Tapes, Transcripts and Trials: The Routine Contamination of Police Interview Evidence
Haworth, The International Journal of Evidence & Proof, 2018
Haworth contends that the way police interviews are conducted, recorded, transcribed, and then used as evidence in court fundamentally alters the original content and meaning of the interview. This contaminated version, when used in court, is often to reinforce or contradict oral testimony based on the exact wording of the transcript. With the reliance of the prosecution on the written record, this contaminated evidence could influence jurors inappropriately or change the possible outcome of a case. Haworth’s UK study, where she compares and analyzes official police transcripts with her own transcripts, as well as a high profile case where both full audio recordings of the interview and the official transcript were made available, examines four areas common to the process of interview production that will result in errors in meaning or content; audibility, transcription, editing, and courtroom presentation.
Haworth notes that the original context of the interview can never fully be captured; the surroundings, environment, emotional content, and body language cues cannot really be articulated in an audio recording and properly portraying spoken language in a written format can be challenging. Editing for, and the actual production of, the transcript can also omit valuable information.
Audio recordings may be of overall poor quality, with some departments still using cassette recording equipment, and just through circumstances of the interview, some recorded portions may be inaudible. Haworth notes it is rare that transcripts indicate inaudible portions of the recording and references research where transcribers may just insert “what they thought they heard” in such instances, resulting in errors. Haworth also notes that some countries don’t even bother with audio recordings, relying instead on the interviewer recall in writing the interview summary, while some locations don’t provide an actual transcript but a monologue in first or third person that presents a synopsis of the interviewees’ information rather than a back and forth exchange.
The transcription process also is likely not to include the para and extralinguistic features people utilize in interpreting meaning in face to face interactions. Features like intonation, emphasis, breathiness, facial expressions, and body and head movements all provide details in face to face communication that help the viewer interpret truthfulness, emotions, and cognitive processes in the speaker but are typically lost in producing a transcription. The transcription process is also prone to “correcting” spoken word by omitting or changing speaking features like false starts, stammering, inarticulations, repetition, interruptions, and correcting sentence structure, syntax, and grammar to improve readability. However, she notes research has shown this correction is typically used for lawyers and police officers but not for witnesses and suspects. The author’s interviews with transcription clerks in her study indicate there is no transcription training and no formalization in transcribing these features, with each transcriber developing there own personal style to the transcription process. But Haworth notes that pauses convey important meaning in communication which isn’t imparted to the reader or listener. Non-response to questions were also handled differently with some transcribers, indicating everything from no indication in the transcript to transcriptions like “no comment” “no audible reply”, “no response”, “defendant remained silent”, or “defendant refused to answer” and each of the indicators present a slightly different interpretation of the suspects’ reaction following questions.
The transcription process of a police interview also involves substantial editing and typically produces a summary, (for example, just a simple text statement indicating that the suspect was informed about the charges or circumstances and not transcribing the actual interaction) with only certain parts fully transcribed to ease time and resource costs. These civilian employees of the police department were given guidance sheets on different offenses so they knew what particular points or elements of the crime, known as “points to prove”, to focus on instead of full transcription. For example, in regards to theft, points to prove would involve showing dishonesty, appropriation, property, that belonged to another, with intent to permanently deprive. However this relies upon the individual transcribers being knowledgeable enough determine what is relevant enough to be transcribed. This transcription process is also geared toward prosecution. There is no indication that transcribers were given guidance to include full transcription in areas regarding duress, legal justifications, or exculpatory admissions. In summarizing interactions, transcribers can also inadvertently introduce value judgments or interpretations based on their perspective. For example, in summarizing an interaction transcribers can indicate behavioral motivations that are just assumed but not actually stated, or fail to transcribe communication indicating the detailed information and emotional content of responses that removes the tone and attitude of an interaction.
The transcript is typical the only evidence of the interview used in court as opposed to the audio recording of the interview. This evidence presentation system is further contaminated as transcripts are not provided to the jurors but are typically presented orally by prosecutors and police witnesses, thus adding another layer of interpretation to a document that had already gone through interpretive changes. This oral presentation allows for the presenter to produce their own intonation, emotion, inflections, and pauses, and either purposely, or inadvertently, present this evidence with a bias toward producing a conviction. Even when jurors are provided with a copy they are likely to be highly influenced by the oral court presentation which may be biased toward the prosecution. Haworth noted in a particularly high-profile case, that punctuation errors in transcription can change the meaning of the words related orally and that mistakes in reading the transcript aloud or prosecutors attempts to “correctly” use the right intonation or emphasis change the meaning of the transcript presented.
