Dr. Eric Piza on using bodycam video to determine use of force predictors
"We did find that verbal antagonism was actually associated with a lower likelihood of force occurring during a police-civilian interaction."
Download this week's episode on Apple Podcasts, Amazon Music, Stitcher, Spotify or via RSS feed.
While the majority of police-civilian interactions resolve peacefully, a small number of situations end with use of force as police respond to subject resistance.
In this episode of Policing Matters, host Jim Dudley speaks with Dr. Eric Piza about his analysis of body-worn video to determine the factors that contribute to whether or not force is used during a police-civilian interaction.
Dr. Piza is a professor of criminology and criminal justice and director of crime analysis initiatives at Northeastern University. He served as the GIS Specialist for the Newark (New Jersey) Police Department where he was responsible for day-to-day crime analysis and program evaluation activities of the agency. A summary of his latest study, "Situational factors and police use of force across micro-time intervals: A video systematic social observation and panel regression analysis," can be found here.
Connect with and learn more about Dr. Piza's research here. A transcript of this episode is available below.
This episode of Policing Matters is sponsored by Utility. Utility provides a universe of intuitive solutions for effectively capturing, analyzing, managing, and sharing video evidence. Technologies include a variety of cameras, sensors, and devices, as well as situational awareness software solutions for law enforcement, first responders, transportation agencies, and utility providers. To learn more about Utility and its technology solutions, visit utility.com.
Jim Dudley: Welcome back. You are listening to Policing Matters on Police1. I'm your host, Jim Dudley. It is safe to say that the majority of police and civilian interactions end without the use of force, yet some situations may become violent with resisting subjects and result in police use of force. Today's guest set out to study video footage from body-worn cameras in order to see what contributed to situations that ended with force while other situations did not. Our guest today is Dr. Eric Piza. He is a professor of criminology and criminal justice and director of crime analysis initiatives at Northeastern University. Dr. Piza served as the GIS specialist of the Newark (New Jersey) Police Department responsible for the day-to-day crime analysis and program evaluation activities of the agency. He has a Ph.D. from Rutgers University and has published over 60 peer-reviewed journal articles and three books. Dr. Piza's research agenda focuses on the spatial analysis of crime patterns, evidence-based policing, crime control technology, and the integration of academic research and police practice. Welcome to the show, Dr. Piza.
Dr. Eric Piza: Jim, thank you for having me.
Jim Dudley: Well, you know, I found your articles fascinating. Talks a lot about situations before, during, and after police interactions primarily reviewing body-worn camera footage. Of course, we can only see it from the perspective of the officer to the subject. And so I want to talk to you a little bit about that. But also about the subject or offender in some cases what their behavior was observed that might have led to the interactions and use of force as well. You analyzed hundreds of hours of Newark Police Department body camera footage in order to try to understand what factors predict and when use of force occurs during officer civilian interactions. How were these incidents chosen? Were they randomly were they presented to you or did you just go through random footage?
Dr. Eric Piza: So this partnership started with the Newark Police Department soon after they started working with the Department of Justice on the mandated consent decree. One of their primary policies as part of the consent decree was to deploy cameras to all of their officers. So when my colleagues and I thought of this study we approached the Newark Police Department and explained what we were interested in doing. We explained how we thought it may benefit the agency. They agreed to partner with us. All of the incidents that we included in our analysis were all the use of force incidents captured on body cameras during the initial rollout of body cameras in Newark. So essentially everything that was captured in their entire entirety in the year 2018, which was the pilot phase for the agency.
Jim Dudley: So what spurred you in particular to conduct the study? What kind of hypothesis did you have in mind?
Dr. Eric Piza: What spurred the interest in this study was the fact that body cameras have become very widespread, very standard in policing and the entire public discussion around body cameras. And to be honest, a lot of the academic discussion around body cameras as well focused on whether or not they prevented negative outcomes, right? So do they prevent citizen complaints against officers? Do they prevent officer use of unnecessary force? And all of those are very important questions to ask. Right? So I have no problem with researchers looking at those outcomes. At the same time, my colleagues and I, when we thought about this, we realized that we really don't have a lot of good measures of police-citizen interactions in policing, right? We only have written reports or secondhand accounts of events. That's primarily how we learn about on-the-ground, face-to-face interactions between officers and civilians. So what we thought was that the body camera allowed us as researchers for the first time to really look at precisely what happened during police-citizen interactions. And we thought that we could measure quite a bit of very rich data from the video that we couldn't measure from any other data source. So that essentially motivated our interest in this topic.
Jim Dudley: In your research you describe the 91 officer-citizen encounters that you coded the content, cutting the footage into five-second intervals and recording different variables. These included calm commands, shouted commands, non-compliance by the citizen, verbally antagonistic behavior on the part of the citizen or the officer, and different degrees of use of force. What did you discover from these reviews?
