tabular and flextable

[This article was first published on R on ArData, and kindly contributed to R-bloggers]. (You can report issue about the content on this page here)

Want to share your content on R-bloggers? click here if you have a blog, or here if you don’t.

The package ‘flextable’ (Gohel and Skintzos 2023) provides a method as_flextable() to benefit from
table objects created with package ‘tables’ (Murdoch 2023).

Function tables::tabular() is a powerful tool that let users easily
create simple and complex cross tables. Function as_flextable() is simple to use, it
transforms ‘tables’ objects into ‘flextable’ objects and let you
enrich the table with extra information and or formats.

The goal of package ‘tables’ is to compute and display complex tables of summary statistics.
It is described by his author as:

The tables package provides a formula-driven
interface for computing the contents of tables and formatting them.
It was inspired by SAS PROC TABULATE, but is not compatible with it.

The user computes a table object by specifying a formula, with the left-hand
side giving the rows, and the right-hand side giving the columns;
the formula describes the summary functions to apply and how to organize them. […]

The combination of these two packages offers the possibility to create cross tabulations
very easily, to stack or filter them and then to format them with flextable and to
benefit from a whole range of formatting functionalities as well as the possibility
to export them in the output formats supported by ‘flextable’.

tabular example

Let’s create an crosstable of aggregations based on
the penguins dataset provided by package ‘palmerpenguins’ (Horst, Hill, and Gorman 2022):

library(tables)
library(palmerpenguins)

Mean <- function(x) mean(x, na.rm = TRUE)

Sd <- function(x) sd(x, na.rm = TRUE)

tab <- tabular(
  table = (species + 1) * island ~ 
    (n = 1) + Format(digits = 2) * 
      (bill_depth_mm + body_mass_g) * (Mean + Sd), 
  data = penguins)
print(tab)
#>                                                                   
#>                          bill_depth_mm         body_mass_g        
#>  species   island    n   Mean          Sd      Mean        Sd     
#>  Adelie    Biscoe     44   18.37          1.19 3709.66      487.73
#>            Dream      56   18.25          1.13 3688.39      455.15
#>            Torgersen  52   18.43          1.34 3706.37      445.11
#>  Chinstrap Biscoe      0     NaN            NA     NaN          NA
#>            Dream      68   18.42          1.14 3733.09      384.34
#>            Torgersen   0     NaN            NA     NaN          NA
#>  Gentoo    Biscoe    124   14.98          0.98 5076.02      504.12
#>            Dream       0     NaN            NA     NaN          NA
#>            Torgersen   0     NaN            NA     NaN          NA
#>  All       Biscoe    168   15.87          1.82 4716.02      782.86
#>            Dream     124   18.34          1.13 3712.90      416.64
#>            Torgersen  52   18.43          1.34 3706.37      445.11

turn it to a flextable

A simple call to as_flextable() produces a flextable
quite similar to the tabular outputs:

as_flextable(tab)

species	island		bill_depth_mm		body_mass_g
species	island	n	Mean	Sd	Mean	Sd
Adelie	Biscoe	44	18.37	1.19	3709.66	487.73
	Dream	56	18.25	1.13	3688.39	455.15
	Torgersen	52	18.43	1.34	3706.37	445.11
Chinstrap	Biscoe	0	NaN	NA	NaN	NA
	Dream	68	18.42	1.14	3733.09	384.34
	Torgersen	0	NaN	NA	NaN	NA
Gentoo	Biscoe	124	14.98	0.98	5076.02	504.12
	Dream	0	NaN	NA	NaN	NA
	Torgersen	0	NaN	NA	NaN	NA
All	Biscoe	168	15.87	1.82	4716.02	782.86
	Dream	124	18.34	1.13	3712.90	416.64
	Torgersen	52	18.43	1.34	3706.37	445.11

It uses flextable default settings when possible, we recommended to
define them once per session.

set_flextable_defaults(
    font.family = "Open Sans", 
    font.color = "#333333",
    digits = 2,
    big.mark = "'", 
    na_str = "", nan_str = "",
    theme_fun = theme_booktabs)