To eliminate or mitigate these problems, Haworth suggests that departments move to digital audio recording as well as incorporate video recording to capture the all important context. She also suggests, for use in legal purposes, the audio recording should be preferentially referred to over the transcript. The transcription process should be standardized with transcription marks indicating emphasis, the length of a pause, and interruption notations. Standardized training in transcription and understanding spoken and written word features is also recommend as is ongoing training in regard to legal changes and how if may affect their transcription process. These recommendation will bolster the evidentiary value of transcripts for their use in legal proceedings by providing more accurate and useful evidence.
Police Use of Force and Injury: Multilevel Predictors of Physical Harm to Subjects and Officers
Hickman, Strote, Scales, Parkin, & Collins, Police Quarterly, 2020
With closer scrutiny on police use of force, the authors sought to make determinations on the factors that lead to increased likelihood of injury to subjects and officers in police/subject encounters at an individual and agency level. They recognize that the police in order to do their job must have the ability to use coercive force in order to gain compliance or effect arrests and that with the use of weapons, injury to either subject or officer, will be inevitable in some of the encounters.
The authors hypothesized that demographically, females and Blacks will have a lower likelihood of injury and that as age increased for subjects their risk of injury will increase, but as officer age increases, their risk of injury will decrease. They also hypothesized that when mental illness or drugs/alcohol were involved, the likelihood of injury to both officer and subject will be higher. Resistance and weapon use will likely increase injury to subjects but not officers but the greater numbers of officers on scene increase the risk of injury to both subjects and officers. The duration of the incident is also known to be related to injury, and they anticipate the longer force is used, the likelihood of injury to both subjects and officers will also increase. The authors also hypothesized that officer knowledge of more serious underlying offenses prior to the use of force, greater threats to officers/self/others, higher levels of subject resistance, warrants, and both attempted and actual flight, will result in higher likelihoods of injury to both subjects and officers. The authors also consider exploratory hypotheses that agency size and type may show some variation as well as the anecdotal hypotheses that agency and state variation on their training for use of force, and emphasis on different or preferred types of force, may also account for variation.
The authors used the Police Force Analysis System and Network, a regional database of law enforcement incident reports and officer narratives. It has served as a police management tool to assess trends and patterns in use of force as well as identify high-risk officers. The analysis utilized 10,564 subject-incidents drawn from 81 agencies across eight states. Most were small to medium municipal agencies serving small to medium towns. Two thirds were from Washington, a fifth from Wisconsin, as well as some from California and a few other states. Their analyses denoted both descriptive statistics as well as using multivariate logistic regression to examine the incidence and risk of injury to subjects and officers.
Across all agencies and incidents, in use of force incidents, subjects were injured 52% of the time and officers were injured 16% of the time, similar to previous research rates. The large majority of subject injuries were minor with, cuts, scrape, bruises, punctures, and Taser probes making up 79% of the injuries while gun and knife wounds and fatalities made up only .7% each of the injuries. 49% received hospital treatment and 35% were treated at the scene, while 16% refused or not offered treatment. While the race of subjects in use of force incidents involved 44% White, 27% Black and 22% Hispanic, as well as other races, the injury rate ranged from 47.5% for Blacks and 50% for Whites to 57% for Hispanics and higher than 60% for Asians and Native Americans.
