I try to be very careful with statistics. First, there is the possibility that someone has actually misrepresented the underlying data. Second, even if the underlying data is correct, it is easy for a statistician with an agenda to use mathematical techniques to show results that don’t really give a correct context for the data.
When it comes to levels of crime committed by different racial groups, it’s really hard to find reliable data. (My theory on that is because most left-leaning academics know what it will show, and they don’t want the public to know it.)
I knew that black people commit a disproportionate amount of violent crime. (That is, they are only about 13% of the population, but they commit more than 13% of the violent crimes.) What I was having trouble finding was numbers on interracial crime -that is the number of black on white violent crimes and the number of white on black violent crimes.
Thanks to an article , I found good credible data from the FBI on interracial murders. So, for instance, in 2015, of the 3,167 white people murdered, 500 of those victims were murdered by a black person. In 2015, of the 2,664 black people murdered, 229 of them were murdered by a white person. (Intra-racial murder is much more prevalent for both groups. In other words, most whites are murdered by other whites, and most blacks are murdered by other blacks. But, that is not what I am focusing on here.)
So, there were 500 murders of white people committed by a black person in 2015 and there were 229 murders of a black person committed by a white person. Right away, that number is clearly disproportional from the number of black people in the United States. Since black people make up only about 13% of the population, the fact 500 white people were murdered by blacks, while only 229 black people were murdered by whites is remarkable.
But, I don’t know exactly how to compare these two numbers. My probability and statistics skills are not that strong. One possible way to compare them is to take the total number of black on white murders (500) and divide that by the total number of black people in America. (Roughly, 40.5 Million). Similarly, take the number of white on black murders (229) and divide that by the number of white people in America. (Roughly 233.7 Million). This will give you:
500/40.5 Million= 0.00001235 and 229/233.7 Million = 0.00000098
Then, you turn both of these into percentages by multiplying them by 100, which is: 0.001235% and 0.000098%
My thinking is that this is like rolling a six-sided die. The probability that you will roll a 6 is 1/6. The probability that you will roll a 5 or a 6 on one roll is 1/3.
You can think of the statistics on homicide above as like rolling two different dice. One die is a 40.5 Million-sided die. This is the black on white crime die. There are 500 “sides” of that die that would be a black on white crime, while the rest of the sides are all the black people who don’t murder white people. Similarly, there is another 233.7 Million-sided die, and on that one, there are 299 sides that are white people who commit a white on black murder. If you make that assumption, then the way to compare the two probabilities seems like it would be to take the first percentage, 0.001235% and divide it by the second percentage, 0.000098%. The result is 0.001235%/0.000098% = 12.6. To me, this means the probability of a black on white murder is 12.6 times as great as a white on black murder.
However, I am not sure this reasoning makes sense entirely. I am assuming that *every* black on white murder and *every* white on black murder is committed by different people. In other words, I’m assuming that of the 500 black on white murders, each murder was committed by a different black person. It’s possible that the *same* black person committed all of the 500 black on white murders in 2015. You cannot tell from these numbers alone. (Same for the white on black murders.)
It’s also possible that I don’t understand probability calculations very well (which I don’t), and I’m making some other mistake. I wish more people would discuss this matter, and try to come up with good numbers, but the data, much less an analysis of it, seems to be almost completely non-existent. (Like I said, I think most people know what the results of the analysis would be, and they don’t like to think about it, or they have a political agenda.)
