Pell Grants – Unlocking the Power of Data Visualization with R

library(tidyverse)

pell <- readr::read_csv('https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2022/2022-08-30/pell.csv')

head(pell)

usdata::state_stats%>%names

51 abbr

abbr <- usdata::state_stats%>%
  count(abbr) %>%
  select(-n)

59 STATES

pell_abbr <- pell%>%
  count(abbr=STATE) %>%
  select(-n)

missing <- setdiff(pell_abbr,abbr)%>%
  unlist()

year_id <- pell%>%
  count(YEAR) %>%
  select(-n)%>%
  mutate(year_id=seq(1,length(YEAR),1))

df <- pell%>%
  arrange(YEAR) %>%
  filter(STATE%in%missing)%>%
  arrange(STATE) %>%
  mutate(NAME=case_when(NAME=="American Samoa Cmnty College"~"American Samoa Community College",
         TRUE~NAME)) %>%
  mutate(NAME_long=case_when(STATE=="PR"~"Puerto Rico",
                             STATE=="AS"~"American Samoa",
                             STATE=="GU"~"Guam",
                             STATE=="MH"~"Marshall Islands",
                             STATE=="PW"~"Palau",
                             STATE=="VI"~"Virgin Islands",
                             STATE=="MP"~"Marianas Islands",
                             STATE=="FM"~"Micronesia",
                             TRUE~STATE)) %>%
  count(STATE,NAME_long,AWARD,RECIPIENT,YEAR) %>%
  group_by(STATE) %>%
  mutate(n=n(),avg_single_aw=AWARD/RECIPIENT) %>%
  ungroup() %>%
  left_join(year_id,by="YEAR")

df

p <- df %>%
  ggplot(aes(YEAR,avg_single_aw,group=YEAR))+
  geom_boxplot()


p
plotly::ggplotly(p)

library(giscoR)
world <- gisco_get_countries()


states <- df %>%
  count(STATE) %>%
  select(-n)


small_countries <-world%>%
  filter(CNTR_ID%in%states$STATE)%>%
  left_join(df,by=c("CNTR_ID"="STATE"))

library(dplyr)
library(ggplot2)
library(sf)


cropped_world<- st_crop(world, 
        xmin = 120, xmax = 172,
        ymin = -14, ymax = 40)





small_countries_feat <- small_countries %>%
  st_drop_geometry() %>%
  #filter(CNTR_ID=="PR")%>%
  group_by(YEAR,CNTR_ID,NAME_ENGL)%>%
  summarise(avg_single_aw,
            mean_val=mean(avg_single_aw),.groups="drop")%>%
  ungroup() %>%
  count(YEAR,CNTR_ID,NAME_ENGL,mean_val) %>%
  group_by(CNTR_ID)%>%
  mutate(diff=c(0,diff(mean_val)),
         diff_prop=paste0(round(diff/mean_val*100,2),"%"),
        # diff_w=round(c(0,diff(mean_val))*1/n,2),
         avg=round(mean(round(diff/mean_val*100,2)),2))%>%
  ungroup()
  
small_countries_feat

names <- small_countries%>%
  count(CNTR_ID,NAME_ENGL,avg_single_aw)


subworld <- small_countries %>% 
  group_by(CNTR_ID) %>%
  # Mock the data field
  summarise(data=n())%>%
  ungroup()%>%
  mutate(id=seq(1,8,1))

text <- subworld %>% 
  #select(geometry) %>%
  st_cast("MULTIPOLYGON") %>%
 # st_coordinates() %>%
  st_centroid() %>%
  st_coordinates() %>%
  cbind(names) %>%
  select(-n)


text_full <- text %>%
  left_join(small_countries_feat,by=c("CNTR_ID","NAME_ENGL")) %>%
  count(X,Y,NAME_ENGL,avg) %>%
  mutate(text=paste0(NAME_ENGL,"\n",avg,"%\n"))

library(ggforce)

circles <-
data.frame(
x0 = text_full$X,
y0 = text_full$Y,
r = text_full$avg) #seq(min(text_full$avg), max(text_full$avg), length.out = 8))

library(extrafont)

ggplot() +
  geom_sf(data = cropped_world, fill="gray")+
  geom_sf(data = subworld, aes(color=CNTR_ID)) +
  geom_text(data = text_full, aes(x=X,y=Y,label=text),family = "Roboto Condensed") +
  geom_circle(data = circles, aes(x0 = x0, y0 = y0, 
                                  r = r,
                                  fill=r,
                                  alpha=r),
              color="gray",size=0.2) +
  scale_fill_gradient(low = "white",high = "blue")+
  scale_alpha_continuous(range = c(0.1,0.3))+
  coord_sf(xlim = c(120,171.8428), 
           ylim = c( -14.36381, 40.59467),
           crs = 4326) +
  ggthemes::theme_fivethirtyeight() +
  theme(text=element_text(family = "Roboto Condensed"),
        panel.grid = element_line(size=0.08),
        legend.position = "none")

ggsave("map2.png")

library(tidyverse)
library(cowplot)
ggdraw() +
  draw_image("map2.png",scale=0.99) +
  draw_image("globe.png",
             scale=0.2,
             x=0.409,y=0.4) +
  draw_label("Pell Awards",
             x=0.53,y=0.95,
             fontfamily = "Roboto Condensed",
             hjust=0,
             color="grey30",
             size=30) +
    draw_label("Pell Awards",
             x=0.53,y=0.95,
             fontfamily = "Roboto Condensed",
             hjust=0,
             color="#a5bec9",
             size=29.5) +
      draw_label("Basic Educational Opportunity Grants\nby the U.S. Department of Education",
             x=0.47,y=0.88,
             fontfamily = "Roboto Condensed",
             hjust=0,
             color="#24424f",
             size=12) +
        draw_label("Western Pacific Ocean Islands yearly variation+ (average values %) 1990-2017",
               x=0.03,y=0.06,angle = 90,
             #x=0.35,y=0.78,
             fontfamily = "Roboto Condensed",
             hjust=0,
             color="#24424f",
             size=14) +
  
    draw_label("DataSource: #TidyTuesday 2022 week35\nU.S. Department of Education",
             x=0.125,y=0.06,angle = 90,
             fontfamily = "Roboto Condensed",
             hjust=0,size=8) +
      draw_label("DataVisualization: Federica Gazzelloni - @fgazzelloni",
             x=0.6,y=0.015,angle = 0,
             fontfamily = "Roboto Condensed",
             hjust=0,size=9) 

ggsave("w35_pell.png",bg="#b1d2e0") #grey65 Saving 7 x 7 in image