Identification of Confirmatory Factor Analysis Models of Different Levels of Invariance for Ordered Categorical Outcomes

Wu, Hao; Estabrook, Ryne

doi:10.1007/s11336-016-9506-0

Cited by 325 publications

(350 citation statements)

References 36 publications

Supporting

Mentioning

342

Contrasting

Unclassified

Order By: Relevance

“…required normalisation is imposed on the variances of the error terms (Ψ) or on the diagonal of the covariance matrix of the measures (Σ). The MT parameterisation instead proceeds by identifying parameters in one group first, and then imposing cross-group equality constraints to identify parameters in other groups (Wu and Estabrook, 2016).…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Inequality in socio-emotional skills: a cross-cohort comparison

Attanasio¹,

Blundell²,

Conti³

et al. 2020

View full text Add to dashboard Cite

We examine changes in inequality in socio-emotional skills very early in life in two British cohorts born 30 years apart. We construct socio-emotional scales comparable across cohorts for both boys and girls, using two validated instruments for the measurement of child behaviour. We identify two dimensions of socio-emotional skills for each cohort: 'internalising' and 'externalising', related to the ability of children to focus their concentration and to engage in interpersonal activities, respectively. Using recent methodological advances in factor analysis, we establish comparability in the inequality of these early skills across cohorts, but not in their average level. We document for the first time that inequality in these early skills has increased across cohorts, especially for boys and at the bottom of the distribution. We also document changes in conditional skills gaps across cohorts. We find an increase in the socioemotional skills gap in the younger cohort for children born to mothers with higher socioeconomic status (education and employment), and to mothers who smoked during pregnancy. The increase in inequality in early socio-emotional skills is particularly pronounced for boys. On the other hand, we find a decline in the skills gradient for children without a father figure in the household. Lastly, we document that socio-emotional skills measured at a much earlier age than in most of the existing literature are significant predictors of outcomes both in adolescence and adulthood, in particular health and health behaviours. Our results show the importance of formally testing comparability of measurements to study skills differences across groups, and in general point to the role of inequalities in the early years for the accumulation of health and human capital across the life course.

show abstract

Section: Resultsmentioning

confidence: 99%

“…In the framework of factor analysis, measurement invariance is a formally testable property. In this paper, we follow the recent identification methodology by Wu and Estabrook (2016). The configural model defined in the previous section serves as the starting point.…”

Section: Measurement Invariancementioning

confidence: 99%

Inequality in socio-emotional skills: a cross-cohort comparison

Attanasio¹,

Blundell²,

Conti³

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Phase 4: Analysis of measurement invariance. An initial Multigroup CFA [49] was conducted to check if the same configural structure would hold for all sex, age, and educationbased groups-i.e., this was done to check whether configural invariance could be confirmed with the data at hand. The χ 2 , CFI and RMSEA and their previously described cut-off points were used to evaluate configural invariance.…”

Section: Discussionmentioning

confidence: 99%

Development and initial psychometric assessment of the race-related attitudes and multiculturalism scale in Australia

et al. 2020

View full text Add to dashboard Cite

AimThe present study aims to develop the Race-related Attitudes and Multiculturalism Scale (RRAMS), as well as to perform an initial psychometric assessment of this instrument in a national sample of Australian adults. MethodsThe sample comprised 2,714 Australian adults who took part in the 2013 National Dental Telephone Interview Survey (NDTIS), which includes a telephone-based interview and a follow-up postal questionnaire. We used Exploratory Factor Analysis (EFA) to evaluate the RRAMS' factorial structure (n = 271) and then proceeded with Confirmatory Factor Analysis (CFA) to confirm the proposed structure in an independent sample (n = 2,443). Measurement invariance was evaluated according to sex, age and educational attainment. Construct validity was assessed through known-groups comparisons. Internal consistency was assessed with McDonald's Ω H and ordinal α. Multiple imputation by chained equations was adopted to handle missing data. ResultsEFA indicated that, after excluding 4 out of the 12 items, a two-factor structure provided a good fit to the data. This configural structure was then confirmed in an independent sample by means of CFA (χ 2 (19) = 341.070, p<0.001, CFI = 0.974, RMSEA = 0.083; 90% CI [0.076, 0.091]). Measurement invariance analyses suggested that the RRAMS items can be used to compare men/women, respondents with/without tertiary education and young/older participants. The "Anglo-centric/Assimilationist attitudes" (Ω H = 0.83, α ORDINAL = 0.85) and "Inclusive/Pluralistic attitudes" subscales (Ω H = 0.77, α ORDINAL = 0.79) showed adequate PLOS ONE

show abstract

“…The analyses were run according to the guidelines provided by Chen and colleagues, Rudnev and colleagues, and Wu and Estabrook [26][27][28]. First, configural invariance was tested by imposing the model's threshold to be equal across groups.…”

Section: Measurement Invariancementioning

confidence: 99%

“…Weak -1 st ) and then (b) all the factor loadings (Weak -Full). Finally, strict invariance was first tested by applying equality constraints to the first-order factors' variances (the variances of the indicators were fixed to 1 in the configural model) [28].…”

Section: Measurement Invariancementioning

confidence: 99%

Psychometric Properties of the Montreal Cognitive Assessment (MoCA): A Comprehensive Investigation

Sala¹,

Inagaki²,

Ishioka³

et al. 2019

Preprint

View full text Add to dashboard Cite

BackgroundThe Montreal Cognitive Assessment (MoCA) is a common tool for the assessment of global cognition in older adults. Despite its popularity among clinicians and researchers, the test's psychometric properties are still uncertain. ObjectiveTo examine fundamental psychometrics properties of the MoCA that have not been established so far (i.e., factorial structure, general factor saturation, and measurement invariance). DesignCohort study. SettingPopulation-based invitation-type survey in city and rural areas in Hyogo prefectures and Tokyo Metropolitan, Japan. SubjectsIndividuals (N = 2,408) aged 69 to 91 clustered in three cohorts (69-71-year-olds, 79-81year-olds, and 89-91-year-olds). Methods Exploratory Factor Analysis and Confirmatory Factor Analysis. ResultsThe MoCA shows an overall stable hierarchical factorial structure and a satisfactory general factor saturation. Also, measurement invariance occurs across participants with different age, educational level, economic status, and gender. ConclusionThis comprehensive investigation upholds the idea that the MoCA is a psychometrically valid tool for the assessment of global cognition in older adults. Key points The Montreal Cognitive Assessment (MoCA) is a common test for evaluating global cognition in older adults. PSYCHOMETRICS OF THE MOCA The MoCA has never undergone a thorough analysis to test its psychometric properties.  The study evaluates the test's dimensionality, factorial structure, general factor saturation, and invariance.  The results show that the test's items perform well in all the analyses.  The MoCA is thus a reliable tool for assessing global cognition in older adults.

show abstract

Identification of Confirmatory Factor Analysis Models of Different Levels of Invariance for Ordered Categorical Outcomes

Cited by 325 publications

References 36 publications

Inequality in socio-emotional skills: a cross-cohort comparison

Inequality in socio-emotional skills: a cross-cohort comparison

Development and initial psychometric assessment of the race-related attitudes and multiculturalism scale in Australia

Psychometric Properties of the Montreal Cognitive Assessment (MoCA): A Comprehensive Investigation

Contact Info

Product

Resources

About