Police agencies are considering ways to improve police legitimacy and relations with the community. Police2Peace (“P2P”) is a nonprofit, nonpartisan corporation that provides trainings to change the culture of policing. This includes changing leader and officer mindset from law enforcement officers to “peace officers,” and emphasizing a policing model that is effective, empathetic, and just. This study is the first known study to evaluate the perceived effectiveness of P2P training, examining both officer and public perceptions of the training. Findings from a pre-test post-test survey design of officers found strong support for the training among law enforcement personnel. Results from an experimental survey of residents indicated promising potential for P2P training to enhance community perceptions of the police vis P2P training. As P2P emerges throughout the country, we discuss potential implications of these early evaluative findings.
Police2Peace (P2P), a national nonprofit, nonpartisan organization, has emerged as a key player in reshaping police culture through training programs intending to change police mindset from law enforcement officers to “peace officers.” The ultimate goal is to emphasize effective, empathetic, and just policing. Despite growing interest in the P2P approach, empirical assessments of its effectiveness remain sparce, with no known study rigorously examining perceptions of officers who experience the training, and the community they serve. As a first step, understanding various stakeholders’ perceptions of P2P lays a valuable foundation before assessing field outcomes (e.g., use of force, citizen complaints).
This study used a pre-test and post-test survey of officers as well as an informational provision survey of the community in Parsons, Kansas, a P2P training site. Our findings suggest strong officer support for the training, while community perceptions vary with significant positive effects when informed about the P2P initiative. These preliminary findings provide valuable insights into P2P training's potential to enhance police-community relations and foster peace, trust, and legitimacy in policing. As the first known evaluation of P2P, this study offers a foundation for future research evaluating a potentially emerging endeavor.
Unlike traditional police training programs that focus on tactical skills and enforcement strategies, P2P redefines the role of officers as "peace officers," shifting their mindset and operational philosophy (Police2Peace, 2024). Through specialized training, P2P promotes effective, empathetic, and just policing by encouraging officers to approach their roles as peacekeepers and community partners. This proactive model fosters trust and collaboration to address the root causes of crime and social disorder (Police2Peace, 2024).
Rather than isolating skills like de-escalation or crisis intervention, P2P integrates these into a holistic framework prioritizing cultural and psychological transformation. Officers are encouraged to prioritize peacebuilding over enforcement, empathy over authority, and justice over punitive measures. This shift is vital as public trust in law enforcement declines, and the demand for police legitimacy grows (Tyler, 2004).
P2P calls for greater police accountability and transparency by promoting equitable strategies and human rights-respecting practices (Morgan, 2023). It bridges the gap between law enforcement and communities through collaboration rather than external imposition (Tyler, 2006). By emphasizing empathy and justice, P2P aligns with societal trends toward humane and socially responsible governance. In a diverse society, officers trained in this approach are better equipped to engage with communities compassionately and fairly, contributing to a more just and peaceful society (Birch, 2024).
The present study is the first known study to evaluate the effectiveness of P2P training. We implemented a pre-test and post-test survey of trained officers, asking officers’ perceptions of the training. Next, we implement an information provision survey experiment of the community, randomly assigning information about the training to residents to assess their support of officers completing this training and its effect on perceived police legitimacy by the community. This study lays an empirical framework for future evaluations of P2P. We seek to answer the below research questions:
RQ1: How do police officers perceive the effectiveness of Police2Peace training in terms of its impact on their approach to community relations and peacekeeping?
RQ2: To what extent does providing information about Police2Peace training to community members influence their support for the program and their perceptions of police legitimacy?
This study was conducted in Parsons, Kansas. The residential population is 9,362 and is predominantly White (82.9%), with a Black population of around 7.8%. Roughly 90.5% of residents have attained at least a high school diploma. The median household income is $48,990. The Parsons Police Department (PPD) has 30 sworn officers and 10 staff members. In 2023, the department handled approximately 30,000 calls for service.
To assess baseline perceptions of P2P training, a pre-test survey was distributed to all officers and staff of the Parsons Police Department on July 30th, 2024. Officers then received four-hour training while command staff received an eight-hour training. An “exit” post-test survey was administered after each training. Thirty-three responses were received for the pre-test survey (RR 82.5%). Thirty-one responses were received for the post-training survey (RR 77.5%). The pre-test survey consisted of six items about P2P training effectiveness. The post-test survey presented the same six items, followed by post-test only questions regarding officer perceptions of the utility of P2P training. The six repeated items from pre/post-surveys were 5-point Likert-scale questions asking officers specific questions about the training and its utility along with a mean scale of all six items combined. The full list of questions can be found in the Appendix table A1. The sample of officers were 75% male, 79% White, between the ages of 35-54, with some college education.
