I have a slightly different take on this
Via Gunfreezone: Underpants Gnomes Statistics & Research.
Methods. We conducted a negative binomial regression analysis of panel data from the Centers for Disease Control and Prevention’s Web-Based Injury Statistics Query and Reporting Systems database on gun ownership and firearm homicide rates across all 50 states during 1981 to 2010. We determined fixed effects for year, accounted for clustering within states with generalized estimating equations, and controlled for potential state-level confounders.
Taken literally, this would indicate that the CDC Web-based Injury Statistics Query and Reporting System (WISQARS) had data on gun ownership. I examined both the Fatal Injury Data system there and National Violent Death Reporting System and no data is provided on gun ownership at all.
Now, if we assume that Miguel is correct, and they examining the databases of registered firearms possessed by the individual states (those that actually exist) then all they actually demonstrated is a correlation between “firearm homicide rates” and registration of firearms.
However, they mention that they also use a proxy for levels of household firearm ownership: the percentage of suicides committed with a firearm. The problem here is that an assumption is made that the percent of gun owners who commit suicide is a constant no mater what state they are in despite it being known that suicide rates between the states vary wildly (Alaska’s is double Ohio’s for example). They also don’t say if they assume that percentage of suicides is the exact percentage of households with guns or modify it somehow to make it useful to compare to data “measured directly”. The proxy is of dubious value.
This study isn’t in the current issue of the American Journal of Public Health so I will have to wait until it actually sees print to find out what they actually did.
There is also the usual issue that they were only concerned with “the firearm homicide rate” and not “the homicide rate”. Being 1% less likely to die by gun but 1% more likely to die overall (for example) is not a good trade. Gun control advocates not only try to ignore externalities, they try to hide them from everyone else so they don’t realize they are making a fool’s bargain.