The study seems to suggests only that cannabis use is prevalent among violent people, not that cannabis use causes people to be violent. Title is misleading
>However, the specific impact of cannabis use on the risk of violence remains unclear
Isn't 'associated' here a synonym for 'correlated'? Meaning that while they found elevated levels in this population, they haven't demonstrated a causal relationship?