The community survey began with questions about public perceptions of PPD, trust and confidence in police, procedural justice, and willingness to obey. Next, respondents entered the experimental component of the survey. Surveys were distributed beginning on July 17, 2024, and data collection concluded on July 29, 2024, before the training of officers began. A total of 142 residents completed the survey.
We used an information provision survey experimental design (Boehme et al. 2023) to assess whether information about P2P training improved residents’ perceptions of the police. Upon entering the experimental component of the survey, respondents were randomized into either treatment or control conditions based on SurveyMonkey’s randomization function. To contextualize the questions, those assigned to the control condition were provided the baseline statement below to provide context to the Likert-scale questions about P2P that were asked after the baseline statements. Questions about P2P training:
The Parsons Police Department is currently undergoing “Police2Peace” training, in which every police officer is required to take the training. Police2Peace training is intended to change the culture so that police officers view themselves as “peace officers” instead of law enforcers.
If randomly assigned the treatment condition, respondents received the same baseline statement with additional context about the training and its goals (below). Respondents were then asked the same Likert-scale questions about the P2P training.
The Parsons Police Department is currently undergoing “Police2Peace” training, in which every police officer is required to take the training. Police2Peace training is intended to change the culture so that police officers view themselves as “peace officers” instead of law enforcers. Importantly, the training is intended to train officers to be more compassionate, empathetic, and just towards citizens, build confidence and trust with the community, proactively engage with the community, improve police leadership, and enhance police officer health and wellness. Unlike other police trainings, these trainings are intended for officers and leaders to consistently engage in in-person and online trainings throughout their career.
The dependent variables consist of the below list of 8 statements, in which respondents were asked “How much do you agree or disagree with the following statements about Police2Peace training?”:
I support Police 2 Peace training
Police 2 Peace training will build my trust in police
Police 2 Peace training will improve community-police relations
Police 2 Peace training will make me feel safer interacting with police.
Police 2 Peace training is a good use of officer time.
Police 2 Peace training enhances the way I view the police.
All police agencies should receive Police 2 Peace training.
I will be more likely to call the police in a time of need.
The main independent variable is the binary treatment variable that indicates whether respondents were assigned the treatment or condition. Willingness to Obey the police is an unstandardized mean score of four items (alpha = 0.88), see appendix for survey items.
1) people should obey the law even if it goes against what they think is right
2) I always try to follow the law even when I think it is wrong
3) you should do what the police tell you even if you disagree
4) you should accept police decisions even if you think they are wrong.
Trust and confidence in police were also a combined mean score of four items asking respondents how much they agreed with the following statements:
1) police protect peoples basic rights
2) the police are generally honest
3) most police officers do their jobs well
4) the police can be trusted to do what’s right for my community.
A balance table of all covariates by experimental condition can be found in Appendix Table A2.
We analyzed the pre-test and post-test repeated measures using independent group t-tests to assess survey items mean differences from pre to post training. We pooled responses from the pre- and post-test groups together into one aggregate group since we were unable to match respondents from pre- to post-test surveys.1 We present summary statistics of the post-test survey-only survey items. Results are discussed in the results below and can be found in Appendix Table A2.
