The Effect of Latent and Error Non-Normality on Measures of Fit in Structural Equation Modeling

2021 ◽  
pp. 001316442110462
Author(s):  
Lisa J. Jobst ◽  
Max Auerswald ◽  
Morten Moshagen
Keyword(s):  
Structural Equation Modeling ◽  
Root Mean Square ◽  
Latent Variables ◽  
Structural Equation ◽  
Generalized Least Squares ◽  
Estimation Methods ◽  
Equation Modeling ◽  
Analytic Structure ◽  
Mean Square ◽  
Test Statistic

Prior studies investigating the effects of non-normality in structural equation modeling typically induced non-normality in the indicator variables. This procedure neglects the factor analytic structure of the data, which is defined as the sum of latent variables and errors, so it is unclear whether previous results hold if the source of non-normality is considered. We conducted a Monte Carlo simulation manipulating the underlying multivariate distribution to assess the effect of the source of non-normality (latent, error, and marginal conditions with either multivariate normal or non-normal marginal distributions) on different measures of fit (empirical rejection rates for the likelihood-ratio model test statistic, the root mean square error of approximation, the standardized root mean square residual, and the comparative fit index). We considered different estimation methods (maximum likelihood, generalized least squares, and (un)modified asymptotically distribution-free), sample sizes, and the extent of non-normality in correctly specified and misspecified models to investigate their performance. The results show that all measures of fit were affected by the source of non-normality but with varying patterns for the analyzed estimation methods.

The Adult-Oriented Sport Coaching Survey: An Instrument Designed to Assess Coaching Behaviors Tailored to Adult Athletes

2020 ◽  
Vol 42 (5) ◽  
pp. 368-385
Author(s):  
Scott Rathwell ◽  
Bradley W. Young ◽  
Bettina Callary ◽  
Derrik Motz ◽  
Matt D. Hoffmann ◽  
...  
Keyword(s):  
Factor Analysis ◽  
Structural Equation Modeling ◽  
Root Mean Square ◽  
Structural Equation ◽  
Equation Modeling ◽  
Mean Square ◽  
Root Mean Square Residual ◽  
Confirmatory Factor ◽  
Masters Athletes ◽  
Fit Index

Adult sportspersons (Masters athletes, aged 35 years and older) have unique coaching preferences. No existing resources provide coaches with feedback on their craft with Masters athletes. Three studies evaluated an Adult-Oriented Coaching Survey. Study 1 vetted the face validity of 50 survey items with 12 Masters coaches. Results supported the validity of 48 items. In Study 2, 383 Masters coaches completed the survey of 50 items. Confirmatory factor analysis and exploratory structural equation modeling indicated issues with model fit. Post hoc modifications improved fit, resulting in a 22-item, five-factor model. In Study 3, 467 Masters athletes responded to these 22 items reflecting perceptions of their coaches. Confirmatory factor analysis (comparative fit index = .951, standardized root mean square residual = .036, and root mean square error of approximation = .049) and exploratory structural equation modeling (comparative fit index = .977, standardized root mean square residual = .019, and root mean square error of approximation = .041) confirmed the model. The resultant Adult-Oriented Sport Coaching Survey provides a reliable and factorially valid instrument for measuring adult-oriented coaching practices.

scholarly journals Erfassung der Emotionsregulation bei Jugendlichen anhand des „Affective Style Questionnaire – Youth (ASQ-Y)“

Diagnostica ◽  
2019 ◽  
Vol 65 (1) ◽  
pp. 49-59
Author(s):  
Johannes Graser ◽  
Christiane Heimlich ◽  
Augustin Kelava ◽  
Stefan G. Hofmann ◽  
Ulrich Stangier ◽  
...  
Keyword(s):  
Structural Equation Modeling ◽  
Root Mean Square Error ◽  
Root Mean Square ◽  
Structural Equation ◽  
Equation Modeling ◽  
Mean Square ◽  
Affective Style ◽  
Root Mean Square Residual ◽  
Fit Index ◽  
Tucker Lewis Index

