Rumination is a maladaptive emotion regulation strategy linked to depression. An intriguing question is why some people tend to use it. The metacognitive model proposes that people having positive metacognitive beliefs (“ruminating helps me cope”) are more prone to engage in ruminative processes. On the other hand, negative metacognitive beliefs (“rumination is uncontrollable”) lead to negative appraisals about rumination, increasing depressive symptomatology. Two scales assess both of these core beliefs: the Positive and Negative Beliefs about Rumination Scales (PBRS and NBRS). While they have been widely used, few studies have examined their factor structure and psychometric properties. Preliminary support has also been provided for the metacognitive model of rumination and depression, but contrary results exist about the specific contributing negative beliefs. In this study, we aimed, first, to add new evidence of the factor structure and psychometric properties of the PBRS and NBRS in a non-clinical population and, second, to test the metacognitive model using structural equation modeling (SEM). We also add the Spanish version of these scales. The participants included 427 individuals from the general population. We confirmed the one-factor structure of the PBRS and found evidence for a three-factor structure of the NBRS (uncontrollability, harm and social consequences), differing from the two-factor structure found in previous studies. SEM analyses showed that positive beliefs were associated to rumination and negative beliefs about social consequences partially mediated the relations among rumination and depression. Our findings show new data about the validity of the two scales and suggest the utility of considering a three-factor model of the NBRS.