In isubirana/compareGroups: Descriptive Analysis by Groups

Several types of statistics (mean, standard deviation, quantiles or frequencies) are displayed according to the nature of each variable (categorical, continuous and normal-distributed or other quantitative type distribution). Also, p-value to test equality between groups is computed using the appropiate test.

knitr::opts_chunk$set(comment="",message=FALSE,warning=FALSE)

Step 1. Install the package

Install the compareGroups package from CRAN and then load it by typing:

install.packages("compareGroups")
library(compareGroups)

library(compareGroups)

Step 2. Load data

Load the REGICOR example data available in compareGroups package:

data(regicor)

Step 3. Compute descriptives and tests

Compute all descriptives and tests from selected variables by using the compareGroups function.
Note the use of formula argument as usual in R, so "." indicates all variables in the dataset while "-" sign indicates removing, and variable on left side of "\~" sign indicates the groups (if descriptives of whole datset without separating by groups is desired left it in blank). Store the results in an object that can be used afterwards to perform plots or bivariate table itself.
By the argument method we set triglycerides (triglyc), days to cardiovascular event (tocv) and days until death (todeath) variables to be reported as median and quartiles instead of mean and standard deviation.

res <- compareGroups(year ~ . - id, data = regicor, 
                     method=c(triglyc=2, tocv=2, todeath=2))

Step 4. Create the descriptive table

Use the creaTable function passing the previous object computed by compareGroups function (res). Using this function you can customize how categorical variables are displayed (only percertage or absolute frequencies or both) by type argument or whether standard deviation appears inside brackets or separated by plus/minus symbol by sd.type argument.
Also note the use of hide.no category which is useful to hide "no" level for those binary variables.
If you only want to show "Female" category use hide argument for sex variable indicating which category is going to be hiden. This argument also applies to categorical variables with more than two categories. To specify the number of decimal digits to show use digitsarguments. In this example height have no decimals and for weight with three digits. Finally, if you want to show how many individuals have non-missing values in each described variable, set show.n argument to TRUE.

restab <- createTable(res, digits = c(height=0, weight=3), type = 1, sd.type = 2, 
                      hide = c(sex = "Male"), hide.no = "no", show.n = TRUE)

Step 5. Print or export the descriptive table

Print on R console

The descriptive table can be printed in the R console using the method print, i.e. just typing the name of the object:

restab

If you want to change some bivariate table header, such "p-value" instead of "p.overall" use header.labels argument:

print(restab, header.labels = c(p.overall = "p-value"))

Exporting to different formats

• Word

Export to a Word.

export2word(restab, file = "example1.docx", header.labels = c(p.overall = "p-value"))

• PDF

If you want to create a PDF document with the table in a publish-ready format

export2pdf(restab, file = "example1.pdf", header.labels = c(p.overall = "p-value"))

Note: To create tables in PDF you must have some LaTeX compiler installed on your computer.

• Excel

If you want export the descriptive table to an Excel file:

export2xls(restab, file = "example1.xlsx", header.labels = c(p.overall = "p-value"))

Note: You must have writexlR package installed.

• Markdown

You can translate to Markdown code and insert it in a R-markdown chunk to create reproducible reports.

export2md(restab, header.labels = c(p.overall = "p-value"))

• LaTeX

Similar to Markdown file, it is possible to insert LaTeX code in a .tex.

export2tex(restab, header.labels = c(p.overall = "p-value"))

isubirana/compareGroups documentation built on Jan. 31, 2024, 9:19 p.m.