Publication-ready tables for Word documents
February 7, 2026
The default R output for tables looks unprofessional in Word documents — it appears as console output with monospace font. In this chapter, you will learn how to create professional, formatted tables using the flextable package that integrate seamlessly into Word documents.
There are several R packages for tables (kable, gt, huxtable, flextable), but for Word output, flextable is the best choice:
The gt package is excellent for HTML output, but its Word support is more limited. For Word documents, I recommend flextable.
The simplest way to create a flextable:
adelie %>%
head(5) %>%
select(island, bill_length_mm, bill_depth_mm, body_mass_g) %>%
flextable()island |
bill_length_mm |
bill_depth_mm |
body_mass_g |
|---|---|---|---|
Torgersen |
39.1 |
18.7 |
3,750 |
Torgersen |
39.5 |
17.4 |
3,800 |
Torgersen |
40.3 |
18.0 |
3,250 |
Torgersen |
36.7 |
19.3 |
3,450 |
Torgersen |
39.3 |
20.6 |
3,650 |
This already looks much better than print(df)! But the column widths are not optimal yet.
With autofit(), flextable automatically adjusts column widths:
adelie %>%
head(5) %>%
select(island, bill_length_mm, bill_depth_mm, body_mass_g) %>%
flextable() %>%
autofit()island |
bill_length_mm |
bill_depth_mm |
body_mass_g |
|---|---|---|---|
Torgersen |
39.1 |
18.7 |
3,750 |
Torgersen |
39.5 |
17.4 |
3,800 |
Torgersen |
40.3 |
18.0 |
3,250 |
Torgersen |
36.7 |
19.3 |
3,450 |
Torgersen |
39.3 |
20.6 |
3,650 |
autofit() should generally be placed at the end of the flextable pipeline, after all other formatting has been applied.
The automatic column names from the dataframe are often not ideal for a report:
adelie %>%
head(5) %>%
select(island, bill_length_mm, bill_depth_mm, body_mass_g) %>%
flextable() %>%
set_header_labels(
island = "Island",
bill_length_mm = "Bill Length (mm)",
bill_depth_mm = "Bill Depth (mm)",
body_mass_g = "Body Mass (g)"
) %>%
autofit()Island |
Bill Length (mm) |
Bill Depth (mm) |
Body Mass (g) |
|---|---|---|---|
Torgersen |
39.1 |
18.7 |
3,750 |
Torgersen |
39.5 |
17.4 |
3,800 |
Torgersen |
40.3 |
18.0 |
3,250 |
Torgersen |
36.7 |
19.3 |
3,450 |
Torgersen |
39.3 |
20.6 |
3,650 |
For scientific tables, you often need a specific number of decimal places:
adelie %>%
head(5) %>%
select(island, bill_length_mm, bill_depth_mm, body_mass_g) %>%
flextable() %>%
set_header_labels(
island = "Island",
bill_length_mm = "Bill Length (mm)",
bill_depth_mm = "Bill Depth (mm)",
body_mass_g = "Body Mass (g)"
) %>%
colformat_double(j = c("bill_length_mm", "bill_depth_mm"), digits = 1) %>%
colformat_double(j = "body_mass_g", digits = 0) %>%
autofit()Island |
Bill Length (mm) |
Bill Depth (mm) |
Body Mass (g) |
|---|---|---|---|
Torgersen |
39.1 |
18.7 |
3,750 |
Torgersen |
39.5 |
17.4 |
3,800 |
Torgersen |
40.3 |
18.0 |
3,250 |
Torgersen |
36.7 |
19.3 |
3,450 |
Torgersen |
39.3 |
20.6 |
3,650 |
adelie %>%
head(5) %>%
select(island, bill_length_mm, body_mass_g) %>%
flextable() %>%
font(fontname = "Arial", part = "all") %>%
fontsize(size = 10, part = "body") %>%
fontsize(size = 11, part = "header") %>%
autofit()island |
bill_length_mm |
body_mass_g |
|---|---|---|
Torgersen |
39.1 |
3,750 |
Torgersen |
39.5 |
3,800 |
Torgersen |
40.3 |
3,250 |
Torgersen |
36.7 |
3,450 |
Torgersen |
39.3 |
3,650 |
adelie %>%
head(5) %>%
select(island, bill_length_mm, body_mass_g) %>%
flextable() %>%
align(j = 1, align = "left", part = "all") %>%
align(j = 2:3, align = "center", part = "all") %>%
autofit()island |
bill_length_mm |
body_mass_g |
|---|---|---|
Torgersen |
39.1 |
3,750 |
Torgersen |
39.5 |
3,800 |
Torgersen |
40.3 |
3,250 |
Torgersen |
36.7 |
3,450 |
Torgersen |
39.3 |
3,650 |
adelie %>%
head(5) %>%
select(island, bill_length_mm, body_mass_g) %>%
flextable() %>%
border_remove() %>%
