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

Original


Source: The State of The North 2018 by IPPR North .


Objective

I have taken this chart from a report titled ‘the State of The North 2018’ published by the Institute for Public Policy Research (IPPR) based in Edinburgh in December 2018. The authors of this report are Luke Raikes who is a senior research fellow at IPPR North, Leah Milward was a Q-step intern at IPPR North for the summer of 2018 and Sarah Longlands is director of IPPR North. This is an official report is about reprioritising the northern powerhouse. IPPR, the Institute for Public Policy Research, is the UK’s leading progressive think tank whose purpose is to conduct and promote research into and the education of the targeted readers from the general public, professional and researchers in social science and those who are interested in the development and current situation of the Northern Powerhouse.

In Figure 4.7, the authors want to illustrate that people live longer and will spend more time with health problems, and on average northerner are already in poor health by the time they retire.

The visualisation chosen had the following three main issues:

  • Initially I would like to point out that, there is no label for the chart, like region on the x axis and life expectancy on the y axis and how the bar charts are denoted by gender as a legend, which made the chart really confusing.
  • Please note that the labels on the horizontal axis are hard to read. Since the original illustration is plotted with female and male in different color, there is a risk of repetition by adding label for female and male in a same region separately, which is not necessary.
  • Moreover, it is barely possible for the viewer to figure out what do the floating bars mean, especially the lower bound of the bar. Viewers might mistake the value of blue horizontal line for zero. From these zig-zag bars, it is hard to recognize the trend that people become unhealthy earlier in northern regions than in south.

Reference

Code

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

getwd()
## [1] "C:/Users/Srikanth/Desktop/Grad certificate in Analytics with RMIT/Data Visualisation/Assignment 2"
library(ggplot2)
df <- data.frame(region=rep(c("North East", "North West", "Yorkshire and Humber","East Midlands","West Midlands","East","London","South East","South West"), each=2),
                  sex=rep(c("Women", "Men"), 9),
                  
                  life_expectancy=c(81.8,77.5,
                                    82.2,77.8,
                                    82.6,78.2,
                                    83,78.9,
                                    82.8,78.0,
                                    83.3,80.2,
                                    84,80.2,
                                    83.7,80.3,
                                    83.7,80.1),
                  
                  healthy_life_expectancy=c(60.6,59.8,
                                            62.1,61.1,
                                            61.6,61.4,
                                            63,63.1,
                                            63.5,62.9,
                                            65.2,65.1,
                                            64.8,63.7,
                                            66.3,66.1,
                                            65,64.9)
)

df$region<-factor(df$region, levels=c("North East", "North West", "Yorkshire and Humber","East Midlands","West Midlands","East","London","South East","South West"))

d1<-ggplot(data = df,
            aes(x=region)) +
  geom_bar(aes(y=healthy_life_expectancy,colour=sex),
           stat="identity",
           position="dodge",
           alpha=1
  ) +
  geom_bar(aes(y=life_expectancy,group=sex, colour=sex),
           stat="identity",
           position="dodge",
           alpha=.4) +
  scale_fill_discrete() + 
  geom_hline(yintercept = 65,color="blue") +  coord_cartesian(ylim = c(30,85))  + labs(title="Healthy Life Expectancy by Sex in North of England",y="Life Expectancy (years)", x="Region") +
  ggsave("geom_bar_healthy_life_expectancy by sex in North of England.jpg", width = 18, height = 12, units = "cm")

Data Reference

Reconstruction

The following plot fixes the main issues in the original.