Visualization Exercise

This data comes from FiveThirtyEight’s article https://projects.fivethirtyeight.com/college-fight-song-lyrics/. It contains stats on the fight songs of 65 schools – all those in the Power Five conferences (the ACC, Big Ten, Big 12, Pac-12 and SEC), plus Notre Dame– including duration, beats per minute, and mentions of typical themes of a fight song. I am going to try to recreate the graph of fight song beats per minute versus duration. The original graph is below (with UGA selected); it is interactive, allowing users to select a university to see where they fall on the fight song bpm vs duration scale as well as how many “fight song cliches” it contains.

Here I load all the packages I need to recreate this graph and load the data provided by FiveThirtyEight.

#load packages
library(tidyverse)

── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
✔ dplyr     1.1.0     ✔ readr     2.1.4
✔ forcats   1.0.0     ✔ stringr   1.5.0
✔ ggplot2   3.4.1     ✔ tibble    3.1.8
✔ lubridate 1.9.2     ✔ tidyr     1.3.0
✔ purrr     1.0.1     
── 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(ggplot2)
library(plotly)

Attaching package: 'plotly'

The following object is masked from 'package:ggplot2':

    last_plot

The following object is masked from 'package:stats':

    filter

The following object is masked from 'package:graphics':

    layout

#load the data
fight_songs <-read_csv("data/fight-songs.csv")

Rows: 65 Columns: 23
── Column specification ────────────────────────────────────────────────────────
Delimiter: ","
chr (19): school, conference, song_name, writers, year, student_writer, offi...
dbl  (4): bpm, sec_duration, number_fights, trope_count

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.

The original graph includes intersecting lines with the average bpm and duration, so I took the mean of each column to determine those exact values so I could include them in my recreation.

fight_songs %>% pull(bpm) %>% mean()

[1] 128.8

fight_songs %>% pull(sec_duration) %>% mean()

[1] 71.90769

Now to the graphing. I decided to create this graph in ggplot first and then make it interactive using the ggplotly() function from plotly.

graph <- fight_songs %>% 
  #color code each point by school name and include number of tropes
  ggplot(aes(x=sec_duration, y=bpm, color=school, text=trope_count)) +
  #make a scatter plot
  geom_point(size=3, alpha=0.5) + 
  #include the average lines using the means I calculated above.
  geom_vline(xintercept = 71.90769) + 
  geom_hline(yintercept = 128.8) + 
  #omit the legend 
  theme(legend.position="none") + 
  #shift scale
  scale_x_continuous(limits= c(0,180), breaks = seq(0,180,20))+
  scale_y_continuous(limits= c(50,200), breaks= c(0,60,80,100,120,140,160,180,200)) +
  #match the quadrant labels 
  annotate("text", x = 20, y = 65, label = "Slow but short") + 
  annotate("text", x = 20, y = 190, label = "Fast and short") + 
  annotate("text", x = 135, y = 190, label = "Fast but long") + 
  annotate("text", x = 135, y = 65, label = "Slow and long") + 
  #match the axes labels and title
  labs(x="Duration", y="Beats per Minute") +
  ggtitle("How Fight Songs Stack Up")

ggplotly(graph)

There were two main things I struggled with in this recreation. I searched and searched for some “university color palette for R” but came up empty. The only options I was left with were using default ggplot colors or making my own using hex codes for each school. I decided that I would use default colors to save myself some sanity (and each school has its own color, just not its school color). I also could not figure out how to include a separate chart of sorts for the number of tropes in each song; after some searching, I found that the aes(text= ) in ggplot() mapping could be used to display the trope number in the hover menu after converting using ggplotly(). You can see in the final that Georgia does indeed have 1 fight song cliche in my recreation (the number at the bottom of the hover menu).

I played around with ggvis for a few hours on my first attempt and, while I found it interesting with cool outputs, I had a hard time figuring out how to add all the elements that I needed to recreate this graph. Upon reading, I learned that ggvis is a sort of work-in-progress, and everything in ggplot does not have a direct translation to ggvis. I think ggvis could be used to make some really cool visualizations, but I was unsuccessful in using it for this exercise.

Overall, this recreation is not an exact replica of the original but I think it turned out pretty good and I really enjoyed getting to play around with the interactive features of plotly().