An important new study examines officer-involved shootings from a different perspective, focusing not on what police bring to these encounters but on certain behavioral characteristics of the people they most often use deadly force against.
The research, based on the shooting experiences of one large sheriff’s department in California, shows that subjects who are under the influence of drugs or alcohol and/or have a history of violence are far more likely to be on the receiving end of police gunfire.
Specifically, among subjects the sheriff’s personnel responded to with deadly force, those under the influence of drugs were 3 times more likely to be shot or shot at by officers than those who weren’t; intoxicated suspects 3.4 times more likely than those who were sober; and people with previous arrests for violent crimes 3.7 times more likely than those without that history.
“This is the first major study of its kind,” says Dr. Bill Lewinski, executive director of the Force Science Research Center at Minnesota State University-Mankato. “It supplies really important data that will help us more clearly understand the dynamics of force interactions. The more we know about the factors involved in these confrontations, the better we can help officers face the challenges that arise out of them.”
“Most research on police use of force fails to look at the suspect’s actions or behavior,” writes Lt. James McElvain of the Riverside County (CA) Sheriff’s Dept., who conducted the study.
Typically, studies on police shootings explore their frequency, the impact of policy, the officers’ decision-making, and the race or ethnicity of the cops and suspects involved. Also typically they refer to the subjects who get shot in these encounters as the “victims.”
One prominent academic researcher has gone so far as to conclude that in cases where the legitimacy of force is challenged, “it appears that in every instance harm could have been averted by exercising some other options.” In other words, better policies and officer decisions could prevent police shootings.
This approach, McElvain notes, “overlooks the fact that the citizen also is making decisions that lead up to the point at which the officer fires his or her weapon.”
Lewinski agrees that past deadly force research too often has reflected “a biased view and doesn’t give us a clear picture of the encounter. In reality, it is very clear from most investigations, grand jury proceedings, review board hearings, trials and so on that most officer-involved shootings in the U.S. are fully justified and result from the officer shooting in self defense because he or she is victimized by an actual or threatened assault by the subject.”
McElvain’s study, titled “Shots Fired: An Examination of Police Shootings and Citizen Behaviors,” was successfully submitted last December as his dissertation for a PhD in sociology from the University of California-Riverside.
McElvain, 42, now a patrol lieutenant with 21 years’ experience in law enforcement, has not personally been involved in using deadly force against a human subject, but he has investigated police shootings in a previous assignment with internal affairs. During the course work toward his degree, he took a class on alcohol, drugs, and violence and, reflecting on his investigative experiences, began to wonder what role these factors might play in officer confrontations.
“I grabbed 5 years’ of data from records at the Sheriff’s Dept. and did a quick calculation of percentages,” he told Force Science News recently. He found that about 70% of the civilians in officer shootings were under some kind of chemical influence.”
With the approval and encouragement of Sheriff Bob Doyle, he ended up examining 15 years’ of data–all instances of officers on the department delivering gunfire at human beings from 1990-2004, including toxicological reports and criminal histories. In all, he analyzed 186 shootings, involving 314 officers and 190 civilians. (The agency currently has some 1,200 sworn personnel on the street and polices a socio-economically diverse population of more than 500,000.)
Each element of McElvain’s study–drugs, alcohol, and violent background–showed a significantly higher correlation with being shot or shot at by the police when measured independently against subjects of shootings who did not have those characteristics. “In combination,” he found, “citizens with prior violent criminal arrest records and who are under the influence of an intoxicant provide the strongest association with police shootings.”
These correlations proved to be far more significant than race or gender on either side of the shooting relationship, McElvain reports.
His findings do not surprise him, McElvain says. Obviously both alcohol and drugs can “disinhibit a person from coherent thinking,” and if not spur aggressive behavior at least contribute to noncompliance that “an officer can interpret as a threat to his/her immediate safety or that of another.” Sober or drug-free, the subject might “have realized the grave circumstances he/she was creating, and in turn, cooperated with the officer, which would have prevented the shooting.
“Arguably, a person who engages in criminal conduct as a matter of routine and is comfortable with using violence as a means to further his/her activities is also less likely to be intimidated by the police when confronted.”
McElvain’s research is complemented by an FBI study recently published by the federal Bureau of Justice Assistance under the title “Violent Encounters.” This study, by Dr. Anthony Pinizzotto, Edward Davis, and Charles Miller III, analyzed 40 attacks by 43 offenders on 50 officers.
About 35% of those offenders reported using alcohol within 2 hours before committing their assaults; in fact, they had consumed an average of 10 drinks each in that time period. More than 75% said they routinely used illicit drugs, on average twice a week; nearly half had used drugs within 2 hours before assaulting an officer. Of 13 gang members included in that study, only 1 indicated no alcohol or drug use prior to the incident being evaluated, and this was a regular drug and alcohol user who didn’t abuse substances as usual that day because he wanted to be “sharp” while robbing a bank.
A significant portion of the offenders in the FBI study had a history of committing violent crimes, including prior assaults on LEOs.
“Both these studies,” says Lewinski, “show that officers in deadly force situations are commonly dealing with individuals who are very difficult to deal with. The challenge is to try to come up with things that can help officers ‘read’ these situations more quickly and then influence subjects who we know can be only minimally influenced at best to reduce their threatening behavior.
“More research will be necessary before effective training methods can be established, but these studies are major steps in broadening our understanding of the dynamics of dangerous encounters. They also can help the civilian community understand how complex and difficult force confrontations can be.”
McElvain sees the possibility of some immediate practical applications of his findings. For example, “If we can identify citizens who are under the influence and have a history of violence, we may be able to approach them differently,” he told Force Science News. “It may be helpful in those instances to get a second officer on the scene, armed with less-lethal force.”
Dispatchers can play a vital role in conveying important information by probing complainants about the sobriety status of suspects and by running record checks on criminal history and prior contacts when an offender’s name is known, he says.
Advanced training programs may also be able to help officers better pick up cues to an offender’s mental state. “But when you talk about training, you’re talking about money,” he says. In agencies where armed encounters are rare, administrators may not feel this problem represents a training priority.
Lewinski points out, however, “If we can’t figure out better ways for officers to deal with drunk, drugged, and violence-prone subjects, it not only is going to be dangerous for those citizens but also for officers who are victimized by the subjects’ impulsiveness and altered state.”
Meanwhile, McElvain has plans to mine his research database for more fresh findings. Among other things, he is currently exploring how officers’ education, age, military experience, gender, race, and prior shooting involvement may correlate to uses of deadly force, and he wants to map out how police shootings relate to neighborhood types. “I think there are 5 or 6 different studies to come off of this data,” he predicts.
[Our thanks to Tom Aveni, a member of FSRC’s Technical Advisory Board, for alerting us to Lt. McElvain’s research project.]