Click the Original, Code and Reconstruction tabs to read about the issues and how they were fixed.

Original


Source: Feller & Wendt (2019).


Objective

This visualisation was created by a law firm in Salt Lake City, USA, “to help educate drivers and make the roads of Salt Lake City a safer place.” The audience is residents of Salt Lake City, focusing on those who use road transport, which is generally everyone. It would be a reasonable assumption that people who look at a law firms website are searching for a lawyer to represent them, so the audience would be potential clients.

The intention of this visualisation is partly to inform road users of the risks associated with road transport and partly as advertising material for their business. Informing road users of the risk of fatalities will help them risk manage their activities on the roads.

The visualisation chosen had the following three main issues:

  1. Six data series are presented over the top of each other. For example, bicycle fatalities are hidden behind other data lines at the bottom of the visualisation. The impact of this is that it makes the visualisation impossible to read

  2. Colours are not suitable for colour blind people The colours chosen include green and green-blue mixed with red and red-blue hues. The impact of this is that identifying different types will not be possible for a small percentage of the population.

  3. The chart type does not clearly represent the discrete variable of fatality The choice of chart and axis in this visualisation does not clearly distinguish the discrete variable of fatality. The lines imply a continuous variable, and the Y-axis includes a decimal implying partial fatalities. The impact of this is that it makes the visualisation difficult to interpret without careful thought. The audience cannot glance at this visualisation and know what is going on.

Reference

Code

The following code was used to fix the issues identified in the original.

library(ggplot2)
library(lubridate)

fatalitiesData <- read.csv(file = "CarCrashSaltLakeCounty.csv", stringsAsFactors = TRUE)

p1 <- ggplot(data = fatalitiesData, mapping = aes(group = 1, x = dmy(MonthYear), y = Fatalities)) +
  geom_point(shape = 21, stroke = 0, size = 2.2, color = "skyblue3", fill = "lightblue") +
  geom_line(color = "lightblue") +
  facet_wrap(vars(Type), ncol = 2) +
  labs(title = "Salt Lake City Car Accident Statistics Over Five Years") +
  labs(x = "Months")

Data Reference

Reconstruction

The following plot fixes the main issues in the original.

The reconstruction fixes the original problems by separating each category into its own facet and by using a combination of points and lines to represent the data.

Points are large and include a darker border to accentuate the visual cue, and lines are used to create a visual connection between points, thus avoiding confusion that would occur with scattered data.