)) %>%
filter(!visits %in% c('No ho sap', 'No contesta')) %>%
group_by(visits) %>%
summarise(avg = mean(simpatia, na.rm = TRUE),
n = n(),
n_no_na = length(which(!is.na(simpatia))),
n0 = length(which(simpatia == 0))) %>%
mutate(p0 = n0 / n_no_na * 100)
pd$visits <- factor(pd$visits,
levels = unique(c('No, mai',
levels(pd$visits))))
pd$visits <- factor(pd$visits,
levels = levels(pd$visits),
labels = gsub(' vegad', '\nvegad', c('Mai', levels(pd$visits)[2:length(levels(pd$visits))])))
if(!ca){
pd$visits <- factor(pd$visits,
labels = c('Never', 'Just\nonce',
'2-5 times',
'6-10 times',
'More than\n10 times'))
the_labs <- labs(x = 'How many times have you travelled to Catalonia?',
y = "Percentage who say 'I don't like them at all'",
title = "Extreme dislike of Catalans among Spaniards\nand times travelled to Catalonia",
caption = "Data source: Survey 'Percepció sobre el debat territorial a Espanya' from the Centre d'Estudis d'Opinió.\nChart: @joethebrew")
} else {
the_labs <- labs(x = 'Quantes vegades ha viatjat a Catalunya?',
y = "Percentatge qui diu que els catalans\n'em cauen molt malament'",
title = "Aversió extrema dels catalans i\nvegades que ha viatjat a Catalunya",
caption = "Font de dades: Enquesta 'Percepció sobre el debat territorial a Espanya' del Centre d'Estudis d'Opinió.\nGràfic: @joethebrew")
}
library(databrew)
ggplot(data = pd,
aes(x = visits,
y = p0)) +
# geom_bar(stat = 'identity') +
# geom_area(alpha = 0.5, color = NA, fill = 'darkorange', group = 1) +
geom_point(size = 10) +
geom_segment(aes(xend = visits,
yend = 5),
alpha = 0.8,
# size = 2,
lty = 3) +
geom_line(alpha = 0.2, group = 1) +
geom_text(aes(label = round(p0, digits = 1)),
# nudge_y = 0.2,
# alpha = 0.8,
color = 'white',
size = 3.6) +
the_labs +
theme_simple(black = T, base_size = 16) +
theme(plot.title = element_text(size = 22)) +
ylim(3.5, 12)
}
plot_support_referendum <- function(ca = FALSE){
# Viatges a Cat i suport pel referendum
pd <- pt %>%
filter(`Comunitat autònoma` != 'Catalunya') %>%
mutate(visits = `I d’aquestes, en concret, em podria dir quantes vegades a viatjat a Catalunya?`) %>%
filter(!is.na(visits)) %>%
filter(!visits %in% c('No ho sap', 'No contesta')) %>%
mutate(referendum = as.character(referendum)) %>%
mutate(referendum = ifelse(referendum %in% c('Nc', 'Nhs'),
'NS/NC', referendum)) %>%
group_by(visits, referendum) %>%
tally %>%
group_by(visits) %>%
mutate(p = n / sum(n) * 100) %>%
ungroup
pd$visits <- factor(pd$visits,
levels = unique(c('No, mai',
levels(pd$visits))))
pd$visits <- factor(pd$visits,
levels = levels(pd$visits),
labels = gsub(' vegad', '\nvegad', c('Mai', levels(pd$visits)[2:length(levels(pd$visits))])))
pd$referendum <- as.character(pd$referendum)
pd$referendum <- ifelse(pd$referendum %in% c('Nhs', 'Nc'),
'NS/NC', pd$referendum)
pd$referendum <- factor(pd$referendum,
levels = c("Molt d’acord",
"Més aviat d’acord",
"NS/NC",
"Ni d’acord ni en desacord",
"Més aviat en desacord",
"Molt en desacord"))
pd$referendum <- factor(pd$referendum,
levels = rev(levels(pd$referendum)))
if(!ca){
pd$visits <- factor(pd$visits,
labels = c('Never', 'Just\nonce',
'2-5 times',
'6-10 times',
'More than\n10 times'))
pd$referendum <- factor(pd$referendum,
labels = rev(c("Strongly agree",
"Agree",
"Don't know/No answer",
"Neither agree nor disagree",
"Disagree",
"Strongly disagree")))
the_labs <- labs(x = 'How many times have you travelled to Catalonia?',
y = 'Percentage',
