Jews, Episcopalians, and Puerto Ricans

jews_pr_epis.R

cc16 <- cces16 %>% 
  mutate(group = frcode(religpew_protestant == 7 ~ "Episcopalian",
                        religpew == 5 ~ "Jewish",
                        Hispanic_origin_4 == 1~ "Puerto Rican")) %>% 
  mutate(age = 2016 - birthyr) %>% 
  select(group, pew_churatd, educ, age, gender, pew_religimp, pew_bornagain, vote = CC16_410a, race, weight = commonweight_vv, income = faminc) %>% 
  mutate(year = 2016)

cc18 <- cces18 %>% 
  mutate(group = frcode(religpew_protestant == 7 ~ "Episcopalian",
                        religpew == 5 ~ "Jewish",
                        CC18_354a_4 == 1~ "Puerto Rican")) %>% 
  mutate(age = 2018 - birthyr) %>% 
  select(group, pew_churatd, educ, age, gender, pew_religimp, pew_bornagain, vote = CC18_317, race, weight = commonpostweight, income = faminc_new) %>% 
  mutate(year = 2018)

cc <- bind_rows(cc16, cc18) %>% 
  as_tibble()

cc <- cc %>% 
  mutate(trump = case_when(vote == 1 ~ 1, 
                           vote == 2 ~ 0)) %>% 
  mutate(male = case_when(gender == 1 ~ 1, 
                          gender == 2 ~ 0)) %>% 
  mutate(white = case_when(race == 1 ~ 1, 
                           race != 1 ~ 0)) 

graph <- cc %>% 
  ct(group, wt = weight)

graph %>% 
  mutate(n = round(n, 0)) %>% 
  filter(group != "NA") %>% 
  ggplot(., aes(x = group, y = n, fill = group)) +
  geom_col(color = "black") +
  fill4_3() +
  theme_gg("Jost") +
  geom_text(aes(y = n + 75, label = n), position = position_dodge(width = .9), size = 5.5, family = "font") +
  labs(x = "", y = "", title = "Sample Size of the Three Groups") +
  ggsave("E://jews_pr_epis.png", type = "cairo-png", width = 5)

graph <- cc %>% 
  group_by(group, year) %>% 
  mutate(vote = frcode(vote == 1 ~ "Trump",
                       vote == 2 ~ "Clinton",
                       vote == 3 | vote == 4 | vote == 5 | vote == 8 ~ "Someone Else")) %>% 
  ct(vote, wt = weight, show_na = FALSE)



one <- graph %>% 
  filter(group == "Episcopalian") %>% 
  ggplot(., aes(x = 1, y = pct, fill = fct_rev(vote))) +
  geom_col(color = "black") + 
  coord_flip() +
  facet_wrap(~ year, ncol =1, strip.position = "left") +
  scale_fill_manual(values = c("forestgreen","dodgerblue3", "firebrick3")) +
  theme_gg("Abel") +
  # theme(legend.position = "bottom") +
  scale_y_continuous(labels = percent) +
  theme(strip.text.y.left = element_text(angle=0)) +
  guides(fill = guide_legend(reverse=T, nrow = 1)) +
  theme(axis.title.y=element_blank(), axis.text.y=element_blank(), axis.ticks.y=element_blank()) +
  theme(panel.grid.minor.y=element_blank(), panel.grid.major.y=element_blank()) +
  geom_text(aes(label = ifelse(pct >.05, paste0(pct*100, '%'), '')), position = position_stack(vjust = 0.5), size = 4, family = "font", color = "black") +
  labs(x = "", y = "", title = "Vote Choice for President in 2016", subtitle = "Episcopalians", caption = "") +
  ggsave("E://epis_vote.png", width = 5, height = 3)


two <- graph %>% 
  filter(group == "Jewish") %>% 
  ggplot(., aes(x = 1, y = pct, fill = fct_rev(vote))) +
  geom_col(color = "black") + 
  coord_flip() +
  facet_wrap(~ year, ncol =1, strip.position = "left") +
  scale_fill_manual(values = c("forestgreen","dodgerblue3", "firebrick3")) +
  theme_gg("Abel") +
  # theme(legend.position = "bottom") +
  scale_y_continuous(labels = percent) +
  theme(strip.text.y.left = element_text(angle=0)) +
  guides(fill = guide_legend(reverse=T, nrow = 1)) +
  theme(axis.title.y=element_blank(), axis.text.y=element_blank(), axis.ticks.y=element_blank()) +
  theme(panel.grid.minor.y=element_blank(), panel.grid.major.y=element_blank()) +
  geom_text(aes(label = ifelse(pct >.05, paste0(pct*100, '%'), '')), position = position_stack(vjust = 0.5), size = 4, family = "font", color = "black") +
  labs(x = "", y = "", title = "", subtitle = "Jewish", caption = "") +
  ggsave("E://jewish_vote.png", width = 5, height = 3)


