Bias, Distrust, and Trauma

Jump to Section

• Key Takeaways
• Conclusion
• Methodology
• Appendix: Additional Charts
• Endnotes

Key Takeaways

1. Black Bostonians report various types of police harassment at much higher rates than non-Black Bostonians.

Among Bostonians who reported having had any contact with law enforcement, we asked if they had experienced any of the following: accused of having or selling drugs, assumed to be a thief, treated unfairly because of how they were dressed, verbally abused, physically abused, followed for no reason while they were walking or driving, pulled over for no reason while driving, stopped and searched for no reason, and arrested for something they didn’t do.

Substantial differences exist across the board, but gaps between Black and non-Black Bostonians were largest for being followed while walking or driving, pulled over for no reason while driving, stopped and searched, and assumed to be a thief. For all but physical violence, racial differences reported are statistically significant.[2]

On average, Black Bostonians reported experiencing 1.7 of the nine types of police harassment experiences listed above (chart not shown). The figures for Latino, AAPI, and White Bostonians are significantly lower at .8, .8, and 1.0, respectively. 

To put this another way, whereas the vast majority of White, AAPI, and Latino Bostonians reported that they had not experienced any of the types of police harassment listed above (67%, 73%, and 77%, respectively), 50% of Black Bostonians had.[3]

Further, a much higher percentage of Black Bostonians experienced an excessive number of harassment types – between 6 and 9. Whereas 4.0%, 5.1%, and 5.6% of Latino, White, and AAPI Bostonians, respectively, reported experiencing an excessive number of harassment types, Black Bostonians experienced an excessive number at 2-3 times those rates – 11.9%.

Importantly, while we often imagine that the targets of police bias are primarily young, lesser educated and low-income Black males from disadvantaged, predominantly Black neighborhoods, findings from our survey of Bostonians reveal that, with few exceptions, Black Bostonians experience disparate treatment by law enforcement within categories of gender, age, educational attainment, neighborhood of residence, and income status. For instance, among residents with advanced degrees, 24%, 27%, and 31% of Latino, AAPI, and White residents, respectively, reported at least one type of harassment, but 58% of Black residents with advanced degrees did. Similarly, while 41%, 35%, and 31% of Latino, AAPI, and White residents, respectively, with annual incomes above $150K reported experiencing at least one type of harassment, 59% of Black residents did. And in all but one neighborhood category, roughly 50% of Black residents experienced at least one type of harassment by law enforcement. This compares to roughly 20% of Latino residents, 30% of AAPI residents, and 40% of White residents.[4] In other words, no matter their gender, age, educational attainment, neighborhood of residence, and income, Black Bostonians experience a greater number of police harassment types, on average, when compared to their non-Black counterparts.

A significantly higher percentage of Black Bostonians also perceive that they are being harassed because of their racial background. At much higher rates, they reported that they or others in their neighborhood had experienced harassment because they were Black. While 38% of Black Bostonians reported that police had treated them unfairly in their own neighborhood specifically because of their race, 63% reported that others in their neighborhood had felt the same. Furthermore, while 51% of Black Bostonians reported that police stopped or followed them because of their racial background, 67% shared that others in their neighborhood had felt the same. Significantly lower percentages of Latino, AAPI, and White Bostonians offered similar responses. For instance, only 10% of White residents had ever felt that police had treated them unfairly in their own neighborhood because of their race, and only 8% had ever felt that the police stopped or followed them just because of their racial background.[5] 

Here again, with few exceptions, these disparities persisted no matter the gender, age, educational attainment, neighborhood of residence, and income status categories considered. For instance, Black men reported all four forms of racially motivated harassment at rates between 3-6 times greater than non-Black men. And Black residents with advanced degrees reported at least one type of racially motivated police harassment at twice the rate of non-Black residents – 73% versus 38%, 31%, and 34% of Latino, AAPI, and White residents, respectively. Indeed, multivariate analysis (not shown) reveals that even after controlling for demographic characteristics, prior arrest status, and neighborhood of residence, Black Bostonians experienced a greater number of police harassment types and more racially motivated police harassment. 

2. In contrast to non-Black Bostonians, Black Bostonians feel a deep distrust towards law enforcement, and their distrust is strongly associated with experiences of police harassment.

We asked Boston residents about the extent to which they agreed with nine statements indicative of race-based trust in law enforcement. Responses bring into sharp relief not only the degree to which Black Bostonians distrust law enforcement, but also how much their levels of distrust differ from those of Latino, AAPI, and White Bostonians. A minority of Black residents agreed with statements indicative of trust in law enforcement, often in contrast with residents of other racial and ethnic groups. For instance, proportionately few Black residents agreed to the following statements: People of my racial group are treated the same as people of other racial groups (20% versus 40% of non-Black residents); in most courts, people of different racial groups receive the same kind of treatment from the judge (23% versus 39% of non-Black residents); officers have the best interest of people of my racial group in mind (23% versus 54% of non-Black residents); and people of my racial group can trust officers (30% versus 62% of non-Black residents).

