ct_selection_untitled-32D56CBC.R

over <- gss %>% 
  filter(year == 1982 | year == 1987) %>% 
  group_by(year) %>% 
  ct(reltrad, wt = oversamp)

## This is the weight for the rest of the sample ####
wtss <- gss %>% 
  filter(year <= 2018) %>% 
  group_by(year) %>% 
  ct(reltrad, wt = wtssall)

##Removing the two years that used the overweight ####
wtss <- wtss %>% 
  filter(year != 1982) %>% 
  filter(year != 1987)


last <- gss %>% 
  filter(year == 2021) %>% 
  ct(reltrad, wt = wtssps) %>% 
  mutate(year = 2021)

## Bind them both together ####
graph <- bind_rows(over, wtss, last)

graph <- graph %>% 
  mutate(reltrad = frcode(reltrad == 1 ~ "Evangelical",
                          reltrad == 2 ~ "Mainline",
                          reltrad == 3 ~ "Black Prot.",
                          reltrad == 4 ~ "Catholic",
                          reltrad == 5 ~ "Jewish",
                          reltrad == 6 ~ "Other Faith",
                          reltrad == 7 ~ "No Religion", 
                          TRUE ~ "Unclassified")) 


graph %>% 
  filter(reltrad == "Catholic") %>% 
  ggplot(., aes(x = year, y = pct)) +
  geom_line() +
  geom_point(stroke = 1, shape = 21, fill = "white") + 
  geom_smooth(se = FALSE, linetype = "twodash", color = "black") +
  scale_y_continuous(labels = percent, limits = c(0, .29)) +
  theme_rb() +
  labs(x = "", y = "", title = "Share of Americans Who Identify as Catholic", caption = "@ryanburge\nData: General Social Survey, 1972-2021")
save("cath_gss21.png")


over <- gss %>% 
  filter(year == 1982 | year == 1987) %>% 
  filter(reltrad == 1 | reltrad == 2 | reltrad == 3 | reltrad == 4) %>% 
  mutate(att = case_when(attend == 6 | attend == 7 | attend == 8 ~ 1, 
                         attend <= 5 ~ 0)) %>% 
  group_by(year, reltrad) %>% 
  mean_ci(att, wt = oversamp)

## This is the weight for the rest of the sample ####
wtss <- gss %>% 
  filter(year <= 2018) %>% 
  filter(reltrad == 1 | reltrad == 2 | reltrad == 3 | reltrad == 4) %>% 
  mutate(att = case_when(attend == 6 | attend == 7 | attend == 8 ~ 1, 
                         attend <= 5 ~ 0)) %>% 
  group_by(year, reltrad) %>% 
  mean_ci(att, wt = wtssall)

##Removing the two years that used the overweight ####
wtss <- wtss %>% 
  filter(year != 1982) %>% 
  filter(year != 1987)


last <- gss %>% 
  filter(year == 2021) %>% 
  filter(reltrad == 1 | reltrad == 2 | reltrad == 3 | reltrad == 4) %>% 
  mutate(att = case_when(attend == 6 | attend == 7 | attend == 8 ~ 1, 
                         attend <= 5 ~ 0)) %>% 
  group_by(year, reltrad) %>% 
  mean_ci(att, wt = wtssps) %>% 
  mutate(year = 2021)

## Bind them both together ####
graph <- bind_rows(over, wtss, last)

graph <- graph %>% 
  gss_reltrad(reltrad)

graph %>% 
  ggplot(., aes(x = year, y = mean, color = reltrad, group = reltrad)) +
  geom_point(stroke = .5, shape = 21, alpha = .45) +
  geom_labelsmooth(aes(label = reltrad),  method = "lm", formula = y ~ x, family = "font", linewidth = 1, text_smoothing = 30, size =6, linewidth = 1, boxlinewidth = 0.3, hjust = .65) +
  y_pct() + 
  scale_color_manual(values = c(met.brewer("Johnson", 4))) +
  theme_rb() + 
  labs(x = "Year", y  = "", title = "Share Attending Services Nearly Every Week or More", caption = "@ryanburge\nData: General Social Survey, 1972-2021")
save("wk_att_reltrad.png")
  



