Rethinking fit assessment in structural equation modelling: A commentary and elaboration on Barrett (2007)

By C. Mcintosh. 2007

230 citations recorded by the source corpus.

Subfields: Management Science and Operations Research

View Paper: https://www.semanticscholar.org/paper/1052594b2f95039113403cc35bdf740f4b552b26

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Management Science and Operations Research

Citations in the corpus

RMSEA, CFI, and TLI in structural equation modeling with ordered categorical data: The story they tell depends on the estimation methods (2018, methodology): “Strong arguments against the application of RMSEA, CFI, and TLI and their conventional cutoff values have been raised in the SEM literature (e.g., Barrett, 2007; Marsh et al., 2004; McIntosh, 2007).”
On doing better science: From thrill of discovery to policy implications (2017, background): “models (Ropovik, 2015)—such models have biased estimates (Bollen, Kirby, Curran, Paxton, & Chen, 2007; Hayduk, 2014; McIntosh, 2007); indexes of fit (CFI, RMSEA) should not be trusted because they depend on idiosyncrasies of the model (Chen, Curran, Bollen, Kirby, & Paxton, 2008; Marsh, Hau, & Wen, ”
Perceived Labor Productivity in Small Firms—The Effects of High–Performance Work Systems and Group Culture through Employee Retention (2012, methodology): “The data used in the analysis were collected by Michie (2006). The sampling frame was derived from a stratified sample of firms in the UK from the Dunn and Bradstreet database.”
Values Matter, Don’t They? – Combining Theory of Planned Behavior and Personal Values as Predictors of Social Entrepreneurial Intention (2018, background): “Furthermore, as the test is very sensitive to violations of multivariate normality (as seen in our sample) there is an even higher probability that models will be rejected regardless of their suitability to meet the data structure (McIntosh 2007).”
On making causal claims: A review and recommendations (2010, unknown): “…test will always be significant (see Antonakis, House, Rowold & Borgmann, submitted for 1103J. Antonakis et al. / The Leadership Quarterly 21 (2010) 1086–1120 publication; Hayduk, Cummings, Boadu, Pazderka-Robinson, & Boulianne, 2007; Kline, 2010; Marsh et al., 2004; McIntosh, 2007; Shipley, 2000).”
Current Methodological Considerations in Exploratory and Confirmatory Factor Analysis (2011, methodology): “, factor loadings) when using global fit indexes, such as the χ(2), because models with irrelevant misspecifications might be rejected if the sample is large, and models with substantial misspecifications might not be rejected if the sample is small (McDonald & Marsh, 1990; McIntosh, 2007; Saris et ”
On the adoption of partial least squares in psychological research: Caveat emptor (2015, background): “…Willaby et al. were left unaware that their model of the drivers of GPA was in fact misspecified, which in turn implies that the parameter estimates are untrustworthy (Hayduk, Cummings, Boadu, PazderkaRobinson, & Boulianne, 2007; Kolenikov, 2011; McIntosh, 2007; Yuan, Marshall, & Bentler, 2003).”
Instrumental leadership: Measurement and extension of transformational–transactional leadership theory (2014, unknown): “Despite the large sample size and power to reject a wrong model, the fit was excellent: χ 2 (70) = 65.80, p > .05 (CFI = 1.00, RMSEA = .00).”
Evaluating Model Fit:a Synthesis of the Structural Equation Modelling Literature (2008, background): “Firstly, this test assumes multivariate normality and severe deviations from normality may result in model rejections even when the model is properly specified (McIntosh, 2006).”