Table 1. T-Test Assessing Aggregate Mean Differences between Pre- and Post Survey | |||||
---|---|---|---|---|---|
Pre-Test Mean (S.D.) | Pre-Test Mean (S.D.) | Difference | One-Tailed P | Two-Tailed P | |
Majority non-enforcement duties | 3.970 (0.127) | 4.323 (0.117) | +0.353 | 0.023 | 0.046 |
P2P training might have merit | 3.667 (0.120) | 4.065 (0.131) | +0.398 | 0.014 | 0.029 |
Community wants empathetic/just police | 4.182 (0.124) | 4.355 (0.099) | +0.173 | 0.154 | 0.307 |
Interested in P2P mindset training | 3.848 (0.131) | 4.065 (0.131) | +0.216 | 0.124 | 0.249 |
All LEO should receive peace officer training | 3.879 (0.129) | 4.000 (0.117) | +0.121 | 0.246 | 0.492 |
Peace officer empathetic increases legitimacy | 3.727 (0.146) | 4.065 (0.139) | +0.337 | 0.050 | 0.100 |
Summed alpha scale of six items | 3.879 (0.089) | 4.151 (0.096) | +0.272 | 0.021 | 0.042 |
Notes: S.D. = standard deviation, Difference = change in mean from pre- to post. |
Table 1 shows findings from the independent sample t-tests. Across all survey items and the combined mean scale, officer perceptions of P2P training positively increased from pre- to post-survey. However, these increases were only statistically significant between three items and the combined scale. That is, there was a significant increase in officer awareness that majority of their duties are non-enforcement, that P2P may have merit for both officers and the community, and the combined mean scale of all six items.2
Turning to the post-test only survey items, we found positive sentiment towards the training. Over 75% of respondents agreed with each item asked. For example, 100% of respondents felt the training was not burdensome, 90% agreed that P2P training makes sense for policing, 87% agreed that P2P makes sense for elected officials who oversee policing, 87% agreed that the police should commit to P2P. Additionally, over 80% agreed that their agency should adopt P2P as an organizational philosophy, that P2P will help them do their job, that P2P gave them conceptual tools to interact with the community, and that they would recommend P2P training to others. The lowest support was that P2P training helped think differently about policing (add percentage here in parentheses).
We first descriptively compare the average responses to each item that vary by treatment or control group. Next, we estimate a series of logistic regression models on single items after collapsing the 5-point Likert-scales into dichotomous variables to assess any agreeableness with the statements that consist our dependent variables (Jeong and Lee 2016). We incorporate pre-existing perceptions of police presented above as well as socio-demographics to control for any covariate imbalance and covariate regression adjustment (Johansson and Nordin 2022; Yang, Qu, and Li 2023)(Johansson and Nordin 2022; Yang, Qu, and Li 2023), particularly with a small sample size (Hennessy et al. 2016).
Table A4 in the Appendix reports differences in the means of those of the collapsed variables between treatment and control group. A greater percentage in the treatment group agreed with the various benefits of P2P. Other than the statement of “all police agencies should receive Police 2 Peace training,” two-tailed t-tests did not indicate significant differences at the P < 0.05 level. However, differences were significant for one-tailed t-tests on the following items: “I support P2P training”, “all police agencies should receive P2P training,” and the combined mean scale of all eight items.
Table 2 Experimental regression results | |||||||||
---|---|---|---|---|---|---|---|---|---|
Support P2P | Increase Trust | Community Relations | Interactions Safer | Good Use of Time | View LEO Positively | All LEO Need P2P | Encouraged 2 Call LEO | Combined Item Scale | |
AME (S.E.) | AME (S.E.) | AME (S.E.) | AME (S.E.) | AME (S.E.) | AME (S.E.) | AME (S.E.) | AME (S.E.) | AME (S.E.) | |
Treatment | 0.155 (0.08)* | 0.163 (0.07)* | 0.100 (0.08) | 0.161 (0.07)* | 0.131 (0.08) | 0.138 (0.08) | 0.201 (0.08)** | 0.089 (0.08) | 0.144 (0.06)* |
Obey | 0.049 (0.06) | -0.016 (0.06) | -0.023 (0.06) | -0.060 (0.07) | 0.062 (0.06) | 0.095 (0.06) | 0.019 (0.06) | -0.011 (0.07) | 0.016 (0.05) |
Trust/Confidence | -0.029 (0.07) | -0.061 (0.08) | -0.074 (0.07) | -0.040 (0.08) | -0.132 (0.08) | -0.102 (0.08) | 0.053 (0.08) | 0.016 (0.08) | -0.045 (0.06) |
Procedural Justice | -0.104 (0.07) | -0.139 (0.07)* | -0.049 (0.07) | -0.122 (0.07) | 0.012 (0.08) | -0.146 (0.072) | -0.173 (0.07) | -0.125 (0.08) | -0.111 (0.06) |
Age | -0.023 (0.04) | -0.036 (0.04) | -0.010 (0.04) | 0.039 (0.04) | 0.019 (0.04) | 0.008 (0.04) | 0.004 (0.04) | 0.044 (0.04) | 0.010 (0.032) |
Non-White | -0.055 (0.10) | -0.115 (0.10) | -0.0195 (0.09)* | -0.037 (0.11) | -0.253 (0.09)** | -0.005 (0.11) | -0.206 (0.10)* | -0.008 (0.113) | -0.109 (0.08) |
