Abstract Background Recreational cannabis legalization marked a significant policy shift in Canada, but has been difficult to evaluate because of the absence of a control group. Although it is unfeasible to evaluate legalization using a randomized controlled trial design, sophisticated statistical techniques can employ quasi-experimental designs using natural experiments. This study evaluates the impact of cannabis legalization in a longitudinal cohort of Canadian emerging adults by comparing changes in cannabis use frequency and related consequences over time to changes in a similar cohort in a United States jurisdiction where no policy change took place. Methods Two samples of emerging adults from Hamilton, Ontario, and...