The UPPS-P measures impulsivity as a five-factor construct (lack of premeditation, lack of perseverance, positive urgency, negative urgency and sensation seeking). Drawing on a number of theoretical considerations and alternative conceptions of impulsivity, the current study used confirmatory factor analysis (N = 1635) and multiple regression to evaluate and test alternative models comprising three, five, and a hierarchical model containing latent factors. The five factor and hierarchical models were shown to be valid and of near identical fit, whereas the three-factor model fit the data poorly. The current findings suggest that both the five factor and hierarchical models are useful applications of the UPPS-P. Depending on the purpose of future research, both models demonstrate utility in both risk assessment and treatment development. Multiple regression analysis revealed that positive urgency predicted problem gambling, which supports the predictive utility of impulsivity as a five-factor construct. While the latent factors of the hierarchical model are consistent with emerging theory, those using the UPPS-P should not overlook the unique contributions of the five factors. As the current study found meaningful predictive distinctions between positive and negative urgency, utilizing all five factors may increase measurement sensitivity and predictive utility.