Dr. Eric Piza: What we found by looking at the footage and breaking it down into those five-second intervals is that a number of situational factors do have a significant influence on whether and when police officers use physical for restoring police-citizen interactions in a certain respect. A lot of what we found echoes prior research in this area. So, for example, suspect resistance, suspects being in possession of a weapon – in particular firearms – that as you would expect, kind of increased the chance that force would occur. I think that's understandable, but what we also found was that a lot of police officer actions also were very predictive of exactly when during an interaction police used force. So, for example, force tend to happen quickly after police officers issue shout commands, when police officers use what we classified as verbally antagonistic behavior force. We found that when police officers adhered to what we refer to in the literature as procedural justice, that the time to force actually lengthened. In other words, when police officers did things like maintaining respectful tones throughout an encounter, explaining explicitly why individuals were being detained, and offering civilians the opportunities to ask questions about their detainment, when officers did those things, the need for force actually reduced during police-citizen encounters. Right? So essentially we were able to identify specific measures that we argue couldn't be measured in a valid way from existing police reports, and we're able to identify factors that influence physical force during police-citizen encounters,
Jim Dudley: You're well versed in police use of force, you know, about the legal conditions and descriptions. To a layman though, they might not understand that police sometimes get into a situation that's already accelerated, right? You already have an offender or a suspect who's already maybe assaulted someone or they're fleeing a crime or something like that. And when we talk about use of force in academic settings, I know sometimes we still hear people talking about a continuum of force when we know there's no continuum that officers go into a situation, apply whatever force is necessary, take whatever off the shelf works for that situation. Did you consider this in these interactions where, yeah, it would be great to be able to explain the situation to the offender, but sometimes that's just not possible?
Dr. Eric Piza: What we found was that in most cases officers actually did a decent job of adhering to the use of force continuum, right? So in most cases, physical force was preceded by some type of verbal command. In most cases, more severe levels of physical force was preceded by lower levels of force. So police officers in our study at least actually did seem to adhere to a continuum at the same time. And that's a great point you make, Jim. At the same time, what we found was that other factors may explain and highlight why strict adherence to a physical force continuum may not always be realistic in a lot of cases. Again suspects attempting to flee a situation. We found that when that happens, the likelihood of force occurring pretty quickly increases substantially. So in such a case where a suspect is trying to flee a scene of a crime, strict adherence to a continuum is not practically feasible. So you, you make a great point there. And generally speaking, our findings do support that perspective.
Jim Dudley: I've been to some conferences recently where there's the debate of whether or not an officer should use obscenities in commands. And I know there are some who feel very strongly about it. I know that there's at least one campaign that's called "swear not to swear." But it's to bring down you know, the level and bring up civility and bring up professionalism. Did you touch on that at all? That when obscenities are shouted, did they tend to calm the situation or did they accelerate it?
Dr. Eric Piza: In the most recent study that was published in "Criminology," we did find that verbal antagonism was actually associated with lower likelihood of force occurring during a police-citizen interaction, which was interesting, right? Because as you said, on one hand, theories like procedural justice suggest that officers should minimize their use of profanities as much as possible. On the other hand, we found that at least in our sample, the seconds after an aggressive verbal command was given, and profanities were captured that the likelihood of force actually went down. So what we posed in the discussion section of that article was, this is probably very context specific when we look at things like officer use of profanity.
So for example, if an officer pulls a car over and immediately starts dropping F-bombs on the driver, I think we could consider that as probably inappropriate action. On the other hand, we think that aggressive officer tones may be part of the continuum of force that we mentioned. So if essentially following a respectful shout command the need for physical force is still likely, I think we have to ask the question of maybe an officer being a little bit more verbally aggressive in such situations. And using profanities maybe is not the worst thing in the world if it actually prevents the need for physical force. Personally, I'd rather have an officer swear at me than punch me, kick me point his weapon at me. So again, we think this is really context specific. I think we all could agree there are situations where profanity would be highly inappropriate, but maybe there are situations where it is not as inappropriate, and frankly speaking, maybe there are some benefits if that actually de-escalates the situation.
Jim Dudley: How was the compliance factor in the outcomes of the officer's decision whether or not to use force?
Dr. Eric Piza: We found that non-compliance to officer commands was one of the top predictors of when force is more likely to occur, which again makes sense. Prior research suggests that citizen resistance, both physical and verbal resistance, is related to use of force and our findings support that perspective. So essentially non-compliance to verbal commands and also what we categorized as physical civilian resistance, were both highly related to the use of force during police-citizen encounters.
Jim Dudley: Yeah. Oftentimes in review especially by professional standards units or internal affairs or offices of citizens complaints that part is often discounted. That doesn't matter what the individual did or said, there are still procedures to follow before using force. Did your study come out with findings or recommendations to law agencies or investigative agencies that look at these kinds of situations?
