What you don’t see can’t hurt you? Panel data analysis and the dynamics of unobservable factors
[…] a statistical model of crime and drug-taking behaviour by people aged 10-19 observed over a four-year period 2003-6 we show that, even if cannabis has no causal impact on the risk of initiation into other drugs, conventional statistical methods tend to show a spurious positive effect of cannabis. Our conclusion is that many research […]