Reflections on 'Statistical Power and Structural Equation Models in Business Research'

Autor:	Shaun McQuitty
Rok vydání:	2018
Předmět:	Computer science 05 social sciences 050401 social sciences methods 01 natural sciences Statistical power Structural equation modeling Power (physics) 010104 statistics & probability 0504 sociology Business Research Econometrics 0101 mathematics Statistic Statistical hypothesis testing
Zdroj:	Journal of Global Scholars of Marketing Science. 28:272-277
ISSN:	2163-9167 2163-9159
DOI:	10.1080/21639159.2018.1434806
Popis:	This article reflects on McQuitty’s 2004 paper, which describes the importance of statistical power for hypothesis testing and, more specifically, the evaluation of structural equation models (SEMs). McQuitty also explains the use of what was at the time a new method developed by MacCallum, Browne, and Sugawara for estimating the power associated with the RMSEA fit statistic, and then applies the method to SEMs used in marketing journals. A sizeable portion of the published models had too little power to reject false models (a Type II error) or too much power, which leads to the over rejection of correct models (a Type I error). A brief discussion of the impact of McQuitty and ongoing research of power for SEMs follows.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::821ae710701b5b5202c6dc8cd915d49e https://doi.org/10.1080/21639159.2018.1434806 Zobrazit plný text záznamu