In a recent project, I was looking to plot data from different variables along the same time axis. The difficulty was, that some of these variables I wanted to have as point plots, while others I wanted as box-plots.

Because I work with the tidyverse, I wanted to produce these plots with ggplot2. Faceting was the obvious first step but it took me quite a while to figure out how to best combine facets with point plots (where I have one value per time point) with and box-plots (where I have multiple values per time point).

The reason why this isn’t trivial is that box plots require groups or factors on the x-axis, while points can be plotted over a continuous range of x-values. If your alarm bells are ringing right now, you are absolutely right: before you try to combine plots with different x-axis properties, you should think long and hard whether this is an accurate representation of the data and if its a good idea to do so! Here, I had multiple values per time point for one variable and I wanted to make the median + variation explicitly clear, while also showing the continuous changes of other variables over the same range of time.

So, I am writing this short tutorial here in hopes that it saves the next person trying to do something similar from spending an entire morning on stackoverflow. ;-)

For this demonstration, I am creating some fake data:

library(tidyverse)
dates <- seq(as.POSIXct("2017-10-01 07:00"), as.POSIXct("2017-10-01 10:30"), by = 180) # 180 seconds == 3 minutes
fake_data <- data.frame(time = dates,
                        var1_1 = runif(length(dates)),
                        var1_2 = runif(length(dates)),
                        var1_3 = runif(length(dates)),
                        var2 = runif(length(dates))) %>%
  sample_frac(size = 0.33)
head(fake_data)
##                  time    var1_1    var1_2    var1_3       var2
## 1 2017-10-01 08:33:00 0.4208415 0.2589455 0.3786275 0.80532017
## 2 2017-10-01 08:42:00 0.4853185 0.4949028 0.9104159 0.25552958
## 3 2017-10-01 09:42:00 0.4144495 0.6314172 0.5832432 0.74209701
## 4 2017-10-01 09:54:00 0.9315311 0.8266359 0.1509052 0.55146543
## 5 2017-10-01 08:36:00 0.1212433 0.3228635 0.5638170 0.43761903
## 6 2017-10-01 09:21:00 0.2826186 0.8656590 0.8774104 0.07265883

Here, variable 1 (var1) has three measurements per time point, while variable 2 (var2) has one.

First, for plotting with ggplot2 we want our data in a tidy long format. I also add another column for faceting that groups the variables from var1 together.

fake_data_long <- fake_data %>%
  gather(x, y, var1_1:var2) %>%
  mutate(facet = ifelse(x %in% c("var1_1", "var1_2", "var1_3"), "var1", x))
head(fake_data_long)
##                  time      x         y facet
## 1 2017-10-01 08:33:00 var1_1 0.4208415  var1
## 2 2017-10-01 08:42:00 var1_1 0.4853185  var1
## 3 2017-10-01 09:42:00 var1_1 0.4144495  var1
## 4 2017-10-01 09:54:00 var1_1 0.9315311  var1
## 5 2017-10-01 08:36:00 var1_1 0.1212433  var1
## 6 2017-10-01 09:21:00 var1_1 0.2826186  var1

Now, we can plot this the following way:

  • facet by variable
  • subset data to facets for point plots and give aesthetics in geom_point()
  • subset data to facets for box plots and give aesthetics in geom_boxplot(). Here we also need to set the group aesthetic; if we don’t specifically give that, we will get a plot with one big box, instead of a box for every time point.
fake_data_long %>%
  ggplot() +
    facet_grid(facet ~ ., scales = "free") +
    geom_point(data = subset(fake_data_long, facet == "var2"), 
               aes(x = time, y = y),
               size = 1) +
    geom_line(data = subset(fake_data_long, facet == "var2"), 
               aes(x = time, y = y)) +
    geom_boxplot(data = subset(fake_data_long, facet == "var1"), 
               aes(x = time, y = y, group = time))

sessionInfo()
## R version 4.0.2 (2020-06-22)
## Platform: x86_64-apple-darwin17.0 (64-bit)
## Running under: macOS Catalina 10.15.6
## 
## Matrix products: default
## BLAS:   /Library/Frameworks/R.framework/Versions/4.0/Resources/lib/libRblas.dylib
## LAPACK: /Library/Frameworks/R.framework/Versions/4.0/Resources/lib/libRlapack.dylib
## 
## locale:
## [1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8
## 
## attached base packages:
## [1] stats     graphics  grDevices utils     datasets  methods   base     
## 
## other attached packages:
## [1] forcats_0.5.0   stringr_1.4.0   dplyr_1.0.2     purrr_0.3.4    
## [5] readr_1.3.1     tidyr_1.1.2     tibble_3.0.3    ggplot2_3.3.2  
## [9] tidyverse_1.3.0
## 
## loaded via a namespace (and not attached):
##  [1] tidyselect_1.1.0 xfun_0.17        haven_2.3.1      colorspace_1.4-1
##  [5] vctrs_0.3.4      generics_0.0.2   htmltools_0.5.0  yaml_2.2.1      
##  [9] blob_1.2.1       rlang_0.4.7      pillar_1.4.6     glue_1.4.2      
## [13] withr_2.2.0      DBI_1.1.0        dbplyr_1.4.4     modelr_0.1.8    
## [17] readxl_1.3.1     lifecycle_0.2.0  munsell_0.5.0    blogdown_0.20.1 
## [21] gtable_0.3.0     cellranger_1.1.0 rvest_0.3.6      evaluate_0.14   
## [25] labeling_0.3     knitr_1.29       fansi_0.4.1      broom_0.7.0     
## [29] Rcpp_1.0.5       scales_1.1.1     backports_1.1.10 jsonlite_1.7.1  
## [33] farver_2.0.3     fs_1.5.0         hms_0.5.3        digest_0.6.25   
## [37] stringi_1.5.3    bookdown_0.20    grid_4.0.2       cli_2.0.2       
## [41] tools_4.0.2      magrittr_1.5     crayon_1.3.4     pkgconfig_2.0.3 
## [45] ellipsis_0.3.1   xml2_1.3.2       reprex_0.3.0     lubridate_1.7.9 
## [49] assertthat_0.2.1 rmarkdown_2.3    httr_1.4.2       rstudioapi_0.11 
## [53] R6_2.4.1         compiler_4.0.2