title = "Times travelled to Catalonia\nand support for referendum",
subtitle = "Agreement with phrase: 'A referendum should take place so Catalans\ncan decide the relationship they want between Catalonia and the Spanish State'",
caption = "Data source: survey 'Percepció sobre el debat territorial a Espanya' from the Centre d'Estudis d'Opinió.\nChart: @joethebrew")
legend_title <- ''
} else {
the_labs <-
labs(x = 'Quantes vegades ha viatjat a Catalunya?',
y = 'Percentatge',
title = "Vegades que ha viatjat a Catalunya i\nsuport al referèndum",
subtitle = "Grau d'acord amb la frase: 'S'hauria de fer un referèndum a Catalunya perquè els catalans i les\ncatalanes decidissin quina relació volen que hi hagi entre Catalunya i l'Estat espanyol'",
caption = "Font de dades: Enquesta 'Percepció sobre el debat territorial a Espanya'. Centre d'Estudis d'Opinió.\nGràfic: @joethebrew")
legend_title <- ""
}
library(databrew)
library(RColorBrewer)
cols <- brewer.pal(n = 6, 'Spectral')
cols[3] <- grey(c(0.5))
ggplot(data = pd,
aes(x = visits,
y = p)) +
geom_bar(stat = 'identity',
aes(fill = referendum),
alpha = 0.9) +
# geom_text(aes(label = round(simpatia, digits = 1)),
# # nudge_y = 0.2,
# # alpha = 0.8,
# color = 'white',
# size = 3.6) +
the_labs +
theme_simple(black = T, base_size = 16) +
theme(plot.title = element_text(size = 22),
plot.caption = element_text(hjust = 0),
axis.text.x = element_text(size = 8),
plot.subtitle = element_text(size = 10)) +
theme(legend.position = 'right') +
scale_fill_manual(name = legend_title,
values = cols) +
geom_hline(yintercept = 50, lty = 2, alpha = 0.7)
}
# Grau de simpatia i suport pel referendum
plot_simpatia_referendum <- function(ca = TRUE){
pd <- pt %>%
filter(`Comunitat autònoma` != 'Catalunya') %>%
mutate(simpatia = `Grau de simpatia cap als habitants de cadascuna de les diferents comunitats autònomes: Catalans`) %>%
mutate(simpatia = simpatia_cleaner(simpatia)) %>%
mutate(referendum = as.character(referendum)) %>%
mutate(referendum = ifelse(referendum %in% c('Nc', 'Nhs'),
'NS/NC', referendum)) %>%
group_by(referendum, simpatia) %>%
tally %>%
group_by(simpatia) %>%
mutate(p = n / sum(n) * 100) %>%
ungroup
pd$referendum <- as.character(pd$referendum)
pd$referendum <- ifelse(pd$referendum %in% c('Nhs', 'Nc'),
'NS/NC', pd$referendum)
pd$referendum <- factor(pd$referendum,
levels = c("Molt d’acord",
"Més aviat d’acord",
"NS/NC",
"Ni d’acord ni en desacord",
"Més aviat en desacord",
"Molt en desacord"))
pd$referendum <- factor(pd$referendum,
levels = rev(levels(pd$referendum)))
if(!ca){
pd$referendum <- factor(pd$referendum,
labels = c("Strongly agree",
"Agree",
"Don't know/No answer",
"Neither agree nor disagree",
"Disagree",
"Strongly disagree"))
the_labs <- labs(x = 'How much do you like Catalans (0-10)?',
y = 'Percentage',
title = "Sympathy/antipathy for Catalans\nand support for referendum",
subtitle = "Agreement with phrase: 'A referendum should take place so Catalans\ncan decide the relationship they want between Catalonia and the Spanish State'",
caption = "Data source: survey 'Percepció sobre el debat territorial a Espanya' from the Centre d'Estudis d'Opinió.\nChart: @joethebrew")
legend_title <- ''
} else {
the_labs <- labs(x = 'Grau de simpatia (escala 0 a 10)',
y = 'Percentatge',
title = "Grau de simpatia pels catalans i\nsuport al referèndum",
