A man smokes a marijuana joint at a party celebrating weed Wednesday, April 20, 2016, in Seattle. (Elaine Thompson/AP)

0.32 grams.

That's how much marijuana is in the typical American joint, according to a rigorous statistical analysis recently published in the journal Drug and Alcohol Dependence by drug policy researchers Greg Ridgeway of the University of Pennsylvania and Beau Kilmer of the Rand Corp.

The amount of weed in the typical joint is actually an important quantity for researchers and policymakers to know. Marijuana is often sold by the joint in both legal and illegal markets. If you want to know how much weed people are consuming — which is highly important for things such as tax projections and public health studies — you need to know how much weed is in those joints.

In the past, researchers have relied on some fairly comical methods to figure out how much weed was in the typical joint. In one study, scientists had marijuana users measure out quantities of oregano that they felt were equivalent to a typical joint, but that didn't work — the estimates were all over the place, and "the relative density of oregano to marijuana is unknown," as Ridgeway and Kilmer dryly note.

Other studies have asked marijuana users to assess the size of their typical joint by comparing it to the size of common household items such as coins and rulers. But those don't seem to be terribly accurate either, according to Ridgeway and Kilmer.

So Ridgeway and Kilmer took a different approach for the current paper. They gathered high-quality data on over 10,000 people arrested for marijuana possession in various U.S. cities between 2000 and 2010. This data came from a now-defunct federal program known as ADAM, which interviewed arrestees in jail about their drug use. It contains detailed information on the quantities of marijuana people were caught with, as well as the price they reported paying for it.

"Some arrestees report the weight of loose marijuana purchased and the purchase price, while other arrestees report the number of joints purchased and the purchase price," Ridgeway and Kilmer write (emphasis added). You can mash these two parts of the dataset together, correcting for things such as different prices in different cities, bulk discounts, prices changes over time and whatnot.

I'm oversimplifying quite a bit here, but when you crunch all the numbers together you get a robust estimate of the weight of marijuana in the typical joint. That weight is 0.32 grams, plus or minus 0.03 grams.

If that number's accurate, it underscores just how cheap it is to get stoned on weed, relative to the cost of intoxication via other means such as alcohol. Highly potent weed — 18 percent THC content — is being sold legally in Washington at about $95 dollars an ounce. There are 28 grams in an ounce. So that ounce could be turned into roughly 84 joints, each enough to "get three naïve users wrecked out of their gourds (if you’ll allow me the use of technical terminology) for about three hours each," according to drug policy expert Mark Kleiman.

That 0.32 gram figure is actually kind of surprising, as it turns out. Many marijuana users who care to measure out their quantities will tell you that their joints are two or three times that size. A survey of nearly 3,000 High Times readers suggested that the typical joint size was 0.75 grams, with nearly 3 in 10 respondents saying their joints weighed a full gram or more.

Why the discrepancy? A couple  of suggestions seem plausible. There could be something of a "Cheech and Chong" effect, when marijuana users are tempted to overstate the size of their joints in the same way, say, fishermen tell tall tales of their latest catch.

On the other hand, it's also quite likely that the people who read High Times and respond to online surveys about marijuana use are a distinctly different set of people than those who get arrested on the street for marijuana possession.

At any rate, Ridgeway and Kilmer's analysis make it possible for policymakers and other researchers to derive more accurate estimates of marijuana consumption across the United States.

"These estimates can be incorporated into drug policy discussions to produce better understanding about illicit marijuana markets, the size of potential legalized marijuana markets, and health and behavior outcomes," they conclude.