# Knihovna pro utilities theme, a save_plot
source("c:/users/matous/desktop/cp oddluzeni data/kódy/utils.R")
## ── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
## ✔ dplyr 1.1.4 ✔ readr 2.1.5
## ✔ forcats 1.0.0 ✔ stringr 1.5.1
## ✔ ggplot2 3.5.1 ✔ tibble 3.2.1
## ✔ lubridate 1.9.3 ✔ tidyr 1.3.1
## ✔ purrr 1.0.2
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag() masks stats::lag()
## ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errors
library("tidyverse")
library("AER")
## Warning: package 'AER' was built under R version 4.4.3
## Loading required package: car
## Loading required package: carData
##
## Attaching package: 'car'
##
## The following object is masked from 'package:dplyr':
##
## recode
##
## The following object is masked from 'package:purrr':
##
## some
##
## Loading required package: lmtest
## Loading required package: zoo
##
## Attaching package: 'zoo'
##
## The following objects are masked from 'package:base':
##
## as.Date, as.Date.numeric
##
## Loading required package: sandwich
## Loading required package: survival
options(scipen=10)
vysledky <- read_csv2("PS2025data20251005_csv/csv/pst4p.csv")
## ℹ Using "','" as decimal and "'.'" as grouping mark. Use `read_delim()` for more control.
## Rows: 179877 Columns: 51── Column specification ────────────────────────────────────────────────────────
## Delimiter: ";"
## dbl (51): ID_OKRSKY, TYP_FORM, OPRAVA, CHYBA, OKRES, OBEC, OKRSEK, KC_1, KST...
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
vysledky
## # A tibble: 179,877 × 51
## ID_OKRSKY TYP_FORM OPRAVA CHYBA OKRES OBEC OKRSEK KC_1 KSTRANA POC_HLASU
## <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 1 2 0 0 7204 500011 1 6 3 2
## 2 1 2 0 0 7204 500011 1 6 4 1
## 3 1 2 0 0 7204 500011 1 6 6 41
## 4 1 2 0 0 7204 500011 1 6 8 5
## 5 1 2 0 0 7204 500011 1 6 10 4
## 6 1 2 0 0 7204 500011 1 6 11 114
## 7 1 2 0 0 7204 500011 1 6 12 5
## 8 1 2 0 0 7204 500011 1 6 14 1
## 9 1 2 0 0 7204 500011 1 6 15 5
## 10 1 2 0 0 7204 500011 1 6 16 42
## # ℹ 179,867 more rows
## # ℹ 41 more variables: KC_2 <dbl>, HLASY_01 <dbl>, HLASY_02 <dbl>,
## # HLASY_03 <dbl>, HLASY_04 <dbl>, HLASY_05 <dbl>, HLASY_06 <dbl>,
## # HLASY_07 <dbl>, HLASY_08 <dbl>, HLASY_09 <dbl>, HLASY_10 <dbl>,
## # HLASY_11 <dbl>, HLASY_12 <dbl>, HLASY_13 <dbl>, HLASY_14 <dbl>,
## # HLASY_15 <dbl>, HLASY_16 <dbl>, HLASY_17 <dbl>, HLASY_18 <dbl>,
## # HLASY_19 <dbl>, HLASY_20 <dbl>, HLASY_21 <dbl>, HLASY_22 <dbl>, …
kandidati <- read_csv2("PS2025reg20251005_csv/csv/psrk.csv", locale = locale("cs", encoding = "windows-1250"))
