First of all that research is correlation study, so equally well we can argue that increasing number of homicides in some area made people to buy more guns to defend themselves.
Secondly, this research looks only on guns vs. homicides, I would say that there could be a few other factors that might be more stronger then guns (poverty level, drugs abuse level, local society structure, etc.) - this is speculation, but I don't see that researches even looked on possible other factors that can potentially explain higer/lower homicide rate.
> so equally well we can argue that increasing number of homicides in some area made people to buy more guns to defend themselves.
And taking that reasoning one step further I'm going to argue that in places where guns are prevalent/accepted, people might think a gun can be used for self defense, so more guns actually lead to even more guns.
Whereas in places where guns are scarce and unaccepted, they are less likely to be seen as tools for self defense, so a neighborhood having more guns does not lead to others buying guns.
So I'm arguing both things happen. If you have a culture of gun ownership then more guns will lead to both more homicides and more guns.
But if there is no cultural belief in guns as self protection, then that correlation doesn't occur.
Secondly, this research looks only on guns vs. homicides, I would say that there could be a few other factors that might be more stronger then guns (poverty level, drugs abuse level, local society structure, etc.) - this is speculation, but I don't see that researches even looked on possible other factors that can potentially explain higer/lower homicide rate.