THE BLOG

An Apology to Kevin Sabet Over a Meaningless Marijuana Statistic

05/28/2013 01:39 pm ET | Updated Jul 28, 2013

Kevin Sabet asked for an apology over my recent post, Project SAM's Kevin Sabet Claims 39% Washington Teens Use Dispensary Pot, Reality is 9.4%. Basically, I accused him of adding up columns of percentages for 9th, 10th, 11th, and 12th graders to come up with a scary-sounding -- and completely inaccurate -- claim that "39% of HS students in Washington state report getting their marijuana from a 'medical' marijuana dispensary." (Kevin is required to put "medical" in scare quotes, despite thousands of years of human medical use, tens of thousands of published studies on the medical efficacy of cannabinoids, and closing in on 21 U.S. states that have recognized these scientific and historical facts. Because, you know, the FDA hasn't approved it. Like Vioxx.)

So here it is: Kevin, I apologize for questioning the means by which you came up with a completely irrelevant statistic. You obviously didn't add up percentages -- Oxford taught you better than that. You merely reported on Seattle Schools reporting of a phenomenon called "sharing."

Before we get too deep into the numbers, let's remember for a moment Kevin initially claimed it was "Washington state HS students" not just the teens in Seattle this report covers. And that for all the teens in Seattle, the figure is 9.4%. And for readers not savvy enough to understand "got their marijuana" means the 39% refers only to the subset that smoke marijuana, not all Washington State high school students, the initial tweet was fear-mongering and deception at their finest. And that within this same report, Seattle Schools writes, "current marijuana use rates in all grades have remained statistically unchanged."

Here is the detailed table from the Seattle Schools report:

2013-05-25-Q97Detail.png

Right off the bat, something jumped out at me. When did 1,185 become 75.4% of 1,608? I know Kevin and these researchers have doctorates, but my Excel tells me the figure is 73.69%, and at best you can round that to 73.7%, at least in public school.

Then I checked some of the other numbers. What happened to one of the 15-and-younger, 16 and 17, and 18 and older teens who didn't smoke pot, who add up to 1,184? What happened to 11 of the 9th, 10th, 11th, and 12th graders, who add up to 1,597?

Whatever. Let's play with the published numbers. Given those N = 1,608 figures above, let's just concentrate on the 423 teens who did use marijuana. My Excel tells me this is how those figures break down:

2013-05-25-Q97Excel.png

OK, so 25 out of 423 teens surveyed in Seattle usually got the marijuana they used from a dispensary. That's about 5.9%, assuming Seattle Schools doesn't make some of them disappear in the data. Now, for comparison's sake, I thought it would be interesting to look at the data for alcohol acquisitions by teens in the past thirty days, as also reported in the survey:

2013-05-25-Q46Detail.png

Now again, my Excel doesn't seem to agree exactly with Seattle Schools' numbers, but here I see 34 out of 445 teens who drank bought booze from stores, restaurants, and events (or as I'd call them, recreational alcohol dispensaries). That's over 7.6% compared to the 5.9% for marijuana dispensaries.

But then the Seattle Schools ask Q98, which changes the focus from "the marijuana they usually got came from a dispensary" to "used marijuana that came from a dispensary."

2013-05-25-Q98Detail.png

Again, it's weird that now 1,187 teens don't smoke pot, but let's run with it. In this data, we find that 160 of the 421 (not 423) who used marijuana in the past month say it came from a dispensary. That's 9.4% of all high school students in Seattle and a shocking 38% of the teen pot smokers using dispensary pot! Or as Seattle Schools has hastily added:

2013-05-25-QAddendum.png

Now, here's why this is a meaningless statistic. Sharing and lying.

Recall that 5.6% of our pot-smoking teens usually got his pot from a dispensary. Consider that pot smoking is an activity that happens in social circles. Given 423 pot smokers 25 of them "usually" get it from dispensaries, how big does your pot smoking circle have to be in order to have a 39% chance that month of smoking some dispensary weed from somebody?

Pot smokers share. If one teen gets some dispensary marijuana, he's going to share, either passing a joint or selling an eighth, and now there's a bunch of teens who "smoked dispensary weed" but only one who "usually got weed from dispensary." (Seattle Schools data showed a 2/3 "got it from friends.")

Also, people lie. If some teen is growing and selling, he'll upsell his product by saying, "Naw, man, this is some MEDICAL shit, brah," and every teen who believed that is now checking "used marijuana that came from a dispensary" on the survey.

Finally, go back up to that alcohol table. Every column that is "Did not drink" is a data point of teenagers who used alcohol that came (eventually) from a recreational alcohol dispensary. One hundred percent, unless some of them are fermenting or distilling their own ethanol. Yet Kevin Sabet is silent about the need to repeal the 21st Amendment for the sake of the children.