{"title":"Omega over alpha for reliability estimation of unidimensional communication measures","authors":"Alan K. Goodboy, Matthew M. Martin","doi":"10.1080/23808985.2020.1846135","DOIUrl":null,"url":null,"abstract":"ABSTRACT Cronbach’s alpha (coefficient α) is the conventional statistic communication scholars use to estimate the reliability of multi-item measurement instruments. For many, if not most communication measures, α should not be calculated for reliability estimation. Instead, coefficient omega (ω) should be reported as it aligns with the definition of reliability itself. In this primer, we review α and ω, and explain why ω should be the new ‘gold standard’ in reliability estimation. Using Mplus, we demonstrate how ω is calculated on an available data set and show how preliminary scales can be revised with ‘ω if item deleted.’ We also list several easy-to-use resources to calculate ω in other software programs. Communication researchers should routinely report ω instead of α.","PeriodicalId":36859,"journal":{"name":"Annals of the International Communication Association","volume":"132 1","pages":"422 - 439"},"PeriodicalIF":0.0000,"publicationDate":"2020-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"57","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Annals of the International Communication Association","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1080/23808985.2020.1846135","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"Social Sciences","Score":null,"Total":0}
引用次数: 57
Abstract
ABSTRACT Cronbach’s alpha (coefficient α) is the conventional statistic communication scholars use to estimate the reliability of multi-item measurement instruments. For many, if not most communication measures, α should not be calculated for reliability estimation. Instead, coefficient omega (ω) should be reported as it aligns with the definition of reliability itself. In this primer, we review α and ω, and explain why ω should be the new ‘gold standard’ in reliability estimation. Using Mplus, we demonstrate how ω is calculated on an available data set and show how preliminary scales can be revised with ‘ω if item deleted.’ We also list several easy-to-use resources to calculate ω in other software programs. Communication researchers should routinely report ω instead of α.