In a previous post, I discussed the difference between firearm and violence skill. I want to expand on this topic here. The driver is the article “Hitting (or missing) the mark: An examination of police shooting accuracy in officer-involved shooting incidents,” for which the citation is at the bottom. This document costs money to download, so I’ll summarize the basics here. Also, I would like to put in my two cents on variables I believe are missed in these studies. Lastly, I want to explore the ramifications for those who carry concealed firearms (CCF).
For those interested in more detail, please download the study for your review. The study has an excellent bibliography for future research as well.
Background
Before we get into the study itself, I will selectively quote from it to establish the background of police training and accuracy with a firearm in Officer-Involved Shootings (OIS).
- “Vila and Morrison (1994) note that although the amount – and quality – of firearms training received by officers over the last century has increased considerably, there appears to have been little improvement in shooting accuracy.”
- “[…] [B]elow-average police shooting accuracy has been consistently established in the literature […]”
In short, despite significant improvements in the quality of firearms training, there has been no measurable improvement in police skill when discharging their firearm in an OIS. Unfortunately, police are often below average, sometimes significantly so. This fact is very troubling, especially for civilians who CCF because the police will train more and have more sophisticated training than civilians. If police are inaccurate after such training, civilians will likely be as well.
Study Results
Here are a couple of notable quotes from the study.
- “Therefore, while the [Dallas] police were accurate in 54 percent of shooting incidents, they were only accurate 35 percent of the time with respect to the total number of rounds discharged.”
- “[…] [Dallas] officers were less accurate when discharging their firearm at non-white suspects, during nighttime hours and when the suspect’s weapon increased in dangerousness. Additionally, in analyzing hit rates, an increased level of visual threat increased the shooting accuracy.”
I must confess to being a little confused why there is so much emphasis on race and ethnicity over other variables. For example, Chinese people may be a minority in America, but they certainly are not when considering the world’s population. Chinese people may be well represented in California and even better in San Francisco. Further, there are certain areas in San Francisco where they are no doubt the majority. If the OIS occurred in such an area, what is the point of saying listing race and ethnicity (i.e., non-white, minority) if we aren’t going to provide a critical context? I would much rather see the study describe a variable that indicated whether the suspect was moving or stationary when the firearm was discharged. That variable is way more important to understand. It may be that race and ethnicity correlate to a suspect moving or not, but who knows for sure. I don’t mean this as a criticism of this study but around the gathering of data.
In the study, the number of OIS incidents is much higher at night, at around two-thirds, compared to during the day. I certainly feel that night shooting is much more difficult; the study reflects that difference. This fact is also common sense. Certainly, distance plays a role as well, but it needed to be captured in the data.
Ultimately, if I were going to capture accuracy statistics about OIS incidents, I would require the following data:
- Distance to suspect. This metric can also be influenced by whether the suspect and officer were moving.
- Running, walking, or stationary suspect. This metric can be complex as the suspect may have started stationary, then ran. Alternatively, they may have run and then were cornered.
- Running, walking, or stationary officer. Furthermore, was the officer tired or exhausted from chasing a suspect when they tried to shoot.
- Day or night.
- Presence of innocents.
- Officer’s weapon type.
- Shots hit / shots fired.
- Previous Accuracy Standard (PAS).
- Violence Skill Factor (VSF).
I’ll explain PAS and VSF in the next section.
Hidden Factors Hypothesis
In some studies, there is an attempt to capture an officer’s experience. However, I feel that there is more to it than that. Here is a quote from the study:
- “Although existing literature suggests that the likelihood of an officer discharging their firearm throughout their professional career is relatively low (e.g., Smith, 2004) […].”
If that is true, we have a situation where the OIS incidents may constantly involve officers who had previously never fired their gun at a suspect. This fact would be very similar for those who CCF. However, it is essential to know that there are officers who typically fire one round, hit the suspect, and never shoot again at the suspect. In short, we know that accuracy is obtainable in these situations.
What could account for this discrepancy? Well, I think there are two things:
- An officer isn’t going to be more accurate in a violent situation than they were in their firearms training class. We need to know how accurate they were before the OIS.
