The ramblings of an Eternal Student of Life
. . . still studying and learning how to live

Latest Rambling Thoughts:
Sunday, August 24, 2014
Current Affairs ... Society ...

In the wake of the Michael Brown killing in Ferguson, MO two weeks ago, various articles have pointed out how difficult it is to get good stats on the number of police killings in the US per year, and regarding the circumstances behind them. The most commonly quoted stats are from the FBI’s Supplemental Homicide Report, from its Uniform Crime Reporting program. The basic number is about 400 non-accidental fatalities per year stemming from police actions. However, many point out that this number probably undercounts the real level of fatal police activity.

There is a dataset, however, from the US Bureau of Justice Statistics regarding all deaths in police custody or attempted custody (as with Micheal Brown) from 2003 to 2009, which breaks down the circumstances including race of the victim. The raw numbers from this study are fairly close to the FBI estimate, averaging about 443 homicide-like killings per year by police, ranging from 375 to 497 (with a rough upward trend seen over the 7 year period). The FiveThirtyEight web site still thinks this is too low, but it appears to me to be about the best data available for the years involved. So, I will use this data to get a rough view on whether the police are more apt to shoot a black person versus a white person during an arrest or attempted apprehension action.

OK, the BJS stats say that between 2003 and 2009 (inclusive), police killed 1,233 white people and 937 African-American people non-accidentally during a police action, whether or not an arrest had or had not been made of the victim (obviously, no arrest had been made of Michael Brown; although Officer Wilson arguably would have made a formal arrest had he wounded but not killed Brown in the incident). What can we compare this against that allows an “apples-to-apples” comparison?

Well, I think that the best we can do is to add up FBI UCR arrests for these years. In some ways, this is an apples-to-oranges situation, in that many police killings do not involve an arrest (because the victim is dead before an arrest can be made). Arguably, a lot of police shootings, fatal and otherwise, come during police stops for questioning, before grounds for arrest have been established (as with Michael Brown). And the UCR stats do not count stops for questioning, just actual arrest incidents.

However, I think we can reasonably assume that in most instances where the police decide to use their weapons against someone, grounds for an arrest have been established in their minds. Simply the act of doing something that an officer perceives to be an intentional threat to him or her is grounding for an arrest. If police officers shoot, we can assume they perceived such a threat, or at least would claim that.

I know what the next objection might be. I.e., that in racial situations, police often shoot without there being an actual threat to them or anyone else from the victim. OK, that may be true in many instances. But, if a police officer takes a shot, he or she is probably going to CLAIM that exigent and threatening circumstances existed regarding the victim, and thus they probably would have made an arrest (rightly or wrongly) had the victim been subdued without dying. E.g., a police shooting victim can be arrested and charged in the hospital after being treated.

So, I believe we can reasonably say that when a police officer shoots or otherwise fatally injures someone (intentionally), they believe themselves to have grounds for an arrest (and that potentially lethal force is necessary to effect that arrest).

Most of the homicide-type killings by police are probably not counted in the UCR arrest data, because they probably occur before a formal arrest was made. Thus, in getting the ratio of deaths to the sum of intended and actual arrest actions, the denominator is arguably too small. However, in practice that isn’t much of a problem, because the number of fatalities is quite small relative to the total number of arrests. Here are the total UCR arrest numbers from 2003 to 2009: for whites, 50,350,406 arrests; for blacks, 20,033,160. Adding another thousand or so to these numbers in the denominator is not going to make much difference.

So let’s do the ratios: for the 7 year period 2003-2009, there were 2.45 white people killed by police non-accidentally for every 100,000 white arrests. By comparison, there were 4.68 black people killed for every 100,000 black arrests. For sake of comparison, if we look at ALL deaths occurring in police custody or pursuit (including accidents, suicide, intoxication, natural causes, and the dubious “unknown” category, along with police homicide-like killings), there were 4.02 white people who died during a custodial police action for every 100,000 white arrests, and 7.63 black people killed for every 100,000 black arrests.

So, overall, the ratio of deaths relative to serious police actions is fairly low. By comparison, there were 13.6 deaths per 100,000 motor vehicles in the US in 2012. Thus, you are probably safer when a cop attempts to stop and arrest you than you would be by using a car for one year. BUT, the relative frequency of deaths for blacks is about 90% greater during a police action involving custody or attempted custody than for whites.

In an attempt to not jump to conclusions here, let’s consider what might be involved, besides the obvious notion that cops blindly have bad attitudes about blacks, attitudes that they grew up with. Well, first off, black are arrested more frequently per capita than for whites; if we estimate that the average white population was 236,925,000 between 2003 and 2009 and the average black population was 38,464,000, then there were 3.0 arrests per year for every 100 whites, and 7.4 arrests per year for every 100 blacks. This does NOT directly explain or justify the death rate difference upon arrest or attempted arrest, of course; but it does give us some hints about the attitude that makes for the hair-trigger usage of police weapons when black actors are involved.