Zusammenfassung. Zur Erfassung der 3 Emotionsregulationsstrategien Unterdrücken, Anpassen / Neubewerten und Akzeptieren wurde der Affective Style Questionnaire für Jugendliche (ASQ-Y) adaptiert und an einer entsprechenden Stichprobe (N = 1 092) validiert. Die Dimensionalität des englischen Originalfragebogens und der deutschen Version für Erwachsene konnte auch für Jugendliche bestätigt werden. Während der Analyse kam das ESEM-Verfahren (Exploratory Structural Equation Modeling) zum Einsatz, die Kennwerte bewegten sich im akzeptablen bis sehr guten Bereich. Der Comparative Fit Index (CFI) erreichte einen akzeptablen Wert von .938, ebenso der Tucker–Lewis Index (TLI) mit einem Wert von .911. Der Root Mean Square Error of Approximation (RMSEA) lag bei einem sehr guten Wert von .050, das Standardized Root Mean Square Residual (SRMR) erreichte einen guten Wert von .030. Die internen Konsistenzen der 3 Skalen (Unterdrücken: α = .77; Anpassen / Neubewerten: α = .76; Akzeptieren: α = .76) erreichten (vergleichbar mit dem englischen Original und der deutschen Erwachsenenstichprobe) zufriedenstellende Werte. Die Subskalen zeigten hypothesenkonforme diskriminante und konvergente Zusammenhänge mit etablierten Verfahren des Forschungsbereichs Emotionsregulation, was für die Konstruktvalidität spricht. Insgesamt ist der ASQ-Y als Messinstrument zur Erfassung von verschiedenen Emotionsregulationsstrategien bei Jugendlichen geeignet und ökonomisch in seiner Anwendung. Der ASQ-Y kann in der Allgemeinbevölkerung und in der Prävention eingesetzt werden. Nach entsprechender Validierung ist der Einsatz auch im klinischen Setting möglich.

scholarly journals The Effect of Estimation Methods on SEM Fit Indices

2019 ◽  
Vol 80 (3) ◽  
pp. 421-445 ◽  
Author(s):  
Dexin Shi ◽  
Alberto Maydeu-Olivares
Keyword(s):  
Least Squares ◽  
Root Mean Square ◽  
Structural Equation ◽  
Weighted Least Squares ◽  
Estimation Method ◽  
Estimation Methods ◽  
Equation Modeling ◽  
Model Parameters ◽  
Mean Square ◽  
Fit Indices

We examined the effect of estimation methods, maximum likelihood (ML), unweighted least squares (ULS), and diagonally weighted least squares (DWLS), on three population SEM (structural equation modeling) fit indices: the root mean square error of approximation (RMSEA), the comparative fit index (CFI), and the standardized root mean square residual (SRMR). We considered different types and levels of misspecification in factor analysis models: misspecified dimensionality, omitting cross-loadings, and ignoring residual correlations. Estimation methods had substantial impacts on the RMSEA and CFI so that different cutoff values need to be employed for different estimators. In contrast, SRMR is robust to the method used to estimate the model parameters. The same criterion can be applied at the population level when using the SRMR to evaluate model fit, regardless of the choice of estimation method.

Chinese Americans’ Family History of Colorectal Cancer Communication With Primary Care Physicians

2021 ◽  
pp. 109019812110097
Author(s):  
Yu-Lyu Yeh ◽  
Ming Li ◽  
Oi-Man Kwok ◽  
Ping Ma ◽  
Lei-Shih Chen
Keyword(s):  
Colorectal Cancer ◽  
Primary Care ◽  
Structural Equation Modeling ◽  
Family History ◽  
Root Mean Square ◽  
Primary Care Physicians ◽  
Structural Equation ◽  
Chinese Americans ◽  
Equation Modeling ◽  
Mean Square

