In Search of Golden Rules: Comment on Hypothesis-Testing Approaches to Setting Cutoff Values for Fit Indexes and Dangers in Overgeneralizing Hu and Bentler's (1999) Findings

By Herbert W. Marsh, K. Hau, Zhonglin Wen. 2004

6,670 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Management Science and Operations Research

View Paper: https://www.semanticscholar.org/paper/a509556484bfae1c07049fe5620d6afa19396252

How researchers see this paper

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

In Artificial Intelligence

In Management Science and Operations Research

Citations in the corpus

The Process by Which Relative Autonomous Motivation Affects Intentional Behavior: Comparing Effects Across Dieting and Exercise Behaviors (2006, background): “Recently, however, Marsh, Hau, and Wen (2004) recommended that researchers treat such cut-off criteria as guidelines rather than criteria to be used in conjunction with a strict hypothesis-testing approach.”
Small-Sample Robust Estimators of Noncentrality-Based and Incremental Model Fit (2009, neutral): “" Rules of thumb, however, are always subject to criticism and should be interpreted with caution and tolerance (e.g., Chen et al., 2008; Marsh, Hau, & Wen, 2004).”
Relationships between perceived parental involvement in homework, student homework behaviors, and academic achievement: differences among elementary, junior high, and high school students (2015, methodology): “In addition to a chi-squared (χ(2)) test and its associated probability (p), we used two related indexes: the comparative fit index (CFI) and the Tucker-Lewis index (TLI) (Hu and Bentler 1999); a close-fit parsimony-based index—the root mean square error of approximation (RMSEA), including 90 % conf”
Sensitivity of Fit Indexes to Misspecified Structural or Measurement Model Components: Rationale of Two-Index Strategy Revisited (2005, methodology): “As discussed in Marsh et al. (2004) , the majority of the misspecified models considered in Hu and Bentler (1998, 1999) were actually acceptable models with very minor degrees of misspecification (acceptable misspecified models).”
The influence of future time perspective on work engagement and job performance: the role of job crafting (2017, positive): “The conventional cut-off values of these fit indices were used to assess model fit (i.e., CFI, TLI, IFI > .90, and RMSEA < .08 to indicate acceptable fit, and CFI, TLI, IFI > .95, and RMSEA < .05 to indicate good fit; Hu & Bentler, 1999; Marsh, Hau, & Wen, 2004).”
How to factor-analyze your data right: do’s, don’ts, and how-to’s. (2010, background): “Perhaps some criteria may be loosened without causing overly drastic consequences; for example, RMSEA smaller than .08 should be considered acceptable under most circumstances and so is CFI/TLI/RNI greater than .90 (see Fan et al., 1999; Marsh et al., 2004).”