4
Training Methods and Delivery
While the previous chapter examined the knowledge and skills needed for police to promote the rule of law (ROL) and protect the population, in this chapter we address the second part of our task by answering the question, What is known about mechanisms (e.g., basic and continuing education or other capacity building programs) for developing the core skills needed for police to promote the ROL and protect the population? The committee sought to find answers to the “how to teach” questions that should be considered when developing police training. One of the papers commissioned by the committee for this report (Herold, 2021) demonstrated that little robust evidence is available on the effectiveness of different methods of police training in any context, let alone evidence on how effectiveness of the same methods might vary in different countries with different levels of prior education and experience.
The committee recognizes that there is additional need for more research and knowledge on the effectiveness of different police training methods. This need is just as important as the need to understand the effects on officer behavior of specific knowledge imparted by training (see Chapter 3), since what is taught and how it is taught are intricately linked.
This chapter examines both the training environment and instructors and draws attention to current challenges that exist with common forms of police training. It briefly summarizes learning theories and best practices for training design. It also identifies five studies that compared different methods of police training across varying contexts and countries. Our central conclusion is that the way police officers are trained likely matters as much as the skills and knowledge on which they are trained.
Yet the evidence for what might constitute effective training methods remains limited. In order to build a more robust evidence base and inform future training efforts, training approaches that donor agencies like the U.S. Department of State, Bureau of International Narcotics and Law Enforcement Affairs support should be tracked and evaluated with regard to implementation and effectiveness in achieving specific outcomes (such as protecting the population and supporting human rights and the ROL).
SETTINGS FOR POLICE TRAINING
An international review of police recruit training programs reveals that 17 of the 24 programs studied separated training into two primary settings and separate “standalone blocks”—residential/academy and field training (Belur et al., 2020). Police recruits in U.S. municipal police departments, for example, receive an average of approximately six months of basic academy training and approximately four months of field officer training (Reaves, 2016). In less common instances, field training may be followed by a final block of learning within an academy setting (Belur et al., 2020). Officers in other parts of the world (e.g., Estonia, Croatia, and Germany) often receive longer periods of basic training or, as in Finland and Norway, are required to attend three-year police universities (Council on Criminal Justice Task Force on Policing, 2021a). There is also a variation in who pays for the training: in the U.S. setting, it is common for police agencies to cover a recruited officers’ pay and training costs, and in other countries the officer may be responsible for this. Depending on the institutional constraints on how much officers can be paid, and the nature of the supply and demand for police officer positions, differences in the incidence of the cost of training could affect how much training officers receive. The frequency and dosages of additional in-service and specialized training not only vary but, for some, neither is available at all. In some agencies or nations, specialized training may only be available to a privileged few.
Recruits tend to first learn within an academy, which provides a controlled learning setting that involves classroom instruction. Classroom instruction can vary widely; some agencies use only lecture-based instruction, while others may use role-playing and mock situations.
Academy training is typically followed by experiential learning within a field-training setting, often without a return to the classroom. These two basic settings are most likely to be sequential, not alternating or combined. The symbolic overtones of those locations and their separation are substantial, especially given the usual sequence of classroom training followed by in-the-field training. This sequence gives rise to the widespread cliché allegedly voiced by the field training officers (FTOs): “Forget what they told you in the academy; we’ll show you how police work is really done,”
thus creating a “stigma” against content taught in the recruit academy (Hundersmarck, 2009). The sequential separation of training, moving from the classroom to the streets, appears to leave recruits particularly vulnerable to the predilections, and possible failings, of either their academy instructors or their FTOs. Sequencing academy training to be followed by experiential or field training—without some reinforcement between the two—may undermine or allow recruits to disregard knowledge learned in one setting and not the other.
In addition, police might receive in-service training after they are fully certified or serving as full-time police officers. In-service training can take place in a variety of settings, including informally during an officer’s shift by a supervisor requesting adjustments to an officer’s behavior, or during daily roll calls to update officers on new policies, laws, or equipment (see a creative approach to roll call training in Box 4-1). In addition, in-service training could involve officers returning to an academy setting for a few hours or days to obtain certain certifications, such as in firearms use or in responding to specific situations, or else to acquire specialized new knowledge (Martin, 2020), such as in evidence-based policing, procedural justice, de-escalation training, or learning about a new reporting or analytic system. A variety of teaching methods have been used across these various needs for in-service training (e.g., lecture/classroom-based, web-based, simulations, role-play scenarios, supervisory meetings).
