Correlation % gun ownership and % Black vs murder rate
Correlation is not causation, but recently the anti-gun faction made a big deal about correlation between party control of states and murder rates.
Well, % gun ownership had a NEGATIVE correlation with murder rate but % Black population had a very strong POSITIVE correlation with murder rate.
Can facts be racist?
Tidewater News and Opinion Forum 
Are you saying that Black Americans are not Americans?
Does the graph describe mixed race such as mulattos, quadroons, one drop?
Does it include Black doctors, investment bankers, singers, ministers, police…
LikeLiked by 2 people
I said you weren’t going to like it.
Those things might move the correlation a percent or 2 but the point it that you thought it was significant when red states had higher murder rates, now you’re upset when states with more Blacks have a higher murder rate.
You can’t have it both ways.
Besides, more gun ownership correlated negatively with murder rates. So much for “more guns equals more deaths”
LikeLike
So Americans are different than Blacks.
I know you are trying hard to blame most of our crime on minorities. Crime in the inner cities is nothing new. It was true in the 1800’s, the 1900’s and true today. The difference today, of course, are the lethal weapons by the hundreds of millions.
So who put those charts together about a decade ago? I could barely make out much from the tweet.
LikeLiked by 2 people
Click on it and it enlarges
So Americans are different that Republicans?
I’m not trying to do anything. You and Paul thought it was a big deal that there was a correlation between party and murder rates at the state level. You have raised the issue many times.
Well, here is a much better correlation based on race, and a NEGATIVE correlation for gun ownership.
You thought correlation was a big deal when it supported your claims that GOP gun laws were a problem but now you don’t think it means anything.
LikeLike
“Can facts be racist?”
No one here has ever disputed that the poorest communities have the highest rates of gun violence. What has been disputed is the idea that this fact means we do not really have a gun violence problem. That is the racist element in the gun debate.
As far as negative correlation between gun ownership and murder rates, I find that very, very dubious. Dishonest even. And so does just about everyone who ever studied the question.
https://www.hsph.harvard.edu/hicrc/firearms-research/guns-and-death/
https://ajph.aphapublications.org/doi/full/10.2105/AJPH.2013.301409
LikeLiked by 1 person
Still clinging to that state and even country level correlation?
We established last week that county level correlation provided a very different picture.
But you think correlation proves causation when it suits your purposes, and then explain it away when ot doesn’t
LikeLike
The chart on the left shows exactly what has been pointed out by other studies. Look at the cluster of states at the top right – high gun ownership and high murder rates and redder than red.
LikeLiked by 1 person
And yet the correlation line has a negative slope.
LikeLike
It’s a three dimensional graph with population rendered as dot size.
The correlation is calculated, the graph is only a presentation.
LikeLike
Understood. The point is that the calculated correlation tells a different story from the subjective visual interpretation Mr. Murphy thinks is important.
LikeLike
Only if you don’t understand the meaning of the do size.
But in any case, the calculated value s what matters.
LikeLike
The data is per 100K, Note that CA’s dot is much large than MS’s
Position on the graph is not in isolation. population matter too.
LikeLike
This presentation is completely bogus. There is no valid reason for population size of a state to weight the comparison. As the saying goes . . . “There are lies. There are damn lies. And there are statistics. Higher gun ownership is positively correlated with gun death rates. And the reddest of red states cluster where the gun ownership and the deaths are highest.
LikeLiked by 2 people
Also, this bogus presentation uses a bogus vertical scale. Whoever put this together is not honest.
LikeLiked by 1 person
How is the scale invalid?
LikeLike
How is the scale invalid?
Just look at it. The distance from 0 to 2 is the same as the distance from 2 to 9. There is no valid reason for anything but a straight linear vertical scale. It is visually deceptive. Not honest.
LikeLiked by 1 person
It’s the same scale in both graphs,
LikeLike
“It’s the same scale in both graphs,”
Yes, clearly. So both graphs are visually deceptive. If you have another explanation for such visual skewing, let’s hear it.
LikeLiked by 1 person
Did you have any problem seeing the difference in rates?
If it doesn’t deceive anyone, how is it deceptive?
Compressing the higher end visually simply makes the graph easier to scan. If it were straight linear, how tall would it have to be compared to the x axis?
