The authors investigate the empirical generalizability of Aaker and Keller's model of how consumers evaluate brand extensions. Various replications have reported different results. Using a comprehensive data set containing the data from the original study and seven replications conducted around the world, the authors undertake a secondary analysis to understand what generalizations emerge. The study has implications for the understanding of how brand extensions are evaluated and how empirical generalizations are made. For brand extensions, Aaker and Keller's model hypothesizes that evaluations of brand extensions are based on the quality of the original brand, the fit between the parent and extension categories, and the interaction of the two. The authors find support for this full model despite published results, including Aaker and Keller's own, that support only some of the hypotheses. The authors find evidence that the level of contribution of each of these components varies by brand and culture. With respect to empirical generalizations, the key implication is that it is premature to make firm conclusions about theory on the basis of only one study.