On the Computation of the RMSEA and CFI from the Mean-And-Variance Corrected Test Statistic with Nonnormal Data in SEM

A new type of nonnormality correction to the RMSEA has recently been developed, which has several advantages over existing corrections. In particular, the new correction adjusts the sample estimate of the RMSEA for the inflation due to nonnormality, while leaving its population value unchanged, so that established cutoff criteria can still be used to judge the degree of approximate fit. A confidence interval (CI) for the new robust RMSEA based on the mean-corrected (“Satorra-Bentler”) test statistic has also been proposed. Follow up work has provided the same type of nonnormality correction for the CFI (Brosseau-Liard & Savalei, 2014). These developments have recently been implemented in lavaan. This note has three goals: a) to show how to compute the new robust RMSEA and CFI from the mean-and-variance corrected test statistic; b) to offer a new CI for the robust RMSEA based on the mean-and-variance corrected test statistic; and c) to caution that the logic of the new nonnormality corrections to RMSEA and CFI is most appropriate for the maximum likelihood (ML) estimator, and cannot easily be generalized to the most commonly used categorical data estimators.     

A scaled difference chi-square test statistic for moment structure analysis

A family of scaling corrections aimed to improve the chi-square approximation of goodness-of-fit test statistics in small samples, large models, and nonnormal data was proposed in Satorra and Bentler (1994). For structural equations models, Satorra-Bentler's (SB) scaling corrections are available in standard computer software. Often, however, the interest is not on the overall fit of a model, but on a test of the restrictions that a null model sayM ₀ implies on a less restricted oneM ₁. IfT ₀ andT ₁ denote the goodness-of-fit test statistics associated toM ₀ andM ₁, respectively, then typically the differenceT _d =T ₀−T ₁ is used as a chi-square test statistic with degrees of freedom equal to the difference on the number of independent parameters estimated under the modelsM ₀ andM ₁. As in the case of the goodness-of-fit test, it is of interest to scale the statisticT _d in order to improve its chi-square approximation in realistic, that is, nonasymptotic and nonormal, applications. In a recent paper, Satorra (2000) shows that the difference between two SB scaled test statistics for overall model fit does not yield the correct SB scaled difference test statistic. Satorra developed an expression that permits scaling the difference test statistic, but his formula has some practical limitations, since it requires heavy computations that are not available in standard computer software. The purpose of the present paper is to provide an easy way to compute the scaled difference chi-square statistic from the scaled goodness-of-fit test statistics of modelsM ₀ andM ₁. A Monte Carlo study is provided to illustrate the performance of the competing statistics. This research was supported by the Spanish grants PB96-0300 and BEC2000-0983, and USPHS grants DA00017 and DA01070.     

Assessing Approximate Fit in Categorical Data Analysis

A family of Root Mean Square Error of Approximation (RMSEA) statistics is proposed for assessing the goodness of approximation in discrete multivariate analysis with applications to item response theory (IRT) models. The family includes RMSEAs to assess the approximation up to any level of association of the discrete variables. Two members of this family are RMSEA₂, which uses up to bivariate moments, and the full information RMSEA_n. The RMSEA₂ is estimated using the M₂ statistic of Maydeu-Olivares and Joe (2005, 2006), whereas for maximum likelihood estimation, RMSEA_n is estimated using Pearson's X² statistic. Using IRT models, we provide cutoff criteria of adequate, good, and excellent fit using the RMSEA₂. When the data are ordinal, we find a strong linear relationship between the RMSEA₂ and the Standardized Root Mean Squared Residual goodness-of-fit index. We are unable to offer cutoff criteria for the RMSEA_n as its population values decrease as the number of variables and categories increase.     

Identifying the Source of Misfit in Item Response Theory Models

When an item response theory model fails to fit adequately, the items for which the model provides a good fit and those for which it does not must be determined. To this end, we compare the performance of several fit statistics for item pairs with known asymptotic distributions under maximum likelihood estimation of the item parameters: (a) a mean and variance adjustment to bivariate Pearson's X², (b) a bivariate subtable analog to Reiser's (1996) overall goodness-of-fit test, (c) a z statistic for the bivariate residual cross product, and (d) Maydeu-Olivares and Joe's (2006) M₂ statistic applied to bivariate subtables. The unadjusted Pearson's X² with heuristically determined degrees of freedom is also included in the comparison. For binary and ordinal data, our simulation results suggest that the z statistic has the best Type I error and power behavior among all the statistics under investigation when the observed information matrix is used in its computation. However, if one has to use the cross-product information, the mean and variance adjusted X² is recommended. We illustrate the use of pairwise fit statistics in 2 real-data examples and discuss possible extensions of the current research in various directions.     

