Numerous research studies have found that marijuana is on balance less of a public road hazard than alcohol. Various accident surveys have found that over half of fatal drivers have alcohol in their blood, as opposed to 7 - 20% with THC, the major psychoactive component of marijuana (a condition usually indicative of having smoked within the past 2 - 4 hours). However, the great majority (70% - 90%) of THC positive drivers also have alcohol in their blood. There is accordingly little evidence that marijuana use by itself is a major public safety hazard. The second NHTSA study, "Marijuana and Actual Driving Performance," concluded that the adverse effects of cannabis on driving appear "relatively small" and are less than those of drunken driving. The study, conducted in the Netherlands, examined the performance of drivers in actual freeway and urban driving situations at various doses of marijuana. It found that marijuana produces a moderate, dose related decrement in road tracking ability, but is "not profoundly impairing" and "in no way unusual compared to many medicinal drugs." It found that marijuana's effects at the higher doses preferred by smokers never exceed those of alcohol at blood concentrations of .08%, the minimum level for legal intoxication in stricter states such as California. The study found that unlike alcohol, which encourages risky driving, marijuana appears to produce greater caution, apparently because users are more aware of their state and able to compensate for it (similar results have been reported by other researchers).