80 other geoms 2
Written by Ananya Jha and last updated on 5 February 2022
80.1 Introduction
Ggplot2 is powerful package that offers a wide variety of functions and options to produce graphics. So far, we have covered some important and useful individual geoms. In this lesson we will be talking about some other individual geoms that are used to create the basic layers of data.
In this lesson, you will learn how to:
- Create interesting graphics such as density plots, polygons, violin plots, and rectangles using
geom_density(),geom_polygon(),geom_violin(), andgeom_rect(). - Use
geom_polygon()to create basic maps using the data from maps package - Differentiate between and know the correct use cases of the three different rectangle functions:
geom_raster(),geom_rect(), andgeom_tile().
Prerequisites:
- Having the most updated version of the
ggplot2andmapspackage installed and loaded - Basic understanding of
ggplot2and experience working with the basic functions and aesthetics such as: fill, factor() etc.
Let’s first try to understand what these geoms are and take a quick look at their function definitions:
80.2 Density Plots
A Density plot(or Kernel density plot) is used to display the distribution of a numeric variable and can be created using the geom_density() function in R. It is similar to the histogram for continuous data but gives a smoothed version of the graph.
geom_density(mapping = NULL, data = NULL, stat = “density,” position = “identity,” …, na.rm = FALSE, orientation = NA, show.legend = NA, inherit.aes = TRUE, outline.type = “upper”)
Where:
| Argument | Details |
|---|---|
| mapping | Set of aesthethic mappings created by aes(). By default, it is combined with the mapping specified in the top layer of the plot. |
| data | The data to be dispayed. If not specified, inherited directly from top layer. |
| stat | The statistical transformation to use on the data for this layer, as a string. |
| position | The position adjustment. Is “identity” by default for density plots. |
| … | Other optional arguments(like aesthetics such as color/size etc.) |
| na.rm | Can be used to remove missing values without a warning. Is FALSE by default. |
| orientation | By default, orientation would follow from aesthetic mapping but to change manually you can set this to “x” or “y” |
| show.legend | Set this to TRUE to include legend. Is NA by default which woulde include if aesthetics are mapped. |
80.3 Violin Plots
Violin plots can be thought of as a hybrid of the box plot and kernel density plot. They are used with numerical data to display the distribution and the summary statistics. In R, violin plots can be created with the geom_violin() function.
geom_violin(mapping = NULL, data = NULL, stat = “ydensity,” position = “dodge,” …, draw_quantiles = NULL, trim = TRUE, scale = “area,” na.rm = FALSE, orientation = NA, show.legend = NA, inherit.aes = TRUE)
Where:
| Argument | Details |
|---|---|
| mapping | Set of aesthethic mappings created by aes(). By default, it is combined with the mapping specified in the top layer of the plot. |
| data | The data to be displayed. If not specified, inherited directly from top layer. |
| stat | The statistical transformation to use on the data for this layer, as a string. |
| position | The position adjustment. Is “dodge” by default for violin plots |
| draw_quantiles | Set to not(NULL) to draw lines to indicate quantiles |
| trim | Set to FALSE to not trim the tails to match the range of data |
| scale | By default all violins will have the same area. Set to “count” to scale to number of observations or to “width” to make all violins have same max width |
| … | Other optional arguments(like aesthetics such as color/size etc.) |
| na.rm | Can be used to remove missing values without a warning. Is FALSE by default. |
| orientation | By default, orientation would follow from aesthetic mapping but to change manually you can set this to “x” or “y” |
| show.legend | Set this to TRUE to include legend. Is NA by default which woulde include if aesthetics are mapped. |
| inherit.aes | Set this to FALSE to override default aesthetics inherited from the first layer |
80.4 Polygons
To draw polygons, which can also be thought of as filled paths(more on this later), we use the geom_polygon() function. The function generally requires 2 data frames- one for the polygon coordinates(positions) and one for the values of each polygon(values). Polygons are considered very useful when working with map/spatial data or when you wish to layer simple polygons over other plots.
geom_polygon(mapping = NULL, data = NULL, stat = “identity,” position = “identity,” rule = “evenodd,” …, na.rm = FALSE, show.legend = NA, inherit.aes = TRUE)
Where:
| Argument | Details |
|---|---|
| mapping | Set of aesthetic mappings created by aes(). By default, it is combined with the mapping specified in the top layer of the plot. |
| data | The data to be displayed. If not specified, inherited directly from top layer. |
| stat | The statistical transformation to use on the data for this layer, as a string. |
| position | The position adjustment. Is “identity” by default for polygons. |
| … | Other optional arguments(like aesthetics such as color/size etc.) |
| na.rm | Can be used to remove missing values without a warning. Is FALSE by default. |
| show.legend | Set this to TRUE to include legend. Is NA by default which woulde include if aesthetics are mapped. |
| inherit.aes | Set this to FALSE to override default aesthetics inherited from the first layer. |
80.5 Rectangles
Rectangles are mainly used to draw surfaces on plots but sometimes we might want to draw random rectangles as well. As mentioned previously, there are 3 different functions in R that can be used to graph rectangles. We will be exploring the differences between these later.
geom_raster(mapping = NULL, data = NULL, stat = “identity,” position = “identity,” …, hjust = 0.5, vjust = 0.5, interpolate = FALSE, na.rm = FALSE, show.legend = NA, inherit.aes = TRUE)
geom_rect(mapping = NULL, data = NULL, stat = “identity,” position = “identity,” …, linejoin = “mitre,” na.rm = FALSE, show.legend = NA, inherit.aes = TRUE)
geom_tile(mapping = NULL, data = NULL, stat = “identity,” position = “identity,” …, linejoin = “mitre,” na.rm = FALSE, show.legend = NA, inherit.aes = TRUE)
Where:
| Function(s) | Argument | Details |
|---|---|---|
| geom_raster(), geom_rect(), geom_tile() | mapping | Set of aesthetic mappings created by aes(). By default, it is combined with the mapping specified in the top layer of the plot. |
| geom_raster(), geom_rect(), geom_tile() | data | The data to be displayed. If not specified, inherited directly from top layer. |
| geom_raster(), geom_rect(), geom_tile() | stat | The statistical transformation to use on the data for this layer, as a string. |
| geom_raster(), geom_rect(), geom_tile() | position | The position adjustment. Is “identity” by default for rectangles. |
| geom_raster(), geom_rect(), geom_tile() | … | Other optional arguments(like aesthetics such as color/size etc.) |
| geom_raster() | hjust, vjust | Horizontal and Vertical justification for the plot(can be between 0-1). 0 would mean left-justified and 1 would be right-justified |
| geom_raster() | interpolate | Set this to TRUE to get a smoothed output |
| geom_raster(), geom_rect(), geom_tile() | linejoin | The line join style preferred |
| geom_raster(), geom_rect(), geom_tile() | na.rm | Can be used to remove missing values without a warning. Is FALSE by default. |