Evaluation of Model Fit in Cognitive Diagnosis Models

Cognitive diagnosis models (CDMs) estimate student ability profiles using latent attributes. Model fit to the data needs to be ascertained in order to determine whether inferences from CDMs are valid. This study investigated the usefulness of some popular model fit statistics to detect CDM fit including relative fit indices (AIC, BIC, and CAIC), and absolute fit indices (RMSEA2, ABS(fcor) and MAX(χ²_jj′)). These fit indices were assessed under different CDM settings with respect to Q-matrix misspecification and CDM misspecification. Results showed that relative fit indices selected the correct DINA model most of the times and selected the correct G-DINA model well across most conditions. Absolute fit indices rejected the true DINA model if the Q-matrix was misspecified in any way. Absolute fit indices rejected the true G-DINA model whenever the Q-matrix was under-specified. RMSEA2 could be artificially low when the Q-matrix was over-specified.     

Empirically Corrected Rescaled Statistics for SEM with Small N and Large p

Survey data often contain many variables. Structural equation modeling (SEM) is commonly used in analyzing such data. With typical nonnormally distributed data in practice, a rescaled statistic T_rml proposed by Satorra and Bentler was recommended in the literature of SEM. However, T_rml has been shown to be problematic when the sample size N is small and/or the number of variables p is large. There does not exist a reliable test statistic for SEM with small N or large p, especially with nonnormally distributed data. Following the principle of Bartlett correction, this article develops empirical corrections to T_rml so that the mean of the empirically corrected statistics approximately equals the degrees of freedom of the nominal chi-square distribution. Results show that empirically corrected statistics control type I errors reasonably well even when N is smaller than 2p, where T_rml may reject the correct model 100% even for normally distributed data. The application of the empirically corrected statistics is illustrated via a real data example.     

The Problem with Having Two Watches: Assessment of Fit When RMSEA and CFI Disagree

The root mean square error of approximation (RMSEA) and the comparative fit index (CFI) are two widely applied indices to assess fit of structural equation models. Because these two indices are viewed positively by researchers, one might presume that their values would yield comparable qualitative assessments of model fit for any data set. When RMSEA and CFI offer different evaluations of model fit, we argue that researchers are likely to be confused and potentially make incorrect research conclusions. We derive the necessary as well as the sufficient conditions for inconsistent interpretations of these indices. We also study inconsistency in results for RMSEA and CFI at the sample level. Rather than indicating that the model is misspecified in a particular manner or that there are any flaws in the data, the two indices can disagree because (a) they evaluate, by design, the magnitude of the model's fit function value from different perspectives; (b) the cutoff values for these indices are arbitrary; and (c) the meaning of “good” fit and its relationship with fit indices are not well understood. In the context of inconsistent judgments of fit using RMSEA and CFI, we discuss the implications of using cutoff values to evaluate model fit in practice and to design SEM studies.     

Limited‐information goodness‐of‐fit testing of hierarchical item factor models

In applications of item response theory, assessment of model fit is a critical issue. Recently, limited‐information goodness‐of‐fit testing has received increased attention in the psychometrics literature. In contrast to full‐information test statistics such as Pearson’s X² or the likelihood ratio G², these limited‐information tests utilize lower‐order marginal tables rather than the full contingency table. A notable example is Maydeu‐Olivares and colleagues’M₂ family of statistics based on univariate and bivariate margins. When the contingency table is sparse, tests based on M₂ retain better Type I error rate control than the full‐information tests and can be more powerful. While in principle the M₂ statistic can be extended to test hierarchical multidimensional item factor models (e.g., bifactor and testlet models), the computation is non‐trivial. To obtain M₂, a researcher often has to obtain (many thousands of) marginal probabilities, derivatives, and weights. Each of these must be approximated with high‐dimensional numerical integration. We propose a dimension reduction method that can take advantage of the hierarchical factor structure so that the integrals can be approximated far more efficiently. We also propose a new test statistic that can be substantially better calibrated and more powerful than the original M₂ statistic when the test is long and the items are polytomous. We use simulations to demonstrate the performance of our new methods and illustrate their effectiveness with applications to real data.     

