This paper attempts to update the analysis that utilizes a regression discontinuity design to examine the effect of increased availability of legal alcohol at age 21, caused by the minimum legal drinking age in the United States, on the consumption of marijuana (Crost and Guerrero, 2012). Unfortunately, results from public access data from the NSDUH show no significant effect of increased access to legal alcohol on marijuana consumption, unlike the analysis performed by Crost and Guerrero (2012) that found a decrease in marijuana consumption. The differences in the results between the two analyses are likely due to a less precise measure of age in the public access data compared to the private data utilized in the previous analysis.

Read the full paper here.