## ℹ Using "','" as decimal and "'.'" as grouping mark. Use `read_delim()` for more control.
## Rows: 4475 Columns: 19── Column specification ────────────────────────────────────────────────────────
## Delimiter: ";"
## chr (9): JMENO, PRIJMENI, TITULPRED, TITULZA, POVOLANI, BYDLISTEN, PLATNOST...
## dbl (10): VOLKRAJ, KSTRANA, PORCISLO, VEK, PSTRANA, NSTRANA, POCHLASU, SKRUT...
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
kandidati
## # A tibble: 4,475 × 19
## VOLKRAJ KSTRANA PORCISLO JMENO PRIJMENI TITULPRED TITULZA VEK POVOLANI
## <dbl> <dbl> <dbl> <chr> <chr> <chr> <chr> <dbl> <chr>
## 1 1 1 1 Miloslav Zientek <NA> <NA> 49 IT exper…
## 2 1 1 2 Michal Kotrba <NA> <NA> 46 manažer …
## 3 1 1 3 Adam Křikava <NA> <NA> 51 majitel …
## 4 1 1 4 Miroslav Stejskal <NA> <NA> 46 manažer …
## 5 1 1 5 Tereza Sodomková <NA> <NA> 38 manažerk…
## 6 1 1 6 Dominika Švarcová <NA> <NA> 27 marketin…
## 7 1 3 1 Vladimír Štěpán Ing. <NA> 74 expert v…
## 8 1 3 2 Josef Smítka <NA> <NA> 59 jednatel…
## 9 1 3 3 Jiří Hudec <NA> <NA> 66 důchodce
## 10 1 3 4 Radek Vaisocher Bc. <NA> 51 podnikat…
## # ℹ 4,465 more rows
## # ℹ 10 more variables: BYDLISTEN <chr>, PSTRANA <dbl>, NSTRANA <dbl>,
## # PLATNOST <chr>, POCHLASU <dbl>, POCPROC <chr>, MANDAT <chr>,
## # SKRUTINIUM <dbl>, PORADIMAND <dbl>, PORADINAHR <dbl>
strany <- read_csv2("PS2025reg20251005_csv/csv/psrkl.csv", locale = locale("cs", encoding = "windows-1250"))
## ℹ Using "','" as decimal and "'.'" as grouping mark. Use `read_delim()` for more control.
## Rows: 26 Columns: 13── Column specification ────────────────────────────────────────────────────────
## Delimiter: ";"
## chr (8): NAZEVCELK, NAZEV_STRK, ZKRATKAK30, ZKRATKAK8, SLOZENI, STAVREG, PLA...
## dbl (4): KSTRANA, VSTRANA, POCSTRVKO, POCMANDCR
## lgl (1): SLOZNEPLAT
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
strany
## # A tibble: 26 × 13
## KSTRANA VSTRANA NAZEVCELK NAZEV_STRK ZKRATKAK30 ZKRATKAK8 POCSTRVKO SLOZENI
## <dbl> <dbl> <chr> <chr> <chr> <chr> <dbl> <chr>
## 1 1 1706 "\"Rebelov… "\"Rebelo… "\"Rebelo… Rebelové 1 1706
## 2 2 1190 "Moravské … "Moravské… "Moravské… MZH 1 1190
## 3 3 1282 "Jasný Sig… "Jasný Si… "Jasný Si… JaSaN 1 1282
## 4 4 1704 "VÝZVA 202… "VÝZVA 20… "VÝZVA 20… VÝZVA 1 1704
## 5 5 1061 "SMS – Stá… "SMS – St… "SMS – St… SMSka 1 1061
## 6 6 1114 "Svoboda a… "Svoboda … "Svoboda … SPD 1 1114
## 7 7 759 "ČSSD – Če… "ČSSD – Č… "Česká su… ČSSD 1 759
## 8 8 1245 "PŘÍSAHA o… "PŘÍSAHA … "PŘÍSAHA … PŘÍSAHA 1 1245
## 9 9 74 "Levice" "Levice" "Levice" Levice 1 074
## 10 10 1283 "Česká rep… "Česká re… "Česká re… ČR1 1 1283
## # ℹ 16 more rows
## # ℹ 5 more variables: STAVREG <chr>, PLAT_STR <chr>, SLOZNEPLAT <lgl>,
## # POCMANDCR <dbl>, NAZEVPLNY <chr>
kraje <- read_csv2("PS2025ciselniky20251005_csv/csv/psvolkr.csv", locale = locale("cs", encoding = "windows-1250"))
