title	author	date	output	vignette
How to make a Nightingale's rose plot with ggplotAssist	Keon-Woong Moon	2017-11-15	rmarkdown::html_vignette	%\VignetteIndexEntry{RosePlot} %\VignetteEngine{knitr::rmarkdown} %\VignetteEncoding{UTF-8}

In 1858 nurse, statistician, and reformer Florence Nightingale published Notes on Matters Affecting the Health, Efficiency, and Hospital Administration of the British Army. Founded Chiefly on the Experience of the Late War. Presented by Request to the Secretary of State for War. This privately printed work contained a color statistical graphic entitled "Diagram of the Causes of Mortality in the Army of the East" which showed that epidemic disease, which was responsible for more British deaths in the course of the Crimean War than battlefield wounds, could be controlled by a variety of factors including nutrition, ventilation, and shelter. The graphic, which Nightingale used as a way to explain complex statistics simply, clearly, and persuasively, has become known as Nightingale's "Rose Diagram." (http://www.historyofinformation.com/expanded.php?id=3815)

Prerequisite

You have to install the developmental version of R package editData from github.

#install.packages("devtools")
devtools::install_github("cardiomoon/editData")

Install package

You can install ggplotAssist package from github.

#install.packages("devtools")
devtools::install_github("cardiomoon/ggplotAssist")

What is the rose plot?

A rose plot is a stacked bar plot with polar coordinate. With ggplot2, you can make the rose plot easily.

Data

I will use the rose data included in package ggiraphExtra. The rose data is a phony dataset representing rose sales.

library(ggiraphExtra)
head(rose)

# A tibble: 6 x 3
     Month group value
    <fctr> <chr> <dbl>
1  January     G    10
2 February     G     5
3    March     G     7
4    April     G     9
5      May     G    11
6     June     G    13

Alternatively, you can download the rose.csv file from my github using the following code.

library(readr)
url <- "https://raw.github.com/cardiomoon/ggplotAssistFigures/master/rose.csv"
rose=read_csv(url)

Parsed with column specification:
cols(
  Month = col_character(),
  group = col_character(),
  value = col_double()
)

rose$Month=factor(rose$Month,levels=month.name)

R code for rose plot

1. Make a bar plot

First, make a bar plot. In this plot the height of bar represent values in the data.

require(ggplot2)
ggplot(data=rose,aes(x=Month,y=value,fill=group))+geom_bar(stat="identity")

2. Adjust options for bar graph

The default bar width is 0.9(90% of the resolution of data). To remove the spaces between the bars, set the width of bar 1 and the boundary color "black" and the thickness of line 0.1.

ggplot(data=rose,aes(x=Month,y=value,fill=group))+
    geom_bar(stat="identity",width=1,color="black",size=0.1)

3. Apply polar coordinate

To make a rose plot, apply polar coordinate with coord_polar(). Apply "Reds" palette with scale_fill_brewer().

ggplot(data=rose,aes(x=Month,y=value,fill=group))+
    geom_bar(stat="identity",width=1,color="black",size=0.1)+
    coord_polar()+
    scale_fill_brewer(palette="Reds",direction=-1)

Make a rose plot with ggplotAssist - Standard method

1. In RStudio, Select File ... New File...R script. To use data rose, highlight the data name rose(1). Execute ggplotAssist addin(arrow), to interactively manipulate it.

2. To mapping the variables, select x among aesthetics and Month among mapping. Select y among aesthetics and value. Select fill(1) and group(arow) to map the fill variable.

3. Press geom button(1). Select geom_bar(2). Select identity as stat(3). To set the color argument of geom_bar(), select color(4) and enter black(5) and select addquote checkbox(6). The R code for this layer is shown(7) and you can see the plot preview at lower right corner(8).

4. Set the size(9) of bar 0.1(10). To remove the spaces between the bars, set the width of bar 1(11). You can see the R code for this layer(scarlet rectangle). Press Add Layer button to add this layer.

5. To apply Reds palette to this plot, press scale button(12). Select scale_fill_brewer(13) and select Reds palette(14) and set the direction of palette -1(15). Press Add layer button to add this layer(16).

6. To apply ploar coordinate, press coord button(17). Select coord_polar(18) and press Add Layer button(19).

7. You can see the plot layer by layer. Press the Layer by layer button(20) and press the arrow head(arrow). You can see the plot layer by layer.

Make a rose plot with ggplotAssist - A shortcut

1. After maaping the variables, press the geom button(1). Select geom_bar(2) among geoms and set the stat identity(3). You can see the R code for this layer(4) and the plot preview(5). Press the Make a Rose Plot button(6) and you can get the rose plot.

2. Press the Done button(arrow).

3. The code for the rose plot will be emitted at the cursor position(scarlet rectangle).