hline_top(border = fp_border(width = 2), part = "header") %>%
hline_bottom(border = fp_border(width = 1), part = "header") %>%
hline_bottom(border = fp_border(width = 2), part = "body") %>%
autofit()island |
bill_length_mm |
body_mass_g |
|---|---|---|
Torgersen |
39.1 |
3,750 |
Torgersen |
39.5 |
3,800 |
Torgersen |
40.3 |
3,250 |
Torgersen |
36.7 |
3,450 |
Torgersen |
39.3 |
3,650 |
adelie %>%
head(5) %>%
select(island, bill_length_mm, body_mass_g) %>%
flextable() %>%
bold(part = "header") %>%
autofit()island |
bill_length_mm |
body_mass_g |
|---|---|---|
Torgersen |
39.1 |
3,750 |
Torgersen |
39.5 |
3,800 |
Torgersen |
40.3 |
3,250 |
Torgersen |
36.7 |
3,450 |
Torgersen |
39.3 |
3,650 |
For our penguin report, we create a descriptive statistics table:
summary_table <- adelie %>%
summarise(
n = n(),
`Bill Length (mm)` = mean(bill_length_mm),
`SD` = sd(bill_length_mm),
`Bill Depth (mm)` = mean(bill_depth_mm),
`SD ` = sd(bill_depth_mm),
`Body Mass (g)` = mean(body_mass_g),
`SD ` = sd(body_mass_g)
)
summary_table %>%
flextable() %>%
colformat_double(digits = 1) %>%
colformat_double(j = "n", digits = 0) %>%
set_header_labels(n = "N") %>%
bold(part = "header") %>%
autofit()N |
Bill Length (mm) |
SD |
Bill Depth (mm) |
SD |
Body Mass (g) |
SD |
|---|---|---|---|---|---|---|
146 |
38.8 |
2.7 |
18.3 |
1.2 |
3,706.2 |
458.6 |
A table with statistics per island:
adelie %>%
group_by(island) %>%
summarise(
N = n(),
`Bill Length` = mean(bill_length_mm),
`Body Mass` = mean(body_mass_g),
.groups = "drop"
) %>%
flextable() %>%
set_header_labels(island = "Island") %>%
colformat_double(j = c("Bill Length"), digits = 1) %>%
colformat_double(j = c("Body Mass"), digits = 0) %>%
bold(part = "header") %>%
hline_top(border = fp_border(width = 2), part = "header") %>%
hline_bottom(border = fp_border(width = 1), part = "header") %>%
hline_bottom(border = fp_border(width = 2), part = "body") %>%
autofit()Island |
N |
Bill Length |
Body Mass |
|---|---|---|---|
Biscoe |
44 |
39.0 |
3,710 |
Dream |
55 |
38.5 |
3,701 |
Torgersen |
47 |
39.0 |
3,709 |
To add a table caption, use the chunk option tbl-cap:
The label must start with tbl- for Quarto to recognize it as a table and enable cross-references (see Chapter 7).
For correct display in Word documents, the chunk option output: asis is often no longer needed (current flextable versions detect the format automatically). If the table does not appear correctly, you can add it:
Here is a complete, publication-ready table:
adelie %>%
group_by(island, sex) %>%
summarise(
N = n(),
`Bill Length (mm)` = mean(bill_length_mm),
`Body Mass (g)` = mean(body_mass_g),
.groups = "drop"
) %>%
flextable() %>%
set_header_labels(
island = "Island",
sex = "Sex"
) %>%
colformat_double(j = "Bill Length (mm)", digits = 1) %>%
colformat_double(j = "Body Mass (g)", digits = 0) %>%
font(fontname = "Arial", part = "all") %>%
fontsize(size = 10, part = "all") %>%
bold(part = "header") %>%
align(align = "center", part = "header") %>%
align(j = 1:2, align = "left", part = "body") %>%
align(j = 3:5, align = "right", part = "body") %>%
border_remove() %>%
hline_top(border = fp_border(width = 1.5), part = "header") %>%
hline_bottom(border = fp_border(width = 0.75), part = "header") %>%
hline_bottom(border = fp_border(width = 1.5), part = "body") %>%
autofit()Island |
Sex |
N |
Bill Length (mm) |
Body Mass (g) |
|---|---|---|---|---|
Biscoe |
female |
22 |
37.4 |
3,369 |
Biscoe |
male |
22 |
40.6 |
4,050 |
Dream |
female |
27 |
36.9 |
3,344 |
Dream |
male |
28 |
40.1 |
4,046 |
Torgersen |
female |
24 |
37.6 |
3,396 |
Torgersen |
male |
23 |
40.6 |
4,035 |
tbl-cap
Now we can create professional tables. In Chapter 6, we will learn how to optimally integrate ggplot2 graphics into Quarto documents — with the right size, resolution, and captions.