Background Colorectal cancer (CRC) is the third most common cancer for Chinese Americans. Family history (FH) plays an important role in clinical practice for CRC prevention. Nevertheless, Chinese Americans’ FH of CRC communication with primary care physicians (PCPs) are still unknown. Aims This study examined Chinese Americans’ behavior and the underlying psychological factors for FH of CRC communication with PCPs. Method A total number of 742 Chinese Americans completed a survey developed based on the health belief model, the theory of planned behavior, and the social cognitive theory. Data were analyzed using structural equation modeling. Results Majority of the Chinese American participants (75.3%) had never discussed FH of CRC with their PCPs. Lack of inquiries from the PCPs was the main barrier. Structural equation modeling results suggested a good model fit between our theoretical model and the survey data (comparative fit index [CFI] = .946, root mean square error of approximation [RMSEA] = .070, and standardized root mean square residual [SRMR] = .020). Participants’ FH of CRC communication with PCPs was positively associated with their intention (β = .30, p < .001), which was positively correlated to attitudes (β = .29, p < .001) and self-efficacy in discussing FH of CRC with PCPs (β = .57, p < .001). Their attitudes were positively associated with perceived susceptibility to CRC (β = .08, p < .05) and the perceived benefits of communicating FH of CRC (β = .52, p < .001). Conclusions Given that most Chinese Americans in this study did not communicate their FH of CRC with their PCPs, it is important to promote such behavior among Chinese Americans. Our structural equation modeling findings can guide future interventions and education for this underserved racial/ethnic minority group.

scholarly journals Fixed Effect Meta-Analytic Structural Equation Modeling (MASEM) Estimation Using Generalized Method of Moments (GMM)

Symmetry ◽  
2021 ◽  
Vol 13 (12) ◽  
pp. 2273
Author(s):  
Rahmawati Erma Standsyah ◽  
Bambang Widjanarko Otok ◽  
Agus Suharsono
Keyword(s):  
Structural Equation Modeling ◽  
Method Of Moments ◽  
Structural Equation ◽  
Generalized Method Of Moments ◽  
Estimation Method ◽  
Generalized Least Squares ◽  
Fixed Effect ◽  
Least Square ◽  
Estimation Methods ◽  
Equation Modeling

The fixed effect meta-analytic structural equation modeling (MASEM) model assumes that the population effect is homogeneous across studies. It was first developed analytically using Generalized Least Squares (GLS) and computationally using Weighted Least Square (WLS) methods. The MASEM fixed effect was not estimated analytically using the estimation method based on moment. One of the classic estimation methods based on moment is the Generalized Method of Moments (GMM), whereas GMM can possibly estimate the data whose studies has parameter uncertainty problems, it also has a high accuracy on data heterogeneity. Therefore, this study estimates the fixed effect MASEM model using GMM. The symmetry of this research is based on the proof goodness of the estimator and the performance that it is analytical and numerical. The estimation results were proven to be the goodness of the estimator, unbiased and consistent. To show the performance of the obtained estimator, a comparison was carried out on the same data as the MASEM using GLS. The results show that the estimation of MASEM using GMM yields the SE value in each coefficient is smaller than the estimation of MASEM using GLS. Interactive GMM for the determination of the optimal weight on GMM in this study gave better results and therefore needs to be developed in order to obtain a Random Model MASEM estimator using GMM that is much more reliable and accurate in performance.

Meta-Accuracy and Perceived Reciprocity From the Perception-Meta-Perception Social Relations Model

2019 ◽  
Vol 50 (1) ◽  
pp. 24-37
Author(s):  
Ben Porter ◽  
Camilla S. Øverup ◽  
Julie A. Brunson ◽  
Paras D. Mehta
Keyword(s):  
Structural Equation Modeling ◽  
Social Relations ◽  
Latent Variables ◽  
Structural Equation ◽  
Multilevel Structural Equation Modeling ◽  
Equation Modeling ◽  
Complete Understanding ◽  
Social Relations Model ◽  
Flexible Framework ◽  
Multilevel Structural Equation

Abstract. Meta-accuracy and perceptions of reciprocity can be measured by covariances between latent variables in two social relations models examining perception and meta-perception. We propose a single unified model called the Perception-Meta-Perception Social Relations Model (PM-SRM). This model simultaneously estimates all possible parameters to provide a more complete understanding of the relationships between perception and meta-perception. We describe the components of the PM-SRM and present two pedagogical examples with code, openly available on https://osf.io/4ag5m . Using a new package in R (xxM), we estimated the model using multilevel structural equation modeling which provides an approachable and flexible framework for evaluating the PM-SRM. Further, we discuss possible expansions to the PM-SRM which can explore novel and exciting hypotheses.