## ℹ Using "','" as decimal and "'.'" as grouping mark. Use `read_delim()` for more control.
## Rows: 14 Columns: 4── Column specification ────────────────────────────────────────────────────────
## Delimiter: ";"
## chr (1): NAZVOLKRAJ
## dbl (3): VOLKRAJ, KRAJ, MAXKAND
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
kraje
## # A tibble: 14 × 4
## VOLKRAJ NAZVOLKRAJ KRAJ MAXKAND
## <dbl> <chr> <dbl> <dbl>
## 1 1 Hl. m. Praha 1100 36
## 2 2 Středočeský 2100 34
## 3 3 Jihočeský 3100 22
## 4 4 Plzeňský 3200 20
## 5 5 Karlovarský 4100 14
## 6 6 Ústecký 4200 26
## 7 7 Liberecký 5100 17
## 8 8 Královéhradecký 5200 20
## 9 9 Pardubický 5300 19
## 10 10 Vysočina 6100 20
## 11 11 Jihomoravský 6200 34
## 12 12 Olomoucký 7100 23
## 13 13 Zlínský 7200 22
## 14 14 Moravskoslezský 8100 36
obce <- read_csv2("PS2025ciselniky20251005_csv/csv/pscoco.csv", locale = locale("cs", encoding = "windows-1250")) %>% select(OBEC, VOLKRAJ)
## ℹ Using "','" as decimal and "'.'" as grouping mark. Use `read_delim()` for more control.
## Rows: 6389 Columns: 10── Column specification ────────────────────────────────────────────────────────
## Delimiter: ";"
## chr (1): NAZEVOBCE
## dbl (9): KRAJ, OKRES, CPOU, ORP, OBEC, VOLKRAJ, MINOKRSEK1, MAXOKRSEK1, OBEC...
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
zahrnute_strany <- c(6, 11, 16, 20, 22, 23, 25)
nejvyssi_titul <- function(titul) {
if(is.na(titul)) {
return(0)
}
else if(str_detect(titul, "prof.")) {
return(5)
}
else if(str_detect(titul, "(doc.|Csc.)")) {
return(4)
}
else if(str_detect(titul, "Ph.D.")) {
return(3)
}
else if(str_detect(titul, "(Ing.|Mgr.|MA.|MBA|LL.M.|MUDr.|JUDr.|PhDr.|RnDr.|MgA.|MVDr.|PharmDr.|PaeDr.|MSc.)")) {
return(2)
}
else if(str_detect(titul, "(Bc.|BcA.|DiS.)")) {
return(1)
}
else {
return(0)
}
}
vysledky_clean <- vysledky %>%
select(OBEC, KSTRANA, POC_HLASU, starts_with("HLASY_")) %>%
inner_join(obce, by = "OBEC") %>%
filter(KSTRANA %in% zahrnute_strany) %>%
summarise(.by = c(VOLKRAJ, KSTRANA), across(c(POC_HLASU, starts_with("HLASY_")), sum)) %>%
pivot_longer(starts_with("HLASY_"), names_to = "PORCISLO", values_to = "preferencni_hlasy", names_transform = \(x) as.integer(str_remove(x, "HLASY_"))) %>%
mutate(procento_preferencnich_hlasu = preferencni_hlasy / POC_HLASU) %>%
inner_join(kandidati %>% select(VOLKRAJ, KSTRANA, PORCISLO, JMENO, PRIJMENI, TITULPRED, TITULZA, VEK, PSTRANA), by = c("VOLKRAJ", "KSTRANA", "PORCISLO")) %>%
inner_join(strany %>% select(KSTRANA, STRANA = ZKRATKAK8), by = "KSTRANA") %>%
mutate(
nejvyssi_titul = map2_int(TITULPRED, TITULZA, \(x, y) max(nejvyssi_titul(x), nejvyssi_titul(y))),
pohlavi = factor(if_else(str_detect(PRIJMENI, "(ová|á|Demetrashivili|Decroix|Rázga|Crkvenjaš)$"), "F", "M"), levels = c("M", "F") )
) %>%
mutate(.by = c(VOLKRAJ, KSTRANA, pohlavi), pohlavi_na_kandidatce = row_number(PORCISLO))