Finally, police officers and other police supervisors and commanders may attend specialized in-service training programs that are not offered to everyone, such as INL’s International Law Enforcement Academies training or other seminars or training offered by external organizations, nongovernmental organizations, or other nations.
PROBLEM-SOLVING TECHNIQUES
Transferring the core knowledge and skills outlined in Chapter 3 will require training methods that activate critical thinking, problem-solving, communication and conflict resolution, and community organization skills (Birzer and Tannehill, 2001; Bradford and Pynes, 1999). While the evidence is mixed concerning recruit and trainer perceptions of the effectiveness of problem-based learning, at least one study found that those exposed to problem-based learning techniques acquired thought processes that better supported and aligned with community policing strategies than those trained using traditional style methods (see McGinley et al., 2019).
Some recruit training has adopted problem-based learning strategies, including scenario-based training, in efforts to develop higher-level problem-solving and decision-making skills (see Werth, 2011). One quasi-experimental study reported that police recruit academy trainees
believed problem-based learning produced better problem-solving and critical thinking skills than lecture-based academy training (Vander Kooi and Palmer, 2014). The integration of other learning theories, including cognitive load theory (stressing knowledge retention and skills acquisition), have been suggested (Mugford et al., 2013). The degree to which these types of integration have occurred in police academies is currently unknown.
Relatedly, training methods also need to take into account the abilities of trainees. For instance, in countries where basic literacy levels may be low, a combination of written, pictorial, and verbal dissemination of knowledge is likely required to share news or policies in ways that are accessible to all officers. Similarly, as Eric Beinart noted in a presentation to
the committee, where access to reliable Internet or cell service is challenging, the use of radio for training can provide more reliable communication to support consistent training goals. While radio is not a secured means of relaying information, it is valuable for reaching a large number of people across a wide area. Where Internet or WiFi is more reliable, social media groups and apps may also present opportunities for training that can be disseminated through a messenger platform. However, social media groups can often be peer-based with little senior management participation and little to no oversight, which can make sharing (and fact-checking) information through typical face-to-face training processes more challenging. See Box 4-2 for further discussion about the importance of adapting training to local realities.
INSTRUCTORS
Box 4-3 provides an example of a local training that had global impact—where people with related experience from a peer country were employed as peer trainers to conduct trainings. It is not known how frequently this is done and whether other outcomes are as successful as in this example. The present state of research fails to provide a comprehensive picture of the characteristics of police personnel providing police training instruction. Even less information is available about the skills of people typically tasked with designing training curriculum content and selecting training methods. Much more information and knowledge is needed about not only who is training police officers, but also who is best suited to carry out this vital task and with what training background.
The trustworthiness of agency leadership and instructors appears to influence training outcomes (MacQueen and Bradford, 2017). Instructors with a particular prior police experience, or with certain dispositions,
may not be the best trainers for all training purposes. A tendency to select instructors with certain characteristics may lead to the underutilization of highly effective female, younger, or minority officers. Instructors with a penchant for sharing their experiences (e.g., “war stories”) in an attempt to engage officers may actually create barriers to the effective delivery of a problem-solving training curriculum (see Chappell and Lanza-Kaduce, 2010). Further, such training methods have been argued to misrepresent actual police working conditions and duties (Belur et al., 2020), further reinforcing distorted perceptions concerning the dangers or mandates of police work and shifting focus away from problem-solving or community-centric policing. Moreover, as Grace Longe, Assistant Commissioner of Police of the Nigerian Police Force stated: some commanders may send officers to act as trainers as a punishment or a demotion, negatively impacting how the training is viewed by the trainer, the trainee, and the larger organization.