LikeLike
“Compressing the higher end visually simply makes the graph easier to scan. If it were straight linear, how tall would it have to be compared to the x axis?”
That is completely lame. The graph does not have to be any taller. It simply should be linear. And, the same is true of the horizontal axes of these charts. Why are they skewed? The more you actually look at this chart of 2014 data the more suspect it becomes.
As for comparisons with Louisiana – it is in a class by itself. It combines – apparently – the most violent aspects of both urban and Southern culture with almost non-existent gun control. It has lead the nation in gun homicides for decades.
LikeLiked by 1 person
And Maryland, in third place, has some of the most Draconian gun laws in the nation and has nothing in common with Louisiana except a 30% Black population.
LikeLike
RE: “There is no valid reason for population size of a state to weight the comparison.”
It doesn’t. The dot size in the plot tells you something about the values on the x and y axes. For example, a larger population (sample size) means that the calculated gun murder rate on the y axis is more “realistic.” But notice that each of the state dots has the same vertical position in both plots.
The correlation lines in both plots would be the same if the state dots were all the same size. That’s the whole point of the two graphs.
LikeLike
Here is an honest, multi-year, and verifiable summary of the red state murder problem.
https://tinyurl.com/sy95xsmh
LikeLiked by 1 person
Where is the murder rate?
LikeLike
“Where is the murder rate?”
Huh?
It is shown in the text and in the graphs
LikeLiked by 1 person
The article states that it resolves the blue city problem by subtracting the worst county in each state.
That doesn’t do much. For example, New Orleans includes Orleans Parrish, but also the populous parts of Jefferson, St Bernard and St Charles parishes.
Simply subtracting Orleans parish is only half the job.
LikeLike
RE: “Here is an honest, multi-year, and verifiable summary of the red state murder problem.”
No. We have already leaned (https://tinyurl.com/mvfkzvsn) that the “red state murder problem” is an illusion. Specifically, red states that appear to have a murder problem are actually red states that have blue counties that have a murder problem.
LikeLike
“Specifically, red states that appear to have a murder problem are actually red states that have blue counties that have a murder problem.”
The link I provided tested this big city excuse . . . “Even when murders in the largest cities in red states are removed, overall murder rates in Trump-voting states were 12% higher than Biden-voting states across this 21-year period and were higher in 18 of the 21 years observed.”
Even without the biggest cities there is still a red state murder problem. And whatever the “contribution” of cities, the laxer gun laws make the problems worse in both urban and rural areas.
LikeLiked by 1 person
And again, the strongest correlation is race.
Note that in the left graph, Iowa and Louisiana have about the same gun ownership rate, but the murder rate is almost 10 times higher in LA.
Then look at the right graph and you will see that Iowa is 3% Black while Louisiana is 30% Black.
Does it start to make uncomfortable sense now?
LikeLike
“Does it start to make uncomfortable sense now?”
Lax control of guns makes it easier for people of all races to get their hands on weapons. That is why red states over decades have consistently higher gun ownership and gun deaths than blue states.
You keep pushing that it is the black community that suffers the most from gun violence. I am not aware of anyone disputing that. I certainly don’t. But, unlike some, I do not take that fact to mean that we do not have a gun problem.
LikeLiked by 1 person
And yet Iowa and Louisiana have the same gun ownership rate.
Clearly guns are not the problem.
LikeLike
And NY State has almost half of its population in one of the largest Democratic controlled cities in the country. Yet, the state, and NYC BTW, has one of the lowest gun violence rates in the country.
LikeLiked by 2 people
RE: “Even without the biggest cities there is still a red state murder problem.”
Only if you insist on using the arbitrary state-level aggregation as your point of reference. But we know from the more granular county-level analysis that it is illogical to insist on the state-level aggregation, because it falsifies reality.
LikeLike
“Only if you insist on using the arbitrary state-level aggregation as your point of reference. ”
When comparing the impact of gun controls, the state-level is not arbitrary.
LikeLiked by 1 person
Sure. Illinois has the same gun laws statewide.
Outside Chicago, it’s as safe as any other place in the country.
LikeLike
https://www.americanprogress.org/article/gun-violence-in-rural-america/
LikeLiked by 2 people
What the hell is an “age adjusted homicide”?
So, there are a few rural counties scattered among the list of cities. I’m going to guess there is something unusual about them, meth or the like.
They don’t change the rule.
LikeLike