obs.agree (version 1.0)

IBMD: Information-based measure of disagreement

Description

Calculates the Information-Based Measure of Disagreement (IBMD) coefficient on a continuous measure.

Usage

IBMD(x, conf.levels = 0.95)

Arguments

x

n*m matrix or dataframe with n subjects and m observers. If the observer number differ for each subject missing values should be represented by the NA symbol.

conf.levels

confidence level of the interval. Must be a single number between 0 and 1.

Value

A list containing the following components:

Subjects

number of subjects.

Observers

maximum number of observers.

IBMD

the information based measure of disagreement coefficient and the respective confidence interval.

Details

The IBMD was proposed (Costa-Santos, 2010) on the basis of Shannon's notion of entropy (Shannon, 1948), described as the average amount of information contained in a variable. In 2013 (Henriques, 2013) was generalized to measure the disagreement among measurements obtained by several observers, allowing different number of observations in each case. It is appropriate for ratio-scale variables with positive values and ranges from 0 (no disagreement) to 1. The confidence interval is estimated using a bootstrap procedure.

References

Costa-Santos, C, Antunes, L., Souto, A. and Bernardes, J. (2010) Assessment of disagreement: a new information-based approach. Ann Epidemiol, 20(7):555-61.

Shannon, C.E. (1948) A mathematical theory of communication. Bell System Technical Journal, 27:379-423 and 623-656.

Henriques, T., Antunes, L., Bernardes, J., Matias, M., Sato, D. and Costa-Santos, C. (2013) Information-based measure of disagreement for more than two observers: a useful tool to compare the degree of observer disagreement. BMC Medical Research Methodology. 13(1):47.

Carpenter J. and Bithell J. (2000) Bootstrap confidence intervals: when, which, what? A practical guide for medical statisticians. Stat Med, 19(9):1141-1164.

Examples

data(gymnasts)
head(gymnasts)

## Not run: 
# 	IBMD(gymnasts[1:20,-1])
# 	IBMD(gymnasts[21:40,-1])## End(Not run)