Harvard study finds NO bias against blacks in police shootings

[madoka]

i have often wondered if the race of the cops more closely matched the areas they patrolled, if it would have an effect in negating some of these issues

You might have a large increase in cop shootings.

Black and Hispanic police officers are more likely to fire a gun at blacks than white officers. This is according to a Department of Justice report in 2015 about the Philadelphia Police Department, and is further confirmed that by a study conducted University of Pennsylvania criminologist Greg Ridgeway in 2015 that determined black cops were 3.3 times more likely to fire a gun than other cops at a crime scene.

 

[woolfe9998]

It definitely contradicts prior research, as that research found that blacks were many times more likely to be shot by police. As for what is more accurate, each data set has its own problems. When you use things other than police reports you have a less comprehensive picture, but when you do use police reports you have an obviously self-interested actor who is likely to present a less than truthful account of the situation.

It's not so much a question of which is best, it is that together they provide a more complete picture. Before this latest study the evidence strongly suggested bias in police shootings. Now it is less clear if that's the case. One thing we shouldn't overlook however is that all the studies seem to agree that police disproportionately employ physical violence against black people, which is still horrifying.



And of course we frequently have 'armed and attacking' black suspects who turned out to be nothing of the sort. Both datasets have problems with reliability.

Yeah, it's a bit difficult to parse the data we have at this point. So for example, FBI crime data does show blacks are more likely to be shot when "not attacking."

http://www.vox.com/2016/7/11/12148452/police-shootings-racism-study

Yet, if I'm not mistaken, the FBI data is also taken from police reports, like this study.

Clearly, more data is needed.

 

[WHAMPOM]

Just one observation, is there one case of a cop obviously lying under oath and getting prosecuted for it?

 

[ivwshane]

Just one observation, is there one case of a cop obviously lying under oath and getting prosecuted for it?

Like this?

http://www.nbc-2.com/story/31998123/cape-cop-lies-under-oath-previous-cases-could-be-overturned

 

[werepossum]

I guess you don't follow your own thread.

It's not from Harvard. Its not a study.

You people cannot seriously be this stu-

Wait, what am I saying?

http://www.snopes.com/2016/07/15/harvard-study-officer-involved-shootings/

The article pertained to research published by Harvard University professor of Economics Roland G. Fryer, Jr. and was described by the Times as a "study." However, the Times linked to a "working paper" issued by the National Bureau of Economic Research (NBER), which is not a peer-reviewed journal but rather an economy-based research collective:

From the Times article:

“It is the most surprising result of my career,” said Roland G. Fryer Jr., the author of the study and a professor of economics at Harvard. The study examined more than 1,000 shootings in 10 major police departments, in Texas, Florida and California.

The working paper IS published research. There IS a study, whose results are summarized in the working paper. The STUDY ITSELF has not yet been published, and is presumably in peer review. (Possibly working toward that end, although Fryer is his follow-up certaily speaks of his study as finished.) The working paper is Fryer publishing a summary of his results. So given that it is a study by a Harvard professor, saying it is a Harvard study seems pretty reasonable.

The Times did a follow-up where Fryer answers questions raised.
http://www.nytimes.com/2016/07/13/u...r-questions-about-his-police-force-study.html

Yeah, it's a bit difficult to parse the data we have at this point. So for example, FBI crime data does show blacks are more likely to be shot when "not attacking."

http://www.vox.com/2016/7/11/12148452/police-shootings-racism-study

Yet, if I'm not mistaken, the FBI data is also taken from police reports, like this study.

Clearly, more data is needed.

I did not take it that this study indicates that blacks are not more likely to be shot when not attacking, but rather that when correcting for all possible factors, there was no bias in the shooting rates. Blacks are more likely to be shot when police are responding to a crime or when the shooting victim is belligerent; that does not indicate that they were attacking or even that the shooting is in any way justified. It only means that given identical circumstances and behavior, blacks are slightly less likely (but not statistically significantly so) to be shot when not attacking. Blacks can still be more likely to be shot without cause without invalidating this study.

Note that Fryer is hardly singing from the police hymnal. His study also found significantly rates of non-lethal force used against blacks than has previous research.

 

[woolfe9998]

I did not take it that this study indicates that blacks are not more likely to be shot when not attacking, but rather that when correcting for all possible factors, there was no bias in the shooting rates. Blacks are more likely to be shot when police are responding to a crime or when the shooting victim is belligerent; that does not indicate that they were attacking or even that the shooting is in any way justified. It only means that given identical circumstances and behavior, blacks are slightly less likely (but not statistically significantly so) to be shot when not attacking. Blacks can still be more likely to be shot without cause without invalidating this study.

Note that Fryer is hardly singing from the police hymnal. His study also found significantly rates of non-lethal force used against blacks than has previous research.

I'm not questioning the objectivity of the study at all. I think it's another valuable data point in assessing the issue. I am concerned about it being limited to 10 cities in 4 states, however. There is no way to know if this is truly a representative sampling, and even if a broader national survey would yield similar results, it still doesn't rule out regional problems, i.e. problems in a specific state or area of the country, or just in a specific police department. So I'm saying I'd like to see more data.

I don't know if the study being based on police reports is problematic or not. I suspect we'll have a better idea when we've seen more study using this data source and comparing them to studies based on other data sources, i.e media reports.

The publicity on this issue I think has been very inciting in many ways, because people react emotionally to these stories. The only silver lining I see is that all of this publicity is causing some people and organizations to study the issue more closely. If only people paid more attention to the studies and less to the individual stories but unfortunately studies are "boring."

 

[werepossum]

I'm not questioning the objectivity of the study at all. I think it's another valuable data point in assessing the issue. I am concerned about it being limited to 10 cities in 4 states, however. There is no way to know if this is truly a representative sampling, and even if a broader national survey would yield similar results, it still doesn't rule out regional problems, i.e. problems in a specific state or area of the country, or just in a specific police department. So I'm saying I'd like to see more data.

I don't know if the study being based on police reports is problematic or not. I suspect we'll have a better idea when we've seen more study using this data source and comparing them to studies based on other data sources, i.e media reports.

The publicity on this issue I think has been very inciting in many ways, because people react emotionally to these stories. The only silver lining I see is that all of this publicity is causing some people and organizations to study the issue more closely. If only people paid more attention to the studies and less to the individual stories but unfortunately studies are "boring."

I would venture to guess that limiting the study to ten cities certainly limits its applicability, since these large cities are more likely to be proactive about counseling and training to keep cops from wrongfully shooting black people. Nonetheless it is quite interesting.

Next step for me would be investigating why blacks have so many more interactions with law enforcement, even in areas where the police force is majority black. Are they truly aggressively looking for serious crimes, or merely being more sophisticated mega-Fergusons?