“How Well Does Your Structural Equation Model Fit Your Data?”: Is Marcoulides and Yuan’s Equivalence Test the Answer?

Peugh, James; Feldon, David F.

doi:10.1187/cbe.20-01-0016

Cited by 73 publications

(58 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Previous researches pointed out model fit indices can be affected by sample size [ 45 ], degree of freedom (df) [ 46 ] and numbers of variables analyzed [ 47 ]. The denominator of the formula for RMSEA calculation contains both sample size and model df, which means the RMSEA value in complex model with high df estimated with large sample size can be decreased [ 48 ]. Accordingly, more participants can be recruited to calculate the model fit indices again especially RMSEA value in future studies.…”

Section: Discussionmentioning

confidence: 99%

Evaluating the validity and reliability of the Chinese entrapment scale and the relationship to depression among men who have sex with men in Shanghai, China

Tsamlag

et al. 2021

BMC Psychiatry

View full text Add to dashboard Cite

Background Perception of entrapment can emerge when someone feels trapped in an aversive situation and incapable of escape. Depression is closely related to the construct of entrapment. In China, men who have sex with men (MSM) have a high prevalence of depression; therefore, a tool to evaluate entrapment in this population is needed. We evaluated the validity and reliability of the Chinese version of the entrapment scale (ES) and the relationship to depression among MSM in Shanghai, China. Methods We recruited 304 MSM from four districts in Shanghai, China. Participants completed health behavior questionnaires that included baseline information and psychological measurements such as the ES and Patient Health Questionnaire (PHQ-9). The sample was randomly divided into two groups for exploratory factor analysis (n = 143) and confirmatory factor analysis (n = 161). Criterion validity was tested to explore the correlation between the ES and PHQ-9 scores. The reliability of the ES was evaluated with internal consistency reliability (Cronbach’s α coefficient) and split-half reliability (Spearman-Brown coefficient). We performed hierarchical regression analysis to determine the variance explained of entrapment to predicting depressive symptoms after adjusting for sociodemographic factors. Finally, receiver operator characteristic curve analysis was performed to measure the optimal ES cut-off value for predicting depression. Results Factor analysis showed the ES had one principal component, and one-dimensional scale had more acceptable model fit indices than two-dimensional model. The correlation coefficient between the ES and PHQ-9 scores was 0.756 (P < 0.01). The Cronbach’s α coefficient was 0.970 and the Spearman-Brown coefficient was 0.976. ES scores significantly predicted an additional 45.1% of depressive symptoms after controlling for sociodemographic characteristics in the MSM population (β = 0.689, P < 0.001). The optimum cut-off value was 23, which had a sensitivity of 70% and a specificity of 85.4%. Conclusions The Chinese version of the ES has good validity and reliability in the MSM population in Shanghai, and can be used to evaluate perception of entrapment among MSM. The findings confirmed an association between entrapment and depression.

show abstract

Section: Discussionmentioning

confidence: 99%

Evaluating the validity and reliability of the Chinese entrapment scale and the relationship to depression among men who have sex with men in Shanghai, China

Tsamlag

et al. 2021

BMC Psychiatry

View full text Add to dashboard Cite

show abstract

“…Model fit is classically determined by a nonsignificant χ 2 value. Yet, the χ 2 test is sample size dependent in that smaller differences are more likely to detect with larger sample sizes, whereas larger differences are less likely to detect with smaller sample sizes (Peugh & Feldon, 2020; Tucker & Lewis, 1973). Due to the issue associated with the χ 2 test, we adopted the following recommendation regarding multigroup invariance; a more parsimonious model is supported if ΔCFI < .01 and if ΔRMSEA < .015 (Chen, 2007; Cheung & Rensvold, 2002; Lang et al, 2011).…”

Section: Methodsmentioning

confidence: 99%

Math homework purpose scale: Confirming the factor structure with high school students

2021

Psychology in the Schools

View full text Add to dashboard Cite

The present investigation aims to test the validity of scores on the math homework purpose scale (MHPS) for high school students. Factorial results found that the MHPS contained three subscales: academic purpose, self‐regulatory purpose, and approval‐seeking purpose. Furthermore, the MHPS had a sufficient level of measurement and latent factor mean invariance across gender and grade level. Additionally, the MHPS had adequate coefficient alpha and omega coefficients. Regarding its validity evidence, consistent with theoretical predictions, the MHPS was associated positively with homework effort and completion. In summary, our investigation presents strong evidence showing that the MHPS is a valid tool for use with high school students.

show abstract

“…Statistical significance of the model parameter estimates (e.g., regression slopes) were examined following acceptable model fit (Thompson, 2004). Some researchers have recently argued that a nonsignificant χ 2 statistic and the model fit criteria suggested by Hu and Bentler (1999) have several limitations for assessing model fit (for a detailed discussion of both benefits and limitation, see Yuan et al, 2016;Marcoulides and Yuan, 2017;Peugh and Feldon, 2020). Therefore, we additionally evaluated model fit using equivalence testing outlined by Marcoulides and Yuan (2017) and Peugh and Feldon (2020).…”

Section: Data Analytic Plan Model Fit and Statistical Assumptionsmentioning

confidence: 99%

“…Some researchers have recently argued that a nonsignificant χ 2 statistic and the model fit criteria suggested by Hu and Bentler (1999) have several limitations for assessing model fit (for a detailed discussion of both benefits and limitation, see Yuan et al, 2016;Marcoulides and Yuan, 2017;Peugh and Feldon, 2020). Therefore, we additionally evaluated model fit using equivalence testing outlined by Marcoulides and Yuan (2017) and Peugh and Feldon (2020). Equivalence testing estimates a "T-size" CFI and RMSEA denoting the size of the misspecification for these two statistics.…”

Section: Data Analytic Plan Model Fit and Statistical Assumptionsmentioning

confidence: 99%

Content and Quality of Science Training Programs Matter: Longitudinal Study of the Biology Scholars Program

Estrada

Young

Flores

et al. 2021

LSE

View full text Add to dashboard Cite

show abstract

“How Well Does Your Structural Equation Model Fit Your Data?”: Is Marcoulides and Yuan’s Equivalence Test the Answer?

Cited by 73 publications

References 43 publications

Evaluating the validity and reliability of the Chinese entrapment scale and the relationship to depression among men who have sex with men in Shanghai, China

Evaluating the validity and reliability of the Chinese entrapment scale and the relationship to depression among men who have sex with men in Shanghai, China

Math homework purpose scale: Confirming the factor structure with high school students

Content and Quality of Science Training Programs Matter: Longitudinal Study of the Biology Scholars Program

Contact Info

Product

Resources

About