On Chi-Square Difference and z Tests in Mean and Covariance Structure Analysis when the Base Model is Misspecified

By Ke-Hai Yuan, Peter M. Bentler. 2004

251 citations recorded by the source corpus.

Subfields: Management Science and Operations Research, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/6f9dac7f179853dca1d8797d54b59b04eca0b0a5

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

In Statistics and Probability

Citations in the corpus

Testing Measurement Invariance Using MIMIC (2012, methodology): “As explained by Yuan and Bentler (2004), the LR tests between a misspecified baseline model and a misspecified unconstrained model did not yield the same power in detecting DIF items as the LR tests did with only one DIF item.”
Motivational and Skills, Social, and Self-Management Predictors of College Outcomes: Constructing the Student Readiness Inventory (2005, background): “Recently, Yuan and Bentler (2004) showed that the nested chi-square test is not reliable when the base model is not correctly specified.”
Goodness-of-Fit Testing (2010, background): “When more than one substantive model is under consideration , researchers are also interested in a relative model fit (i.e., the discrepancy between two models; see Yuan and Bentler, 2004; Maydeu-Olivares and Cai, 2006).”
The interplay of perceived stress, self-determination and school engagement in adolescence (2014, methodology): “The first model was considered as the (1) baseline model; it was compared to the (2) less restrictive model (partial mediation) by means of the 2-difference test (Geiser, 2010; Yuan & Bentler, 2004).”
Using ΔGoodness-of-Fit Indexes in Assessing Mean Structure Invariance (2009, background): “Because other researchers (eg, Satorra & Bentler, 2001 ; Yuan & Bentler, 2004 ) have done considerable work related to the use of the δχ 2 test in multisample invariance analysis, we did not address the issues related to the δχ 2 test, but focused on δgoodness-of-fit indexes instead.”
Multilevel Latent Class Analysis: Parametric and Nonparametric Models (2014, background): “In addition, other types of latent models do differ in terms of outcomes (Type I error and power) when misspecification is present (e.g., French & Finch, 2011; Yuan & Bentler, 2004).”
Using Modification Indexes to Detect Turning Points in Longitudinal Data: A Monte Carlo Study (2010, background): “Several sources suggest that misspecification of the covariance matrix of the residuals primarily affects the variance 236 KWOK, LUO, WEST components and the standard errors (Kwok, West, & Green, 2007; Verbeke & Lesaffre, 1997; Yuan & Bentler, 2004).”
The Rediscovery of Bifactor Measurement Models (2012, background): “Several scholars have advised that only if the least restricted model (in this case the bifactor) is judged to fit the data (Yuan & Bentler, 2004) is it appropriate to consider whether applying a more restricted, nested model significantly degrades that fit.”
Current Methodological Considerations in Exploratory and Confirmatory Factor Analysis (2011, methodology): “Although the LRT is used to compare models for factor selection, it should not be used to compare models when the base model is misspecified because it can result in inflated Type 1 and Type 2 errors (Yuan & Bentler, 2004).”