Reliability analysis for continuous measurements: Equivalence test for agreement
- 5 November 2007
- journal article
- research article
- Published by Wiley in Statistics in Medicine
- Vol. 27 (15) , 2816-2825
Abstract
In tandem with the rapid development of medical technology, methods for assessing intrarater and interrater reliability or agreement across tools for continuous measurements have become an increasingly important research topic. Thus far, a number of reliability assessment methods have been proposed. Among them, the limits of agreement and repeatability coefficients were found to be the most useful tools for assessing reliability when measurements are on a continuous scale. However, both are considered as descriptive methods. The concepts of consistency or conformity require an equivalence test without which judgment would be subjective. In this paper we will extend the repeatability coefficient approach and propose an equivalence test that can be used to confirm the agreement between two or more measurement tools or assess interrater and intrarater reliability. Using this approach, a formula to calculate sample size will also be suggested and examples will be provided to illustrate the method. Copyright © 2007 John Wiley & Sons, Ltd.Keywords
This publication has 11 references indexed in Scilit:
- Methods for assessing reliability and validity for a measurement tool: a case study and critique using the WHO haemoglobin colour scaleStatistics in Medicine, 2004
- Assessing intrarater, interrater and test–retest reliability of continuous measurementsStatistics in Medicine, 2002
- Measuring agreement in method comparison studiesStatistical Methods in Medical Research, 1999
- Sample size and optimal designs for reliability studiesStatistics in Medicine, 1998
- A critical discussion of intraclass correlation coefficientsStatistics in Medicine, 1994
- A note on the use of the intraclass correlation coefficient in the evaluation of agreement between two methods of measurementComputers in Biology and Medicine, 1990
- A Concordance Correlation Coefficient to Evaluate Reproducibility1989
- Statistical evaluation of agreement between two methods for measuring a quantitative variableComputers in Biology and Medicine, 1989
- Sample size requirements for reliability studiesStatistics in Medicine, 1987
- STATISTICAL METHODS FOR ASSESSING AGREEMENT BETWEEN TWO METHODS OF CLINICAL MEASUREMENTThe Lancet, 1986