Dr. Eric Piza: Yeah, certainly. So one of our recommendations here is for agencies to use body cameras more as a performance evaluation tool rather than just a tool to prevent negative outcomes. Because again, a lot of things that we're interested in having officers do, my co-authors and I argue that we really can't measure those any other way. So, as an example, a lot of police agencies require in their policies for police officers to announce to everyone on scene that they are indeed wearing a camera. I can't imagine any other way where we could actually measure officer compliance with that policy directive, right? And in this example, we were able to demonstrate, unfortunately, that only about 40% of incidents involved police officers explicitly announcing that they were wearing a camera. So essentially the good news though is that police supervisors, sergeants, lieutenants, etc., now are armed with the information needed to go back and reemphasize to the officers under their command exactly why that policy directed is so important, and to kind of support officers in maybe doing a little bit better job of making sure that they are indeed announcing the presence of a body camera when, when possible.
Jim Dudley: The study looked at encounters between December 2018 and 2019 well before the George Floyd incident and before de-escalation became a household word. Did your study help reshape the deescalation policy for the Newark Police Department?
Dr. Eric Piza: We believe it did. Again, this partnership started at least in part as a result of the Department of Justice consent decree in Newark. So as we were working on this, before any of our research results were published, they went directly to the police leadership in Newark. We had several meetings and presentations with the command staff over there as the project would go on. So they were always the first to know everything that we found. And to be frank, too, they also helped us as outside academics understand and interpret the results in a practical fashion. I know for a fact that the Newark Police Department has started to incorporate reviews of body camera footage. Essentially they're now randomly selecting body camera footage of specific events and looking into police officer interactions. I like to think at least that our study kind of helped motivate that policy decision in the agency.
Jim Dudley: I know we were offered free body cameras at the San Francisco Police Department as part of a study I believe it was 2010. And you know the predictions about the effectiveness were wild, right? That they were going to prove that the police were mistreating citizens across the board. And then the diametric opposition to that was it's going to show the police are doing everything, right? In general, what have you seen?
Dr. Eric Piza: In my experience, I think hesitancy oftentimes surfaces around body cameras from police officers for the first reason. Everyone kind of assumes that body cameras are being deployed as a gotcha, for lack of a better term. And then over time, in my experience, as officers see that perhaps an increase of complaints against officers are not substantiated because of the video evidence, they come around to actually appreciating the benefits of their body camera. One police officer I know from a mid-western agency actually told me that she loves her body camera so much, she wished she had a back camera so that all angles could be covered while she was out on patrol. From a larger perspective, though, I think this dichotomy of it's either going to record all bad things or it's going to show how great police are kind of misses the point of exactly why body cameras are beneficial, right?
To me, I don't think the footage can help inform only in cases where officers do something blatantly illegal or blatantly against policy. It could help inform officers how to make small adjustments or emphasize maybe some of the good things they're already doing in a way to mitigate the situational threats that we identified in our research as increasing the likelihood that force is going to be necessary. One analogy I like to draw is between how we use body camera footage and policing as compared to how football analysts use video of actual games. Essentially, when someone's trying to understand something about football they look at very specific things. They look at the routes that the wide receiver ran, they look at whether or not the quarterback did a sufficient job of looking at all of the different pass options. They look at how long the offensive lineman holds the block before moving upfield. If football coaches used their footage similar to how most police departments use body camera footage, all they would look for would be players either doing something blatantly illegal or making sure players don't step outta bounds all of the time. They won't care about any of those nuances, right? I think we're guilty of doing that too much as a field, right? We're reviewing body camera footage just to see if officers blatantly violated policy or to see if officers are doing something illegal. All of this nuance about exactly how to engage with the public about effective strategies in dealing with uncooperative suspects, about how to manage best large crowds that may gather around a potential use of force incident, all of those little perspectives and learning something from those little nuances is the real value of reviewing body camera footage and policing, in my opinion.
I'm a strong believer in having officers along with their supervisors regularly review their own body camera footage. And again, not for any discipline purposes. We have a perfectly legal event that with some small changes to how officers responded to situational dynamics could have been improved. So for that reason, I am an advocate of officers being given direct access to their own body camera footage so that they can learn a little bit about maybe how to adjust their techniques or demeanor during police-citizen interactions.
I realize I've been talking about footage review in the context of negative events, there's also probably a whole lot we can learn about positive events. Incidents where officers do successfully de-escalate a situation and avoid the need for force. Situations where officers do a good job of communicating with the public in a manner that does not agitate an already stressful situation. If we systematically look at those types of events, there are probably a lot of positive benefits and, and positive behaviors that we could highlight and start to distribute to the field a little bit more readily.
RATE AND REVIEW THE POLICING MATTERS PODCAST
Enjoying the show? Please take a moment to rate and review us on Apple Podcasts. Contact the Policing Matters team at email@example.com to share ideas, suggestions and feedback.