vysledky_clean
## # A tibble: 2,396 × 16
## VOLKRAJ KSTRANA POC_HLASU PORCISLO preferencni_hlasy procento_preferencnich…¹
## <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 13 6 27784 1 3259 0.117
## 2 13 6 27784 2 1871 0.0673
## 3 13 6 27784 3 785 0.0283
## 4 13 6 27784 4 1089 0.0392
## 5 13 6 27784 5 685 0.0247
## 6 13 6 27784 6 435 0.0157
## 7 13 6 27784 7 310 0.0112
## 8 13 6 27784 8 2253 0.0811
## 9 13 6 27784 9 552 0.0199
## 10 13 6 27784 10 439 0.0158
## # ℹ 2,386 more rows
## # ℹ abbreviated name: ¹procento_preferencnich_hlasu
## # ℹ 10 more variables: JMENO <chr>, PRIJMENI <chr>, TITULPRED <chr>,
## # TITULZA <chr>, VEK <dbl>, PSTRANA <dbl>, STRANA <chr>,
## # nejvyssi_titul <int>, pohlavi <fct>, pohlavi_na_kandidatce <int>
model <- tobit(
procento_preferencnich_hlasu ~ I(cut(PORCISLO, c(1, 3, 5, 10, 15))) + I(VEK <= 30)*STRANA + nejvyssi_titul + pohlavi*STRANA,
data = vysledky_clean %>% filter(PORCISLO <= 15),
left = 0
)
summary(model)
##
## Call:
## tobit(formula = procento_preferencnich_hlasu ~ I(cut(PORCISLO,
## c(1, 3, 5, 10, 15))) + I(VEK <= 30) * STRANA + nejvyssi_titul +
## pohlavi * STRANA, left = 0, data = vysledky_clean %>% filter(PORCISLO <=
## 15))
##
## Observations: (98 observations deleted due to missingness)
## Total Left-censored Uncensored Right-censored
## 1365 2 1363 0
##
## Coefficients:
## Estimate Std. Error z value
## (Intercept) 0.0456026 0.0024427 18.669
## I(cut(PORCISLO, c(1, 3, 5, 10, 15)))(3,5] -0.0207586 0.0020745 -10.007
## I(cut(PORCISLO, c(1, 3, 5, 10, 15)))(5,10] -0.0363358 0.0017450 -20.823
## I(cut(PORCISLO, c(1, 3, 5, 10, 15)))(10,15] -0.0446298 0.0017664 -25.265
## I(VEK <= 30)TRUE 0.0041220 0.0063788 0.646
## STRANAAUTO 0.0005756 0.0024613 0.234
## STRANAPiráti 0.0173375 0.0026742 6.483
## STRANASPD 0.0034914 0.0024901 1.402
## STRANASPOLU 0.0281675 0.0024588 11.456
## STRANAStačilo! 0.0155291 0.0024754 6.273
## STRANASTAN 0.0122896 0.0025904 4.744
## nejvyssi_titul 0.0028495 0.0005638 5.054
## pohlaviF 0.0034035 0.0033186 1.026
## I(VEK <= 30)TRUE:STRANAAUTO 0.0045219 0.0082317 0.549
## I(VEK <= 30)TRUE:STRANAPiráti 0.0261316 0.0076116 3.433
## I(VEK <= 30)TRUE:STRANASPD 0.0045764 0.0097586 0.469
## I(VEK <= 30)TRUE:STRANASPOLU 0.0009624 0.0078684 0.122
## I(VEK <= 30)TRUE:STRANAStačilo! -0.0053179 0.0088262 -0.603
## I(VEK <= 30)TRUE:STRANASTAN 0.0226573 0.0077453 2.925
## STRANAAUTO:pohlaviF -0.0014557 0.0051763 -0.281
## STRANAPiráti:pohlaviF 0.0304413 0.0044703 6.810
## STRANASPD:pohlaviF -0.0013954 0.0047059 -0.297
## STRANASPOLU:pohlaviF 0.0038348 0.0048932 0.784
## STRANAStačilo!:pohlaviF 0.0082897 0.0049128 1.687
## STRANASTAN:pohlaviF 0.0105757 0.0045604 2.319
## Log(scale) -3.8888377 0.0191607 -202.959
## Pr(>|z|)
## (Intercept) < 2e-16 ***
## I(cut(PORCISLO, c(1, 3, 5, 10, 15)))(3,5] < 2e-16 ***
## I(cut(PORCISLO, c(1, 3, 5, 10, 15)))(5,10] < 2e-16 ***
## I(cut(PORCISLO, c(1, 3, 5, 10, 15)))(10,15] < 2e-16 ***
## I(VEK <= 30)TRUE 0.518151
## STRANAAUTO 0.815100
## STRANAPiráti 0.00000000008975 ***
## STRANASPD 0.160869
## STRANASPOLU < 2e-16 ***
## STRANAStačilo! 