over <- gss %>% 
  filter(year == 1982 | year == 1987) %>% 
  filter(reltrad == 4) %>% 
  mutate(pid7 = partyid + 1) %>% 
  cces_pid3(pid7) %>% 
  mutate(att = case_when(attend == 6 | attend == 7 | attend == 8 ~ 1, 
                         attend <= 5 ~ 0)) %>% 
  group_by(year, reltrad, pid3) %>% 
  mean_ci(att, wt = oversamp)

## This is the weight for the rest of the sample ####
wtss <- gss %>% 
  filter(year <= 2018) %>% 
  filter(reltrad == 4) %>% 
  mutate(pid7 = partyid + 1) %>% 
  cces_pid3(pid7) %>% 
  mutate(att = case_when(attend == 6 | attend == 7 | attend == 8 ~ 1, 
                         attend <= 5 ~ 0)) %>% 
  group_by(year, reltrad, pid3) %>% 
  mean_ci(att, wt = wtssall)

##Removing the two years that used the overweight ####
wtss <- wtss %>% 
  filter(year != 1982) %>% 
  filter(year != 1987)


last <- gss %>% 
  filter(year == 2021) %>% 
  filter(reltrad == 4) %>% 
  mutate(pid7 = partyid + 1) %>% 
  cces_pid3(pid7) %>% 
  mutate(att = case_when(attend == 6 | attend == 7 | attend == 8 ~ 1, 
                         attend <= 5 ~ 0)) %>% 
  group_by(year, reltrad, pid3) %>% 
  mean_ci(att, wt = wtssps) %>% 
  mutate(year = 2021)

## Bind them both together ####
graph <- bind_rows(over, wtss, last) %>% filter(pid3 != "NA")

graph %>% 
  ggplot(., aes(x = year, y = mean, color = pid3, group = pid3)) +
  geom_point(stroke = .5, shape = 21, alpha = .45) +
  geom_labelsmooth(aes(label = pid3),  method = "loess", formula = y ~ x, family = "font", linewidth = 1, text_smoothing = 30, size =5, linewidth = 1, boxlinewidth = 0.3, hjust = .65) +
  pid3_color() +
  y_pct() + 
  theme_rb() +
  labs(x = "Year", y = "", title = "Share of Catholics Attending Mass Nearly Every Week or More", caption = "@ryanburge\nData: General Social Survey, 1972-2021")
save("cath_pid3_attend.png")  


ggg <- cces %>% 
  filter(year == 2008 | year == 2012 | year == 2016 | year == 2020 | year == 2022) %>% 
  cces_pid3(pid7) %>% 
  filter(race == 1 & religion == 2) %>% 
  mutate(att2 = frcode(pew_attendance == 6 | pew_attendance == 5 ~ "Never/Seldom", 
                       pew_attendance == 1 | pew_attendance == 2 ~ "Weekly")) %>% 
  group_by(year, att2) %>% 
  ct(pid3, wt = weight, show_na = FALSE) %>% 
  filter(att2 != "NA")



ggg %>% 
  mutate(lab = round(pct, 2)) %>% 
  ggplot(., aes(x = 1, y = pct, fill = fct_rev(pid3))) +
  geom_col(color = "black") + 
  coord_flip() +
  facet_grid(year ~ att2, switch = "y") +
  pid3_fill(rev = TRUE) +
  theme_rb(legend = TRUE) + 
  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 >.08, paste0(lab*100, '%'), '')), position = position_stack(vjust = 0.5), size = 6, family = "font", color = "black") +
  labs(x = "", y = "", title = "Partisanship of White Catholics by Church Attendance", subtitle = "", caption = "@ryanburge\nData: Cooperative Election Study, 2008-2022") 
save("att_cath_pid3_ces2022.png", wd = 9, ht = 4)