Dr. Volkow speaking: The survey studies teenagers that are in school 8th, 10th, and 12th grade. And the survey is for approximately 44,000 kids through 360 schools across all of the states in the United States. So, we are looking at asking questions about their patterns of drug use but also very much about their attitudes towards drug use and this is done on a yearly basis. So, what are the limitations of the survey? It is a large survey so it's actually quite representative, but it is representative of teenagers that are in school, but what we know is that teenagers that engage in heavy drug taking are at much higher risk of dropping out and so it has been estimated for example when we are addressing the rate of use of opioid drugs which are there of such great concern nowadays that the rates that we would observe in teenagers if we were to actually survey those that have dropped out maybe 1.7 times higher than those that we are recording among teenagers in schools. And another limitation in the survey is that these are cross-sectional in that you survey one kid one year, but you do not know what in that kid may be the patterns and that would be incredibly valuable because you will be able to monitor if you initiate a given drug whether it that continues to go into increasing levels into other drugs, or whether it stops. When you are relying on self-report there's always an uncertainty but the way that we actually compensate for that is that we evaluate not just the prevalence estimates that we're getting through Monitoring the Future, but we also evaluate them against other surveys such as the one from SAMHSA and or the one from the CDC. And what's interesting is that while the absolute exact… the numbers are not exactly the same, they are very consistent and certainly the trends go in parallel so they are they are similar, which leads us to think basically, that in fact there is robustness to the data that is of turn with Monitoring the Future.