George Floyd and Minneapolis Police UoF - Rioting

[jpmuscle]

That chart is purposely misleading, per million of the person being murdered race would make more sense. It’s always going to be skewed the way it is because of over 250M people in the US are considered white and around 44M are considered black. It doesn’t make sense to say what percent/factor of white people are murdering black people because your denominator is always going to skew things in the direction the person who created it wanted. I’m not saying the proper way to show the data would be any different in the overall conclusion but as someone who specializes in data analysis and has done extensive work in the data visualization space that’s just dishonest.

That just makes BLM cause even more statistically invalid.


Sent from my iPhone using Tapatalk

[Crow Hunter]

That chart is purposely misleading, per million of the person being murdered race would make more sense. It’s always going to be skewed the way it is because of over 250M people in the US are considered white and around 44M are considered black. It doesn’t make sense to say what percent/factor of white people are murdering black people because your denominator is always going to skew things in the direction the person who created it wanted. I’m not saying the proper way to show the data would be any different in the overall conclusion but as someone who specializes in data analysis and has done extensive work in the data visualization space that’s just dishonest.

Are you talking about Ram's chart?

That one is per million.

[WillBrink]

Candice pulls no punches....

[Life's a Hillary]

Are you talking about Ram's chart?

That one is per million.

Yes, it’s per million of the murders race, basic math says that’s always going to skew the numbers because the denominator for millions of white people is much larger. It’s not normalizing the data like you think it is but it makes you think that’s happening.

[Crow Hunter]

Yes, it’s per million of the murders race, basic math says that’s always going to skew the numbers because the denominator for millions of white people is much larger. It’s not normalizing the data like you think it is but it makes you think that’s happening.

Where does it say that it is just per million of people murdered? Not saying you are wrong, I just don't see an indication of that.

It just says per 1,000,000 of the members race. That implies that it is counting the whole. But maybe you see something that I don't.

ETA: I think I maybe understand what you are getting at but I am not sure. Can you elaborate?

[PracticalRifleman]

Yes. I called the guys medical history and toxicology and it’s been verified.
But that picture looks horrible. And they blew him off when he was unresponsive for a few minutes.

If they simply had him proned out with a HAND lying on the back of his neck, totally effective to control him if he tried to squirt or fight, and a sentence or two saying “OK man, the ambulance should be here any minute.” Or made an effort to check or roll him over when he went full Fred Sanford, cities may not have burned.

Or if he died sitting in the back of a cruiser.

Perception is reality.

All they had to do was manage perception.

Now the fabric of our society if further torn over some drugged out dude with bad heart disease breaking the law that already had a rap sheet and a nickel in prison under his belt. Which I believe includes armed robbery and punching a pregnant woman in the stomach. And the media is surprisingly silent about the appearance and pictures of the other 3 LEOs. Because it looks like a cop purposefully dug his knee into his neck to choke him to death and kept it there an extra few minutes. And that’s how a ton of people perceive it.

I was going to grab some more recent data, but someone had already made a graphic that expresses it. (And keep in mind of that 0.77/million, only a fraction are by LE. And that the numbers are tainted as last I check, a Hispanic perpetrator is recorded as white, but a hispanic victim is not counted as white.)



I have been a harsher critic of LEO actions than most here on many occasions, and on at least one occasion more supportive than most. But by any objective measure, the number of black men killed by LE, let alone the tiny fraction of that that may not have been justified, is not the actual issue.

Current politics, the COVID scare, these riots, I can’t express how insane and twisted I find it. Tons of African Americans are running around in LE, in politics, as doctors, as lawyers, as judges, running businesses, etc. I have not met a single individual trying to stop or prevent any of that.
Not a single person I interact with wants that community to disproportionately be unemployed, committing crime, using drugs, having medical problems, etc. None of my ancestors were even Americans until like 70 years after slavery ended. And they were all dirt poor. I can’t wrap my mind around why the media narrative is so different.

Perception is very, very bad. And I agree with all which you’ve stated. I actually looked up some similar stats myself through the justice department page and that info graphic is spot. There is even officer demographic available as well.

I also found some info on the amount of officer injuries by race of offender, frequency of officers encountering a weapon by race of offender, etc. it’s eye opening.


Sent from my iPhone using Tapatalk

[Life's a Hillary]

Where does it say that it is just per million of people murdered? Not saying you are wrong, I just don't see an indication of that.

It just says per 1,000,000 of the members race. That implies that it is counting the whole. But maybe you see something that I don't.

ETA: I think I maybe understand what you are getting at but I am not sure. Can you elaborate?

In the title of the chart it says in parentheses, “per 1,000,000 members of the murderer’s race.” This means it’s saying how many black people are murdered per 1M white people. Well there are around 250M white people so the denominator is going to be 250 while the opposite bar is going to have a denominator of 44. Black on black and white on white are 1 to 1.

It waters down the number of black people murdered by white people just due to simple division.

I’m not even saying a better representation of the data would show some drastically different story but I would not look to that chart to see anything meaningful as you cannot compare the bars to one another.

[Diamondback]

In the title of the chart it says in parentheses, “per 1,000,000 members of the murderer’s race.” This means it’s saying how many black people are murdered per 1M white people. Well there are around 250M white people so the denominator is going to be 250 while the opposite bar is going to have a denominator of 44. Black on black and white on white are 1 to 1.

It waters down the number of black people murdered by white people just due to simple division.

I’m not even saying a better representation of the data would show some drastically different story but I would not look to that chart to see anything meaningful as you cannot compare the bars to one another.

Precisely. A chart showing B-on-B, W-on-W, etc as percentages of total homicides would be better.

[Crow Hunter]

In the title of the chart it says in parentheses, “per 1,000,000 members of the murderer’s race.” This means it’s saying how many black people are murdered per 1M white people. Well there are around 250M white people so the denominator is going to be 250 while the opposite bar is going to have a denominator of 44. Black on black and white on white are 1 to 1.

It waters down the number of black people murdered by white people just due to simple division.

I’m not even saying a better representation of the data would show some drastically different story but I would not look to that chart to see anything meaningful as you cannot compare the bars to one another.

I see it now. I missed the "per murderer's race". What I thought I was reading wasn't what I was, just like you said.

[Life's a Hillary]

I see it now. I missed the "per murderer's race". What I thought I was reading wasn't what I was, just like you said.

The only reason I noticed is I literally use data to tell stories for a living pertaining to businesses so I have lots of experience picking apart improperly used data. It’s very easy to change little things and drastically change the story, I have to be very vigilant that I’m doing things the right way because that’s the fastest way to lose credibility in my world.