- How experienced an officer is at violence may be the issue. This experience is different from the number of years served.
The first one is just a matter of data collection. The second one is more complex. Here is how I would approach it.
- The number of arrests the officer has made (NA).
- The number of arrests where:
- The suspect was a previously arrested violent criminal. (VCA)
- The suspect resisted arrest (VC-RA)
- The officer used physical force to make the arrest. (VCRA-PF)
- The officer used weapon force to make the arrest. (VCRA-WF)
We could calculate the Violence Skill Factor using coefficients (CE) to weigh the categories properly. Here is an example formula:
- VSF = (VCA*CE1 + VCA-RA*CE2 + VCRA-PF*CE3 + VCRA-WF*CE4) / NA
I would also add: If VSF > 1, then VSF = 1.
Here is an example: NA = 500; VCA = 100; VC-RA = 50; VCRA-PF = 25; VCRA-WF=25; CE1 = 1; CE2=1.5; CE3 = 2 ; CE4 = 3.
- VSF = (100*1 + 50*1.5 + 25*2 + 25*3)/500 = 0.6
- 0.6 < 1, so VSF = 0.6
At this point, we will factor in the Previous Accuracy Standard (PAS) and try to get a metric for predicting the officer’s accuracy in an OIS.
- Prediction = PAS * VSF
For this example, we would expect the officer to be about 60 percent as accurate as they were in firearms training.
My idea here would be to work on what values a given police department would have for the coefficients based on historical data. Then, when a new OIS happens, evaluate the prediction against the actual. If we get a reasonable prediction rate, this information may help police departments. On the other hand, if no good predictions can be made, we are missing other factors, or perhaps handgun accuracy cannot be predicted for violent dynamic events.
Difficulties With CCF
Civilians who CCF will likely never discharge their weapons in a Defensive Gun Use (DGU) situation. If they do discharge it, it may be the only time they ever will in their lifetime. Accordingly, if it turns out VSF is relevant to police departments, it won’t be relevant to civilians who CCF. Civilians will not get the experience necessary to excel in these circumstances.
What if VSF is not relevant to police departments? I think it points to a general issue of handgun accuracy, violence, and stress. Handguns are hard to shoot accurately at a distance and at moving targets. Throwing in stress makes it worse. Alternative ways of shooting a handgun may be required to achieve decent accuracy. For example, quickly deployed to a handgun, a separate pistol brace could improve accuracy substantially, turning the pistol into a Short Barrel Rifle (SBR). These options may be more suited to civilians who CCF than police departments. Unfortunately, the unrelenting focus on speed must be overcome to at least try these options out.
Another consideration is that these factors do not matter for most civilian DGUs. As most DGUs don’t result in the firearm being discharged and when the firearm is discharged, our best study shows that the distance between the victim and offender is approximately arms length. In short, it may very well be another difference between police and civilians regarding violent encounters.
Conclusion
For people who believe that police and civilians who CCF have the same threat types, police shooting statistics are bad news. Police have been increasingly well-trained in firearms over the last several decades and experience violent encounters much more often than civilians. Yet, their accuracy in OIS incidents has remained dismal. If I were a betting man, I would bet that civilians against these same threats would be worse.
All is not lost for the police though. Along with properly gathering relevant data, there may be two hidden factors – previous marksmanship and violence experience – that when properly compensated for, may give us a better accuracy prediction rate. More research is needed.
Ultimately, for people who CCF, the combination stress due to violence and the difficulty of handguns may be too potent a combination to overcome. Being open to additional concealable accessories that increase the accuracy of handguns at the cost of speed may lead to better marksmanship. However, our best data shows that civilians aren’t facing the same threats as the police, and because of that, perhaps OIS incidents aren’t a reflection of civilian DGUs where the firearm is discharged.
Citation
Donner, C.M. and Popovich, N. (2019), “Hitting (or missing) the mark: An examination of police shooting accuracy in officer-involved shooting incidents,” Policing: An International Journal, Vol. 42 No. 3, pp. 474-489. https://doi.org/10.1108/PIJPSM-05-2018-0060