@online{schmidt2026,
author = {{Dr. Paul Schmidt}},
publisher = {BioMath GmbH},
title = {5. {Tables} with Flextable},
date = {2026-02-07},
url = {https://biomathcontent.netlify.app/content/quarto/05_tables.html},
langid = {en}
}
---
title: "5. Tables with flextable"
subtitle: "Publication-ready tables for Word documents"
---
The default R output for tables looks unprofessional in Word documents — it appears as console output with monospace font. In this chapter, you will learn how to create professional, formatted tables using the flextable package that integrate seamlessly into Word documents.
```{r}
#| label: qrt-tables-setup
#| include: false
library(tidyverse)
library(palmerpenguins)
library(flextable)
library(officer)
adelie <- penguins %>%
filter(species == "Adelie") %>%
drop_na()
```
# Why flextable?
There are several R packages for tables (kable, gt, huxtable, flextable), but for Word output, **flextable** is the best choice:
- Native Word format (no detour via HTML)
- Full control over formatting
- Actively maintained and well documented
- Part of the "Officeverse" ecosystem
:::{.callout-note}
The **gt** package is excellent for HTML output, but its Word support is more limited. For Word documents, I recommend flextable.
:::
# Basics
## Creating a simple table
The simplest way to create a flextable:
```{r}
#| label: qrt-tables-basic
adelie %>%
head(5) %>%
select(island, bill_length_mm, bill_depth_mm, body_mass_g) %>%
flextable()
```
This already looks much better than `print(df)`! But the column widths are not optimal yet.
## Automatic column widths
With `autofit()`, flextable automatically adjusts column widths:
```{r}
#| label: qrt-tables-autofit
adelie %>%
head(5) %>%
select(island, bill_length_mm, bill_depth_mm, body_mass_g) %>%
flextable() %>%
autofit()
```
:::{.callout-tip}
`autofit()` should generally be placed at the end of the flextable pipeline, after all other formatting has been applied.
:::
# Renaming and formatting columns
## Changing column headers
The automatic column names from the dataframe are often not ideal for a report:
```{r}
#| label: qrt-tables-headers
adelie %>%
head(5) %>%
select(island, bill_length_mm, bill_depth_mm, body_mass_g) %>%
flextable() %>%
set_header_labels(
island = "Island",
bill_length_mm = "Bill Length (mm)",
bill_depth_mm = "Bill Depth (mm)",
body_mass_g = "Body Mass (g)"
) %>%
autofit()
```
## Formatting numbers
For scientific tables, you often need a specific number of decimal places:
```{r}
#| label: qrt-tables-colformat
adelie %>%
head(5) %>%
select(island, bill_length_mm, bill_depth_mm, body_mass_g) %>%
flextable() %>%
set_header_labels(
island = "Island",
bill_length_mm = "Bill Length (mm)",
bill_depth_mm = "Bill Depth (mm)",
body_mass_g = "Body Mass (g)"
) %>%
colformat_double(j = c("bill_length_mm", "bill_depth_mm"), digits = 1) %>%
colformat_double(j = "body_mass_g", digits = 0) %>%
autofit()
```
# Formatting and styling
## Font and size
```{r}
#| label: qrt-tables-font
adelie %>%
head(5) %>%
select(island, bill_length_mm, body_mass_g) %>%
flextable() %>%
font(fontname = "Arial", part = "all") %>%
fontsize(size = 10, part = "body") %>%
fontsize(size = 11, part = "header") %>%
autofit()
```
## Alignment
```{r}
#| label: qrt-tables-align
adelie %>%
head(5) %>%
select(island, bill_length_mm, body_mass_g) %>%
flextable() %>%
align(j = 1, align = "left", part = "all") %>%
align(j = 2:3, align = "center", part = "all") %>%
autofit()
```
## Borders
```{r}
#| label: qrt-tables-borders
adelie %>%
head(5) %>%
select(island, bill_length_mm, body_mass_g) %>%
flextable() %>%
border_remove() %>%
hline_top(border = fp_border(width = 2), part = "header") %>%
hline_bottom(border = fp_border(width = 1), part = "header") %>%
hline_bottom(border = fp_border(width = 2), part = "body") %>%
