OBJECTIVES: For the past several decades, the state-wide legalization of medical marijuana and its effect has been a debated public health issue in the United States. Previous studies rarely controlled for selection biases that may be caused by a state’s political atmosphere as well as marijuana users’ political impetus. Using propensity-score matching to reduce selection bias, this study reexamined the effects of state-wide medical marijuana legalization on general marijuana use among adolescents and adults. METHODS: A retrospective cross-sectional study using the 2013 National Survey on Drug Use and Health was conducted. Adolescents aged 12-18 years (n=13,700) and adults (n =23,992) in the U.S. were extracted for analyses. Propensity score matchings were conducted to ensure the comparability between the treatment and control groups that included subjects in the states that legalized and did not legalize medical marijuana, respectively. After matchings, logistic regressions were performed to estimate the effect of medical marijuana legalization on general marijuana use, controlling for subjects’ sociodemographics. RESULTS: Among adolescents, after controlling for selection biases, adolescents who lived in a state that legalized medical marijuana were more likely to use marijuana (OR=1.74, p<.001), compared to those who lived in a state that medical marijuana was illegal. Among adults, those who lived in a state that medical marijuana was legalized were also more likely to use marijuana (OR=1.59, p<.001), compared to those who lived in a state that medical marijuana was illegal. CONCLUSIONS: This study affirmed that legalization of medical marijuana did increase general marijuana use among adults and adolescents in the U.S., even after controlling for selection biases. Researchers and policy makers should pay special attentions on the long-term effects of medical marijuana as well as its long-term costs and benefits when considering future legislation.