Reply To: THE PUZZLE OF MURDER-BY-GUN STATISTICS

[Unseen]

Simon Mathews wrote:

A relationship would indicate you could in some way use A to predict B. Not just that B necessarily changes (but in some unknown way) as A changes. That kind of relationship doesn’t really tell us anything and is not how we commonly use the word in statistics.

I found that if you look at the broadest possible picture, there is no functional relationship (not even a loose one) between quantity of guns in a country’s populace and the gun homicides. Some countries have very high rates of gun ownership and low rates of gun homicide whereas in other countries the homicide rate is unusually high whereas gun ownership is relatively uncommon.

In the US, while the disparities are not so great, you have, for example, The District of Columbia with an extremely low rate of gun ownership, 3.6%, but the nation’s highest murder rate, 16.5/100,000, whereas Wyoming has the highest rate of gun ownership at 59.7% and one of the lowest rates of gun homicide at .09/100,000. (source)

Even in the US, there is no justification for saying or even implying some sort of functional relationship between ownership rates and gun murder rates.