I agree that the gun ownership data for the USA is not very rigorous. Each study seems to use a different proxy. And they’re using suicide rates too heavily as that proxy, in my opinion, especially if they’re going to be plotting it against data which also includes suicides.