Injury rates were also higher for subjects and officers if mental illness, drugs/alcohol, prior knowledge of past incidents, and warrants were involved. The severity or seriousness of the investigated offense was also positively correlated with injury rates for subject ranging from 35% to 65% while officer risk was mixed with the lowest rate for a violent offense with a weapon at 13% to the highest for disorderly conduct/drugs/trespassing at 18%. The risk of subject injury increased with severity of the threat the subject presented, ranging from verbal threat (39%) to deadly weapon (68%) but officer injury rates were also mixed with the lowest rate involving deadly weapon (11%) but the highest rate involving assault or self-harm, or less than lethal weapons (22%). The range of subject resistance to officers was also correlated with subject injury rates from 37% with no resistance to 75% to those who resist with a deadly weapon. Officer injury rates remained low in all but the top resistance categories with the highest in active physical resistance (31%) and less than lethal weapon (34%). Both attempted and actual flight from officers showed higher rates of injuries for both subjects, 49% and 59% respectively, and officers, 19% and 17% respectively
The greater the number of force sequences, or duration of the incident, also increased injury rates for both officers and subjects. Officer force factor (subject resistance level subtracted from officer force use) indicates that when resistance and officer force are equal (force factor=zero), subject injury rates were low but increased as force factor increased. Officer injuries rates decreased with higher force factors in use. However when force factors are negative, with subject resistance greater than officer force, risks of injury to both officers and subjects are increased. Weapons used is also indicated in injury risk. Two thirds of incidents were physical force only and 12% involved only a weapon. Physical force presented the lowest risk of injury for subjects while incidents that involved a weapon resulted in 75% of subjects injured. For officers, while injury rates were lowest when using a weapon only (3%) injuries rates increased to 15% when using physical force and were highest at 25% when using both physical force and a weapon. Tactics also affected injuries with subjects and officers receiving the most injuries during strikes, lateral neck restraints and wrestling. The risk of injury for subjects was high with impact weapons, like batons, being associated with a 71% injury rate and canines with a 97.5% injury rate, though officer injury rate was high with baton use and low with canines.
Injury rates for officers and subjects were both associated with agency size and jurisdiction size , with risk of injury increasing as the size increases. Rates for injuries of both subjects and officers was slightly higher in Sheriff’s Departments compared to municipal agencies. The authors, while noting the examination of regional differences was limited, found that Midwestern agencies (18, primarily Wisconsin) had lower subject and officer injury rates (20.5% and 10.5% respectively) than non-Midwestern agencies (63) rates (59 and 17 percent respectively).
The authors’ multivariate regression analyses examined both subject injury and officer injury models. The subject injury model predicted lower rates of injuries for female (compared to males) and Blacks (compared to Whites), while increasing age increases the risk of injury. While the level of threat presented was not significantly associated with risk of injury, level of resistance was, and showed an 188% odds increase in subject injury when a deadly weapon is involved, and a 132% odds increase when active physical resistance is used. Actual flight from officers also raised the odds of subject injury by 22%. Officer use of a weapon had a 342% odds risk increase for subject injury while those involving a weapon and physical force only increased the odds 181%. Each additional sequence (or length of force incident ) increased the risk of injury, so minimizing the length of incident will decrease injuries. However, they also found that officers’ increased level of force, which could end the incident more quickly, was also associated with increased subject injury. Subject injury odds were also greater in larger agencies’ but a decrease in odds if the agency is Midwestern.
For the officer injury model, demographics indicated females subjects and Asian and Native American were associated with decreased risk of officer injury, and risk of officer injury decreased with increase in subject age. While drug/trespass/disorderly crimes were significantly related to a 30 percent increase in officer injury odds, none of the threat or resistance levels were significantly related to officer injury odds. However attempted and actual flight significantly increased the odds of officer injury by 26 and 47 percent, respectively. Force incidents in which an officer only used a weapon was associated with a 78% decreased odds of officer injury as well as those involving alcohol or drugs. Duration and force factors were similar to subject injury analysis where decreased duration and higher force factors by officers were significantly associated with a decrease in odds of officer injury. Officer injuries were less likely in midsized agencies compared to small agencies and if the agency is Midwestern, the odds of officer injuries is 36% lower than in non-Midwestern agencies. The authors suggest that exploratory analysis of tactics used may contribute to additional knowledge on officer and subject injury as weapon use was not as common in Wisconsin agencies. The authors suggest that agencies could adopt more standardized training on use of force but ultimately the use of force will reflect community standards. Further research could focus on the systematic evaluation of training methods as well as other more specific incident variables that may also influence subject and officer injury outcomes.