OF COURSE, many commentators claim that the higher arrest rate for blacks is itself a function of pre-existing racism on the part of the police and justice system, more so than an alleged higher level of illegal acts by blacks. I.e, for any actual illegal act that a police officer becomes aware of, the officer is more likely to “let it slide” for a white person than for a black person. I have no doubt there is some truth to this contention. Still, other socioeconomic factors such as poverty and unemployment [which admittedly are exacerbated by on-going racial attitudes amidst whites] probably do drive the reality that there is more crime on average in black communities than in white communities.

A more quantifiable hypothesis is that blacks are more likely to be involved in crime situations where there is heightened danger to a responding police officer. E.g., murders, assaults, sexual assaults, violent arson and possession of weapon situations. Taking a look at the FBI arrest stats for 03-09, we see that for white arrests, such dangerous crimes are involved in about 19.2% of arrests. For blacks, the number is 26.7%. Mathematically, however, this fact alone should not drive a 90% increase in the black arrest rate. Even if we assume that making an arrest for one of these offenses is 10 times more dangerous to the officer than for property or narcotics crimes, the increase in the rate of death for blacks versus whites should not be more than 25%. So, a 90% rate says that a lot more is going on than the fact that blacks are more often arrested for the dangerous types of crime.

Thus, there is some very rough statistical evidence (not on an academic level, admittedly) supporting the suspicion that police are more likely to take stronger measures against a black actor versus a white actor in an otherwise equivalent situation when a criminal act is suspected. But again, this analysis is still rough, and doesn’t consider other factors that might be involved. E.g., are police actions against blacks more likely to occur when more than one actor-suspect is present, e.g. a gang action or flash mob? Or in high-crime neighborhoods versus low-crime environments? According to Bayesian decision-making principles, the officer might be justified in taking action quicker when dealing with a gang, or with crime suspects in high-crime areas, possibly causing more blacks to be shot by police.

Still, the notion that when a policeman enters a situation or responds to a call reporting a possible criminal act, sees a person unknown to him or her, and if that person is black, the officer then automatically sets a higher “first guess probability” in her or his mind regarding the unknown person’s involvement in the crime . . . this notion does not sit well with our theories of justice and human rights in America. We say that everyone is presumed innocent until proven guilty, and we would like to think that our police apply this rule too.

Actually however, they don’t; their job is to presume innocence until some reasonable grounds for suspicion against a person are established (a lesser standard than “proven guilty”). Being black should not be an automatic grounding for such a suspicion . . . although in the real world, it probably enters into the minds of many law enforcement officers. And being a young black male in a high-crime neighborhood no doubt escalates that initial “probability-in-the-mind” even more.

Perhaps in some situations, such as in very high crime areas, such “initial probabilities”, however wrong, are going to have to be tolerated to keep the peace. But when such attitudes enter into a police officer’s decision as to whether to use fatal force (or how much force to use — again with regard to Michael Brown, even if Mr. Brown did threaten Officer Wilson, were six shots including two to the head really necessary to meet that threat?), then there is a need for intervention and oversight, for clearer policing policies, better training, constant monitoring and review, and sanctions when inappropriate force is applied. My own statistics may not be solid enough to trigger such action, but they certainly show the need to look deeper.

◊   posted by Jim G @ 4:28 pm      

  1. Jim, A good statistical report on “Police Killings and Race”. Yet, I find it difficult to leave out entirely the emotional aspect involved in any “police killing” regarding blacks, or whites for that matter. So, I forewarn you that my response will be very unorganized and concentrate less on the intellectual aspects involved.

    It would seem to me that any time a police officer pulls over anyone in a car, even the normal person driving a car is told in any documentary on the subject that the police officer never has any idea of what may happen in any situation where he/she pulls over a car. Thus, it seems to me raising the emotional level involved in both parties involved in such situations.

    I know a black person who was waiting for a ride on a downtown street in the city I live in; he was approached by an officer in a car, searched in the very cold weather we have and thus was exposed to bad weather. Just about that time I pulled up to pick him up, and the situation was resolved. I know the person I picked up was emotionally upset; I doubt the police officer was not also emotionally involved in some way, although in a lesser manner probably.

    It would seem to me that it’s not so much the intellectual stats that affect these situations; it’s the emotional aspects of them that affect the stats. For police officers (as for all first responders) to move *toward* a situation that might be fraught with danger must be most difficult; I myself would prefer not to be in such a situation; but police officers are first responders and deserve respect and honor for that. Yet, for the person being pulled over in a car or being called over for a search, or even just being called over for a “why are you here?” has its emotional effects.

