plotICC: Visualization of Intra Class Correlation

In rmngb: Miscellaneous Collection of Functions for Medical Data Analysis

Description

Visually represents Intra Class Correlation.

Usage

plotICC(x, ...)
## Default S3 method:
plotICC(x, subjects, p = FALSE, ptype = 4, ...)
## S3 method for class 'matrix'
plotICC(x, p = FALSE, ptype = 4, ...)
## S3 method for class 'data.frame'
plotICC(x, p = FALSE, ptype = 4, ...)

Arguments

x	vector with the measurement of interest, or matrix with one line per individual and one column per measurement.
subjects	vector with a unique identifiant per subject, not used if x is a matrix.
p	logical. Should points be added for each unique measurement (useful if there is more than 2 measurements per subject)?
ptype	integer. Plotting symbol for the points, corresponds to pch. See points.
...	additional arguments passed to plot.

Details

The range of values measured within a single subject are represented by a vertical line, so there is one line per subject. For lisibility lines are sorted by the means of the measurements.

Value

Nothing of interest.

Author(s)

Antoine Filipovic Pierucci

Examples

m1 <- rnorm(300)
subject <- rep(1:100, 3)
plotICC(m1, subject) # poor ICC

m2 <- m1 + subject / 5
plotICC(m2, subject) # better ICC
plotICC(m2, subject, TRUE) # with points

Example output

rmngb documentation built on May 29, 2017, 9:22 p.m.