Background: The p-factor is thought to cause a positive manifold in psychopathology data, and many researchers presume that it is a substantive mechanism as opposed to a methodological artifact. Limited research suggests that including completely undiagnosed cases (i.e., cases without a single diagnosis) affects the dimensionality of psychological constructs, so we examined whether empirical support for the p-factor arises with their inclusion in structural models of psychopathology.Methods: We drew on data from three large, nationally representative samples of US community members, the: National Epidemiologic Study on Alcohol and Related Conditions Wave 1 (N=43,093), National Comorbidity Survey (N=5,877), and National Comorbidity Survey- Replication (N=5,692). We systematically culled undiagnosed cases from the data in increments of 10% and re-estimated structural models of psychopathology.Results: The correlation between latent externalizing and internalizing factors varied considerably as a function of the proportion of undiagnosed cases, with the correlation dropping to zero or even slightly negative when all undiagnosed cases were excluded. Also, as undiagnosed cases were removed, general factors of psychopathology explained less variance in psychopathology and weakened in terms of the extent to which it was well-represented by its indicators.Conclusions: Including undiagnosed cases in structural models of psychopathology induces a homogeneous, unidimensional structure. Ultimately, our findings raise questions about the nature of the p-factor, including whether it reflects a methodological artifact or arises due to the inclusion of cases with absence of diagnosed psychopathology.

We used multitrait-multimethod (MTMM) modeling to examine general factors of psychopathology in three samples of youth (ns = 2119, 303, 592) for whom three informants reported on the youth’s psychopathology (e.g., child, parent, teacher). Empirical support for the p-factor diminished in multi-informant models compared with mono-informant models: the correlation between externalizing and internalizing factors decreased and the general factor in bifactor models essentially reflected externalizing. Widely used MTMM-informed approaches for modeling multi-informant data cannot distinguish between competing interpretations of the patterns of effects we observed, including that the p-factor reflects, in part, evaluative consistency bias or that psychopathology manifests differently across contexts (e.g., home vs. school). Ultimately, support for the p-factor may be stronger in mono-informant designs, although it is does not entirely vanish in multi-informant models. Instead, the general factor of psychopathology in any given mono-informant model likely reflects a complex mix of variances, some substantive and some methodological.