    Thus, the intellectual/statistical stats tell us quite a bit about how blacks may be treated versus whites by police officers; yet it is most difficult to factor in the emotional aspects of such interactions. It does seem that an attempt is made to include the emotional aspects by noting the presumption of innocence until proven guilty, etc.; yet that principle of our laws does not really factor in any emotional aspects.

    Perhaps this is a tangential note, but I find myself wondering just what kind of psychological testing police officers have to determine their ability to remain calm and as emotionally uninvolved as possible in interactions with the public, among many other such questions. Then too, I find myself wondering if there *are* situations where an emotional response might be appropriate to include when a police officer responds to some situation.

    Again, tangentially, I recently saw a police program where the SWAT team went to arrest an individual it was pretty obvious had committed a really serious and awful crime. All police involved went to the home where the person lived; it was a very nice place and presumably the home of his parents. The SWAT team called out several times over loud speakers, “We have a warrant, come out with your hands up”. There was no response after several of these announcements. So, they used a heavy-duty hook of some kind, tore off the front screen door, went thru the front door and ransacked the home looking for the offender (who in this case seemed most likely to be the person wanted for this terrible crime). After all kinds of armor and heavy duty gun power, several dozen cops entering the home, etc., it was determined “nobody was home”. I found myself wondering if, even with all this fire power, the police commander had even attempted to make a phone call to the home, ring the doorbell, or use some such other less invasive approach. The person the police were searching for gave up easily when he was found at his girlfriend’s home. I find myself wondering in such a case (altho no one was killed in apprehending this individual but a lot of damage to a nice house was done) how much some serious testosterone flowing and how much emotions were involved in this situation.

    Thus, while intellectual statistics are important and give us some information, I find myself considering that they leave out some other very important aspects of police response to individuals who may not respond “properly” when stopped in some way by a police officer.

    In still another tangential remark, I find that no one has yet found even one picture of Michael Brown where he does not look like a thug. Here again, while a picture of an individual certainly tells us little, if anything, of what may have been involved in the altercation he was included in, still there’s an emotional aspect to seeing a picture of him as a young man who does not look like a thug which would make the world (and me, I admit) more sympathetic to him. It’s a shame to say; yet those emotions *will* find their way into almost every human situation, most especially ones that are so sensitive in our society at this time. MCS

    [MARY: the point of a good, fair statistical analysis like this is to try to eliminate all reasonably possible contributing factors, or adjust for the effects thereof; and if a residual effect is seen after that (with a confidence level significantly above “noise variation”), then you can reasonably say that the effects of human motivation are being seen, and can gauge just how strong that effect is.]

    Comment by Mary S. — August 24, 2014 @ 7:23 pm

  2. Thanks, a good objective analysis …. And DOJ is an excellent source of factual information on this subject and the findings are somewhat different compared to what is spoken and printed on the “numbers” and their meaning

    The approach which has been adopted and used is is devoid of any responsibility as it relates to crime and the acceptance that just maybe there are some in the black community that make it bad for everyone black

    I supported Dr. Martin Luther King and vote for LBJ in 1964 and still believe in equality for all Americans

    Just think the message has changed and I don’t see it as positive for people living in poverty… regardless of race

    Dr. King was successful because of his balanced approach … one could NOT argue against his points on how to treat people and what rights an American has …….. yet he was wise enough to point out the short comings too … The Positives and the Negatives … which is the fair way to do it

    “Do you know that Negroes are 10 percent of the population of St. Louis and are responsible for 58% of its crimes? We’ve got to face that. And we’ve got to do something about our moral standards,” Dr. Martin Luther King Jr. told a congregation in 1961. “We know that there are many things wrong in the white world, but there are many things wrong in the black world, too. We can’t keep on blaming the white man. There are things we must do for ourselves.”“Do you know that Negroes are 10 percent of the population of St. Louis and are responsible for 58% of its crimes? We’ve got to face that. And we’ve got to do something about our moral standards,” Dr. Martin Luther King Jr. told a congregation in 1961. “We know that there are many things wrong in the white world, but there are many things wrong in the black world, too. We can’t keep on blaming the white man. There are things we must do for ourselves.

    Again thanks … enjoyed the read

    Comment by Wiley Wesson — September 10, 2014 @ 6:23 am

RSS feed for comments on this post.

Leave a comment:


To blog is human, to read someone's blog, divine
NEED TO WRITE ME? eternalstudent404 (thing above the 2) gmail (thing under the >) com - THE SIDEBAR - ABOUT ME - PHOTOS - RSS FEED - Atom
Church of the Churchless
Clear Mountain Zendo, Montclair
Fr. James S. Behrens, Monastery Photoblog
Of Particular Significance, Dr. Strassler's Physics Blog
My Cousin's 'Third Generation Family'
Weather Willy, NY Metro Area Weather Analysis
Spunkykitty's new Bunny Hopscotch; an indefatigable Aspie artist and now scolar!

Powered by WordPress