FTOs are thought to play a critical role in socializing officers by demonstrating and reinforcing police agency and community values outside of the academy setting. This socialization process may influence field application of academy training, trainee behaviors, and any misconduct. For example, research demonstrates a significant correlation between the misconduct of FTOs (Getty et al., 2016), peers (Ouellet et al., 2019), and officers. Research also finds that some FTOs actively negate academy learning (Hundersmarck, 2009). Thus, field training assignment—particularly to FTOs who engage in/fail to address officer misconduct as well as assignment to FTOs who do not provide training congruent with academy curriculum—could harm police training outcomes.
The persistent fear that FTOs may not be advancing the ROL or the protection of the public was highlighted by the murder of George Floyd by Dereck Chauvin, a Minneapolis Police Department officer, in 2020. On the day Chauvin killed Floyd, he was serving as an FTO and training two other officers, despite the fact that he had been the subject of several prior complaints, including three shooting incidents.1 Those two recruits were then charged under state law with aiding and abetting second-degree murder and second-degree manslaughter, as well as under federal law against deprivation of civil rights.2 Floyd’s murder reflects the worst-case scenario of an FTO selection process failing to screen out unsuitable instructors. More generally, instructors both for the police academy and field training should be equally qualified, armed with not only the same
___________________
1 See https://edition.cnn.com/2020/06/05/us/minneapolis-officers-background-george-floyd-trnd/index.html, downloaded August 10, 2021.
2 https://www.reuters.com/world/us/after-chauvin-sentencing-charges-remain-police-officers-floyd-case-2021-06-25/, downloaded August 10, 2021.
knowledge and skills to promote the ROL, as well as to protect the population, alongside the ability and moral code to convey, teach, and inculcate this knowledge in others.
The committee found no studies documenting the qualification process or disciplinary records for FTOs in any nation. Nor did it find any studies of the possible differences in training outcomes across FTOs with and without records of disciplinary violations, such as studies examining whether there may be higher levels of recruit misconduct among officers trained by FTOs who themselves had substantial evidence of misconduct. (However, see Getty et al., 2016, which found concentrations of complaints against recruits with a small portion of FTOs in Dallas, Texas). Given the major role that FTOs play in so much training of the police, the lack of attention to the selection and management of the FTO training work is a serious global gap in police knowledge.
Furthermore, instructors’ cultural competency may also impact police training outcomes. To accurately assess situational threats, officers must be able to address personal biases and be able to communicate despite potential religious, ideological, cultural, and identity barriers (Gerspacher et al., 2019). To the extent that instructors are unfamiliar with these barriers within a particular jurisdiction (e.g., U.S. officers training lacking cultural competency within an international context), they may be ill equipped to serve as effective instructors.
TRAINING DESIGN
Many of the most effective forms of professional development have a basis in theories of adult learning (Institute of Medicine, 2010; Salas et al., 2012). Several disciplines—such as adult education, cognitive science, developmental psychology, industrial and organizational psychology, neuroscience, and sociology—have advanced theories of learning and knowledge translation as well as developed science-informed steps for the design and evaluation of training. The perspectives on learning are complex; with adult learning seen as dependent on sociocultural contexts (NASEM 2018a,b). The question of the effectiveness of training design rests on what is attained in knowledge, skills, attitude, and improved job performance as a result of training.
A number of best practice principles are available through the work of both researchers and industries. A fundamental principle is that effective training is developed through a systematic process (Brown and Sitzmann, 2011; Goldstein, 1986, 1991; Salas et al., 2012). Such a process includes conducting a training needs analysis, developing training objectives, selecting training methods, pilot testing the training design, and evaluating the outcomes of training. These fundamental steps of training design
apply to all levels of trainees, from entry-level to management, as well as to instructors themselves in cases where training is aimed at preparing future instructors.
A training needs analysis should include assessments of the physical and cognitive tasks of individuals and teams, drawing on critical incident analyses for tasks with variable approaches. The needs analysis should also include an investigation of the organizational climate and state of the workforce (Harvey, 1991). It is important for training designers to take into account the conditions that might affect training delivery and outcomes, such as whether the organization is ready to support the training and to make use of the knowledge and skills from the training. Research has shown that features of the organizational environment—such as how supportive managers and coworkers are toward integrating new knowledge and skills—influence the application of training to the job (Blume et al., 2010; Ottoson and Patterson, 2000; Rouiller and Goldstein, 1993).
