Background and purpose: The measurement of affect in the context of exercise has become a controversial issue. To help elucidate some of the problems, the conceptual and methodological bases of the Exercise-induced Feeling Inventory (Gauvin, L., & Rejeski, W.J. (1993). The Exercise-induced Feeling Inventory: Development and initial validation. Journal of Sport & Exercise Psychology, 15, 403.) are critiqued, emphasizing deviations from established scale development and validation guidelines.Methods: From a conceptual standpoint, the analysis concentrates on the definition of ''feeling states,'' the demarcation of the content domain, the decision to adopt a categorical conceptualization of affect, the notion of a universal phenomenology of exercise, and the notion of exercise-specific affect. From a methodological standpoint, emphasis is placed on the item selection and content validation methods, the exploratory structural analysis, and the application of structural equation modeling.Results and conclusions: Substantial deficiencies in conceptual groundwork and deviations from established guidelines are identified that may have important implications for the validity and utility of the EFI.