0.00000000035348 ***
## STRANASTAN 0.00000209240244 ***
## nejvyssi_titul 0.00000043296589 ***
## pohlaviF 0.305085
## I(VEK <= 30)TRUE:STRANAAUTO 0.582777
## I(VEK <= 30)TRUE:STRANAPiráti 0.000597 ***
## I(VEK <= 30)TRUE:STRANASPD 0.639098
## I(VEK <= 30)TRUE:STRANASPOLU 0.902649
## I(VEK <= 30)TRUE:STRANAStačilo! 0.546829
## I(VEK <= 30)TRUE:STRANASTAN 0.003441 **
## STRANAAUTO:pohlaviF 0.778534
## STRANAPiráti:pohlaviF 0.00000000000978 ***
## STRANASPD:pohlaviF 0.766837
## STRANASPOLU:pohlaviF 0.433217
## STRANAStačilo!:pohlaviF 0.091532 .
## STRANASTAN:pohlaviF 0.020393 *
## Log(scale) < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Scale: 0.02047
##
## Gaussian distribution
## Number of Newton-Raphson Iterations: 5
## Log-likelihood: 3364 on 26 Df
## Wald-statistic: 1656 on 24 Df, p-value: < 2.22e-16
summary(tobit(
procento_preferencnich_hlasu ~ I(cut(PORCISLO, c(1, 3, 5, 10, 15))) + I(VEK <= 30)*STRANA + pohlavi*nejvyssi_titul + pohlavi*STRANA,
data = vysledky_clean %>% filter(PORCISLO <= 15, STRANA %in% c("Piráti", "STAN")),
left = 0
))
##
## Call:
## tobit(formula = procento_preferencnich_hlasu ~ I(cut(PORCISLO,
## c(1, 3, 5, 10, 15))) + I(VEK <= 30) * STRANA + pohlavi *
## nejvyssi_titul + pohlavi * STRANA, left = 0, data = vysledky_clean %>%
## filter(PORCISLO <= 15, STRANA %in% c("Piráti", "STAN")))
##
## Observations: (28 observations deleted due to missingness)
## Total Left-censored Uncensored Right-censored
## 390 0 390 0
##
## Coefficients:
## Estimate Std. Error z value
## (Intercept) 0.0686965 0.0043477 15.800
## I(cut(PORCISLO, c(1, 3, 5, 10, 15)))(3,5] -0.0255292 0.0044819 -5.696
## I(cut(PORCISLO, c(1, 3, 5, 10, 15)))(5,10] -0.0451459 0.0037580 -12.013
## I(cut(PORCISLO, c(1, 3, 5, 10, 15)))(10,15] -0.0554026 0.0037621 -14.727
## I(VEK <= 30)TRUE 0.0316995 0.0049358 6.422
## STRANASTAN -0.0052068 0.0031955 -1.629
## pohlaviF 0.0310849 0.0050674 6.134
## nejvyssi_titul 0.0043963 0.0014976 2.936
## I(VEK <= 30)TRUE:STRANASTAN -0.0036620 0.0070385 -0.520
## pohlaviF:nejvyssi_titul 0.0009958 0.0025056 0.397
## STRANASTAN:pohlaviF -0.0191228 0.0049896 -3.833
## Log(scale) -3.7492350 0.0358057 -104.710
## Pr(>|z|)
## (Intercept) < 2e-16 ***
## I(cut(PORCISLO, c(1, 3, 5, 10, 15)))(3,5] 0.000000012258 ***
## I(cut(PORCISLO, c(1, 3, 5, 10, 15)))(5,10] < 2e-16 ***
## I(cut(PORCISLO, c(1, 3, 5, 10, 15)))(10,15] < 2e-16 ***
## I(VEK <= 30)TRUE 0.000000000134 ***
## STRANASTAN 0.103221
## pohlaviF 0.000000000855 ***
## nejvyssi_titul 0.003330 **
## I(VEK <= 30)TRUE:STRANASTAN 0.602869
## pohlaviF:nejvyssi_titul 0.691052
## STRANASTAN:pohlaviF 0.000127 ***
## Log(scale) < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Scale: 0.02354
##
## Gaussian distribution
## Number of Newton-Raphson Iterations: 5
## Log-likelihood: 908.8 on 12 Df
## Wald-statistic: 502.3 on 10 Df, p-value: < 2.22e-16
koef <- tibble(
