Riskier Tests of the Validity of the Bifactor Model of Psychopathology

Abstract

We advanced several “riskier tests” of the validity of bifactor models of psychopathology, which included that the general and specific psychopathology factors should be reliable and well represented by their respective indicators and that including a general factor should improve on the correlated factor model’s external validity. We compared bifactor and correlated factors models of psychopathology using data from a community sample of youth ( N = 2,498) whose parents provided ratings on psychopathology and theoretically relevant external criteria (i.e., personality, aggression, antisociality). Bifactor models tended to yield either general or specific factors that were unstable and difficult to interpret. The general factor appeared to reflect a differentially weighted amalgam of psychopathology rather than a liability for psychopathology broadly construed. With rare exceptions, bifactor models did not explain additional variance in first-order psychopathology symptom dimensions or external criteria compared with correlated factors models. Together, our findings call into question the validity of bifactor models of psychopathology and the p factor more broadly.

Criterion Validity and Relationships between Alternative Hierarchical Dimensional Models of General and Specific Psychopathology

“…In contrast, because all factors are orthogonal in bifactor models, one can simultaneously regress measured criterion variables on the general and specific factors to determine if each of these factors independently accounts for unique variance in that variable. Such advantages of bifactor models for discovering the correlates of psychopathology at each level of the hierarchy are not in dispute (Bornovalova et al, 2020;Watts, Poore, & Waldman, 2019). What is in dispute, however, is whether the psychometric properties of all factors in bifactor models are sufficient to realize its logical advantages (Conway, Mansolf, & Reise, 2019;Watts et al, 2019).…”

Multi-Faceted Constructs in Abnormal Psychology: Implications of the Bifactor S - 1 Model for Individual Clinical Assessment

“…A prominent example in abnormal psychology is the general factor of psychopathology (p factor; Caspi and Moffitt 2018;Lahey et al 2012). After reviewing the research on the p factor of psychopathology over the last eight years, Watts et al (2019) concluded that "the precise nature of the p factor is not yet understood" (p. 1285). It is only recently that systematic analyses of applications of the bifactor model revealed that many applications are affected by serious problems (e.g., negative variances of specific factors, vanishing specific factors, irregular loadings patterns; see, for example, Eid et al 2017).…”

“…Moreover, the results of these applications are typically not in line with the theoretical expectations. For example, Watts et al (2019) convincingly stated that from a theoretical point of view all observed variables of a bifactor model should have relatively equal loadings on the common general factor and each facet should be represented by a specific factor. This structure, however, is often not supported (Eid et al 2017;Watts et al 2019).…”

“…For example, Watts et al (2019) convincingly stated that from a theoretical point of view all observed variables of a bifactor model should have relatively equal loadings on the common general factor and each facet should be represented by a specific factor. This structure, however, is often not supported (Eid et al 2017;Watts et al 2019). The many estimation problems and the theoretically not expected results found in many applications show that the bifactor model is not a reasonable model for analyzing multi-faceted constructs in clinical psychology (Heinrich et al 2018;Markon 2019;Watts et al 2019).…”

“…This structure, however, is often not supported (Eid et al 2017;Watts et al 2019). The many estimation problems and the theoretically not expected results found in many applications show that the bifactor model is not a reasonable model for analyzing multi-faceted constructs in clinical psychology (Heinrich et al 2018;Markon 2019;Watts et al 2019). A major reason for the unsatisfactory results in clinical psychology is that the facets of clinical symptoms typically are not interchangeable as it is required for a psychometrically valid application of a bifactor model (Eid et al 2017).…”