Exercise 10

Task 1: Reflection

One feature I find makes a really good dashboard is when the different elements are connected and fit well together. I remember one time seeing a dashboard with two graphs stacked on top of one another. The top was a line graph showing each country’s CO2 emissions by year over the last decade or so. The bottom one was a world map that shaded each country by its CO2 emissions in the most recent year in the data. What was really cool is that when you selected a line it would show the country name and data for that country and only highlight that country on the map. I think using interactivity to make two visualizations work together like this is a great way to make a cool and informing dashboard. Cramming as many graphs as you can onto a dashboard can be overwhelming for the reader so it is best to think about how the separate elements can work together to tell one story.

Put your reflection here

Task 2: Interactive plots

library(tidyverse)
library(plotly)
library(gghalves)
library(viridis)

df = read_csv("data/pgaTourData copy.csv")

df$money <- as.numeric(gsub("[$,]", "", df$Money))

Do the following:

1. Make a plot. Any kind of plot will do (though it might be easiest to work with geom_point()).

df = df %>%  rename(driving_distance = `Avg Distance`) 

df <- df %>% 
  mutate(distance_quantile = cut(driving_distance,
                                 breaks = quantile(driving_distance,
                                                   c(0, 0.25, 0.5, 0.75, 1),
                                                   na.rm = TRUE),
                                 labels = c("0-25th", "26-50th", "50-75th",
                                            "Over 75th")))
 

df2 = df %>% 
  drop_na()



plot1 = ggplot(df2, aes(x = distance_quantile, y = Points))+
  geom_half_point(aes(color = distance_quantile), side = "l", size = 0.5) +
  geom_half_violin(aes(fill = distance_quantile), side = "r")+
  guides(color = "none", fill = "none") +
  coord_flip()+
  labs(title = "Distribution of Season Points by Driving Distance Bracket",
       x = "Driving Distance Percentile",
       y = "Season Points") +
  theme_bw()+
  scale_color_viridis(option = "cividis", discrete = T)+
  scale_fill_viridis(option = "cividis", discrete = T)


plot1

plot2 = ggplot(df2, aes(x = distance_quantile, y = Points, text = Year))+
  labs(title = "Distribution of Season Points by Driving Distance Bracket",
       x = "Driving Distance Percentile",
       y = "Season Points")+ 
  geom_jitter(width = 0.2, height = 0, aes(color = distance_quantile), alpha = 0.7)+
  guides(color = "none")+
  scale_color_viridis(option = "A", discrete = T, begin = 0, end = .7)

plot2

2. Make the plot interactive with ggplotly().

ggplotly(plot2)

3. Make sure the hovering tooltip is more informative than the default.

ggplotly(plot2, tooltip = "text")

df2 = df2 %>% 
  mutate(label = paste0( "Player: ", `Player Name`, "<br>", 
                       "Year: ", Year,"<br>",
                       "Points: ", Points))

plot3 = ggplot(df2, aes(x = distance_quantile, y = Points, text = label))+
  labs(title = "Distribution of Season Points by Driving Distance Bracket",
       x = "Driving Distance Percentile",
       y = "Season Points")+ 
  geom_jitter(width = 0.2, height = 0, aes(color = distance_quantile), alpha = 0.7)+
  guides(color = "none")+
  scale_color_viridis(option = "A", discrete = T, begin = 0, end = .7)+
  theme_bw()


ggplotly(plot3, tooltip = "label")