No, I'm saying that per state capita homicide rates are less meaningful than total homicides given that homicides are not homogenously distributed throughout any state's population. Per capita homicide rates are a deception used by people like the Gifford's Law Center to justify more gun control. Murders are not evenly distributed throughout a state, they are concentrated in first counties, then further in cities and then yet again in neighborhoods.
That would be true generally but when we're
only assessing the effect of gun control laws, it is the
only significant factor. Think of it as an experimental group versus a control group. If we have one group of 100 hypertensive people testing a new blood pressure pill, we compare their blood pressure numbers against another group of 100 people that didn't take the blood pressure pill in order to determine what effect the new medication had. We hypothesize that the 100 people who took the blood pressure pill will have lower blood pressure numbers than the 100 people who didn't. If that turns out to not be the case, we can conclude that the blood pressure medicine was ineffective, that it does not lower blood pressure.
So with that analogy in mind, let's compare the experimental/strong gun control group/city of Oakland, California with its 440,646 people with the control group/lax gun control city of Miami, Florida with its population of 442,241 people.
Oakland had 75 murders in 2018. Miami Florida had 51 murders in 2018. For purposes of evaluating the effect of gun control on homicide, this is a straight forward apples to apples comparison. Two different cities of nearly identical size. One city is in a state that is ranked as having the strictest gun control in the country. The other city is in a state ranked as having the 24th strongest gun control laws in America. The fact that the city of Miami had significantly fewer murders is evidence that A) California's gun control laws don't work and B) per state capita comparisons are meaningless.
All ranking is done by the Gifford's law center. California is ranked as having the toughest gun control laws in America. Maryland is ranked #6. New Jersey is ranked #2.
No, I'm saying that per state capita homicide rates are less meaningful than total homicides given that homicides are not homogenously distributed throughout any state's population.
Why does distribution have anything to do with using a per capita measure? Why does that same argument not apply total numbers as well? You do not have to have a uniform distribution of anything to make per capita numbers meaningful. And certainly, whatever shortcomings per capita numbers have in describing crime (or anything else) because they mask distribution effects is 100% true of just using the total of anything as well.
Example. GDP per capita vs GDP. China has the world's largest GDP, but the US has a higher GDP per Capita. Which one correctly measures the wealth of the average citizen? Clearly GDP per capita. And that does not change even though there are plenty of people in China who are richer than some people in the US.
Per capita homicide rates are a deception used by people like the Gifford's Law Center to justify more gun control.
This is known as the genetic fallacy. The fact that other institutions you disagree with use that statistical measure does not make the measure itself invalid. Plenty of other scholars have used per capita numbers to argue against stricter gun control.
Murders are not evenly distributed throughout a state, they are concentrated in first counties, then further in cities and then yet again in neighborhoods.
No one is disagreeing with that from what I can see, but it has nothing to do with the choice of how to measure homicides. How would taking the total number, which includes no information about distribution, in any way solve for this issue?
we're only assessing the effect of gun control laws, it is the only significant factor. Think of it as an experimental group versus a control group. If we have one group of 100 hypertensive people testing a new blood pressure pill, we compare their blood pressure numbers against another group of 100 people that didn't take the blood pressure pill in order to determine what effect the new medication had. We hypothesize that the 100 people who took the blood pressure pill will have lower blood pressure numbers than the 100 people who didn't. If that turns out to not be the case, we can conclude that the blood pressure medicine was ineffective, that it does not lower blood pressure.
No, that is not how statistical analysis works. Just because it is what you are trying to investigate, does not make it the only significant factor. Any factor which may effect the dependent variable is significant, and needs to be controlled for as much as possible.
To use your example of the drug testing, in the real world when a study like that is done a great deal of information must be collected on both groups. Because you cannot recruit 200 human clones and keep them in the same environment to eliminate confounding variables, statistics would be used to control for the effects of outside variables. Failing to do this would render your study invalid. For example, if the people who take the new medication are on average older, heavier smokers, and heavier drinkers than the control group you would need to control for that in the analysis.
Furthermore, I will add that comparing states is far less straightforward than your clinical trial example. There are relatively large differences between the states, on top of the fact that the sample size is extremely small.
So with that analogy in mind, let's compare the experimental/strong gun control group/city of Oakland, California with its 440,646 people with the control group/lax gun control city of Miami, Florida with its population of 442,241 people.
Oakland had 75 murders in 2018. Miami Florida had 51 murders in 2018. For purposes of evaluating the effect of gun control on homicide, this is a straight forward apples to apples comparison. Two different cities of nearly identical size. One city is in a state that is ranked as having the strictest gun control in the country. The other city is in a state ranked as having the 24th strongest gun control laws in America. The fact that the city of Miami had significantly fewer murders is evidence that A) California's gun control laws don't work and B) per state capita comparisons are meaningless.
Except it is not that straightforward. You have only controlled for population, nothing else. It is not a straight forward "apples to apples" comparison, all you have is two fruit that are the same weight, not the same species.
There are many other differences between the cities. Demographically, they are not comparable, Miami is more homogeneous while Oakland is more diverse. Miami is also older, by almost 4 years on average, than Oakland.
And finally, even to the extent we find a difference between the two cities, we cannot attribute that automatically to firearms laws. Again, a robust analysis would try to first answer the question of what the homicide rate in Oakland would be without those laws in place.
All ranking is done by the Gifford's law center. California is ranked as having the toughest gun control laws in America. Maryland is ranked #6. New Jersey is ranked #2.
Which is fine, but my point is that their methodology is not the only one that could be used, nor inherently the best.