#Load libraries
library(tidyverse)
library(ggthemepark)
library(swissdd)
library(swissgd)
library(sf)
library(glm.predict)
library(readxl)#Import data
mentees <- read_excel("Mentees_Frauenzentrale.xls")
zmer <- read_csv(file.choose())
zmer <- zmer %>%
mutate(
party2=
case_when(
party== "JUSO" ~ "SP",
party== "Junge Mitte" ~ "Die Mitte",
party== "JGLP" ~ "GLP",
party %in% c("SVP", "SP", "EVP", "GLP", "FDP", "Grüne", "parteilos", "Die Mitte") ~ party,
T ~ "other"
),
age= 2022-birthyear,
mun_name= ifelse(mun_id== 21, "Andelfingen", mun_name),
mun_id= ifelse(mun_id== 21, 30, mun_id)
) %>%
group_by(type, municipality, round) %>%
mutate(candidates= n()) %>%
ungroup() %>%
mutate(candidates=ifelse(tacit_election == 0, candidates-1,
candidates),
can_per_seat = candidates / mandates)
geo_data <- get_geodata(geolevel = "municipality")%>%
mutate(mun_id= as.numeric(mun_id)) %>%
filter(mun_id <=298)
geo_data2 <- st_read( "C:\\Users\\nickg\\OneDrive - Universität Zürich UZH\\Documents\\Statistics\\Datenjournalismus\\SHAPEFILE_LV95_LN02\\swissBOUNDARIES3D_1_3_TLM_HOHEITSGEBIET.shp") %>%
filter(BFS_NUMMER <=298) %>%
rename(mun_id= BFS_NUMMER)
geo_data2 <- st_read(file.choose()) %>%
filter(BFS_NUMMER <=298) %>%
rename(mun_id= BFS_NUMMER)
geo_data2 %>%
select(mun_id, NAME)
mun_name <- unique(zmer$mun_name)
mun_id <- unique(zmer$mun_id)
info <- data.frame(name= mun_name, mun_id=mun_id)
geo_data3 <- left_join(geo_data2, info) %>%
select(name, NAME, mun_id)zmer %>%
filter(type=="GE" | type2== "SchP" )%>%
#filter(!(type== "PS" & unified != "unified: opt 1")) %>%
mutate( elected_ge=
case_when(
type2== "GE" & elected==1 ~ 1,
type2== "SchP" & elected_pres==1 & unified == "unified: opt 1" ~1,
T~0
)) %>%
filter(elected_ge ==1)%>%
group_by(mun_name, mun_id)%>%
mutate(sex2 = ifelse(sex=="m", 0, 1)) %>%
summarise(n=sum(elected_ge),fem_quote = mean(sex2))%>%
arrange(fem_quote)%>%
filter(fem_quote== 0)
only_school <- zmer %>%
filter(type=="GE" | type2== "SchP" )%>%
#filter(!(type== "PS" & unified != "unified: opt 1")) %>%
mutate( elected_ge=
case_when(
type2== "GE" & elected==1 ~ 1,
type2== "SchP" & elected_pres==1 & unified == "unified: opt 1" ~1,
T~0
)) %>%
filter(elected_ge ==1, type2== "SchP") %>%
mutate(sex2= ifelse(sex=="m",0,1))%>%
summarise(fem_quote= mean(sex2))
zmer %>%
filter(lname!= "Vereinzelte") %>%
ggplot(aes(x=sex))+
geom_bar()+
theme_bw()+
facet_wrap(~type2)
zmer %>%
filter(elected==1) %>%
group_by(sex) %>%
summarise(n=n())%>%
mutate(perc= n/sum(n))zmer %>%
filter(type=="GE" | type2== "SchP" )%>%
#filter(!(type== "PS" & unified != "unified: opt 1")) %>%
mutate( elected_ge=
case_when(
type2== "GE" & elected==1 ~ 1,
type2== "SchP" & elected_pres==1 & unified == "unified: opt 1" ~1,
T~0
)) %>%
filter(elected_ge ==1)%>%
group_by(mun_name, mun_id)%>%
mutate(sex2 = ifelse(sex=="m", 0, 1)) %>%
summarise(n=sum(elected_ge),fem_quote = mean(sex2)) %>%
mutate(fem_quote_factor=
factor(case_when(
fem_quote < 0.20 ~ "",
fem_quote >= 0.20 & fem_quote < 0.25 ~ "20",
fem_quote >= 0.25 & fem_quote < 0.30 ~ "25",
fem_quote >= 0.30 & fem_quote < 0.35 ~ "30",
fem_quote >= 0.35 & fem_quote < 0.40 ~ "35",
fem_quote >= 0.40 & fem_quote < 0.50 ~ "40",
fem_quote >= 0.50 & fem_quote < 0.55 ~ "50",