three <- graph %>% 
  filter(group == "Puerto Rican") %>% 
  ggplot(., aes(x = 1, y = pct, fill = fct_rev(vote))) +
  geom_col(color = "black") + 
  coord_flip() +
  facet_wrap(~ year, ncol =1, strip.position = "left") +
  scale_fill_manual(values = c("forestgreen","dodgerblue3", "firebrick3")) +
  theme_gg("Abel") +
  theme(legend.position = "bottom") +
  scale_y_continuous(labels = percent) +
  theme(strip.text.y.left = element_text(angle=0)) +
  guides(fill = guide_legend(reverse=T, nrow = 1)) +
  theme(axis.title.y=element_blank(), axis.text.y=element_blank(), axis.ticks.y=element_blank()) +
  theme(panel.grid.minor.y=element_blank(), panel.grid.major.y=element_blank()) +
  geom_text(aes(label = ifelse(pct >.05, paste0(pct*100, '%'), '')), position = position_stack(vjust = 0.5), size = 4, family = "font", color = "black") +
  labs(x = "", y = "", title = "", subtitle = "Puerto Rican", caption = "") +
  ggsave("E://pr_vote.png", width = 5, height = 3)

library(patchwork)

all <- one / two / three

ggsave("E://vote16_pr_ep_jew.png", type = "cairo-png", height = 7, width = 7,  all)


graph <- cc %>% 
  filter(income <= 16) %>% 
  group_by(group) %>% 
  ct(income, show_na = FALSE, wt = weight)

graph %>% 
  filter(group != "NA") %>% 
  ggplot(., aes(x = income, y = pct, fill = group)) +
  geom_col(color = "black") +
  facet_wrap(~ group, ncol = 1) +
  fill4_3() +
  theme_gg("Jost") +
  scale_x_continuous(breaks = c(1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16), labels = c("<$10k", "$10k", "$20k", "$30k", "$40k", 
                                                                                    "$50k", "$60k", "$70k", "$80k", "$100k", 
                                                                                    "$120k", "$150k", "$200k", "$250k", "$350k", ">$500k")) +
  y_pct() + 
  lab_bar(top = TRUE, type = pct, pos = .01, sz = 3) +
  labs(x = "", y = "", title = "Income Distributions of Each Group", caption = "") +
  ggsave("E://income_dist_pr_jew_ep.png", width = 9, type = "cairo-png")


graph <- cc %>% 
  group_by(group) %>% 
  mutate(inc = frcode(income <= 5 ~ "<$50k",
                      income >= 6 & income <= 9 ~ "$50k-$100k",
                      income >= 10 & income <= 16 ~ ">$100k")) %>% 
  ct(inc, wt = weight, show_na = FALSE) %>% 
  na.omit()


graph %>% 
  ggplot(., aes(x = inc, y = pct, fill = group)) +
  geom_col(color = "black") +
  facet_wrap(~ group, ncol = 3) +
  fill4_3() +
  theme_gg("Jost") +
  y_pct() + 
  lab_bar(type = pct, pos = .02, sz = 5.5) +
  labs(x = "", y = "", title = "Income Distribution Among the Three Groups", caption = "") +
  ggsave("E://income_dist_pr.png", dpi = 300, type = "cairo-png", width = 8)


cc <- cc %>% 
  mutate(trump = case_when(vote == 1 ~ 1, 
                           vote == 2 ~ 0)) %>% 
  mutate(male = case_when(gender == 1 ~ 1, 
                          gender == 2 ~ 0)) %>% 
  mutate(att = case_when(pew_churatd == 6 ~ 1, 
                         pew_churatd == 5 ~ 2, 
                         pew_churatd == 4 ~ 3,
                         pew_churatd == 3 ~ 4, 
                         pew_churatd == 2 ~ 5, 
                         pew_churatd == 1 ~ 6)) %>% 
  mutate(income = car::recode(income, "20:99 = NA")) %>% 
  mutate(white = case_when(race == 1 ~ 1, 
                           TRUE ~ 0))

reg1 <- glm(trump ~ income*group + age + educ + white + male + att, family = "binomial", data = cc)

gg <- interact_plot(reg1, pred= income, modx = group, int.width = .76, interval = TRUE) 

gg + 
  fill4_3() + 
  color4_3() + 
  y_pct() + 
  theme_gg("Jost") +
  theme(legend.position = c(.2, .8)) +
  scale_x_continuous(breaks = c(1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16), labels = c("<$10k", "$10k", "$20k", "$30k", "$40k", 
                                                                                    "$50k", "$60k", "$70k", "$80k", "$100k", 
                                                                                    "$120k", "$150k", "$200k", "$250k", "$350k", ">$500k")) +
  labs(x = "", y = "Pr(Trump Vote)", title = "Impact of Income on Likelihood of a Trump Vote") +
  ggsave("E://interact_pr_jews.png", type = "cairo-png", width = 8)