Don’t Use a Bifactor Model Unless You Believe the True Structure Is Bifactor

Decker, Scott L.

doi:10.1177/0734282920977718

Cited by 20 publications

(18 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This includes the use of estimation methods that account for the ordinal and often non‐normal nature of HADS data (Finney & DiStefano, 2006; Li, 2016; Shi et al, 2020). When comparing the latent structure of the HADS, most studies have focused on fit statistics; however, consideration of the theoretical and practical suitability of the model needs to be a priority (Decker, 2021). This is particularly pertinent when investigating bifactor models; in these cases, the presence, strength and reliability of the general factor needs to be considered (Bornovalova et al, 2020; Decker, 2021).…”

Section: Discussionmentioning

confidence: 99%

The structure of the Hospital Anxiety and Depression Scale: Theoretical and methodological considerations

Lloyd

Sugden

Thomas

et al. 2023

British J of Psychology

View full text Add to dashboard Cite

The Hospital Anxiety and Depression Scale (HADS; Zigmond ‐ Snaith, 1983) is widely used; however, its factor structure is unclear, with studies reporting differing unidimensional, two‐factor and three‐factor models. We aimed to address some key theoretical and methodological issues contributing to inconsistencies in HADS structures across samples. We reviewed existing HADS models and compared their fit using confirmatory factor analysis (CFA). We also investigated methodological effects by comparing factor structures derived from Rasch and Principal Components Analysis (PCA) methods, as well as effects of a negative wording factor. An Australian community‐dwelling sample consisting of 189 females and 158 males aged 17–86 (M = 35.73, SD = 17.41) completed the 14‐item HADS. The Rasch Analysis, PCA and CFA all supported the original two‐factor structure. Although some three‐factor models had good fit, they had unacceptable reliability. In the CFA, a hierarchical bifactor model with a general distress factor and uncorrelated depression and anxiety subscales produced the best fit, but the general factor was not unidimensional. The addition of a negative wording factor improved model fit. These findings highlight the effects of differing methodologies in producing inconsistent HADS factor structures across studies. Further replication of model fit across samples and refinement of the HADS items is warranted.

show abstract

Section: Discussionmentioning

confidence: 99%

The structure of the Hospital Anxiety and Depression Scale: Theoretical and methodological considerations

Lloyd

Sugden

Thomas

et al. 2023

British J of Psychology

View full text Add to dashboard Cite

show abstract

“…Orthogonalized bifactor applications are best suited for item-level analysis to assess what indicators have in common, while higher-order models assess what latent factors have in common (Decker, 2021) If a higher-order model's second-order factor is only meaningful in the context of their associations with lower-order factors, a bifactor model should be avoided in nested model comparisons. How interpretable is my latent factor model?…”

Section: Table 1 Comparable Aspects Of Model Qualitymentioning

confidence: 99%

Model fit is a fallible indicator of model quality in quantitative psychopathology research: A reply to Bader and Moshagen.

Greene¹,

Eaton²,

Forbes³

et al. 2022

Journal of Psychopathology and Clinical Science

View full text Add to dashboard Cite

As evidenced by our exchange with Bader and Moshagen (2022), the degree to which model fit indices can and should be used for the purpose of model selection remains a contentious topic. Here, we make three core points. First, we discuss the common misconception about fit statistics' abilities to identify the "best model," arguing that mechanical application of model fit indices contributes to faulty inferences in the field of quantitative psychopathology. We illustrate the consequences of this practice through examples in the literature. Second, we highlight the parsimony-adjacent concept of fitting propensity, which is not accounted for by commonly used fit statistics. Finally, we present specific strategies to overcome interpretative bias and increase generalizability of study results and stress the importance of carefully balancing substantive and statistical criteria in model selection scenarios.

show abstract

“…A case could be made that bi-factor models (Murray & Johnson, 2013) would provide a more equitable distribution of variance between SCA and g indexed as a latent variable representing what is in common among SCA. However, the use of bifactor models is not straightforward (Decker, 2021). Moreover, phenotypic and genomic analyses of SCA.g are likely to use regression-derived SCA.g scores because bifactor models, like Cholesky models, involve latent variables that cannot be converted to phenotypic scores for SCA.g.…”

Section: Discussionmentioning

confidence: 99%

The genetics of specific cognitive abilities

Procopio

Zhou

Wang

et al. 2022

Preprint

View full text Add to dashboard Cite

Most research on individual differences in performance on tests of cognitive ability focuses on general cognitive ability (g), the highest level in the three-level Cattell-Horn-Carroll (CHC) hierarchical model of intelligence. About 50% of the variance of g is due to inherited DNA differences (heritability) which increases across development. Much less is known about the genetics of the middle level of the CHC model, which includes 16 broad factors such as fluid reasoning, processing speed, and quantitative knowledge. We provide a meta-analytic review of 863,041 monozygotic-dizygotic twin comparisons from 80 publications for these middle-level factors, which we refer to as specific cognitive abilities (SCA). Twin comparisons were available for 11 of the 16 CHC domains. The average heritability across all SCA is 55%, similar to the heritability of g. However, there is substantial differential heritability and the SCA do not show the dramatic developmental increase in heritability seen for g. We also investigated SCA independent of g (g-corrected SCA, which we refer to as SCA.g). A surprising finding is that SCA.g remain substantially heritable (53% on average), even though 25% of the variance of SCA that covaries with g has been removed. Our review frames expectations for genomic research that will use polygenic scores to predict SCA and SCA.g. Genome-wide association studies of SCA.g are needed to create polygenic scores that can predict SCA profiles of cognitive abilities and disabilities independent of g. These could be used to foster children's cognitive strengths and minimise their weaknesses.

show abstract

Don’t Use a Bifactor Model Unless You Believe the True Structure Is Bifactor

Cited by 20 publications

References 29 publications

The structure of the Hospital Anxiety and Depression Scale: Theoretical and methodological considerations

The structure of the Hospital Anxiety and Depression Scale: Theoretical and methodological considerations

Model fit is a fallible indicator of model quality in quantitative psychopathology research: A reply to Bader and Moshagen.

The genetics of specific cognitive abilities

Contact Info

Product

Resources

About