Scales = "Free" Works for Facet_Wrap But Doesn't for Facet_Grid

Scales = free works for facet_wrap but doesn't for facet_grid

Referring to this link:

facet_grid split the data into facets by one or two variables that vary on the horizontal and/or vertical direction, while facet_wrap places the facets next to each other, wrapping with a certain number of columns or rows. In other words, facet_wrap only has horizontal dimension.

Therefore, using the example from that link, sp + facet_grid(. ~ sex) would behave the same as sp + facet_grid( ~ sex). In your case, facet_grid(. ~ referencia) and facet_wrap( ~ referencia) should produce the same plot.

For two or more dimensional facets, facet_grid produces a grid of plots based on the parameter (vertical ~ horizontal). facet_wrap, on the other hand, would just stack the plots horizontally. User then can set the layout by specifying the number of columns or rows.

Now, when the scales = "free" argument is added, facets in facet_grid would still be bounded by the grid, therefore plots on the same row cannot have different y-axis. Similarly, there can only single x-axis for each column. Using facet_wrap though, each plot is displayed independently, so it can "free" its x-axis and y-axis.

In my opinion, facet_grid is useful when you want to relatively compare the plots within a category, which can be accomplished by setting the same axis scales. Meanwhile, facet_wrap is more useful for plots that more independent between one another.

Looking to solve problem with facet_wrap/facet_grid that isn't resolved with space = free

You can get this easily with lemon::facet_rep_grid(repeat.tick.labels = T) instead of ggplot2::facet_grid().

library(tidyverse)
library(lemon)

b <- structure(list(Race = c("Asian", "Asian", "Asian", "Asian", "Asian", "Asian", "Asian", "Asian", "Asian", "Asian", "Black", "Black", "Black", "Black", "Black", "Black", "Black", "Black", "Black", "Black"), Symptom = structure(c(10L, 9L, 8L, 7L, 6L, 5L, 4L, 3L, 2L, 1L, 10L, 9L,  8L, 7L, 6L, 5L, 4L, 3L, 2L, 1L), .Label = c("Symptom10", "Symptom9", "Symptom8", "Symptom7", "Symptom6", "Symptom5", "Symptom4", "Symptom3", "Symptom2", "Symptom1"), class = "factor"), Percentage = c(60L, 50L, 20L, 70L, 90L, 100L, 10L, 30L, 40L, 60L, 40L, 20L, 50L, 50L, 85L, 30L, 30L, 20L, 80L, 40L), Group = c("Benign", "Benign", "Warning", "Warning", "Warning", "Fatal", "Fatal",  "Fatal", "Fatal", "Fatal", "Benign", "Benign", "Warning", "Warning", "Warning", "Fatal", "Fatal", "Fatal", "Fatal", "Fatal")), row.names = c(NA, -20L), class = c("tbl_df", "tbl", "data.frame"))

ggplot(data=b, aes(x=Percentage, y= Symptom, fill = Race)) +
  geom_bar(stat="identity", position=position_dodge()) +
  facet_rep_grid(Group~., scales = "free", space = "free", repeat.tick.labels = T)

^{Created on 2022-01-18 by the reprex package (v2.0.1)}

scales = free has no effect on facet_grid

facet_grid enforces all facet plots to have the same y axis per row and the same x axis per column. This is why it's called a grid. You have to use facet_wrap(Season ~ year, scales = "free") instead to allow having separate y axes for each facet.

facet_wrap(scales=free_y) not producing desired result

You'll have to remove this from your ggplot code

scale_y_continuous(limits=c(0, max(dat$y)))

because it is 'fixing' the range of y-axis across all the facets

How to change the facet scales using ggplot2?

An alternative is to use facet_wrap() instead of facet_grid(), e.g.

library(tidyverse)

Source <- c(rep("Water", 12), rep("Oil", 12))
Range <- rep((c(rep("First", 4), rep("Second", 8))),2)
Xaxis <- c(0,1,2,5,0,1,2,5,10,20,30,40,
           0,1,2,5,0,1,2,5,10,20,30,40)
Yaxis <- c(0,1,2,5,0,1,2,5,10,20,30,40,
           0,1,2,5,0,1,2,5,10,20,30,40)

DF <- data.frame(Source, Range, Xaxis, Yaxis)

ggplot(data = DF, aes(x = Xaxis, y = Yaxis)) +
  geom_smooth(method = "lm") +
  geom_point() +
  facet_wrap(Range~Source,
             scales = "free")

ggplot: create a facet grid with free scales

The dev/github version of ggh4x has facet_grid2() which allows these independent axes within a grid layout. (Disclaimer: I'm the author of ggh4x.) If you find any bugs or unclear documentation, feel free to leave an issue on the github since I don't think this has been field-tested a lot.

library(ggplot2)
library(ggh4x) # devtools::install_github("teunbrand/ggh4x")

# Create data
nX <- 3
nY <- 3
nVal <- 2
df <- data.frame(M = sort(rep(paste0("M",1:nX), nVal * nY)),
                 n = rep(sort(paste0("n",rep((1:nY)-1, nVal))),nX),
                 G = rep(1:2,nX * nY),
                 val = c(100,300,20,10,1,2,10,25,NA,NA,0.1,0.2,25,27,2,5,0.5,0.6))

# Delete observations to resemble my data
df <- subset(df, !is.na(val))

# scales does not work in facet_grid(), thus obscuring trends for low values
ggplot(df, aes(x = G,
               y = val)) +
  geom_line()+
  ylim(0,NA)+
  facet_grid2(n ~ M,scales = "free_y", independent = "y")

^{Created on 2021-06-09 by the reprex package (v1.0.0)}

ggplot2, facet_grid, free scales?

Perhaps it's because you have only one y axis, using your way. Did you try something like this?

mt + facet_grid(cyl ~ ., scales="free")

Facet_grid with all panels with a free_y

facet_wrap does this, and is what I use, although it does mean you have the faceting value as a title for each individual facet, which can be a pain.

  ggplot(mpg, aes(displ, hwy, colour = as.factor(cyl))) + 
  geom_point() + 
  facet_wrap(year ~ drv,
             scales = "free_y")

Understanding facet_grid scale=free

Instead of facet_grid use facet_wrap for example,

facet_wrap(reformulate("Name","."), scales = 'free', nrow = 1) +

With facet_grid one can not get both x and y scales free; see here https://github.com/tidyverse/ggplot2/issues/1152

---

Related Topics