variable = names(summary(model)$coefficients[,1]),
estimate = summary(model)$coefficients[,1],
se = summary(model)$coefficients[,2],
z = summary(model)$coefficients[,3],
p = summary(model)$coefficients[,4]
)
crit_val90 <- qnorm(0.95)
crit_val95 <- qnorm(0.975)
crit_val99 <- qnorm(0.995)
save_plot("Kroužky ženy", plot = koef %>%
filter(str_detect(variable, ":pohlavi")) %>%
mutate(strana = replace_na(str_extract(variable, "STRANA([\\w!]+):", 1), "ANO")) %>%
ggplot(aes(x = reorder(strana, desc(estimate)), y = estimate)) +
geom_hline(yintercept = 0) +
geom_pointrange(aes(ymin = estimate - crit_val90*se, ymax = estimate + crit_val90*se), alpha = 0.5, linewidth = 1.25, size = 0.75, colour = "#515") +
geom_pointrange(aes(ymin = estimate - crit_val95*se, ymax = estimate + crit_val95*se), alpha = 0.5, linewidth = 1.25, size = 0.75, colour = "#515") +
geom_pointrange(aes(ymin = estimate - crit_val99*se, ymax = estimate + crit_val99*se), alpha = 0.5, linewidth = 1.25, size = 0.75, colour = "#515") +
scale_y_continuous(labels = scales::percent_format(accuracy = 1), breaks = c(seq(-0.05, 0.05, 0.01))) +
theme_mf() +
labs(
title = "Ženy a mladé kandidáty\\*ky kroužkovali častěji pouze voliči Pirátů a STANu",
subtitle = "",
caption = "Výsledky očištěné od vlivu pořadí na kandidátce, věku a dosaženého vzdělání pomocí Tobit modelu.\nLinky zobrazují statistickou chybu s 90%, 95% a 99% CI.",
x = "",
y = "Boost preferenčních hlasů u žen"
))
pohlavi <- vysledky_clean %>%
filter(PORCISLO <= 15) %>%
summarise(.by = c(STRANA, pohlavi), preferencni_hlasy = sum(preferencni_hlasy), pocet_kandidatu = n()) %>%
pivot_longer(c(preferencni_hlasy, pocet_kandidatu)) %>%
pivot_wider(names_from = pohlavi, values_from = value) %>%
mutate(celkem = `F` + M, pomer_zen = `F` / celkem) %>%
select(-`F`, -M, -celkem)
pohlavi
## # A tibble: 14 × 3
## STRANA name pomer_zen
## <chr> <chr> <dbl>
## 1 SPD preferencni_hlasy 0.253
## 2 SPD pocet_kandidatu 0.268
## 3 SPOLU preferencni_hlasy 0.221
## 4 SPOLU pocet_kandidatu 0.206
## 5 Piráti preferencni_hlasy 0.509
## 6 Piráti pocet_kandidatu 0.388
## 7 AUTO preferencni_hlasy 0.138
## 8 AUTO pocet_kandidatu 0.163
## 9 ANO preferencni_hlasy 0.278
## 10 ANO pocet_kandidatu 0.282
## 11 STAN preferencni_hlasy 0.455
## 12 STAN pocet_kandidatu 0.335
## 13 Stačilo! preferencni_hlasy 0.370
## 14 Stačilo! pocet_kandidatu 0.220
save_plot("Pozice a křížky", plot = vysledky_clean %>%
mutate(STRANA = case_when(STRANA == "Piráti" ~ "Piráti", STRANA == "STAN" ~ "STAN", .default = "Ostatní"),
pohlavi = if_else(pohlavi == "M", "Muži", "Ženy")) %>%