An understanding of the workforce capacities will help determine which training methods and resources best fit the training situation. A best practice for delivery is to use methods and tools that make the training as similar as possible to the tasks on the job (Brown and Sitzmann, 2011; Schmidt and Bjork, 1992). Training designers will also have to consider methods most likely to engage target learners in ways that they comprehend and that best allow them to retain knowledge and skills from the training (Noe, 2010; Salas et al., 2012). For high-risk, high-stake situations, hands-on experiential training can raise the psychological fidelity of the experience over classroom training by providing environmental conditions with simulated difficulties, time pressures, and other aspects similar to the actual situation. Personnel are able to practice managing their emotions while learning to handle high-risk job events (National Research Council, 2013).
Durable, long-term learning is best accomplished by repeated experience with critical tasks and skills. Research supports target learning distributed over multiple training sessions instead of relying on one-time training programs. Single, shorter trainings tend to result in short-term performance that predictably deteriorates over time (Cepeda et al., 2006; Rawson and Dunlosky, 2011; Soderstrom and Bjork, 2015). Additionally, trainings that are interconnected with links to shared knowledge and skills can strengthen long-term memory of key information and functions (NASEM 2018a,b).
In general, people make decisions in two ways. One way is more systematic and analytic and relies heavily on working memory. The other relies more on affective and emotional processes (Kahneman, 2003; Sloman, 1996; Stanovich and West, 2000). In nonstressful situations, working memory works in concert with emotional processing. In stressful situations, one’s capacity to draw on working memory can be compromised, leaving emotional processing to influence decisions (Beilock, 2008, 2010;
Wang et al., 2005). Further, time pressures may limit one’s ability to take in new and important information and make reasoned decisions. Optimal decision-making in stressful situations requires an awareness that emotion-driven decisions occur, and when they are most likely to occur, as well as the appropriate mental models to react quickly. Research has shown that simply making people aware of common internal responses in stressful situations (e.g., sweaty palms, beating heart) can make these responses less distracting (Jameison et al., 2010). Training can be directed at normalizing such physiological responses to lessen the impact on effective reasoning (Mattaralla-Micke et al., 2011). Training can employ realistic, stressful situations and explore the consequences (both positive and negative) of all possible decisions. Such education could also address cultural norms that might drive decisions that are inherently risky and driven by emotions.
One study of learning outcomes at police training academies (Vodde 2009, 2012) illustrated the usefulness of applying insights from adult learning theories to police training. This work emphasized six best practices:
- Explain why the information contained within curriculum is critical knowledge needed to perform their duties;
- Align content with previous knowledge and competencies to increase openness to new concepts and combat mental rigidity;
- Promote methods that include self-directed learning and autonomy;
- Create environments and systems that support and reinforce a continuous learning culture;
- Use application-focused exercises to demonstrate the content’s usefulness to solve current problems; and
- Instill self-determination to learn.
As noted previously, rigorous research on the effectiveness on specific police trainings is largely absent. Much of what is known about training design comes from other work contexts, particularly from health care and medical settings (Institute of Medicine, 2010; Wheller and Morris, 2010). Recent randomized trials, however, have started to examine changes to officer behavior after procedural justice training (Owens et al., 2018; Wheller et al., 2013). One study (Owens et al., 2018) found that officers who participated in the training were less likely to resolve incidents with an arrest and to be involved in incidents where force was used than officers who did not participate in the training. Both studies provide encouraging signs of effectiveness of training in altering officer behavior on the job (NASEM, 2018c).
Training should have clear, specific goals and the training should be piloted and/or evaluated to verify how close the training came to meeting these goals. A strong evaluation moves beyond counting participants and
participants’ reactions to training but is results oriented—the participants’ knowledge, competence, and performance are assessed (e.g., do the participants transfer what they learned to the job?) (Moore et al., 2009). Further, an evaluation may assess the extent to which police interactions with citizens improve or community perspectives of police improve as a result of training and changes in training participants’ behaviors.
