TidyTuesday Week 16: Fatal Car Crashes on DST days

Today we’re exploring the (lack of) connection between fatal car crashes in the United States on Daylight Savings Time

Figure 1

1. R code

# Load the packages in ----------------------------------------------------

if(!require(tidyverse)){install.packages("tidyverse"); library(tidyverse)}
if(!require(patchwork)){install.packages("patchwork"); library(patchwork)}
if(!require(ggplot2)){install.packages("ggplot2"); library(ggplot2)}
if(!require(sugrrants)){install.packages("sugrrants"); library(sugrrants)}
# I stick all my styling into a CUsotm PAckage to tidy up my code and keep it consistent over the time
if(!require(CustomGGPlot2Theme)){devtools::install("CustomGGPlot2Theme"); library(CustomGGPlot2Theme)}

# get the wd - force of habit
wd <- getwd()
#pulling images off internet was taking a lot of time and casing a time out error
options(timeout = 1000) 



font <- "roboto_bold"

das <- readr::read_csv('https://raw.githubusercontent.com/rfordatascience/tidytuesday/main/data/2025/2025-04-22/daily_accidents.csv')

das <- das %>%
  mutate(
    year = lubridate::year(date),
    date_time = as.POSIXct(date, tz = "UTC"), 
    local_time = with_tz(date_time, tzone = "America/New_York"),
    utc_offset = format(local_time, "%z")
  )

das <- das %>%
  arrange(date) %>%
  mutate(
    prev_offset = lag(utc_offset),
    dst_switch = utc_offset != prev_offset
  )

das <- das %>%
  mutate(
    dst_change = case_when(
      is.na(prev_offset) ~ NA_character_,
      utc_offset > prev_offset ~ "Forward",   # Spring: clocks jump ahead
      utc_offset < prev_offset ~ "Backward",  # Fall: clocks go back
      TRUE ~ NA_character_
    )
  )

dst_vs_avg <- das %>%
  filter(year >= 2010) %>%
  group_by(year) %>%
  mutate(avg_fatalities = mean(fatalities_count, na.rm = TRUE)) %>%
  ungroup() %>%
  filter(!is.na(dst_change)) %>%
  group_by(year, dst_change, avg_fatalities) %>%
  summarise(
    dst_fatalities = mean(fatalities_count, na.rm = TRUE),
    .groups = "drop"
  ) %>%
  mutate(
    pct_diff = 100 * (dst_fatalities - avg_fatalities) / avg_fatalities
  )

p1 <- ggplot(dst_vs_avg, aes(x = factor(year), y = pct_diff, fill = dst_change)) +
  geom_col(position = "dodge") +
  geom_hline(yintercept = 0, linetype = "dashed", color = "gray40") +
  scale_fill_manual(values = c("Forward" = "steelblue", "Backward" = "tomato")) +
   geom_text(
    aes(
      label = paste0(round(pct_diff, digits = 1), "%"),
      #had to do this for dynamic position of the 
      vjust = case_when(pct_diff >= 0 ~ -0.5, TRUE ~ 1.5)
    ),
    position = position_dodge(width = 0.9),
    size = 3.5
  ) +
  labs(
    title = str_wrap("Percent Difference in Fatalities on Daylight Savings Change Days (Clocks Forward vs Backward)", 60),
    subtitle = "Compared to annual average fatalities",
    x = "Year", y = "% Difference",
    fill = "Clock Change"
  ) +
  Custom_Style()