I didn't get all the way through the original paper yet (it's a bitch. Meta-analyses have a lot of math...

) but I'll say this: The science is pretty good, but this article is pretty bad at covering it. Which is often true of news articles about research. The only real conclusion that comes from the paper is that *IF* you are already predisposed to psychosis *AND* you smoke weed *AT A YOUNG AGE* you will develop that psychosis on average about 2-3 years earlier than you would have otherwise. That's not the same as "Weed causes psychosis." The only population studied were those who developed psychosis. This didn't compare the rate of psychosis in weed-smokers vs. non-smokers as compared to the general population. In short, if you are already at high risk for schizophrenia and you are in your teens, you should probably lay off the weed; at least until you're older.
There are real problems with the way research is written about in the media. Unless you've had some real training (and a LOT of practice -- everything gets better with reps!) it's not at all obvious how to go about critiquing scientific literature. I have yet to see a news article that includes a proper discussion of methods of critique. We don't even teach critical thinking until college and even then only in fits and starts. This article didn't do a very good job of explaining the methods and results of the paper (how many people here have heard of a meta-analysis before this article much less attempted to critique one?). They largely read the authors' discussion and added in a few tidbits of their own.
It's frustrating because both really good science and some really questionable crap are written about by the same media outlets and are either so "dumbed down" or oversimplified that the only critique most people are able to give is: "That's bullshit." It does a disservice to the people -- like so many you'll find on this forum -- who ARE interested in the topic and actually have the mental wherewithal to READ research. It also does a HUGE disservice to the people who are actually doing very good research. The good stuff gets thrown in with the not-so-good and it all reads the same. It doesn't matter how much fillet mignon you mix with dog shit; it's still gonna taste like dog shit if you lump it all together.
If I can get a non-subscriber link to the original article, I'll post it.