Pairwise likelihood estimation for confirmatory factor analysis models with categorical variables and data that are missing at random

Katsikatsou, Myrsini; Moustaki, Irini; Jamil, Haziq

doi:10.1111/bmsp.12243

Cited by 4 publications

(3 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Subsequently, confirmatory factor analysis (CFA) was employed to further verify the construct validity of the COVID-19 BVS based on Sample 2 ( n = 539) ( 36 ). With reference to the recent simulation studies on the selection of CFA estimator, weighted least square mean and variance adjusted estimators (WLSMV) was used as the estimation method ( 38 , 39 ). We used several fit indices (and the respective cutoff criteria) to determine model fit: CFI > 0.950, RMSEA < 0.05, SRMR < 0.05 ( 40 – 43 ) as well as the χ 2 /df ≤ 3 ( 44 , 45 ).…”

Section: Methodsmentioning

confidence: 99%

Measurement of burnout during the prolonged pandemic in the Chinese zero-COVID context: COVID-19 burnout views scale

Lau

Pang

et al. 2022

Front. Public Health

View full text Add to dashboard Cite

Burnout is an important public health issue at times of the COVID-19 pandemic. Current measures which focus on work-based burnout have limitations in length and/or relevance. When stepping into the post-pandemic as a new Norm Era, the burnout scale for the general population is urgently needed to fill the gap. This study aimed to develop a COVID-19 Burnout Views Scale (COVID-19 BVS) to measure burnout views of the general public in a Chinese context and examine its psychometric properties. A multiphase approach including literature review, expert consultation, and pilot testing was adopted in developing the scale. The scale was administered to a sample of 1,078 of the general public in Hong Kong with an average age of 34.45 years (SD = 12.47). Exploratory and Confirmatory Factor Analyses suggested a 5-item unidimensional model of COVID-19 BVS. The CFA results indicated that the COVID-19 BVS had a good model fit, as χ2 (10.054)/5 = 2.01, SRMR = 0.010, CFI = 0.998, RMSEA = 0.031. Five items were maintained in EFA with high internal consistency in terms of Cronbach's α of 0.845 and McDonald's ω coefficient of 0.87, and the corrected item-to-total correlations of 0.512 to 0.789 are way above the acceptable range. The KMO values of 0.841 and Bartlett's Test of Sphericity (p < 0.01) verified the normal distribution of the EFA and the adequacy of the EFA sampling. The analyses suggest that the COVID-19 BVS is a promising tool for assessing burnout views on the impacts of the epidemic on the Chinese general populations.

show abstract

Section: Methodsmentioning

confidence: 99%

Measurement of burnout during the prolonged pandemic in the Chinese zero-COVID context: COVID-19 burnout views scale

Lau

Pang

et al. 2022

Front. Public Health

View full text Add to dashboard Cite

show abstract

“…In this section, we review pairwise (maximum) likelihood (PML) estimation for IRT models [14][15][16][17]. For item responses X i with IRFs P i (x, θ; γ i ) (i = 1, .…”

Section: Pairwise Likelihood Estimation (Pml)mentioning

confidence: 99%

Pairwise Likelihood Estimation of the 2PL Model with Locally Dependent Item Responses

Robitzsch

2024

Applied Sciences

View full text Add to dashboard Cite

The local independence assumption is crucial for the consistent estimation of item parameters in item response theory models. This article explores a pairwise likelihood estimation approach for the two-parameter logistic (2PL) model that treats the local dependence structure as a nuisance in the optimization function. Hence, item parameters can be consistently estimated without explicit modeling assumptions of the dependence structure. Two simulation studies demonstrate that the proposed pairwise likelihood estimation approach allows nearly unbiased and consistent item parameter estimation. Our proposed method performs similarly to the marginal maximum likelihood and pairwise likelihood estimation approaches, which also estimate the parameters for the local dependence structure.

show abstract

“…Barendse and Rosseel (2020) estimated SEM with the PML estimation method for a mixture of binary and continuous data, by estimating Pearson, tetrachoric, polychoric, and polyserial correlations. Standard errors and missing‐data procedures for the PML estimation method have been developed by Katsikatsou et al (2012) and Katsikatsou and Moustaki (2017).…”

Section: Introductionmentioning

confidence: 99%

Multilevel SEM with random slopes in discrete data using the pairwise maximum likelihood

Barendse

Rosseel

2023

Brit J Math & Statis

View full text Add to dashboard Cite

Pairwise maximum likelihood (PML) estimation is a promising method for multilevel models with discrete responses. Multilevel models take into account that units within a cluster tend to be more alike than units from different clusters. The pairwise likelihood is then obtained as the product of bivariate likelihoods for all within‐cluster pairs of units and items. In this study, we investigate the PML estimation method with computationally intensive multilevel random intercept and random slope structural equation models (SEM) in discrete data. In pursuing this, we first reconsidered the general ‘wide format’ (WF) approach for SEM models and then extend the WF approach with random slopes. In a small simulation study we the determine accuracy and efficiency of the PML estimation method by varying the sample size (250, 500, 1000, 2000), response scales (two‐point, four‐point), and data‐generating model (mediation model with three random slopes, factor model with one and two random slopes). Overall, results show that the PML estimation method is capable of estimating computationally intensive random intercept and random slopes multilevel models in the SEM framework with discrete data and many (six or more) latent variables with satisfactory accuracy and efficiency. However, the condition with 250 clusters combined with a two‐point response scale shows more bias.

show abstract

Pairwise likelihood estimation for confirmatory factor analysis models with categorical variables and data that are missing at random

Cited by 4 publications

References 22 publications

Measurement of burnout during the prolonged pandemic in the Chinese zero-COVID context: COVID-19 burnout views scale

Measurement of burnout during the prolonged pandemic in the Chinese zero-COVID context: COVID-19 burnout views scale

Pairwise Likelihood Estimation of the 2PL Model with Locally Dependent Item Responses

Multilevel SEM with random slopes in discrete data using the pairwise maximum likelihood

Contact Info

Product

Resources

About