A Primer on Reliability via Coefficient Alpha and Omega

  • Miguel Padilla Old Dominion University

Abstract

Reliability is an important property to established for a measurement instrument.  There are several types of reliabilities and each with its own purposes.  However, whether warranted or not, internal consistency via coefficient alpha has been the standard for establishing the reliability of a measurement instrument for over 60 years.  During this time, coefficient omega was proposed as an alternative, but has remained obscured.  Nonetheless, neither coefficient has been further developed since their inception until recent years.  A clearer discussion of the three classical true score model forms and their relationship to reliability, computational methods, and the asymptotic distribution of coefficient alpha provided the foundation for fruitful confidence interval research for each coefficient.  In this respect, the normal theory bootstrap CI is the most optimal for both coefficients.  Even so, there is still more promising research to pursue.

Published
2019-10-24
How to Cite
PADILLA, Miguel. A Primer on Reliability via Coefficient Alpha and Omega. Archives of Psychology, [S.l.], v. 3, n. 8, oct. 2019. ISSN 2573-7902. Available at: <https://archivesofpsychology.org/index.php/aop/article/view/125>. Date accessed: 21 nov. 2019. doi: https://doi.org/10.31296/aop.v3i8.125.

Most read articles by the same author(s)

Obs.: This plugin requires at least one statistics/report plugin to be enabled. If your statistics plugins provide more than one metric then please also select a main metric on the admin's site settings page and/or on the journal manager's settings pages.