Alpha, Omega, and H Internal Consistency Reliability Estimates: Reviewing These Options and When to Use Them

Counseling Outcome Research and Evaluation Pub Date : 2021-07-30 DOI:10.1080/21501378.2021.1940118

Michael T. Kalkbrenner

引用次数: 72

Abstract

Abstract Reliability evidence of test scores is essential in counseling research and program evaluation, as the quality of client care is, in part, based on the proper interpretation of test scores. Cronbach’s coefficient alpha is unquestionably the most frequently reported estimate of internal consistency reliability in counseling research. For over a decade scholars in other disciplines have raised a number of concerns about the utility of coefficient alpha for capturing the reliability of psychological traits, in favor of composite reliability estimates. However, coefficient alpha remains the most dominant reliability index in counseling research. To this end, this article provides a non-technical summary of coefficient alpha, coefficient omega, hierarchical omega, and coefficient H, guidelines for their appropriate usage, and can serve as a reference for counseling practitioners and researchers when conducting outcome research and program evaluation.

查看原文本刊更多论文

Alpha, Omega和H内部一致性可靠性评估:回顾这些选项以及何时使用它们

测试分数的可靠性证据在咨询研究和项目评估中是必不可少的，因为客户服务的质量在一定程度上是基于对测试分数的正确解释。克朗巴赫系数无疑是咨询研究中最常报道的内部一致性信度估计。十多年来，其他学科的学者已经提出了一些关于系数α在捕捉心理特征可靠性方面的效用的担忧，支持复合可靠性估计。然而，在心理咨询研究中，alpha系数仍然是最主要的信度指标。为此，本文对alpha系数、omega系数、hierarchical omega系数和H系数进行了非技术的总结，并给出了适当的使用指南，可供咨询从业者和研究人员在进行结果研究和项目评估时参考。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Counseling Outcome Research and Evaluation Psychology-Psychology (all)

CiteScore

4.50

自引率

0.00%

发文量

期刊介绍： Counseling Outcome Research and Evaluation (CORE) provides counselor educators, researchers, educators, and other mental health practitioners with outcome research and program evaluation practices for work with individuals across the lifespan. It addresses topics such as: treatment efficacy, clinical diagnosis, program evaluation, research design, outcome measure reviews. This journal also serves to address ethical, legal, and cultural concerns in the assessment of dependent variables, implementation of clinical interventions, and outcome research. Manuscripts typically fall into one of the following categories: Counseling Outcome Research: Treatment efficacy and effectiveness of mental health, school, addictions, rehabilitation, family, and college counseling interventions across the lifespan as reported in clinical trials, single-case research designs, single-group designs, and multi- or mixed-method designs.