Giving G a Meaning: An Application of the Bifactor-(S-1) Approach to Realize a More Symptom-Oriented Modeling of the Beck Depression Inventory–II

Abstract

The Beck Depression Inventory–II is one of the most frequently used scales to assess depressive burden. Despite many psychometric evaluations, its factor structure is still a topic of debate. An increasing number of articles using fully symmetrical bifactor models have been published recently. However, they all produce anomalous results, which lead to psychometric and interpretational difficulties. To avoid anomalous results, the bifactor-(S-1) approach has recently been proposed as alternative for fitting bifactor structures. The current article compares the applicability of fully symmetrical bifactor models and symptom-oriented bifactor-(S-1) and first-order confirmatory factor analysis models in a large clinical sample ( N = 3,279) of adults. The results suggest that bifactor-(S-1) models are preferable when bifactor structures are of interest, since they reduce problematic results observed in fully symmetrical bifactor models and give the G factor an unambiguous meaning. Otherwise, symptom-oriented first-order confirmatory factor analysis models present a reasonable alternative.

Factor structure and gender invariance of the Beck Depression Inventory – second edition (BDI-II) in a community-dwelling sample of adults

“…It is also worth noting that although the interpretation of the total score is based on the relevance of the bifactor structure, it cannot be stated that the BDI-II has only a unidimensional structure. Our results showed that the bifactor modelswhich are multidimensionalwere superior to the unidimensional model (Model I), which has been consistently found in similar research with the BDI-II (Brouwer et al, 2013;Heinrich et al, 2018;McElroy et al, 2018;Subica et al, 2014).…”

“…We also analyzed the MI by gender of this model, which was non-invariant. Conventional bifactor models have been considered to have better factor structures in CFA investigations of the BDI-II compared to unidimensional, correlated factors or high order factors (Brouwer et al, 2013;Dere et al, 2015;García-Batista et al, 2018;Heinrich et al, 2018;Subica et al, 2014). We followed an innovative procedure proposed by (Eid et al, 2017) aiming to estimate more reliable parameters for the BDI-II bifactor structure.…”

“…Because of its remarkable impact on health, valid and reliable psychometric More recently, bifactor measurement models have become a popular alternative for testing the structure of the BDI-II. Bifactor analysis tests the hypothesis of a general factor (G-factor) that explains most of the common variance extracted simultaneously from the items under an orthogonal design and its specific factors (García-Batista, Guerra-Peña, Cano-Vindel, Herrera-Martínez, & Medrano, 2018;Heinrich, Zagorscak, Eid, & Knaevelsrud, 2018). Most investigations perform bifactor analysis from two-and three-factor BDI-II models, which means the researchers only added a general factor to the n-factor model.…”

Appropriate Use of Bifactor Analysis in Psychopathology Research: Appreciating Benefits and Limitations

“…Group factors are specified to be orthogonal (uncorrelated) to the general factor, so group factors reflect common variance among indicator subsets that is separable from the general factor. 2 2 Several variations on the bifactor model exist, including a bifactor model with correlated group factors (7) and the S − 1 model (8,9). Correlating group factors, however, changes the interpretation of the latent variables.…”

Alcohol‐Related, Drug‐Related, and Non–Substance‐Related Aggression: 3 Facets of a Single Construct or 3 Distinct Constructs?

“…ARA, alcohol-related aggression; DRA, drug-related aggression; NSRA, non-substance-related aggression; G, general aggression factor. S factor unmodeled, which allows that unmodeled S factor to conceptually define the G factor (Heinrich et al, 2018).…”