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

Original


Source: Alcohol and Drug Use in the FSM (1997).


Objective

The objective of the original data visualisation is to compare which states in the Federated States of Micronesia have higher alcohol consumption based on different age groups. The targetted audience can be everyone. For example, government authorities can use the data to control alcohol abuse and alcoholism in each state. Besides, the liquor company can use the data to find out the target market in each state.

The visualisation chosen had the following three main issues:

  • Briefly explain issue 1

The colours combination that is used in this visualization is terrible, the bright colours irritate our eyes when we look at it for some time. The data visualization should use soft and natural colour to deliver the message instead of using such bright colours to represent different states. Therefore, we use ColorBrewer 2.0 web tool to help us fix the colour themes for this visualisation.

  • Briefly explain issue 2

The use of colours (red and green) in this data visualisation does not consider the people with colour blindness. Some people are not able to distinguish particular colours, the red-green colour blindness types are the most common one(Baglin, 2019). Hence, we use Coblis Colour Blindness Simulator tool to check the visualisation using a red-green colour blindness simulator and avoid using a combination of red and green in the same visualisation.

  • Briefly explain issue 3

The use of a 3D bar graph cause confusion and misleading. It is difficult for the viewer to judge exactly how tall the individual bars are and it is also difficult to make direct comparisons between variables (Wilke, 2019). For example, the alcohol use in the age group of 65+ for Pohnpei and Chunk look the same, but in fact, the alcohol use by Chunk state is slightly higher than Pohnpei State. To fix this issue, we use facet to break the visualisation into subsets based on the States. Hence, the viewer can easily compare alcohol use by different age groups between four different states.

Reference

Code

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

#import data
alcohol <- readxl::read_xlsx("Alcohol.xlsx")

# reconstruct the graph
library(ggplot2)
p1 <- ggplot(data = alcohol, aes(x=Age, y=Alcohol, fill=State))

p1 <- p1 + geom_bar(stat='identity',position="dodge") +labs(
  title = "Alcohol Use by States and Age Cohorts",x = "Age Group",y = "Alcohol Use (Percent)")+
  scale_fill_manual(values=c('#a6cee3','#1f78b4','#b2df8a','#33a02c'))+
  coord_cartesian(ylim=c(0,60))+
  geom_text(aes(x= Age, y= Alcohol, label= paste(Alcohol, "%"), group = State),
            position = position_dodge(width = 1), vjust = -0.5, size = 2) +
  facet_wrap(~State)+ 
  theme(
    plot.title = element_text(size = 12, face = "bold"),
    axis.title.x = element_text(size = 10, face = "bold"),
    axis.title.y = element_text(size = 10, face = "bold"),
    legend.title = element_text(size = 10, face = "bold"),
    strip.text.x = element_text(size = 10, face = "bold"),
    strip.background = element_blank(),
    panel.background = element_blank(),
    panel.border = element_rect(colour = "black", fill=NA, size = 1))

Data Reference

Reconstruction

The following plot fixes the main issues in the original.