Best Practices in the Use of Bifactor Models: Conceptual Grounds, Fit Indices and Complementary Indicators

Abstract

<p>Bifactor models have gained increasing popularity in the literature concerned with personality, psychopathology and assessment. Empirical studies using bifactor analysis generally judge the estimated model using SEM model fit indices, which may lead to erroneous interpretations and conclusions. To address this problem, several researchers have proposed multiple criteria to assess bifactor models, such as a) conceptual grounds, b) overall model fit indices, and c) specific bifactor model indicators. In this article, we provide a brief summary of these criteria. An example using data gathered from a recently published research article is also provided to show how taking into account all criteria, rather than solely SEM model fit indices, may prevent researchers from drawing wrong conclusions.</p>

Towards an Explanatory Model of Suicidal Ideation: The Effects of Cognitive Emotional Regulation Strategies, Affectivity and Hopelessness

“…The bifactor measurement model applied for the automatic ER strategies showed adequate indexes of adjustments (χ 2 = 580.89, df = 88, CFI = .96, TLI = .95, RMSEA = .05). Additional indices were also calculated to evaluate this model properly (Flores-Kanter, Dominguez-Lara, Trógolo, & Medrano, 2018). Among them the H index, an indicator of reliability, indicates how well a series of items represent the latent variable.…”

Testing two competitive models of empathic communication in cancer care encounters: A factorial analysis of the CARE measure

“…Flores‐Kanter, Dominguez‐Lara, Trógolo, and Medrano (2018) specify that it is recommended to use multiple criteria when assessing bifactor models. First, conceptual grounds are needed.…”

Mesleki̇ Karar Envanteri̇ Faktör Yapisina Ai̇t Farkli Ölçme Modelleri̇ni̇n Doğrulayici Faktör Anali̇zi̇ İle İncelenmesi̇

“…In the literature, it is stated that bi-factor model can mostly adapt better than second level model (Beaujean, Parkin ve Parker, 2014;Benson, Hulac ve Bernstein, 2013;Gignac, 2008Gignac, , 2016Gignac ve Watkins, 2013;Murray ve Johnson, 2013;Reise, Kim, Mansolf ve Widaman, 2016;Watkins, 2010). For this reason, in order to validate the measurement model, bi-factor model indicators should be taken into consideration in addition to model-fit indexes (Flores-Kanter et al, 2018). When the significance of t values of the models was examined, t values of some items were not found significant in the bi-factor measurement model.…”