subtitle = "Grau d'acord amb la frase: 'S'hauria de fer un referèndum a Catalunya perquè els catalans i les\ncatalanes decidissin quina relació volen que hi hagi entre Catalunya i l'Estat espanyol'",
caption = "Font de dades: Enquesta 'Percepció sobre el debat territorial a Espanya. Centre d'Estudis d'Opinió.\nGràfic: @joethebrew")
}
cols <- brewer.pal(n = 6, 'Spectral')
cols[3] <- grey(c(0.5))
ggplot(data = pd,
aes(x = simpatia,
y = p)) +
geom_bar(stat = 'identity',
aes(fill = referendum),
alpha = 0.9) +
# geom_text(aes(label = round(simpatia, digits = 1)),
# # nudge_y = 0.2,
# # alpha = 0.8,
# color = 'white',
# size = 3.6) +
the_labs +
theme_simple(black = T, base_size = 16) +
theme(plot.title = element_text(size = 22),
plot.caption = element_text(hjust = 0),
axis.text.x = element_text(size = 8),
plot.subtitle = element_text(size = 10)) +
theme(legend.position = 'right') +
scale_fill_manual(name = '',
values = cols) +
geom_hline(yintercept = 50, lty = 2, alpha = 0.7) +
scale_x_continuous(breaks = 0:10)
}
plot_que_es_catalunya <- function(){
# Que és Catalunya
pd <- pt %>%
filter(`Comunitat autònoma` != 'Catalunya') %>%
mutate(visits = `I d’aquestes, en concret, em podria dir quantes vegades a viatjat a Catalunya?`) %>%
filter(!is.na(visits)) %>%
filter(!visits %in% c('No ho sap', 'No contesta')) %>%
mutate(what_is_cat = as.character(`I parlant de Catalunya, quin terme preferiu per a referir-vos a Catalunya?`)) %>%
mutate(what_is_cat = ifelse(what_is_cat %in% c('No ho sap', 'No contesta', 'Altres (especificar)'), 'NS/NC/Altres', what_is_cat)) %>%
group_by(visits, what_is_cat) %>%
tally %>%
group_by(visits) %>%
mutate(p = n / sum(n) * 100) %>%
ungroup
pd$visits <- factor(pd$visits,
levels = unique(c('No, mai',
levels(pd$visits))))
pd$visits <- factor(pd$visits,
levels = levels(pd$visits),
labels = gsub(' vegad', '\nvegad', c('Mai', levels(pd$visits)[2:length(levels(pd$visits))])))
pd$what_is_cat <- factor(pd$what_is_cat,
levels = c("Un país",
"Una nació",
"Una nacionalitat",
"NS/NC/Altres",
"Una comunitat autònoma",
"Una regió"))
pd$what_is_cat <- factor(pd$what_is_cat, levels = rev(levels(pd$what_is_cat)))
library(databrew)
library(RColorBrewer)
cols <- brewer.pal(n = 6, 'Spectral')
# cols[2:3] <- 'white'
cols[2:3] <- grey(c(0.4, 0.7))
ggplot(data = pd,
aes(x = visits,
y = p)) +
geom_bar(stat = 'identity',
aes(fill = what_is_cat),
alpha = 0.9) +
# geom_text(aes(label = round(simpatia, digits = 1)),
# # nudge_y = 0.2,
# # alpha = 0.8,
# color = 'white',
# size = 3.6) +
labs(x = 'Quantes vegades ha viatjat a Catalunya?',
y = 'Percentatge',
title = "Vegades que ha viatjat a Catalunya i\nopinió sobre què és Catalunya",
subtitle = "xxxyyyzzz",
caption = "Font de dades: Enquesta 'Percepció sobre el debat territorial a Espanya. Centre d'Estudis d'Opinió.\n. Gràfic: @joethebrew") +
theme_simple(black = T, base_size = 16) +
theme(plot.title = element_text(size = 22),
plot.caption = element_text(hjust = 0),
axis.text.x = element_text(size = 8),
plot.subtitle = element_text(size = 10)) +
theme(legend.position = 'right') +
scale_fill_manual(name = "Grau d'acord",
values = cols) +
geom_hline(yintercept = 50, lty = 2, alpha = 0.7)
}
# Grau de simpatia cap als habitants
plot_simpatia_matrix <- function(ca = FALSE, var = 'avg', roundy = 1, reverse_color = FALSE, text_size = 4){
library(RColorBrewer)
pd <- pt %>%
mutate(weight = `Coeficients de ponderació`) %>%
mutate(ccaa = `Comunitat autònoma`) %>%
dplyr::select(contains('Grau de simpatia cap als habitants'),
