Create LaTeX Tables in R

A Step by Step Tutorial

Photo by Scott Graham on Unsplash

This article discusses how to create LaTex tables in R using the KableExtra package. Additionally, I show how to manually export KableExtra tables to MS Word (if you prefer writing documents in MS Word).

In this article, I use the Financial Well-Being Survey 2016 data. This survey was conducted by the Consumer Financial Protection Bureau (CFPB). The dataset is publicly available here:

https://www.consumerfinance.gov/data-research/financial-well-being-survey-data/

In the table, I want to show the summary statistics (count of observations, mean, median, and standard deviation) of financial well-being scores (FWBScore variable) conditional on gender (PPGENDER variable) and educational attainment (PPEDUC variable).

Let’s begin by downloading the data directly from the CFPB website:

data <- read.csv("https://www.consumerfinance.gov/documents/5614/NFWBS_PUF_2016_data.csv")

Now, let’s install (assuming we did not install them before) and load the dplyr and the KableExtra packages:

install.packages("dplyr")
install.packages("kableExtra")
library(dplyr)
library(kableExtra)

Next, we do the following:

1. We prepare the summary statistics using the group_by and summarise functions of the dplyr package.
2. Then, we create a dataframe, which shows the summary statistics.
3. Next, we pass the dataframe inside the kbl function of the kableExtra package.
4. Lastly, we add a caption to the table and select fonts and alignments.

In this dataset, the PPGENDER==1 is male and PPGENDER ==2 is female; and, PPEDUC==1 is less than high school, PPEDUC ==2 is high school degree/GED, PPEDUC ==3 is some college/associate, PPEDUC==4 is bachelor’s degree, and PPEDUC==5 is graduate/professional degree.

Let’s recode the levels of the two variables to help our readers better understand the table:

data$PPGENDER <- recode_factor(data$PPGENDER, "1"="Male",
                                              "2"="Female")
data$PPEDUC <- recode_factor(data$PPEDUC, 
                            "1"="less than high school",
                            "2"="high school degree/GED",
                            "3"="some college/associate",
                            "4"="bachelor's degree",
                            "5"="graduate/professional degree")

Now, let’s calculate the summary statistics of financial well-being scores by gender and education and put the output inside a dataframe called “table”:

table <- data %>% group_by(PPGENDER,PPEDUC) %>%
  summarise(Count=n(),
            Mean.FWBScore=mean(FWBscore),
            Median.FWBScore=median(FWBscore),
            SD.FWBScore=sd(FWBscore)
            )
table

The dataframe looks okay, but we don’t want these many digits after the decimal. Let’s fix this by adding the round function in the previous code to select only one digit after the decimal:

table <- data %>% group_by(PPGENDER,PPEDUC) %>%
  summarise(Count=n(),
            Mean.FWBScore=round(mean(FWBscore),digits=1),
            Median.FWBScore=round(median(FWBscore),digits=1),
            SD.FWBScore=round(sd(FWBscore),digits=1)
  )

Great! Now, let’s convert this dataframe into a HTML table:

table %>%
  kbl(caption="Table 1: Summary Statistics of Financial Well-Being  
               Score by Gender and Education",
       format= "html",
   col.names = c("Gender","Education","Count","Mean","Median","SD"),
                  align="r") %>%
   kable_classic(full_width = F, html_font = "helvetica")

In the above code, we can change alignment by changing the value of align (“r”=right, “l”=left, “c”=center). Also, we can change the font type by changing the value of html_font (e.g., “Arial Narrow”, “Trebuchet MS”, “Source Sans Pro”, etc.). And, if we want to change the table style, in place of kable_classic, we can try: Kable_classic_2, Kable_minimal, Kable_material, Kable_material, Kable_material_dark, and kable_paper.

For example, if we try kable_minimal, this is what we get:

Okay! Now it’s time to convert this table into LaTeX format. We just change format = “html” to format=“latex”. Also, we don’t include “Table 1” in the caption (like we did in the case of the HTML table) because LaTeX editors automatically generate a table number.

table %>%
  kbl(caption="Summary Statistics of Financial Well-Being Score by Gender and Education",
      format="latex",
      col.names = c("Gender","Education","Count","Mean","Median","SD"),
      align="r") %>%
  kable_minimal(full_width = F,  html_font = "Source Sans Pro")

Once we run the code in R, we see the following output:

This is the LaTeX code that we need to paste into any LaTeX editor to create the table. I paste the code into Overleaf (https://www.overleaf.com/) and this is what I get:

Now, if you are a loyal MS Word user and would like to have this table in a word document, do the following:

1. In Viewer, click Zoom.

2. Once you are in Viewer Zoom, press ctrl + A to select everything:

3. Paste the table in your MS Word document:

Now, this requires some editing!

4. Select the table.

5. Click: Insert -> Table -> Insert Table:

Now, the table looks like this:

Finally, I do some simple editing in MS Word and get the following table:

Thanks for following along! If you are interested in more descriptive statistics lessons in R, have a look at some of my previous articles:

Visualizing Continuous Data with ggplot2 in R

Create Line Charts in R using ggplot2

Create Heatmap in R using ggplot2