Type I errors and power of the parametric bootstrap goodness‐of‐fit test: Full and limited information

In sparse tables for categorical data well‐known goodness‐of‐fit statistics are not chi‐square distributed. A consequence is that model selection becomes a problem. It has been suggested that a way out of this problem is the use of the parametric bootstrap. In this paper, the parametric bootstrap goodness‐of‐fit test is studied by means of an extensive simulation study; the Type I error rates and power of this test are studied under several conditions of sparseness. In the presence of sparseness, models were used that were likely to violate the regularity conditions. Besides bootstrapping the goodness‐of‐fit usually used (full information statistics), corrected versions of these statistics and a limited information statistic are bootstrapped. These bootstrap tests were also compared to an asymptotic test using limited information. Results indicate that bootstrapping the usual statistics fails because these tests are too liberal, and that bootstrapping or asymptotically testing the limited information statistic works better with respect to Type I error and outperforms the other statistics by far in terms of statistical power. The properties of all tests are illustrated using categorical Markov models.     

On the asymptotic distribution of Pearson’s X 2 in cross-validation samples

In categorical data analysis, two-sample cross-validation is used not only for model selection but also to obtain a realistic impression of the overall predictive effectiveness of the model. The latter is of particular importance in the case of highly parametrized models capable of capturing every idiosyncracy of the calibrating sample. We show that for maximum likelihood estimators or other asymptotically efficient estimators Pearson's X ² is not asymptotically chi-square in the two-sample cross-validation framework due to extra variability induced by using different samples for estimation and goodness-of-fit testing. We propose an alternative test statistic, X _xval ², obtained as a modification of X ² which is asymptotically chi-square with C−1 degrees of freedom in cross-validation samples. Stochastically, X _xval ² ≤ X ². Furthermore, the use of X ² instead of X _xval ² with a χ _C −1² reference distribution may provide an unduly poor impression of fit of the model in the cross-validation sample. This paper is dedicated to the memory of Michael V. Levine. Requests for reprints should be sent to Albert Maydeu-Olivares, Faculty of Psychology, University of Barcelona, P. Valle de Hebrón, 171, 0835 Barcelona, Spain.     

Abstract: Evaluation of Test Statistics for Robust Structural Equation Modeling With Nonnormal Missing Data

Traditional structural equation modeling (SEM) techniques have trouble dealing with incomplete and/or nonnormal data that are often encountered in practice. Yuan and Zhang (2011a) developed a two-stage procedure for SEM to handle nonnormal missing data and proposed four test statistics for overall model evaluation. Although these statistics have been shown to work well with complete data, their performance for incomplete data has not been investigated in the context of robust statistics.

Focusing on a linear growth curve model, a systematic simulation study is conducted to evaluate the accuracy of the parameter estimates and the performance of five test statistics including the naive statistic derived from normal distribution based maximum likelihood (ML), the Satorra-Bentler scaled chi-square statistic (RML), the mean- and variance-adjusted chi-square statistic (AML), Yuan-Bentler residual-based test statistic (CRADF), and Yuan-Bentler residual-based F statistic (RF). Data are generated and analyzed in R using the package rsem (Yuan & Zhang, 2011b).

Based on the simulation study, we can observe the following: (a) The traditional normal distribution-based method cannot yield accurate parameter estimates for nonnormal data, whereas the robust method obtains much more accurate model parameter estimates for nonnormal data and performs almost as well as the normal distribution based method for normal distributed data. (b) With the increase of sample size, or the decrease of missing rate or the number of outliers, the parameter estimates are less biased and the empirical distributions of test statistics are closer to their nominal distributions. (c) The ML test statistic does not work well for nonnormal or missing data. (d) For nonnormal complete data, CRADF and RF work relatively better than RML and AML. (e) For missing completely at random (MCAR) missing data, in almost all the cases, RML and AML work better than CRADF and RF. (f) For nonnormal missing at random (MAR) missing data, CRADF and RF work better than AML. (g) The performance of the robust method does not seem to be influenced by the symmetry of outliers.     

Limited‐information goodness‐of‐fit testing of diagnostic classification item response models

