http://www.hsph.harvard.edu/hicrc/firea ... and-death/
The sad thing about gun statistics in the USA is that there are none. There is no legal requirement for any authority to retain statistics on how many guns are in civilian possession. This leaves researchers to rely on surveys.
That means that, whatever data is quoted, it is open to suspicion, based on the simple fact that tose with a bias will report their bias. John Lott, for example, who made millions of dollars pushing the NRA line of more guns and less crime, reports opposite results from surveys to researchers from Harvard University. You may argue about who is correct. The man who made the money, or the academics who made no money.
I have seen the results of three surveys on gun ownership over time. One showed an increase, and two showed a fall. Which are correct?
One thing that is known for sure, is that there has been a fall in murder rates throughout the world. That fall is even seen in sub-Saharan Africa. Since it is universal, we can assume the main driver of that reduction in murder rate is independent of the level of gun ownership.
However, within the USA, things are a bit different. Harvard studies (reference above), relying on surveys of which states have the highest gun ownership, show clearly that states with more guns have higher murder rates and higher suicide rates. That does not change the fact that murder rates overall are falling. But it does show that the murders that continue are related to gun ownership.