EVALUATIONS OF TRAINING METHODS
While a growing number of police trainings are subjected to scientific inquiry, the review of the literature in one of the commissioned papers (Herold, 2021) found that most evaluations of training assess training impact in relation to trainees’ attitudes on the training, a few test policing outcomes post training, and even fewer test the effectiveness of methods on learning. Moreover, the evaluations identified in this review were all conducted in the Global North. A prior systematic review of all recruit training evaluation studies (McGinley et al., 2019) found that more than half (61%) have been conducted in the United States, followed by Australia (14%), the United Kingdom (9%), and Canada (8%). It is suspected that the geographic concentration of police training studies limits the generalizability of existing knowledge to places with different political structures, economies, and cultures.
Most of the rigorous police training evaluations have focused on procedural justice training (see, for example, Antrobus et al., 2019; Mazerolle et al., 2012; Murphy et al., 2014; Owens et al., 2018; Rosenbaum and Lawrence, 2017; Sahin et al., 2017; Wheller et al., 2013). Other rigorous police training evaluations have examined implicit bias training (Worden et al., 2020), de-escalation training (Engel et al., 2020), social interaction training (Aremu, 2006; McLean et al., 2020), training to reduce police use of force and improve decision-making (Andersen and Gustafsberg, 2016), officer resilience training (Chitra and Karunanidhi, 2021; McCraty and Atkinson, 2012), and training to improve attitudes toward workplace diversity (Platz et al., 2017).
In general, these studies suggest that training can change targeted police behaviors, justice-related outcomes, and police-public interactions (see references in Chapter 3). However, not all evaluations of police training report positive impacts on all expected outcomes (see La Vigne et al., 2019, for an example of disparate training outcomes across six U.S. cities). The committee thinks more evaluations are needed to identify how and why some police training practices are more effective than others and in what contexts. Further experimentation is needed to accurately attribute programs’ successes or failures to curricula, methods, and/or organizational contexts.
The Herold paper (2021) identified five studies that examined the outcomes of unique police training methods, each systematically comparing a single teaching method with other methods designed to teach the same content. The five studies were conducted in four different Global North countries. Each study focused on a different training topic and assessed a different teaching strategy. Two evaluations were conducted using randomized controlled trials; three were conducted using quasi-experimental designs. Four of the five training method evaluations found evidence of improved learning or performance outcomes in one method compared to the alternative. Given the small number evaluations, the committee is not able to draw strong conclusions about which methods are most effective for particular training topics. We share the description of these studies to illustrate how one might investigate the effectiveness of different approaches to police training. (See also Box 4-4 for ideas for implementing tracking and evaluation of police training.)
Study #1: Training in Ways to Interview Sexual Assault Victims, U.S.
Lonsway (1996) conducted a quasi-experimental study comparing three different ways of teaching police officers how to interview sexual assault victims. The training was conducted in the basic academy offered by the University of Illinois, Police Training Institute for newly appointed municipal officers from across the state. One class of 56 officers was given the traditional sexual assault response police academy training (with one hour devoted to the topic). Two other classes of 56 officers each received different experimental training protocols. One experimental protocol (E1) delivered to a recruit class a nonstop, stand-alone sexual assault simulated victim interview workshop of 3.5 hours’ length, which included 30 minutes of lecture. The other experimental group’s protocol (E2) featured an integrated curriculum interspersing three sessions among other classroom topics, for a total over multiple days of 4 hours of sexual assault response training, comprising 90 minutes of lecture and interactive discussion, 60 minutes of simulated role-play and discussion, with 90 minutes of lecture, discussion, and a videotaped interview of a rape victim.
While all three groups exhibited similar knowledge at entry, after participating in the workshops the group members in condition E1 were more likely to address victim welfare, suspect responsibility, and broader police investigatory options. E2 participants demonstrated greater proficiency in interviewing content and style. The authors concluded that E1 (a single intensive and focused workshop), rather than E2 or the traditional training, was the best of the three approaches. The author suggested that a single intense session would be best to improve learning for specialized topics that recruits cannot yet connect to a broader understanding of police functions.
Study #2: Basic Police Recruit Training at Academies, U.S.