fem_quote >= 0.55 & fem_quote < 0.60 ~ "55",
fem_quote >= 0.60 ~ "60"
), levels = c("", "20", "25", "30", "35", "40", "50", "55", "60")
))%>%
left_join(., geo_data3) %>%
ggplot()+
geom_sf(aes(fill=fem_quote_factor, geometry = geometry), color="white", linewidth=0.5)+
scale_fill_manual(
values = c(
"#680d25", "#8d0613", "#BD2031", "#cf352e", "#f1434a", "#ff9193",
"#91cdff", "#42a2f1", "#105182"
),
drop = F,
name = "Frauenanteil",
guide = guide_legend(
direction = "horizontal",
keyheight = unit(2, units = "mm"),
keywidth = unit(c(25, rep(7, 7), 25), units = "mm"),
title.position = "top",
title.hjust = 0,
label.hjust = 1,
nrow = 1,
byrow = T,
reverse = T,
label.position = "bottom"))+
theme_void()+
theme(legend.position = "bottom")+
annotate(
geom = "curve", x = 2700000, y = 1229000 , xend = 2696900, yend = 1245000,
curvature = -0.15, size=1,
arrow = arrow(length = unit(2, "mm"), type = "closed"))+
annotate(
geom = "text", x = 2700900, y = 1229000, label= "Uster, die Stadt mit dem \nhöchsten Frauenanteil in \nder Exekutive",
size=3, hjust = "bottom")+
annotate(
geom = "curve", x = 2670000, y = 1276000 , xend = 2685300, yend = 1265000,
curvature = 0.15, size=1,
arrow = arrow(length = unit(2, "mm"), type = "closed"))+
annotate(
geom = "text", x = 2670000, y = 1280900, label= "Rorbas, eine von \n5 Gemeinden ohne \nFrauen in der \nExekutive",
size=3, hjust = "bottom")+
labs(title= "Wenige Gemeinden mit Frauenmehrheit im Gemeindevorstand",
subtitle = "Karte der Zürcher Gemeinden und ihrem Frauenanteil")
ggsave("plots/map.jpeg")zmer %>%
filter(elected== 1) %>%
group_by(type2, sex) %>%
tally() %>%
group_by(type2) %>%
mutate(perc= n/ sum(n)) %>%
ggplot(aes(x= type2, y= perc, fill= sex))+
geom_col(position = "dodge")+
geom_text(aes(label= paste0(round(perc,3)*100,"%")), vjust = -0.5,
position = position_dodge(width = 0.9))+
scale_y_continuous(labels = scales::percent, breaks = seq(0,1,0.25))+
scale_x_discrete(labels = c("GE"= "Gemeindeexekutive", "RPK"= "Rechnungsprüfungskommission",
"SchP"= "Schulpflege"))+
scale_fill_manual(values= c("#53A2BE","#357266"),
labels= c("Frauen","Männer"))+
labs(x="", y="",
fill= "Geschlecht",
title = "Massiver Männerüberschuss in Exekutive und RPK, \nkleiner Frauenüberschuss in den Schulpflegen",
subtitle = "Prozent gewählte Personen nach Geschlecht und Behörde")+
theme_minimal()+
theme(legend.position = "right",
legend.title= element_blank(),
panel.grid.major.x = element_blank(),
axis.line.x = element_line())
ggsave("plots/perc.jpeg")model1 <- glm(elected ~ sex*incumbent +type2 + party2+
round + age +
incumbent_pres + can_per_seat,
data = zmer)
summary(model1)
predicts(model1, "F;0;mode;mode;mode;median;0;1") %>%
ggplot(aes(x=sex, y=mean, ymin= lower, ymax=upper))+
geom_errorbar(width=0.1)+
geom_point(size=2)+
scale_x_discrete(labels= c("Frauen", "Männer"))+
scale_y_continuous(labels = scales::percent, breaks = seq(0,1,0.05))+
labs(x="", y="",
title = "Frauen mit hohen Wahlwahrscheinlichkeiten",
subtitle = "Vorhergesagte Wahlwahrscheinlichkeit nach Geschlecht",
caption = "Alle weiteren Prädiktoren wurden auf den Median oder Modus gesetzt")+
theme_bw()+
theme(panel.grid.major.x = element_blank())
ggsave("plots/predicts.jpeg")