Ambush Killings of the Police, 1970–2018: A Longitudinal Examination of the “War on Cops” Debate,
White, Police Quarterly, 2020
The author notes that a few high profile ambush attacks on the police in recent years, which have coincided with high profile deaths of minority suspects in police encounters, have led some observers to conclude there is a war on the police. He notes that although recent research shows significant declines in felonious killings of police officers since 1990, data from the FBI indicate an increasing percentage of those deaths are classified as ambush killings which the International Association of Chiefs of Police (IACP) describes by four characteristics: “an element of surprise; concealment of the assailant, their intentions, or weapon; suddenness of the attack; and a lack of provocation”. IACP data indicates that from 1990 to 2012, the proportion of ambush killings increased by 33%. Other research also indicates increases in ambushes, for example, in 2016, ambush attacks against police reached a 10-year high. The research also indicates varied motivation for attacks on police, from revenge for a perceived transgression to attacks that are designed to avoid arrest or facilitate escape. However, the “war on cops” thesis implies a very specific motivation for an ambush: hatred of police and/or desire to seek vengeance in response to police killings of citizens. They authors, utilizing data from the Officer Down Memorial site, examined felonious killings from 1970 to 2018 (n=3,379) and analyzed 913 ambush style attacks, to determine if they’re has been any trends in these killings. In a time series analysis, they examine two types of ambush; one (pre-encounter, n=206) making up 6.1% of all officer killings, where officers had no knowledge of the suspect prior to the attack, had no opportunity to anticipate the attack, and weren’t actively engaged in “police business (such as sitting in a car eating lunch or writing a report): and the other (anticipation, n=707) making up 20.9% of all officer killings, where officers were responding to a call or other police business and had a general knowledge of the suspect but no contact with him prior to the incident (for example searching an area for a suspicious person when the officer is surprise attacked). These trends were also compared to two types of felonious non-ambush killings of officers, one where an officer is attacked in the very initial stages of contact with a suspect, which made up 22% of all officer killings (n=745), and a second which makes up 50.9% of all officer killings, where the attack comes after the initial contact stage, for example during an interview with a subject (n=1,721). The authors also examined if the rate has changed significantly since 2013 to determine if there is support for the war on cops contention. Rates were standardized to ambushes per 100,000 officers to account for changes in officer population fluctuations and the quasi-experimental time series used to points of intervention. July 2013, date of the formation of Black Lives Matter following the acquittal of George Zimmerman in the Trayvon Martin case and August 2014, the month that Michael Brown was killed in Ferguson, MO and BLM was prominent in the media. (Author’s note: Both relatively early occurrences in the current focus on police brutality and racial justice).
The authors note that in 1970, anticipation ambushes rates (12 per 100k officers) were more than three times higher than that of pre-encounter but through the ’70’s and ’80’s anticipation ambushes dropped rapidly to between 1 and 2 officers per 100K and pre-encounter ambushes dropped as well from a high of almost 3 officers per 100k, to one and sometimes less than one officer per 100k through the ’90’s and beyond. While this rate of decline was statistically significant, the authors also found two 20 year high spikes in both 2011 and 2016 for anticipation ambushes and spikes in excess of 1997 rates in 2002, 2014, and 2016 for pre-encounter ambushes. Initial entry, which in general had a lower rate than information stage attacks, and information stage attacks showed a similar trend with sharp drops through the ’70’s and ’80’s, declining 79 to 88 percent. Like ambushes, there were also spikes relative to 1997. In 2005 and 2011 there were spikes in early stage attacks and spikes in 2002, 2004, 2008 and 2017 from later stage attacks, though there was also a sharp decline in 2013. The rates of the declines were also statistically significant. Examining the time series for the two types of ambushes separately and combined, neither showed any significant increases after 2013, indicating that neither the formation of Black Lives Matter or the death of Michael Brown had any effect on police ambushes. Nor is there support for a war on cops based on an overall downward trend. However, eight of the last 20 years studied showed spikes in excessive of the overall decline shown in 1997. These spikes over the period 2009 to 2013 and 2014 toward 2018 accounted for a 25% increase in pre-encounter ambushes, a 9% increase in anticipation ambushes and a 22% increase in later stage attacks. While the downward trend doesn’t suggest a chronic problem, spikes do warrant observation as they can predict the emergence of a chronic problem.
The authors do note limitations in the study include a lack of known motivation for the attacks, such as revenge, which would inform the hypothesis of a “war on cops”. Also, since non-fatal injuries were not included in the study, changes in rates of officer killing may also be a function of advancements in emergency medical response and treatment, police training, and equipment like body armor for officers and may mask increased rates of potentially deadly attacks on officers. (Author’s note: The study also didn’t examine non-injurious attacks on officers. For example, in 2020, the number of officers in Chicago who were shot doubled compared to 2019 but the number officers shot at increased three times, which also may be masking the effect of an increase in attacks on officers).
