ggplot2: boxplot with colors and text labels mapped to combination of two categorical variables

Question

I would like to create a boxplot with ggplot2 from a dataset densityAGRLKA with 3 categorical variables (species, location, position) on the x-axis.

The following function:

ggplot(densityAGRLKA, aes(species, density, fill=location, alpha=position), dodge=species, position) + 
stat_boxplot(geom ='errorbar') + 
geom_boxplot() 

creates a plot, in which the grouping of the species is fine, but the colours are misleading. I have no idea how to fix this.

I need a plot with the following properties:

• the data is grouped by species
• within the group by location,
• and within the location in the order top and then bottom.

Additionally, it would be great, if location would be written underneath the two boxes that belong together, and position underneath every single box. Or maybe better colouring/shading the boxes and then provide a clear legend?

Sample data:

densityAGRLKA = structure(list(location = structure(c(1L, 1L, 1L, 1L, 1L, 1L, 
    1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 1L, 1L, 
    1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 
    2L, 2L), .Label = c("SF", "SS"), class = "factor"), species = structure(c(1L, 
    1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
    1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 
    2L, 2L, 2L, 2L, 2L, 2L, 2L), .Label = c("AGR", "LKA"), class = "factor"), 
        position = structure(c(1L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 
        2L, 1L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L, 1L, 1L, 1L, 1L, 
        1L, 2L, 2L, 2L, 2L, 2L, 1L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 
        2L), .Label = c("top", "bottom"), class = "factor"), density = c(0.41, 
        0.41, 0.43, 0.33, 0.35, 0.43, 0.34, 0.46, 0.32, 0.32, 0.4, 
        0.4, 0.45, 0.34, 0.39, 0.39, 0.31, 0.38, 0.48, 0.3, 0.42, 
        0.34, 0.35, 0.4, 0.38, 0.42, 0.36, 0.34, 0.46, 0.38, 0.36, 
        0.39, 0.38, 0.39, 0.39, 0.39, 0.36, 0.39, 0.51, 0.38)), .Names = c("location", 
    "species", "position", "density"), row.names = c(NA, -40L), class = "data.frame")

Answer 1

Here are three progressively more involved options for adding text in the way you describe in your question, followed by a different approach using faceting:

Set-up

First, create a couple of utility values we'll use later:

# Color vectors
LocPosCol = c(hcl(0,100,c(50,80)), hcl(240,100,c(50,80)))
LocCol = c(hcl(c(0,240),100,65))

# Dodge width
pd = position_dodge(0.7)

Now create a basic boxplot. We use the interaction function to create a fill aesthetic based on all combinations of location and position:

p = ggplot(densityAGRLKA, 
           aes(species, density, 
               fill=interaction(location, position, sep="-", lex.order=TRUE))) + 
  geom_boxplot(width=0.7, position=pd) + 
  theme_bw() +
  scale_fill_manual(values=LocPosCol)

Now here are five variations on that boxplot. Three based on the request in your question and two alternatives based on faceting:

With legend along bottom

p + labs(fill="Location-Position") +
  theme(legend.position="bottom")

With text identifiers under the boxes

library(dplyr)

p + geom_text(data=densityAGRLKA %>% group_by(species, location, position) %>%
                     summarise(value=unique(paste(location, position, sep="-"))),
              aes(label=value, y=0.29, 
                  color=interaction(location, position, sep="-", lex.order=TRUE)),
              position=pd, size=3.3, fontface="bold") +
  scale_color_manual(values=LocPosCol) +
  guides(color=FALSE, fill=FALSE)

With two levels of text identifiers under the boxes

p + geom_text(data=densityAGRLKA %>% group_by(species, location) %>% 
                     summarise %>% mutate(position=NA),
              aes(label=location, color=location, y=0.29),
              position=pd, size=4.2, fontface="bold") +
    geom_text(data=densityAGRLKA %>% group_by(species, position, location) %>% 
                     summarise,
              aes(label=position, 
                  color=interaction(location, position, sep="-", lex.order=TRUE),
                  y=0.28),
              position=pd, size=3.7, fontface="bold") +
  scale_color_manual(values=c(LocCol[1],LocPosCol[1:2],LocCol[2],LocPosCol[3:4])) +
  guides(color=FALSE, fill=FALSE)

Faceting with standard ggplot facet labels

The code below is relatively straightforward, but I'm not wild about the way faceting results in repeating of labels, rather than using a single spanning label when the same level is repeated two or four times in consecutive facets. Below is the "standard" ggplot faceting. Following that is an example of the (somewhat painful) process of changing the facet labels to span multiple facets.

ggplot(densityAGRLKA, aes("", density)) + 
  geom_boxplot(width=0.7, position=pd) + 
  theme_bw() +
  facet_grid(. ~ species + location +  position) +
  theme(panel.margin=unit(0,"lines"),
        panel.border=element_rect(color="grey90"),
        axis.ticks.x=element_blank()) +
  labs(x="")

Faceting with a single facet label spanning a given category

To change the facet labels so that a single label spans a given category (rather than having the same label repeated for each facet) requires going outside of ggplot and using lower level grid functions to change the facet strip label grobs. Here's an example:

library(gtable)
library(grid)

p=ggplot(densityAGRLKA, aes("", density)) + 
  geom_boxplot(width=0.7, position=pd) + 
  theme_bw() +
  facet_grid(. ~ species + location +  position) +
  theme(panel.margin=unit(0,"lines"),
        strip.background=element_rect(color="grey30", fill="grey90"),
        panel.border=element_rect(color="grey90"),
        axis.ticks.x=element_blank()) +
  labs(x="")

pg = ggplotGrob(p)

# Add spanning strip labels for species
pos = c(4,11)    
for (i in 1:2) {
  pg <- gtable_add_grob(pg, 
                        list(rectGrob(gp=gpar(col="grey50", fill="grey90")),
                             textGrob(unique(densityAGRLKA$species)[i], 
                                      gp=gpar(cex=0.8))), 3,pos[i],3,pos[i]+7,
                        name=c("a","b"))
}

# Add spanning strip labels for location
pos=c(4,7,11,15)
for (i in 1:4) {
    pg = gtable_add_grob(pg, 
                         list(rectGrob(gp = gpar(col="grey50", fill="grey90")),
                              textGrob(rep(unique(densityAGRLKA$location),2)[i], 
                                       gp=gpar(cex=0.8))), 4,pos[i],4,pos[i]+3, 
                         name = c("c", "d"))
}

plot(pg)