autofit()
```
## Bold headers
```{r}
#| label: qrt-tables-bold
adelie %>%
head(5) %>%
select(island, bill_length_mm, body_mass_g) %>%
flextable() %>%
bold(part = "header") %>%
autofit()
```
# Creating a summary table
For our penguin report, we create a descriptive statistics table:
```{r}
#| label: qrt-tables-summary
summary_table <- adelie %>%
summarise(
n = n(),
`Bill Length (mm)` = mean(bill_length_mm),
`SD` = sd(bill_length_mm),
`Bill Depth (mm)` = mean(bill_depth_mm),
`SD ` = sd(bill_depth_mm),
`Body Mass (g)` = mean(body_mass_g),
`SD ` = sd(body_mass_g)
)
summary_table %>%
flextable() %>%
colformat_double(digits = 1) %>%
colformat_double(j = "n", digits = 0) %>%
set_header_labels(n = "N") %>%
bold(part = "header") %>%
autofit()
```
# Grouped tables
A table with statistics per island:
```{r}
#| label: qrt-tables-grouped
adelie %>%
group_by(island) %>%
summarise(
N = n(),
`Bill Length` = mean(bill_length_mm),
`Body Mass` = mean(body_mass_g),
.groups = "drop"
) %>%
flextable() %>%
set_header_labels(island = "Island") %>%
colformat_double(j = c("Bill Length"), digits = 1) %>%
colformat_double(j = c("Body Mass"), digits = 0) %>%
bold(part = "header") %>%
hline_top(border = fp_border(width = 2), part = "header") %>%
hline_bottom(border = fp_border(width = 1), part = "header") %>%
hline_bottom(border = fp_border(width = 2), part = "body") %>%
autofit()
```
# Table captions in Quarto
To add a table caption, use the chunk option `tbl-cap`:
````markdown
```{r}
#| label: tbl-summary
#| tbl-cap: "Descriptive statistics of Adelie penguins"
summary_table %>%
flextable() %>%
autofit()
```
````
The label must start with `tbl-` for Quarto to recognize it as a table and enable cross-references (see Chapter 7).
# Usage in Word documents
For correct display in Word documents, the chunk option `output: asis` is often no longer needed (current flextable versions detect the format automatically). If the table does not appear correctly, you can add it:
````markdown
```{r}
#| label: tbl-example
#| output: asis
my_table %>%
flextable() %>%
autofit()
```
````
# Complete example
Here is a complete, publication-ready table:
```{r}
#| label: qrt-tables-complete
adelie %>%
group_by(island, sex) %>%
summarise(
N = n(),
`Bill Length (mm)` = mean(bill_length_mm),
`Body Mass (g)` = mean(body_mass_g),
.groups = "drop"
) %>%
flextable() %>%
set_header_labels(
island = "Island",
sex = "Sex"
) %>%
colformat_double(j = "Bill Length (mm)", digits = 1) %>%
colformat_double(j = "Body Mass (g)", digits = 0) %>%
font(fontname = "Arial", part = "all") %>%
fontsize(size = 10, part = "all") %>%
bold(part = "header") %>%
align(align = "center", part = "header") %>%
align(j = 1:2, align = "left", part = "body") %>%
align(j = 3:5, align = "right", part = "body") %>%
border_remove() %>%
hline_top(border = fp_border(width = 1.5), part = "header") %>%
hline_bottom(border = fp_border(width = 0.75), part = "header") %>%
hline_bottom(border = fp_border(width = 1.5), part = "body") %>%
autofit()
```
:::{.callout-tip collapse="false"}
## Exercise: Create a summary table
1. Create a table with the number of penguins per island and sex
2. Add a column with the average weight
3. Format the table professionally (font, borders, alignment)
4. Add a table caption with `tbl-cap`
:::
# Further resources
- [flextable book](https://ardata-fr.github.io/flextable-book/) — Comprehensive documentation
- [flextable gallery](https://ardata-fr.github.io/flextable-gallery/) — Examples and inspiration
- [Officeverse](https://ardata-fr.github.io/officeverse/) — The ecosystem around flextable
# What is next
Now we can create professional tables. In Chapter 6, we will learn how to optimally integrate ggplot2 graphics into Quarto documents — with the right size, resolution, and captions.