Police Ethics and Integrity: Keeping the ‘Blue Code’ of Silence
Westmarland and Conway, International Journal of Police Science and Management, 2020
The authors examined police ethics and integrity within a UK police force with a questionnaire presenting ten hypothetical ethical dilemma scenarios. An Us vs Them mentality, the blue code” and the “code of silence” as portrayed in other research suggests that police misconduct may be promoted or go unreported as officers seek to protect their own.. Other research has examined this issue and the current study includes not only sworn police officers but police community support officers, akin to US Community Service Officers, and police support staff. The authors analyzed 1,509 responses from a large non-metropolitan police organization that had both urban and rural policing areas. The survey asked respondents what offenses they found most serious as well as how likely they were to report various types of misdemeanors, what influenced their decision to do so, and their attitudes about reporting offenses. Officers were presented with ethically problematic policing behavior in ten scenarios and, on a 5 point scale, asked to rated them in seriousness and how likely they were to report the behavior, with some scenarios querying whether the officer believed it was against departmental policy and how the officer would go about reporting the offense. Officers were also asked how they felt working with a whistle blower and the degree of confidence in the department’s anonymous messenger system.
Four of the scenarios were regarded as only minor infringements (running a side business unrelated to police work, taking reportedly legal bodybuilding supplements, addressing a disturbance at a party for police staff, and an investigating officer having a romantic relationship with a former crime victim) and were very likely to go unreported, likely because they involved off premise and off duty behavior. Two scenarios were considered major infringements, one involving an officer keeping an expensive watch that was inadvertently received by him, and the other involved an officer taking a sum of cash from a criminal during a search. In regards to the watch, while 85% felt that it was a serious or very serious infringement fewer than half of the respondents stated they would be highly likely to report it, with the authors postulating that keeping the watch may not necessarily be perceived as theft. However, when it came to a clear issue of theft in taking cash from a suspect, 97% responded it was a very serious infringements with 95% of respondents indicating they would be highly likely to report it. Some of the other scenarios also showed some disjuncture between the stated serious and likelihood of reporting. While accessing confidential police data for personal use was considered at the highest level of seriousness by 83% of respondents, only 69% reported they would be highly likely to report the infringement. This was the largest gap evidenced between seriousness levels and other scenarios had notable gaps as well. Witnessing sexual harassment among co-workers was judged as serious to very serious by 93% of respondents though 85% indicated they would be prepared to or likely to report the incident, with a number of responses wanting to gain more information about what they witnessed before reporting. However, opinions were split on the appropriateness of engaging in a romantic relationship with a crime victim by the investigating officer with 39% saying it was acceptable, 26% saying it was inacceptable, while 34% were unsure. While the average response as to whether they would report it was “not very likely”, 40% said they were unlikely to report and only 11% said they were highly likely to report the relationship. Two of the scenarios were regarded as potentially harmful by the researchers, applying extra physical force to a captured fleeing suspect and covering for an off duty officer driving drunk and accordingly, officers viewed these scenarios as very serious (83 and 88 percent respectively) as well most officers being very likely to report the incidents (74 and 77 percent respectively).
The researchers also explored officers’ attitudes about working with whistleblowers and the anonymous messaging system. In regards to the whistleblower question, respondents had the options of choosing more than one response and while some officers had reservation about not wanting to work with the officer, or feeling they couldn’t trust the officer, 42% of officers said they would be on guard around the officer. However two thirds of respondents indicated they would have no reservations working with a whistleblower resulting in some difficulty reconciling officers who both have no reservations about working with a whistle blower, yet feeling they needed to be on guard, or would be mistrustful. The authors suggest that is suggestive of officers commitment to “adhere to formal rules, regulations and ethical principles, while simultaneously recognizing the existence of normative standards which may technically qualify as an infringement”. Confidence in the anonymous messaging system was not high, with respondents’ confidence split mostly evenly between the five levels of confidence, with 17% expressing no confidence to 18% being highly confident with a number of respondents questioning how it could maintain anonymity.
The authors conclude that while research does provide indications of in-group secrecy and solidarity, other research indicates that while new recruits value teamwork and comradeship, the code of silence has been overwritten by the “code of self-protection”. As departments become more accountable to the public and risk aversive, officers fear doing the “wrong thing” more than self-isolation that might come with whistleblowing. The authors’ results indicate that there is some evidence of the blue code with the average gap between level of seriousness and likelihood of reporting at about 10 points, though one departure from this was the high level of very serious assessment of accessing police data but the lower likelihood of reporting it. The authors also note however, that the excessive force and DUI scenarios, which were used in previous research in 2005, showed an over 20% increase in the “highly likely to report” category for both scenarios, suggesting a move toward doing the “right thing” over maintaining a blue code of silence.