Vodde (2009) conducted a quasi-experimental study comparing learning outcomes among recruits between two New Jersey basic police training academies. The first academy generally used traditional militaristic pedagogic-style instruction. The second academy used teaching methods aligned with best andragogical principles intended to facilitate adult learning (Vodde, 2009). The academy using best andragogical principles was found to produce better outcomes. In addition to greater learner satisfaction, problem-based learning assessments revealed that recruits in the andragogical-centered academy showed greater critical thinking, communication, problem-solving, and decision-making competencies than recruits who received traditional academy training methods.
Study #3: Procedural Justice Training, England
Wheller and colleagues (2013) used a randomized controlled trial to evaluate the Greater Manchester Police’s procedural justice training in England. A traditional classroom-based teaching method was compared to two combinations of classroom-based and scenario-based teaching methods. The authors found no statistically significant differences in outcomes across delivery methods but argued that the size of their participant pools might have hindered meaningful training group comparisons.
Study #4: Training in Taking Child Witness Statements, U.K.
Adams and colleagues (2019) conducted a randomized controlled trial to assess a Child Interview Simulator game, which was codesigned with U.K. police forces. The game was designed to help police recruits learn to take initial child witness statements. Assessment outcomes found that recruits who participated in the game learning displayed better tacit understanding, including demonstrations of empathy and attention, compared to those who received the same curriculum in face-to-face training. The recruits taking the face-to-face instruction displayed lower levels of understanding about the importance of gaining child respect through tactics used in the interviewing process.
Study #5: Training in Self-defense Against Knife-wielding Persons, Germany
Koerner and colleagues (2020) used a mixed-method quasi-experimental design to assess differences between two methods for teaching defense techniques against knife-wielding persons to 20 German police recruits. In what the study called “linear” training, learners were taught by a trainer how to execute knife movements and other techniques. In what was called “nonlinear” training, learners were taught by scenario-based, problem-solving discussions that focused on how to execute decisions. The authors report that, while both groups improved their performance in a nine-week post-training retention assessment, those receiving the nonlinear teaching method showed greater increases in problem-solving abilities than those receiving the linear training. In simulated knife attacks, the nonlinear training group were struck less and ended attacks faster and more frequently than the linear training group participants.
Nonetheless, qualitative evaluation found that the nonlinear group was less satisfied with their training, desiring more technique-focused rather than decision-making instruction. Thus, in addition to providing evidence of what the best knife attack training for police is, this experiment further suggests that learner evaluations of teaching quality may be wildly different
from important outcomes of the training. For policing in particular, this study emphasizes the importance of tracking behavioral outcomes, and not just choosing teaching methods based on their popularity with students.
CONCLUSION
Despite some attempts to evaluate police training programs, almost nothing is known about the effects of how training is delivered in the policing context. More scientific work is needed to understand which training methods can best help police to acquire the knowledge and skills necessary to promote the ROL and protect the population. While this is true for both the Global North and the Global South, for the purpose of this report, the opportunity to continue research may be greater in the Global South.
While the policing landscape has shifted substantially over the past two decades (see Cordner and Shain, 2011), much recruit training worldwide is still delivered through traditional militaristic-style instruction. Field-training methods also have characteristics that may undo lessons learned in academy training. In-service training may reinforce old traditions, cultures, and practices that work against sought-after reforms or modernization goals for policing. Few studies have focused on assessing the independent impact of teaching methods, but available evidence suggests that teaching methods selected for curriculum delivery can directly affect officer learning. Importantly, a training instructor’s characteristics and competencies may influence officer learning receptivity and performance outcomes.
Evaluations of training with methods based on best practices for adult learning in other professions are both relevant and encouraging. They suggest that adult learning models may be appropriate to develop the types of knowledge and skills officers need to more effectively promote the rule of law and protect the population. As discussed in Chapter 3, modern police work likely requires skill-building in areas that extend far beyond the basic task-oriented curriculum that has historically dominated police training. Research directly testing the impact of revised training methods is needed to determine if they could improve on traditional lectures, or even on “war story” discussions that use many anecdotes but fail to provide systematic knowledge (Sherman et al., 1998). Ideas for necessary evaluations and future research are discussed further in the next chapter.
