Confidence Interval Methods for Coefficient Alpha on the Basis of Discrete, Ordinal Response Items: Which One, if Any, is the Best?

In this study, the authors aimed to examine 8 of the different methods for computing confidence intervals around alpha that have been proposed to determine which of these, if any, is the most accurate and precise. Monte Carlo methods were used to simulate samples under known and controlled populatio...

Full description

Saved in:
Bibliographic Details
Published inThe Journal of experimental education Vol. 79; no. 4; pp. 382 - 403
Main Authors Romano, Jeanine L., Kromrey, Jeffrey D., Owens, Corina M., Scott, Heather M.
Format Journal Article
LanguageEnglish
Published Washington Taylor & Francis Group 01.01.2011
Taylor & Francis Group, LLC
Routledge
Taylor & Francis Inc
Subjects
coefficient alpha
Computation
Confidence intervals
Effect Size
Evaluation Methods
Fisher's z transformations
Intervals
MEASUREMENT, STATISTICS, AND RESEARCH DESIGN
Measures
Monte Carlo Methods
Monte Carlo simulation
Rating Scales
Reliability
Simulation
Statistical Bias
Studies
Online AccessGet full text

Cover

More Information
Summary:In this study, the authors aimed to examine 8 of the different methods for computing confidence intervals around alpha that have been proposed to determine which of these, if any, is the most accurate and precise. Monte Carlo methods were used to simulate samples under known and controlled population conditions wherein the underlying item distribution is nonnormal and when the items' responses are those of rating scales rather than dichotomous items. Overall, one can conclude that, despite concerns expressed over the use of Fisher's method for coefficient alpha, in general, it actually outperformed the other methods. Larger sample sizes and larger coefficient alphas also resulted in better band coverage, whereas smaller number of items resulted in poorer band coverage.
ISSN:0022-0973
1940-0683
DOI:10.1080/00220973.2010.510859