A Sign of the Crimes: Examining Officers’ Identification of, and Arrest for, Stalking in Domestic Violence Complaints
Brady, Reyns, & Dreke, Police Quarterly, 2020
The authors state that despite stalking as a risk factor for intimate partner homicide, little research has explored the officer decision making process in domestic violence (DV) complaints that involve stalking. Some research on domestic violence indicates that officers may minimize stalking behavior or fail to recognize stalking behavior as actual stalking. Officers have indicated the complexity of the statues, and the need to establish a pattern of behavior that elicits an emotional reaction from the victim, can make investigation and prosecution difficult. To examine the issue, the authors utilize a focal concerns perspective. Accordingly, “criminal justice actors weigh three main considerations when faced with a decision—the blameworthiness of the offender, protection of the community, and any practical constraints or consequences associated with different courses of action or inaction… theoretically, greater criminal justice action (e.g., arrest, charging, sentencing) is positively related to blameworthiness and the need to protect the community, whereas the effects of practical constraints and consequences will vary depending on the organization and how the concept is operationalized”.
The authors’ study examined five years’ worth (230 cases) of intimate partner complaints that met the legal standard of stalking where officers either did or did not identify stalking behaviors in domestic violence complaints, in order to explore legal and extralegal factors in officers’ decisions to arrest suspects for intimate partner stalking. The authors examined three research questions; to what extent are suspects arrested for stalking in DV complaints, which legal and extralegal factors figure into officers’ identification of stalking behavior in DV complaints, and which of those factors are associated with officers making an arrest versus writing an incident report.
Outcome variables included whether officers acknowledged stalking behavior that is, whether the officer indicated that stalking was an observed offense citied in the incident reports or whether the officer articulated stalking behavior but did not identify a specific charge of stalking in the incident report and whether, in the cases where stalking was cited in the incident report, an arrest was made or if only an incident report was made. For the independent variables, in order to examine the focal concerns perspective, the authors operationalized blameworthiness as whether the victim was threatened, total number of offenses articulated in the report, and whether the suspect had violated an active protection order. Community protection (as gauged on the dangerousness to community or victims) was operationalized as whether the subject had been incarcerated, whether they had committed previous physical assaults, and whether the location of the incident was private or public. Practical constraints and consequences included six variables: whether there had been prior police responses involving the parties, whether there was physical evidence, the presence of witnesses, whether the victim or someone else contacted police, whether the victim exhibited fear, and whether the victim was cooperative. The authors also examined the perceptual shorthand used by officers, the cognitive schemas of credibility, dangerousness, cooperation, and risk of recidivism that come from contact with other officers and work experiences that may be influenced by the current relationship status of the parties, age, race, and ethnicity.
Results showed that officers identified stalking behaviors in less than half of DV complaints reported to RI police departments from 2001 to 2005. Of the 94 police-identified stalking cases, however, the majority resulted in an arrest (61.7%). Of the 141 total cases that resulted in an arrest (61.3%), 41.1% were specifically for stalking and 58.9% were for a non-stalking DV-related offense, meaning stalking arrests accounted for 25.2% of all DV cases reported. Officers were significantly more likely to identify stalking behaviors if subjects had committed a greater number of offenses, if the incident occurred in public and if there had been history of prior police responses involving the parties. Officers were significantly more likely to not specifically identify stalking behavior (classifying the incident as a non-stalking DV complaint) when the victim was the complainant, was threatened, or reported prior physical assault by the subject. The logistic regression model isolated the key factors in officer decision-making. Incidents that occurred in public versus a private location were nearly 4 times more likely to be identified. Officers were also two times more likely to identify stalking behaviors if the suspect had committed multiple offense and almost three times more likely if there were prior police responses to previous complaints. Even though officers articulated stalking behavior in the report itself, officers were significantly more likely to classify incidents as a non-stalking DV-related offense when the suspect threatened the victim (OR .41) or the victim had alleged prior physical assaults by the suspect (OR .39). None of the other factors such as the suspect’s relationship to the victim, age, race, prior criminal history, victim fear, who called the police, the availability of witnesses, physical evidence, victim fear, and the presence of an active protection order significantly influenced officers’ identification of stalking. The logistic regression model also examined the significant variables affecting officers decision to arrest versus generate a report in cases they identified as involving stalking but the only factor that had a significant influence on that decision was whether the victim is deemed cooperative by submitting a written statement. No other independent or demographic variable influenced the decision to arrest.
