Disentangling the Measurement Structure of Psychological Flexibility (PF): A Latent Structure Analysis of Four PF Measures

Psychological flexibility (PF) is a prominent construct in modern clinical psychological research, best known for its central role in Acceptance and Commitment Therapy (ACT) and its empirical associations with various health outcomes. However, the PF literature largely relies on the use of different self-report measures that appear to be based off varying definitions and conceptualisations of PF. To investigate this further, the current study jointly examined the latent construct of four prominent PF measures: CompACT, PPFI, Psy-flex and MPFI. The low-order and high-order measurement structure was evaluated through a series of exploratory factor analyses (EFA) and confirmatory factory analyses (CFA). EFA and CFA was conducted using item level data for each measure collected from a student and prolific sample (n = 1210). The hierarchical structure was examined using CFA to test whether the four PF measures examined a construct that aligns with existing conceptualisations of PF. Two theoretical structures were tested, where (1) low-order PF domains load onto a single superordinate global PF construct, and (2) low-order PF domains load onto one of two superordinate PF constructs (trait-level or state-level PF). Results did not support either theoretical model. Instead, both EFA and CFA revealed that the CompACT, PPFI, Psy-flex and MPFI are best explained by a measurement structure with nine low-order domains. These findings suggest that collectively these measures may assess a broader construct than original conceptualisations of PF. This supports a lack of coherence across common PF measurements, likely stemming from conceptual ambiguity. These findings echo concerns regarding PF and the credibility of the current research base. Further research is required to clarify what construct existing measures are assessing. In the meantime, researchers and clinicians are advised to exercise caution in their selection and interpretation of PF measures.