contains('ccaa'),
weight)
names(pd) <- gsub("Grau de simpatia cap als habitants de cadascuna de les diferents comunitats autònomes: ", '', names(pd))
# Make long
pd <- pd %>% gather(key, value, Andalusos:Ceutencs)
pd$value <- simpatia_cleaner(pd$value)
pda <- pd
# Now agg and get averages
pd <- pd %>%
filter(!ccaa %in% c('Ceuta', 'Melilla'),
!key %in% c('Ceutencs')) %>%
group_by(ccaa, key) %>%
summarise(n = n(),
n_no_na = length(which(!is.na(value))),
avg = mean(value, na.rm = TRUE),
avgx = weighted.mean(x = value, w = weight, na.rm = TRUE),
p75 = quantile(value, 0.75, na.rm = TRUE),
p25 = quantile(value, 0.25, na.rm = TRUE),
n_zero = length(which(value == 0)),
n1 = length(which(value <= 1)),
n2 = length(which(value <= 2)),
n3 = length(which(value <= 3)),
n4 = length(which(value <=4)),
n5 = length(which(value <= 5)),
n6 = length(which(value <= 6)),
n10 = length(which(value == 10))) %>%
mutate(p0 = n_zero / n_no_na * 100,
p1 = n1 / n_no_na * 100,
p2 = n2 / n_no_na * 100,
p3 = n3 / n_no_na * 100,
p4 = n4 / n_no_na * 100,
p5 = n5 / n_no_na * 100,
p10 = n10 / n_no_na * 100)
pd <- pd %>% ungroup %>%
mutate(gent_de = fix_levels(ccaa, from_place = TRUE, to_place = TRUE),
cap_a = fix_levels(key, from_place = FALSE, to_place = TRUE))
if(ca){
the_labs <- labs(x = 'Grau de simpatia de les persones d\'aqui',
y = '...cap a les persones d\'aqui',
title = 'Grau de simpatia cap als habitants dels\nhabitants segons comunitat autònoma',
caption = "Font de dades: Enquesta 'Percepció sobre el debat territorial a Espanya' del Centre d'Estudis d'Opinió.\nGràfic: @joethebrew")
} else {
the_labs <- labs(x = 'How people from here feel...',
y = '...about people from here',
title = "How people from one area feel\nabout people from another area",
caption = "Data source: Survey 'Percepció sobre el debat territorial a Espanya' from the Centre d'Estudis d'Opinió.\nChart: @joethebrew")
}
# Define the variable
pd$y <- as.numeric(unlist(pd[,var] ))
# Colors
cols <- colorRampPalette(brewer.pal(n = 9, name = 'Spectral'))(10)
if(reverse_color){
cols <- rev(cols)
}
ggplot(data = pd,
aes(x = gent_de,
y = cap_a)) +
geom_point(size = 7.5,
aes(color = y)) +
# geom_point(size = 5,
# aes(#size = avg,
# # color = p2
# # color = avg
# )) +
# ggthemes::theme_fivethirtyeight() +
scale_color_gradientn(colours = cols) +
the_labs +
theme_simple() +
theme(axis.text.x = element_text(angle = 90,
hjust = 1)) +
geom_text(aes(label = round(y, digits = roundy)), size = text_size) +
theme(legend.position = 'none')
}
# Get average
get_avg <- function(where_from = 'Catalunya',
var = 'avg',
where_to = label_df$ccaa[label_df$ccaa != 'Catalunya']){
pd <- pt %>%
dplyr::select(contains('Grau de simpatia cap als habitants'),
contains('ccaa'),
contains('weight'))
names(pd) <- gsub("Grau de simpatia cap als habitants de cadascuna de les diferents comunitats autònomes: ", '', names(pd))
# Make long
pd <- pd %>% gather(key, value, Andalusos:Ceutencs)
pd$value <- simpatia_cleaner(pd$value)
# Now agg and get averages
pd <- pd %>%
# filter(!ccaa %in% c('Ceuta', 'Melilla'),
# !key %in% c('Ceutencs')) %>%
group_by(ccaa, key) %>%
summarise(n = n(),
n_no_na = length(which(!is.na(value))),
avg = mean(value, na.rm = TRUE),
p75 = quantile(value, 0.75, na.rm = TRUE),
p25 = quantile(value, 0.25, na.rm = TRUE),
n_zero = length(which(value == 0)),
n1 = length(which(value <= 1)),
n2 = length(which(value <= 2)),