The authors discuss that the results may explain some officer decision-making processes. In some ways, officers may treat stalking as an “add-on” offense, something to be added when multiple charges are present but not warranting a sole charge, and may be contingent on the other two factors; public incident and prior police involvement as this may signal the relentlessness of the subject which supports the need to establish a pattern of behavior, and public pursuit may provide the opportunity for more corroborating evidence like witnesses and surveillance footage. While some of the variables fall into the scope of focal concerns in identifying stalking, the authors conclude the perspective is not useful in determining the factors related to establishing an arrest in stalking cases, as the sole, and strong, factor influencing this was the victim’s willingness to cooperate by filing a written statement. This cooperation has been shown in previous criminological research to be positively correlated with arrest in other areas as well.
The authors suggest that training and standard operating procedures be developed and implemented that help officers understand the stalking statutes, and identify stalking behavior, by having them inquire specifically about stalking and the presence of fear to establish probable cause. Officers also need to know and accept the illegality of stalking, instead as just a component of the power and control dynamic present in these types of domestic relationships, thus providing better support for victims.
Attitudinal Changes Toward Body-Worn Cameras: Perceptions of Cameras, Organizational Justice, and Procedural Justice Among Volunteer and Mandated Officers
Huff, Katz, Webb, and Hedberg, Police Quarterly, 2020
The authors examine some issues involving body worn cameras (BWCs) by using a recently implemented BWC mandate in the Phoenix, AZ Police Department to determine the attitudes surrounding camera use and its effect on perceived organizational and procedural justice. They note that previous research on BWCs has yielded a variety of results. There are implications that BWCs could have a deterrent effect on officers, limiting their discretion in citizen encounters and making them more mindful of proper behavior and procedures, but that deterrent effect may not be equally found in all officers.
Many of the previous studies focused on attitudes prior to the implementation of a BWC mandate but did not demonstrate how the attitudes abut BWCs may have changed following the mandates. Some research indicated officers believed BWCs would limit their discretion, would increase adherence to departmental policies, and that it may not affect their behavior to a great degree but that it could have a positive effect on citizen behavior, and could assist in report writing and producing better evidence for prosecution.
However, results after BWC adoption have been mixed. UK officers believed that the cameras had a positive impact in a number of areas and was a means of improving officer effectiveness, but in the US, where adoption has met with more resistance, the suggestion is BWCs were to improve problematic officer behavior, with some officers believing they were necessary as the word of a police officer is not trusted in court anymore. Patrol officers versus those in special units also had differing needs and concerns surround BWCs but the findings generally suggest that officers in departments that have already deployed BWCs have a more positive perception of the technology than those where it had yet to be implemented. Studies that have examined officer attitudes toward BWCs prior to, and following, the adoption of BWCs have largely found that officers become more favorable or remain neutral toward BWCs after BWC programs are implemented. They also found officers believed, as the programs went on, that they improved officers’ ability to protect themselves against citizen complaints and that they improved citizens willingness to talk to officers. However, some studies have found evidence that after implementation officers had no change, or a negative change in attitude to BWCs, believing that they limit officer discretion, and force them to act more legalistic than informal in dealing with citizens, but sometimes these individual attitudes depended on the attitudes of officers in their social network as well.
The authors state that two areas that will be affected by BWCs and a possible deterrent effect are perceptions of organizational justice and procedural justice. Perceptions of organizational justice pertain to officers’ feeling of fairness in implementing, utilizing, and monitoring a BWC program, and the outcomes associated with them, which would be associated with officers’ job performance and commitment to the organization. Research has indicated that when officers have concerns about BWCs generating public disapproval and perceive high monitoring of BWC it can affect their emotional well-being, and when they lack organizational support for BWC officers, has also been linked to burnout. When officers felt their department was more just and fair in its monitoring of BWCs they had more positive perceptions of BWCs. Officers’ positive perceptions and support of BWCs were also linked to perceptions of a strong trust relationship with their supervisor. Research also indicated that officers were concerned about how the footage would be reviewed and used, fearing it would be used to hold officers accountable and get them in trouble over minor issues, and thus limit their proactive contacts.
