SummaryStatsBy: Generate summary statistics

In TalhoukLab/biostatUtil: Utility Functions Used in Biostatistics Projects

Generate summary statistics

Description

Splits the data by two variables, computing relevant statistics for each variable.

Usage

SummaryStatsBy(
  data,
  by1,
  by2,
  var.names,
  stats = c("mean", "sd", "median", "IQR", "range", "missing"),
  digits = 3,
  format = c("raw", "pandoc", "html", "long")
)

Arguments

data	The data.frame containing the data
by1	character string of splitting variable 1
by2	character string of splitting variable 2
var.names	character vector of variables to compute statistics for
stats	statistics to compute for continuous variables
digits	number of digits to round to
format	format to return the table in. Either "raw", "pandoc" (for Word and PDF), "html", or "long" format for graphing and data manipulation using raw values.

Details

The data is split by two variables, by1 and by2, and statistics are computed for continuous variables. Statistics currently supported include ⁠mean, sd, median, IQR, range⁠, and the number of missing cases. For factor variables, the counts, column and row percentages are shown for each of the variable levels.

Note that marginal statistics are also shown for by1, so the order in which you split data matters.

There are four print options for the output: raw gives the output as a character matrix, pandoc gives a Pandoc-friendly output for Word and PDF reports, html gives HTML supported output, and long is a tidy version of raw.

Author(s)

Aline Talhouk, Derek Chiu

Examples

mtcars$vs <- as.factor(mtcars$vs); mtcars$am <- as.factor(mtcars$am)
SummaryStatsBy(mtcars, by1 = "cyl", by2 = "gear", var.names = c("mpg", "vs",
"qsec", "am"))
SummaryStatsBy(mtcars, by1 = "cyl", by2 = "gear", var.names = c("vs",
"qsec"))
SummaryStatsBy(mtcars, by1 = "cyl", by2 = "gear", var.names = c("mpg"))

TalhoukLab/biostatUtil documentation built on April 14, 2025, 4:15 a.m.