Alzheimer's disease (AD) is a progressive dementia caused by degeneration of the central nervous system with a high incidence in the elderly. Resveratrol is a natural compound contained in a wide range of plant species, including grapes. Recent studies have shown positive effects of resveratrol in animal models of AD; however, whether these results justify clinical trials is uncertain. Furthermore, there are multiple theories about the mechanism(s) by which resveratrol works and knowing how it works can suggest targets for future drug development. So far, systematic evaluation of both of these aspects is lacking. In this study, we selected 19 studies describing the efficacy of resveratrol in rodent AD models by electronic and manual retrieval. The method quality of the study were analyzed by the SYRCLE's risk of bias tool and the experimental data were retrieved and meta-analyzed using forest plot. Analysis of these studies demonstrates the consistent neuroprotective effects of resveratrol in AD models and offers insights into the possible pharmacological mechanisms. This information eliminates the bias of each study, providing supporting evidence for the implementation of clinical trials. However, the limits of studies were also noticed: low method quality, lack of sample size calculation and high risks of bias.