The authors also suggest that BWCs can enhance procedural justice during citizen encounters. “Citizens have higher perceptions of procedural justice when they feel they were allowed to contribute to the encounter, when the officer used objective criteria to make decisions, when the citizen felt they were treated with dignity and respect, and when the citizen trusted the officers’ motives in the interaction” and when citizens perceive more procedural justice they are more likely to be compliant and cooperative. Supervisors have the opportunity to monitor or review BWCs to ensure officers are engaging in procedurally just methods. However, research results are mixed with some officers feeling BWCs made them more professional and patient, while others felt it was stifling and limited establishing rapport, while some research found no effect between establishing procedural justice and BWC usage. Associated reductions in complaints with BWC officers may stem from the appearance of more procedurally just behavior as officers narrate their decision making to the camera as well as citizens.
The authors conducted a survey of 467 officers on their attitudes about BWCs in a pre-test phase before the research entered into the BWC wearing phase. A random sample of 177 survey participants was drawn and three groups were established; a group that declined to volunteer to wear a BWC (resistors, n=96), a group that volunteered to wear the cameras (volunteers, n=47) and a third group who were mandated to wear the cameras (mandated, n=34), A random sample of 110 officers were also selected from the pre-test group that had no involvement in the BWC program to serve as a control group. After 6 months, of the 287 officers in the BWC study 237 completed the post-test survey. The survey, identical to the pre-test, contained items related to officer efficacy, officer behavior, citizen reaction, general perception, overall recommendation, and organizational and procedural justice.
Overall, the researchers “identified few statistically significant and only small substantively meaningful changes in officer perceptions of BWCs, organizational justice, and procedural justice over time.” Relative to the control group, BWC officers had more negative perceptions about officer efficacy (i.e. having a more accurate account of the case, obtaining high quality evidence, or assisting in the prosecution of cases), though the authors suggest this may stem from officers not being privy to the benefits derived in the courtroom as research has shown that BWCs were more likely to result in charging, conviction, and a more punitive sentence. Positively though, relative to the control group, volunteer officers showed small declines in their perceptions that BWCs may negatively affect police officer behavior, being less likely to believe that BWCs inhibit contacts with citizens, cause hesitation in making decisions, and feeling they had less discretion. However, both mandated and volunteer officers had lower levels of agreement than the control group of officers that BWCs would increase citizen cooperation, increase citizen respect, decrease citizen resistance, and decrease citizen aggression, mirroring prior research. The authors suggest that a lack of observable effect in this area may stem from citizens not being aware of officers’ BWCs and thus citizens may be unlikely to change their behavior which increased these officers skepticism that BWCs affect citizen behavior positively, as compared to resisters and the control group which did not vary in their views on this issue. Prior research had indicated when citizens were aware of the camera their behavior changed and the authors suggest this result might inform policy issues regarding informing citizens that a BWC is in use. Compared to the control group, mandated officers were also less likely to have positive general perceptions following their field use. These officers were less likely to agree that police and citizens benefit from BWC use, that BWCs are well received by coworkers, and that BWCs improve job satisfaction and performance, training, and officer safety. Similarly, mandated officers reported more negative overall recommendations for BWC use, being less likely to recommend BWCs to other departments or other officers in their department, however the officers mandated wearing status may have generated feelings of loss of autonomy, and affected those results, the researchers surmised.
In contrast to other research, and in contrast to the control group, resistors, mandated, and volunteer officers all held higher perceptions of organizational justice (support and fairness in implementation, monitoring, and outcomes). The authors note this interesting result in comparison to the other study results which indicate BWC wearers did not see them as having an effect on citizen behavior and were less likely to recommend implementation. The researchers suggest that BWCs represent a supervisory program and with being given the opportunity to wear a BWC, perceived a greater capacity in the department for organizational justice. In contrast, there is no significant difference among any of the groups on BWCs prompting more procedural justice behavior. The authors note that the relatively small changes in officer attitudes, regardless of exposure to BWCs, suggests that “efforts to increase officer support for BWCs should be made early in the BWC adoption process. Our results, combined with prior research, highlight the importance of a communication strategy that disseminates information about the benefits and limitations of BWCs prior to their deployment so that officers buy-in to their agency’s BWC program, rather than resist its implementation.”