Female | 0.156 (0.08)* | 0.088 (0.08) | 0.071 (0.08) | 0.060 (0.08) | 0.122 (0.08) | 0.137 (0.08) | 0.110 (0.08) | 0.021 (0.09) | 0.096 (0.07) |
Education | 0.018 (0.03) | -0.004 (0.04) | -0.006 (0.03) | -0.041 (0.03) | -0.026 (0.03) | -0.019 (0.04) | -0.050 (0.03) | -0.061 (0.04) | -0.024 (0.03) |
Crime Victim | 0.035 (0.09) | -0.047 (0.08) | -0.087 (0.08) | -0.112 (0.08) | -0.004 (0.09) | -0.002 (0.08) | 0.004 (0.08) | -0.091 (0.09) | -0.041 (0.07) |
Notes: ** P < 0.010, * P < 0.05, AME = average marginal effects, S.E. = standard errors, P2P = Police 2 Peace, LEO = law enforcement officers. All models report findings from logistic regression models except for the combined item scale column which reports findings from an OLS regression model on the continuous outcome variable |
Turning to logistic regression results (presented in Table 2), findings indicate that the treatment significantly swayed perceptions of P2P training for community residents. Those assigned the treatment were 15.5% more likely to support of P2P training and 16.3% more likely to agree that P2P training will increase trust in police. Further, the treatment group was 16.1% % more likely to agree that P2P training will make interactions with police safer and believe that all police agencies should receive P2P training. Finally, we estimated an OLS regression model on the combined mean scale, which indicated that respondents assigned the treatment were more 14.4% likely to support P2P training.
Sensitivity analytical models assessed robustness of the findings. Coefficients for uncontrolled logistic regressions, which did not include the above discussed covariates were all in positive directions. The treatment variable was largely non-significant without control variables, with the lone exception was increased support for agencies to receive P2P training. The treatment variable, support of P2P training, was marginally significant. Respondent’s also felt that P2P would make interactions with police safer.
The differential findings between the controlled and uncontrolled models warrant further discussion. We acknowledge that the survey of residents resulted in a small sample, which can explain the null findings in the uncontrolled models (Kane 2023; Lovell 2020)(Kane 2023; Lovell 2020). Particularly with small samples in experiments, including covariates provides balance and improves regression precision and adjustment (Berk et al. 2013). However, since our findings were not robust across all model specifications, we urge caution in interpreting our results.
This study provides early evidence regarding the effectiveness of P2P from officers and the community. Our findings reveal two key insights. First, officer perceptions of P2P point to its promise as a tool for cultural transformation within law enforcement. We observed significant increases in officers’ awareness of non-enforcement duties and their beliefs in the P2P’s merit for both police and community. Their positive post-training feedback provides evidence that P2P training resonates with officers. Additionally, it appears to be viewed as a practical approach to enhancing community relations. Notably, officers did not find the training burdensome as part of their daily duties, which is promising given the current staffing crisis (Adams, Mourtgos, and Nix 2023).
Second, our findings suggest that informing community members about Police2Peace training may enhance public support for such reform efforts. When provided with information about the program, community members were approximately 15% more likely to support P2P and express increased trust in police. These results dovetail with previous work on legitimacy (Tyler, 2004) and provide further support for transparency regarding new policing efforts. We note that these effects were only statistically significant when controlling for pre-existing perceptions of police and socio-demographic factors. Therefore, this relationship may be more complex and/or warrants more research with larger samples. Overall, however, our study provides early evidence that P2P training may be perceived positively by officers going through the training as well as the community viewing the training for officers as a positive initiative to improve community-police relations.
Several limitations require discussion. The small sample size in our community survey necessitates caution in interpreting the results. Additionally, our study focused on a single department which limits generalizability. We hope to revisit this research to examine effectiveness across diverse jurisdictions along with a more thorough exploration of its long-term impacts, and the impact of this training on field outcomes.