n3 = length(which(value <= 3)),
n4 = length(which(value <=4)),
n5 = length(which(value <= 5)),
n6 = length(which(value <= 6)),
n10 = length(which(value == 10))) %>%
mutate(p0 = n_zero / n_no_na * 100,
p1 = n1 / n_no_na * 100,
p2 = n2 / n_no_na * 100,
p3 = n3 / n_no_na * 100,
p4 = n4 / n_no_na * 100,
p5 = n5 / n_no_na * 100,
p10 = n10 / n_no_na * 100)
pd <- pd %>%
mutate(gent_de = fix_levels(ccaa, from_place = TRUE, to_place = TRUE),
cap_a = fix_levels(key, from_place = FALSE, to_place = TRUE))
pd$y <- as.numeric(unlist(pd[,var]))
pd %>% ungroup %>%
filter(gent_de %in% where_from,
cap_a %in% where_to) %>%
summarise(avg = mean(y, na.rm = T)) %>%
.$avg
}
# For each ccaa, what do the people there think of catalans compared to what catalans think of them
plot_compare <- function(ca = TRUE){
pd <- pt %>%
dplyr::select(contains('Grau de simpatia cap als habitants'),
contains('ccaa'),
contains('weight'))
names(pd) <- gsub("Grau de simpatia cap als habitants de cadascuna de les diferents comunitats autònomes: ", '', names(pd))
# Make long
pd <- pd %>% gather(key, value, Andalusos:Ceutencs)
pd$value <- simpatia_cleaner(pd$value)
# Rename
pd <- pd %>%
mutate(gent_de = fix_levels(ccaa, from_place = TRUE, to_place = TRUE),
cap_a = fix_levels(key, from_place = FALSE, to_place = TRUE)) %>%
dplyr::select(gent_de,
cap_a, value) %>%
group_by(gent_de, cap_a) %>%
summarise(value = mean(value, na.rm = TRUE))
# Keep only what people think about catalans or vice-versa
pd <- pd %>%
filter(gent_de == 'Catalunya' | cap_a == 'Catalunya')
# Define the ccaa
pd$cap_a <- as.character(pd$cap_a); pd$gent_de <- as.character(pd$gent_de)
pd$ccaa <- ifelse(pd$cap_a == 'Catalunya', pd$gent_de,
ifelse(pd$gent_de == 'Catalunya', pd$cap_a,
NA))
pd <- pd %>%
filter(!(cap_a == 'Catalunya' & gent_de == 'Catalunya')) %>%
filter(!ccaa %in% c('Ceuta', 'Melilla'))
# Make a new var
if(ca){
levs <- c('Valoració mitjana dels habitants d\'aquesta\ncomunitat autònoma cap als catalans',
'Valoració mitjana dels catalans cap als\nhabitants d\'aquesta comunitat autònoma')
the_labs <- labs(x = '',
y = "Valoració mitjana (0-10)",
title = 'Grau de simpatia/antipatia entre catalans\ni espanyols de diferents comunitats autònomes')
} else {
levs <- c('How people from this area feel\nabout Catalans',
'How Catalans feel about people\nfrom this area')
the_labs <- labs(x = '',
y = "Degree to which they 'like' the other (0-10)",
title = 'Degree to which Catalans and Spaniards\nof different regions like/dislike one another')
}
pd$var <- ifelse(pd$cap_a == 'Catalunya', levs[1], levs[2])
pd$var <- factor(pd$var, levels = levs)
fl <- sort(unique(pd$ccaa))
pd$ccaa <- factor(pd$ccaa, levels = ,
labels = gsub('Castella', 'Cas.', gsub('/País Basc','', fl)))
ggplot(data = pd,
aes(x = var,
y = value)) +
geom_line(aes(group = ccaa), alpha = 0.7) +
geom_point(aes(color = var), size = 9) +
facet_wrap(~ccaa) +
theme_simple() +
theme(axis.text.x = element_blank()) +
scale_color_manual(name = '',
values = c('darkorange', 'lightblue')) +
geom_text(aes(label = round(value, digits = 1))) +
the_labs +
ylim(min(pd$value) - 1,
max(pd$value) + 1) +
theme(legend.position = 'top')
}
# Autonomia
# Pel que fa a les relacions entre Catalunya i Espanya, creieu que Catalunya ha assolit...
# pt$`Pel que fa a les relacions entre Catalunya i la resta de l’Estat espanyol, què diríeu que és més necessari?`
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.