‘tables’ comes with many options that can be used to format
the table. The flextable transformation preserves column types, it makes easier
to post format the table with flextable syntax if needed:

as_flextable(tab) |> 
  colformat_double() |>
  theme_vanilla()

species	island		bill_depth_mm		body_mass_g
species	island	n	Mean	Sd	Mean	Sd
Adelie	Biscoe	44	18.37	1.19	3’709.66	487.73
	Dream	56	18.25	1.13	3’688.39	455.15
	Torgersen	52	18.43	1.34	3’706.37	445.11
Chinstrap	Biscoe	0
	Dream	68	18.42	1.14	3’733.09	384.34
	Torgersen	0
Gentoo	Biscoe	124	14.98	0.98	5’076.02	504.12
	Dream	0
	Torgersen	0
All	Biscoe	168	15.87	1.82	4’716.02	782.86
	Dream	124	18.34	1.13	3’712.90	416.64
	Torgersen	52	18.43	1.34	3’706.37	445.11

Use row separator

Groups’ separators can be defined as rows instead of a
first column in order to reduce the width of the table:

as_flextable(tab, spread_first_col = TRUE) |> 
  colformat_double()

island		bill_depth_mm		body_mass_g
island	n	Mean	Sd	Mean	Sd
Adelie
Biscoe	44	18.37	1.19	3’709.66	487.73
Dream	56	18.25	1.13	3’688.39	455.15
Torgersen	52	18.43	1.34	3’706.37	445.11
Chinstrap
Biscoe	0
Dream	68	18.42	1.14	3’733.09	384.34
Torgersen	0
Gentoo
Biscoe	124	14.98	0.98	5’076.02	504.12
Dream	0
Torgersen	0
All
Biscoe	168	15.87	1.82	4’716.02	782.86
Dream	124	18.34	1.13	3’712.90	416.64
Torgersen	52	18.43	1.34	3’706.37	445.11

Enrich with flextable

Now let’s add more customizations to illustrate
some other flextable features that fit well with tabular():

row title formatting with as_paragraph(),
content tabulation with prepend_chunks(as_chunk("\t")),
labels update with labelizor().

ft <- as_flextable(tab, spread_first_col = TRUE,
    row_title = as_paragraph(
      colorize("species: ", color = "#8f2218"),
      colorize(as_b(.row_title), color = "#8f2218")
    )) |> 
  colformat_double() |>
  align(j = 1, align = "left") |>
  prepend_chunks(i = ~ .is_row_title %in% FALSE, j = 1, as_chunk("\t")) |>
  labelizor(part = "header", labels = c(bill_depth_mm = "bill length\n(millimeters)", 
    body_mass_g = "body mass\n(grams)", island = "Island\nin Palmer\nArchipelago",
    n = "Number\nof\npenguins")) |>
  labelizor(part = "body", labels = stringr::str_to_title) |>
  autofit()
ft

Island in Palmer Archipelago		bill length (millimeters)		body mass (grams)
Island in Palmer Archipelago	Number of penguins	Mean	Sd	Mean	Sd
Species: Adelie
Biscoe	44	18.37	1.19	3’709.66	487.73
Dream	56	18.25	1.13	3’688.39	455.15
Torgersen	52	18.43	1.34	3’706.37	445.11
Species: Chinstrap
Biscoe	0
Dream	68	18.42	1.14	3’733.09	384.34
Torgersen	0
Species: Gentoo
Biscoe	124	14.98	0.98	5’076.02	504.12
Dream	0
Torgersen	0
Species: All
Biscoe	168	15.87	1.82	4’716.02	782.86
Dream	124	18.34	1.13	3’712.90	416.64
Torgersen	52	18.43	1.34	3’706.37	445.11

Add into a document

There are simple methods to export a flextable
into a document:

save_as_docx(ft, path = "flextable-tabular.docx")
save_as_rtf(ft, path = "flextable-tabular.rtf")
save_as_html(ft, path = "flextable-tabular.html")
save_as_image(ft, path = "flextable-tabular.png")
save_as_pptx(ft, path = "flextable-tabular.pptx")

files can be downloaded here:

Gohel, David, and Panagiotis Skintzos. 2023. Flextable: Functions for Tabular Reporting. https://CRAN.R-project.org/package=flextable.

Horst, Allison, Alison Hill, and Kristen Gorman. 2022. Palmerpenguins: Palmer Archipelago (Antarctica) Penguin Data. https://CRAN.R-project.org/package=palmerpenguins.

Murdoch, Duncan. 2023. Tables: Formula-Driven Table Generation. https://dmurdoch.github.io/tables/.