Created
January 31, 2018 15:49
-
-
Save joshkatz/c4fd0bdd0e2a06720b71d62d0b94c739 to your computer and use it in GitHub Desktop.
State of the Union WPM chart
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
needs(rvest, tidyverse, zoo) | |
df <- c('minutes', 'words') %>% | |
map(function(var) { | |
doc <- var %>% | |
sprintf(fmt = 'http://www.presidency.ucsb.edu/sou_%s.php') %>% | |
read_html() | |
doc %>% | |
html_node('table tr:nth-of-type(2) table') %>% | |
html_table(fill = T) %>% | |
select(X1:X4) %>% | |
set_names(c('president', 'date', 'type', var)) %>% | |
filter(row_number() > 10) %>% | |
mutate(date = as.Date(date, '%B %d, %Y'), | |
president = ifelse(is.na(date), president, NA) %>% | |
na_if('') %>% | |
na.locf(F)) %>% | |
filter(!is.na(date)) %>% | |
select(-type) | |
}) %>% | |
reduce(left_join) %>% | |
rename(duration = minutes) %>% | |
mutate_at(vars(words), as.numeric) | |
df <- df %>% | |
mutate(minutes = coalesce(duration %>% | |
str_split(':') %>% | |
map(as.numeric) %>% | |
map_dbl(~ 60 * .x[1] + .x[2] + .x[3] / 60), | |
duration %>% | |
str_extract(' ([0-9]+)') %>% | |
as.numeric()), | |
wpm = words / minutes) %>% | |
mutate(last_name = map_chr(str_split(president, ' '), last)) %>% | |
mutate(president = factor(president, levels = unique(.$president))) %>% | |
filter(!is.na(minutes)) | |
colors <- c(gray(.25), gray(.5)) %>% | |
rep_len(nlevels(df$president) - 1) %>% | |
c('red') | |
ggplot(df, aes(date, wpm, color = president)) + | |
geom_line() + | |
geom_point() + | |
geom_text(aes(y = 68, label = last_name), size = 3.5, | |
data = function(df) df %>% | |
group_by(president, last_name) %>% | |
summarize(date = mean(date))) + | |
scale_color_manual(values = colors) + | |
ggtitle('State of the Union: Words per Minute') + | |
theme(axis.title = element_blank(), | |
legend.pos = 'none', | |
plot.margin = margin(80, 40, 80, 40)) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment