Assessing Goodness of Fit: Simple Rules of Thumb Simply Do Not Work

By Christopher D. Nye, Fritz Drasgow. 2011

223 citations recorded by the source corpus.

Subfields: Management Information Systems, Organizational Behavior and Human Resource Management

View Paper: https://www.semanticscholar.org/paper/5f7d862c800d653edb8a647e816947c595dd61a9

How researchers see this paper

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

In Management Information Systems

In Organizational Behavior and Human Resource Management

Citations in the corpus

From Doing Good to Looking Even Better: The Dynamics of CSR and Reputation (2019, methodology): “All of the SEM modeling used maximum likelihood (ML) estimators, which Nye and Drasgow (2011) note, “in large samples, ML estimates are (nearly) unbiased with the smallest possible standard errors” (p. 550).”
An Overview and Practical Guide to IRT Measurement Equivalence Analysis (2015, methodology): “To overcome the problem of using rules of thumb that may not generalize to all scenarios (Nye & Drasgow, 2010), one possibility for assessing overall model-data fit is to use a parametric bootstrap analysis and examine the associated p values (Vermunt & Magidson, 2005).”
Authentic leadership extends beyond work: A multilevel model of work-family conflict and enrichment (2017, methodology): “Fit indices should not be used as golden rules, but to compare different models (Nye & Drasgow, 2011).”
Leader Narcissism Predicts Malicious Envy and Supervisor-Targeted Counterproductive Work Behavior: Evidence from Field and Experimental Research (2018, unknown): “Fit indices should not be treated as golden rules (Nye and Drasgow 2011), but for ‘‘evaluative comparisons of competing a priori models’’ (Goffin 2007, p. 833).”
RMSEA, CFI, and TLI in structural equation modeling with ordered categorical data: The story they tell depends on the estimation methods (2018, result): “The results for the asymptotic values of the DWLS fit indices are consistent with those based on finite samples in Nye and Drasgow (2011). Nye and Drasgow first employed a dataset consisting of 9,292 examinees and showed that DWLS produced smaller RMSEAS and larger CFIS and TLIS values than didML on”
Use of Structural Equation Modeling in Tourism Research (2013, methodology): “Researchers should note that recent developments in SEM suggest that RMSEA may be influenced by sample size, leading to Type I and Type II errors (Nye and Drasgow 2011), SRMSR is most sensitive to nonnormality, while NFI, CFI, and NNFI are most sensitive to model misspecification (Sun, Konold, and…”