Despite the growing popularity of diagnostic classification models (e.g., Rupp et al., 2010, Diagnostic measurement: theory, methods, and applications, Guilford Press, New York, NY) in educational and psychological measurement, methods for testing their absolute goodness of fit to real data remain relatively underdeveloped. For tests of reasonable length and for realistic sample size, full‐information test statistics such as Pearson's X² and the likelihood ratio statistic G² suffer from sparseness in the underlying contingency table from which they are computed. Recently, limited‐information fit statistics such as Maydeu‐Olivares and Joe's (2006, Psychometrika, 71, 713) M₂ have been found to be quite useful in testing the overall goodness of fit of item response theory models. In this study, we applied Maydeu‐Olivares and Joe's (2006, Psychometrika, 71, 713) M₂ statistic to diagnostic classification models. Through a series of simulation studies, we found that M₂ is well calibrated across a wide range of diagnostic model structures and was sensitive to certain misspecifications of the item model (e.g., fitting disjunctive models to data generated according to a conjunctive model), errors in the Q‐matrix (adding or omitting paths, omitting a latent variable), and violations of local item independence due to unmodelled testlet effects. On the other hand, M₂ was largely insensitive to misspecifications in the distribution of higher‐order latent dimensions and to the specification of an extraneous attribute. To complement the analyses of the overall model goodness of fit using M₂, we investigated the utility of the Chen and Thissen (1997, J. Educ. Behav. Stat., 22, 265) local dependence statistic $X_{\mathrm{LD}}^2$ for characterizing sources of misfit, an important aspect of model appraisal often overlooked in favour of overall statements. The $X_{\mathrm{LD}}^2$ statistic was found to be slightly conservative (with Type I error rates consistently below the nominal level) but still useful in pinpointing the sources of misfit. Patterns of local dependence arising due to specific model misspecifications are illustrated. Finally, we used the M₂ and $X_{\mathrm{LD}}^2$ statistics to evaluate a diagnostic model fit to data from the Trends in Mathematics and Science Study, drawing upon analyses previously conducted by Lee et al., (2011, IJT, 11, 144).     

Correcting Too Much or Too Little? The Performance of Three Chi-Square Corrections

This simulation study investigates the performance of three test statistics, T₁, T₂, and T₃, used to evaluate structural equation model fit under non normal data conditions. T₁ is the well-known mean-adjusted statistic of Satorra and Bentler. T₂ is the mean-and-variance adjusted statistic of Sattertwaithe type where the degrees of freedom is manipulated. T₃ is a recently proposed version of T₂ that does not manipulate degrees of freedom. Discrepancies between these statistics and their nominal chi-square distribution in terms of errors of Type I and Type II are investigated. All statistics are shown to be sensitive to increasing kurtosis in the data, with Type I error rates often far off the nominal level. Under excess kurtosis true models are generally over-rejected by T₁ and under-rejected by T₂ and T₃, which have similar performance in all conditions. Under misspecification there is a loss of power with increasing kurtosis, especially for T₂ and T₃. The coefficient of variation of the nonzero eigenvalues of a certain matrix is shown to be a reliable indicator for the adequacy of these statistics.     

Ensuring Positiveness of the Scaled Difference Chi-square Test Statistic

A scaled difference test statistic [(T)\tilde]_d\tilde{T}{}_{d} that can be computed from standard software of structural equation models (SEM) by hand calculations was proposed in Satorra and Bentler (Psychometrika 66:507–514, 2001). The statistic [(T)\tilde]_d\tilde{T}_{d} is asymptotically equivalent to the scaled difference test statistic [`(T)]<sub>d</sub>\bar{T}_{d} introduced in Satorra (Innovations in Multivariate Statistical Analysis: A Festschrift for Heinz Neudecker, pp. 233–247, 2000), which requires more involved computations beyond standard output of SEM software. The test statistic [(T)\tilde]<sub>d</sub>\tilde{T}_{d} has been widely used in practice, but in some applications it is negative due to negativity of its associated scaling correction. Using the implicit function theorem, this note develops an improved scaling correction leading to a new scaled difference statistic [`(T)]_d\bar{T}_{d} that avoids negative chi-square values.     

Modified Distribution-Free Goodness-of-Fit Test Statistic

Covariance structure analysis and its structural equation modeling extensions have become one of the most widely used methodologies in social sciences such as psychology, education, and economics. An important issue in such analysis is to assess the goodness of fit of a model under analysis. One of the most popular test statistics used in covariance structure analysis is the asymptotically distribution-free (ADF) test statistic introduced by Browne (Br J Math Stat Psychol 37:62–83, 1984). The ADF statistic can be used to test models without any specific distribution assumption (e.g., multivariate normal distribution) of the observed data. Despite its advantage, it has been shown in various empirical studies that unless sample sizes are extremely large, this ADF statistic could perform very poorly in practice. In this paper, we provide a theoretical explanation for this phenomenon and further propose a modified test statistic that improves the performance in samples of realistic size. The proposed statistic deals with the possible ill-conditioning of the involved large-scale covariance matrices.     

