Creating Insightful Tables with gt() in R

The gt package is used to create and customize beautiful, high-quality tables for data display and reporting. It offers a wide range of options for formatting, styling, and adding features such as headers, footers, and source notes. The gt() function is ideal for turning raw data into polished, publication-ready tables.

Getting Started

To begin using the gt library, start by installing it on your system:

install.packages("gt")

Importing Libraries & Dataset

library(gt)
library(dplyr)
df <- read.csv("C:/Users/SANKHYA/Downloads/ClinicalData.csv",check.names = FALSE)
head(df)

  PatientID Gender Age  Treatment Smoking_status WeightKg VAS_Score QOL_Score
1      ID_1   Male  42 TreatmentB         Smoker       57         6        79
2      ID_2   Male  69 TreatmentB     Non-Smoker       95         5        59
3      ID_3   Male  39 TreatmentA         Smoker       79         4        63
4      ID_4 Female  42 TreatmentB     Non-Smoker       70         6        57
5      ID_5   Male  49 TreatmentB     Non-Smoker       72         5        84
6      ID_6 Female  63 TreatmentB         Smoker       85         7        56

Description of the dataset

This dataset provides information on patients in a clinical study, including their gender, age, treatment type, smoking status, weight, and scores for Visual Analog Scale (VAS) and Quality of Life (QOL). It aims to assess treatment outcomes and patient-reported measures related to health and well-being.

Let’s explore some functions of the`gt` library:

gt_preview(): It provides a well-formatted preview of a data table.

#Preview the data
gt_preview(df)

	PatientID	Gender	Age	Treatment	Smoking_status	WeightKg	VAS_Score	QOL_Score
1	ID_1	Male	42	TreatmentB	Smoker	57	6	79
2	ID_2	Male	69	TreatmentB	Non-Smoker	95	5	59
3	ID_3	Male	39	TreatmentA	Smoker	79	4	63
4	ID_4	Female	42	TreatmentB	Non-Smoker	70	6	57
5	ID_5	Male	49	TreatmentB	Non-Smoker	72	5	84
6..99
100	ID_100	Male	47	TreatmentB	Non-Smoker	68	5	85

tab_stubhead(): It sets the label for row headers in a table.

md(): It interprets input text as Markdown.

#Non-Smokers by Gender & Treatment
df %>% filter(Smoking_status=="Non-Smoker") %>% group_by(Treatment, Gender) %>%
  summarise(Counts=n()) %>% gt(rowname_col = "Gender", groupname_col = "Treatment") %>%
  tab_stubhead(label = md("**No. of Non-Smokers by Gender & Treatment**"))

No. of Non-Smokers by Gender & Treatment	Counts
TreatmentA
Female	12
Male	17
TreatmentB
Female	13
Male	16

tab_source_note(): It provides source attribution notes.

tab_footnote(): It adds supplementary contextual notes to a table.

#Summary Statistics by Smoking Status
df %>% group_by(Smoking_status) %>% summarise(`VAS Score`=round(mean(VAS_Score),2),
                                              `QOL Score`=round(mean(QOL_Score),2))%>%
  gt() %>% tab_header(title = "Summary Statistics by Smoking Status") %>%
  tab_source_note(source_note = html("Clinical Study")) %>%
  tab_footnote(footnote = md("Statistics presented is the mean"),
               locations = cells_column_labels(columns=vars(`VAS Score`, `QOL Score`)))

Smoking_status	VAS Score¹	QOL Score¹
Summary Statistics by Smoking Status
Non-Smoker	6.66	69.88
Smoker	6.50	73.45
Clinical Study
¹ Statistics presented is the mean

tab_spanner(): It creates a spanning header row in tables.

#Aggregate Information by Smoking Status
df %>% group_by(Smoking_status) %>%
                summarise(Treatment = names(which.max(table(Treatment))),
                          `VAS Score` = names(which.max(table(VAS_Score))),
                          `QOL Score` = names(which.max(table(QOL_Score))),
                           Gender = n()) %>% gt() %>% 
  tab_spanner(label="Aggregate Information by Smoking Status",
              columns=c(Treatment,`VAS Score`, `QOL Score`, Gender)) %>%
  tab_header(title = md("**Summary of Clinical Data by Smoking Status**"),
    subtitle = md("*Most Frequent Treatment, VAS Score, QOL Score, and Gender*"))

Smoking_status	Aggregate Information by Smoking Status
Summary of Clinical Data by Smoking Status
Most Frequent Treatment, VAS Score, QOL Score, and Gender
Smoking_status	Treatment	VAS Score	QOL Score	Gender
Non-Smoker	TreatmentA	8	54	58
Smoker	TreatmentA	6	86	42

Note:

Other functions include as_gt, fmt_number, cols_hide, gt_show, cells_title, tab_options and many more.

Features of `gt`:

Customizable Tables, Markdown Support, Statistical Summaries, Grouping and Spanning, Export Options.

Conclusion:

The gt() function in R is a powerful tool for transforming raw data into polished, publication-ready tables. Its extensive customization options allow for professional presentation and detailed reporting. Using gt, users can effectively communicate complex data insights in a clear and engaging manner, elevating the overall quality of their data reporting and analysis.