Building a Model of Family Caregiving for Children with Emotional Disorders

1997 ◽  
Vol 5 (3) ◽  
pp. 138-148 ◽  
Author(s):  
Thomas P. Mcdonald ◽  
Thomas K. Gregoire ◽  
John Poertner ◽  
Theresa J. Early
Keyword(s):  
Structural Equation Modeling ◽  
Emotional Disorders ◽  
Problem Behaviors ◽  
Family Caregiving ◽  
Latent Variables ◽  
Structural Equation ◽  
Empirical Literature ◽  
Equation Modeling ◽  
Special Challenge ◽  
And Coping

In this article we describe the results of an ongoing effort to better understand the caregiving process in families of children with severe emotional problems. We make two assumptions. First, we assume that these families are essentially like other families but are faced with a special challenge in raising and caring for their special children while at the same time performing the multiple tasks and demands faced by all families. Second, we assume that public policy and programs must be supportive of the care of these children in their own homes and communities whenever possible. The purpose of this article is to present a model of family caregiving that draws broadly from available theory and empirical literature in multiple fields and to subject this model to empirical testing. We use structural equation modeling with latent variables to estimate an empirical model based on the theoretical model. Results of the model testing point to the importance of the child's external problem behaviors and the family's socioeconomic status and coping strategies as determinants of caregiver stress. Other findings highlight difficulties in measuring and modeling the complex mediating process, which includes formal and informal supports, perceptions, and coping behaviors. The use of structural equation modeling can benefit our efforts to support families by making explicit our theories about the important dimensions of this process and the relationship between these dimensions, which can then be subjected to measurement and validation.

scholarly journals Structural Equation Modeling for the Effect of Main Factors on Abortion Issue

2019 ◽  
Vol 7 (1) ◽  
pp. 1-13
Author(s):  
Aras Jalal Mhamad ◽  
Renas Abubaker Ahmed
Keyword(s):  
Structural Equation Modeling ◽  
Latent Variables ◽  
Latent Variable ◽  
Structural Equation ◽  
Close Contact ◽  
Maternity Hospital ◽  
Equation Modeling ◽  
Research Model ◽  
Statistical Tools ◽  
Main Factors

       Based on medical exchange and medical information processing theories with statistical tools, our study proposes and tests a research model that investigates main factors behind abortion issue. Data were collected from the survey of Maternity hospital in Sulaimani, Kurdistan-Iraq. Structural Equation Modelling (SEM) is a powerful technique as it estimates the causal relationship between more than one dependent variable and many independent variables, which is ability to incorporate quantitative and qualitative data, and it shows how all latent variables are related to each other. The dependent latent variable in SEM which have one-way arrows pointing to them is called endogenous variable while others are exogenous variables. The structural equation modeling results reveal is underlying mechanism through which statistical tools, as relationship between factors; previous disease information, food and drug information, patient address, mother’s information, abortion information, which are caused abortion problem. Simply stated, the empirical data support the study hypothesis and the research model we have proposed is viable. The data of the study were obtained from a survey of Maternity hospital in Sulaimani, Kurdistan-Iraq, which is in close contact with patients for long periods, and it is number one area for pregnant women to obtain information about the abortion issue. The results shows arrangement about factors effectiveness as mentioned at section five of the study. This gives the conclusion that abortion problem must be more concern than the other pregnancy problem.

Sign in / Sign up

Export Citation Format

Share Document