means: Calculate summary statistics on a data frame

In RichardHooijmaijers/Rout: Report Functions to Create HTML and PDF Files

Calculate summary statistics on a data frame

Description

This function calculates summary statistics on a data frame. It is intended to use to calculate a standard number of descriptive statistics which are defined in the packs argument

Usage

means(dfrm, variable, by, total = NULL, pack = 1, dig = 2, alpha = 0.1)

Arguments

dfrm	the data frame to calculate the statistics on
variable	character with the variable within the data frame for which the statistics should be calculated
by	character vector of variable(s) within the data frame to stratify the statistics on
total	character vector of variable(s) within the data frame to calculate totals on
pack	numeric indicating number of the pack or name of the function with different descriptive statistics (see details)
dig	the number of digits to use for output statistics (except for N which is always displayed as integer)
alpha	the alpha in case the statistics pack calculates confidence limits

Details

The function calculates multiple statistics of a data frame that can be stratified. Furthermore, the statistics for a total can also be generated. Currently the function has the possibility to calculate 3 different sets of statistics which are commonly used within the field of clinical data analysis:

1. N, Mean, Median, SD, Min, Max

2. Ntot, N, Nmiss, Mean, Median, SD, Min, Max, CLM (different N values are given to identify the number of missing values)

3. N, Mean (SD), Median, Range (Both Mean (SD) and Range are concatenated to generate a dense overview of statistics)

All statistics are calculated on using na.rm set to TRUE, meaning that NA values are removed before calculating the statistics. A predefined set of statistics is chosen and not the possibility to implement user defined descriptive statistics. The reason for this is that there are many other (more simple) options to do this within R.

Value

The function returns a dataframe with calculated descriptive statistics

Examples

data(Indometh)
means(Indometh,"conc","time",total="time",pack=3)

RichardHooijmaijers/Rout documentation built on Sept. 30, 2023, 7:31 p.m.