mutate(.by = c(VOLKRAJ, STRANA, pohlavi), nta_zena_muz = row_number(PORCISLO)) %>%
summarise(.by = c(STRANA, pohlavi, nta_zena_muz), procento_preferencnich_hlasu = mean(procento_preferencnich_hlasu)) %>%
filter(nta_zena_muz <= 10) %>%
ggplot(aes(x = as.factor(nta_zena_muz), y = procento_preferencnich_hlasu, colour = pohlavi, group = pohlavi)) +
geom_point() +
geom_line() +
facet_wrap(~fct_rev(STRANA)) +
#facet_grid(fct_rev(STRANA) ~ pohlavi) +
scale_y_continuous(labels = scales::percent, limits = c(0, NA)) +
scale_colour_manual(values = c("#55a", "#e55")) +
theme_mf() +
labs(title = "Nezdá se, že by voliči Pirátů a STANu kroužkovali jen první dvě / čtyři ženy -- distribuci kroužků mají podobnou jako u můžů",
x = "Pořádí kandidáta*ky daného genderu na kanidátce",
y = "Průměrné procento preferenčních hlasů",
caption = "Průměrné procento preferenčních hlasů podle pořadí na kandidátce napříč všemi volebními kraji",
colour = "") +
theme(legend.position = "top"), save_width = 8, save_height = 5)
pohlavi
## # A tibble: 14 × 3
## STRANA name pomer_zen
## <chr> <chr> <dbl>
## 1 SPD preferencni_hlasy 0.253
## 2 SPD pocet_kandidatu 0.268
## 3 SPOLU preferencni_hlasy 0.221
## 4 SPOLU pocet_kandidatu 0.206
## 5 Piráti preferencni_hlasy 0.509
## 6 Piráti pocet_kandidatu 0.388
## 7 AUTO preferencni_hlasy 0.138
## 8 AUTO pocet_kandidatu 0.163
## 9 ANO preferencni_hlasy 0.278
## 10 ANO pocet_kandidatu 0.282
## 11 STAN preferencni_hlasy 0.455
## 12 STAN pocet_kandidatu 0.335
## 13 Stačilo! preferencni_hlasy 0.370
## 14 Stačilo! pocet_kandidatu 0.220
save_plot("Poměry kandidátek a kroužků", plot =
pohlavi %>%
ggplot(aes(x = reorder(STRANA, desc(pomer_zen)), y = pomer_zen, fill = fct_rev(name))) +
geom_hline(yintercept= 0.5, linetype = "dashed") +
geom_col(position = "dodge") +
theme_mf() +
scale_y_continuous(labels = scales::percent, limits = c(0, 1)) +
scale_fill_manual(values = c("#338", "#a55"), label = c("Procento preferenčních hlasů ženám", "Procento žen na kandidátkách")) +
labs(
title = "I u Pirátů a STANu udělili voliči ženám jen asi polovinu preferenčních hlasů. Každá žena ale dostala více hlasů, protože se hlasy rozdělily mezi méně žen",
x = "",
y = "", #"Procento preferenčních hlasů pro ženy\nProcento žen na kandidátkách",
fill = ""
) +
theme(legend.position = "top"))
save_plot("Kroužky pohlaví", plot =
pohlavi %>%
filter(name == "preferencni_hlasy") %>%
ggplot(aes(x = reorder(STRANA, desc(pomer_zen)), y = pomer_zen)) +
geom_hline(yintercept= 0.5, linetype = "dashed") +
geom_col(fill = "#338") +
theme_mf() +
scale_y_continuous(labels = scales::percent, limits = c(0, 1)) +
labs(
title = "I Piráti a STAN udělili ženám jen asi polovinu preferenčních hlasů. Každá žena ale dostala více hlasů, protože se hlasy rozdělily mezi méně žen",
x = "",
y = "Procento preferenčních hlasů pro ženy",
) )