Despite these limitations, our findings suggest that P2P, complemented by a strategic information sharing campaign, may be a promising approach to enhancing police legitimacy and community trust. As the first known study to evaluate perceptions of P2P of important stakeholders, we provide initial evidence of its success. Future research should evaluate the impact on field outcomes (e.g., use of force, complaints). As departments navigate challenges of reform implementation amid resource constraints, programs like P2P warrant further consideration and evaluation.
Appendix Table A1. Pre/Post-test survey items | |
---|---|
Survey Item | Label |
Officers spend the majority of their time on nonenforcement activities (resolving disputes, keeping the peace, etc.) versus enforcing the law, making arrests, or issuing citations | Majority non-enforcement duties |
Training that transforms organizational mindset, like using the term "Peace Officer" instead of “law enforcement officer” might have the effect of serving both police officers and the communities they protect | P2P training might have merit |
The majority of my community members want policing to be effective, empathetic, and just | Community wants empathetic/just policing |
I am interested in training that helps officers appreciate the mindset of police officers as “Peace Officers | Interested in P2P mindset training |
All members of law enforcement should receive Peace Officer training | All LEO should receive peace officer training |
Being a Peace Officer, as opposed to a law enforcement officer, is important because the peace officer mindset ascribes to the belief that policing can be effective, empathetic, and just while delivering policing services that community members perceive as legitimate and necessary | Peace officer empathetic increases legitimacy |
Table A2. Balance table by experimental condition | |||||
---|---|---|---|---|---|
Treatment Condition | Control Condition | Full Sample | |||
Variable | Mean / % | Mean / % | Mean / % | ||
Age | 4.27 | 4.21 | 4.24 | ||
Non-White | 0.21 | 0.18 | 0.19 | ||
Female | 0.52 | 0.56 | 0.54 | ||
Education | 2.94 | 2.48 | 2.71 | ||
Crime Victim | 0.54 | 0.56 | 0.54 | ||
Notes: T-tests indicated no significant differences between the two groups on all variables. | |||||
Table A3. Post-test only survey items | |||||
Survey Item | % Agreement | ||||
I understood the POF | 97% | ||||
The training was easy to do | 100% | ||||
The POC makes sense for policing | 90% | ||||
My agency should adopt the POF as an organization philosophy | 81% | ||||
The POC makes sense for my community | 84% | ||||
The POC makes sense for elected officials who oversee policing | 87% | ||||
Using the POC will help me do my job | 81% | ||||
The POP makes sense to me | 94% | ||||
The police in my community should commit to making the POP | 87% | ||||
The PO training was presented in a clear and logical manner | 94% | ||||
The PO training helped me think differently about policing | 77% | ||||
The PO training gave me conceptual tools to interact with my community differently | 81% | ||||
I would recommend the PO training to others | 87% | ||||
Notes: PO = Peace Officer, POF = Peace Officer Framework, POC = Peace Officer Concept, POP = Peace Officer Promise. % Agreement = strongly agree or agree = 1; neutral, disagree, and strongly disagree = 0. |
Table A4. Mean differences between treatment and control groups | |||||
---|---|---|---|---|---|
Treatment Group | Control Group | Difference | One-Tailed P | Two-Tailed P | |
Survey Item | % / Mean Agree | % / Mean Agree | |||
I support Police 2 Peace training | 0.765 | 0.615 | +0.150 | 0.032 | 0.063 |
Police 2 Peace training will build my trust in police | 0.552 | 0.431 | +0.121 | 0.083 | 0.165 |
Police 2 Peace training will improve community-police relations | 0.662 | 0.600 | +0.060 | 0.232 | 0.464 |
Police 2 Peace training will make me feel safer interacting with police | 0.493 | 0.354 | +0.139 | 0.054 | 0.109 |
Police 2 Peace training is a good use of officer time | 0.662 | 0.585 | +0.077 | 0.181 | 0.362 |
Police 2 Peace training enhances the way I view the police | 0.525 | 0.400 | 0.094 | 0.187 | |
All police agencies should receive Police 2 Peace training | 0.662 | 0.492 | +0.170 | 0.024 | 0.048 |
I will be more likely to call the police in a time of need | 0.529 | 0.462 | +0.067 | 0.219 | 0.438 |
Combined mean scale of all items | 0.607 | 0.492 | +0.115 | 0.048 | 0.097 |
Notes: % of the sample by condition that either “strongly agreed” or “agreed.” |