A Cautionary Note on Using G2(dif) to Assess Relative Model Fit in Categorical Data Analysis

The likelihood ratio test statistic G²(dif) is widely used for comparing the fit of nested models in categorical data analysis. In large samples, this statistic is distributed as a chi-square with degrees of freedom equal to the difference in degrees of freedom between the tested models, but only if the least restrictive model is correctly specified. Yet, this statistic is often used in applications without assessing the adequacy of the least restrictive model. This may result in incorrect substantive conclusions as the above large sample reference distribution for G²(dif) is no longer appropriate. Rather, its large sample distribution will depend on the degree of model misspecification of the least restrictive model. To illustrate this, a simulation study is performed where this statistic is used to compare nested item response theory models under various degrees of misspecification of the least restrictive model. G²(dif) was found to be robust only under small model misspecification of the least restrictive model. Consequently, we argue that some indication of the absolute goodness of fit of the least restrictive model is needed before employing G²(dif) to assess relative model fit.     

教师组织承诺结构的验证性因素分析

通过文献回顾、半结构访谈、问卷调查及理论分析,提出了教师组织承诺的结构。对278名教师进行测查以考察教师组织承诺的因素结构,验证了教师组织承诺的结构,发现教师组织承诺包含四个维度,即感情承诺、规范承诺、理想承诺和投入承诺。     

An Investigation of the Sample Performance of Two Nonnormality Corrections for RMSEA

The root mean square error of approximation (RMSEA) is a popular fit index in structural equation modeling (SEM). Typically, RMSEA is computed using the normal theory maximum likelihood (ML) fit function. Under nonnormality, the uncorrected sample estimate of the ML RMSEA tends to be inflated. Two robust corrections to the sample ML RMSEA have been proposed, but the theoretical and empirical differences between the 2 have not been explored. In this article, we investigate the behavior of these 2 corrections. We show that the virtually unknown correction due to Li and Bentler (2006) Li, L. and Bentler, P. M. 2006. “Robust statistical tests for evaluating the hypothesis of close fit of misspecified mean and covariance structural models.”. In UCLA Statistics Preprint #506. Los Angeles: University of California..  [Google Scholar], which we label the sample-corrected robust RMSEA, is a consistent estimate of the population ML RMSEA yet drastically reduces bias due to nonnormality in small samples. On the other hand, the popular correction implemented in several SEM programs, which we label the population-corrected robust RMSEA, has poor properties because it estimates a quantity that decreases with increasing nonnormality. We recommend the use of the sample-corrected RMSEA with nonnormal data and its wide implementation.     

How Should We Assess the Fit of Rasch-Type Models? Approximating the Power of Goodness-of-Fit Statistics in Categorical Data Analysis

We investigate the performance of three statistics, R ₁, R ₂ (Glas in Psychometrika 53:525–546, 1988), and M ₂ (Maydeu-Olivares & Joe in J. Am. Stat. Assoc. 100:1009–1020, 2005, Psychometrika 71:713–732, 2006) to assess the overall fit of a one-parameter logistic model (1PL) estimated by (marginal) maximum likelihood (ML). R ₁ and R ₂ were specifically designed to target specific assumptions of Rasch models, whereas M ₂ is a general purpose test statistic. We report asymptotic power rates under some interesting violations of model assumptions (different item discrimination, presence of guessing, and multidimensionality) as well as empirical rejection rates for correctly specified models and some misspecified models. All three statistics were found to be more powerful than Pearson’s X ² against two- and three-parameter logistic alternatives (2PL and 3PL), and against multidimensional 1PL models. The results suggest that there is no clear advantage in using goodness-of-fit statistics specifically designed for Rasch-type models to test these models when marginal ML estimation is used.     

Limited‐information goodness‐of‐fit testing of item response theory models for sparse 2P tables

Bartholomew and Leung proposed a limited‐information goodness‐of‐fit test statistic (Y) for models fitted to sparse 2^P contingency tables. The null distribution of Y was approximated using a chi‐squared distribution by matching moments. The moments were derived under the assumption that the model parameters were known in advance and it was conjectured that the approximation would also be appropriate when the parameters were to be estimated. Using maximum likelihood estimation of the two‐parameter logistic item response theory model, we show that the effect of parameter estimation on the distribution of Y is too large to be ignored. Consequently, we derive the asymptotic moments of Y for maximum likelihood estimation. We show using a simulation study that when the null distribution of Y is approximated using moments that take into account the effect of estimation, Y becomes a very useful statistic to assess the overall goodness of fit of models fitted to sparse 2^P tables.