Furthermore, while a strong majority of Black Bostonians agreed with statements that suggest distrust in law enforcement, with just a few exceptions, these were statements with which only a minority of non-Black residents agreed. For instance, most Black residents agreed that people of my racial group should be suspicious of the criminal justice system (66% versus 43% of non-Black residents); that officers do not take complaints of people of my racial group seriously (64% versus 32% of non-Black residents, although 55% of Latino residents also agreed with this statement); and that, where my racial group is concerned, officers are more likely to act aggressively first and ask questions later (73% versus 32% of non-Black residents, although, here again, 57% of Latino residents also agreed with this statement). Strikingly, half of Black residents agreed that officers treat people of my racial group like animals. Not far behind are Latino residents, among whom 39% agreed. Comparatively speaking, “just” 14% and 23% of White and AAPI residents, respectively, agreed with this very troubling statement.[6] 

Combining the above statements into one composite measure of trust in law enforcement[7] reveals the following statistically significant racial/ethnic differences in trust.

Findings from multivariate analysis (not shown here) indicate that Black-White gaps in trust in law enforcement can be explained, in part, by the fact that law enforcement engages in a greater number of harassment types with Black Bostonians and in police harassment that is racially motivated. Each of these two factors has far greater explanatory power than residents’ demographic characteristics, prior arrest status, and neighborhood of residence combined.

3. More than half of Boston residents report that law enforcement has made their community feel safer, but rates vary by race/ethnicity and are informed by experiences of police harassment and harassment perceived to be racially motivated.

When asked if they agreed with the statement, “law enforcement has made my community feel safer,” more than half of Boston residents agreed. But agreement varied by race/ethnicity, with pronounced differences between Black Bostonians and their non-Black counterparts. Whereas 64%, 67%, and 59% of Latino, AAPI, and White Bostonians, respectively, agreed that law enforcement has made their communities feel safer, just 44% of Black Bostonians agreed. Fully one-third of Black residents disagreed with that statement compared to less than one-quarter of non-Black Bostonians.[8]

While many factors shape whether residents view law enforcement’s impact on their communities positively, experiences of police harassment and the perception that such harassment is racially motivated are linked to Bostonians’ perceptions of law enforcement’s impact on their communities. As shown below, when compared to those who have experienced police harassment, a substantially higher percentage of Black Bostonians who report no harassment agree that law enforcement has made their communities feel safer – 54% and 59% versus 33.8% and 36.6%. But even among Boston residents who exhibit a great deal of trust in law enforcement, the belief that law enforcement has made their communities feel safer plummets with any type of police harassment, especially so when that harassment is perceived to be racially motivated. Among AAPI residents, for instance, 75% and 83% of those who reported no police harassment or racially motivated police harassment agreed that law enforcement had made their communities feel safer – the overwhelming majority. But, among those who experienced any type of police harassment, less than half agreed with this statement, cutting rates of agreement by 40-45%.[9]

4. Among Bostonians, police harassment isn’t just predictive of distrust and feelings of community safety. Police harassment is also predictive of symptoms of trauma, especially so for Boston’s Black men.

Because a growing body of research links mental health vulnerabilities to encounters with police,[10] we sought to determine whether the number of police harassment types and racially motivated police harassment affect individuals’ mental health, with specific attention to symptoms of post-traumatic stress disorder, or PTSD. To address this question, we asked Boston residents to think about a memorable police encounter and consider the extent to which they agreed with a set of seven statements that might indicate trauma. These statements describe experiences including re-experiencing the event (flashbacks, for instance), avoidance, mood changes, and hyperarousal and have been associated with post-traumatic stress disorder (or PTSD). The statements are as follows: 1) Remembering this experience brings back feelings about having been stopped. 2) Other events in my life cause me to think about this experience with the police. 3) I think about having been stopped even when I do not mean to. 4) I try not to remember and think about this incident. 5) My feelings about this incident are kind of numb. 6) I try not to talk about the time I was stopped. 7) Reminders of the time I was stopped cause me to have physical reactions such as sweating, trouble breathing, nausea, or a pounding heart. Although we make no claims here that respondents reporting agreement with any of these statements have PTSD, we do suggest that agreement with one or more of these statements might indicate that individuals left the encounter with police feeling traumatized.

Black Bostonians responded affirmatively to a greater number of these statements.[11] Whereas Latinos, AAPI, and White Bostonians responded affirmatively to 1.1, 1.0, and 1.2 statements, respectively, on average Black residents responded yes to 1.8. Further, it is not just that a significantly lower percentage of Black Bostonians responded “no” to all the trauma statements – 43% relative to 65%, 63%, and 51% of Latino, AAPI, and White residents, respectively – it is also that a significantly higher percentage of Black Bostonians responded “yes” to between 3 and 6 statements – 34% relative to 20% of the other racial/ethnic groups.[12] Importantly, with few exceptions, within categories of gender, age, educational attainment, neighborhood of residence, and income, Black Bostonians agreed with more trauma statements, and with an excessive number of trauma statements. For instance, one striking finding is that Black men agreed with 3-6 trauma statements at a rate roughly two times greater than Latino, AAPI, and White men. Over 40% responded “yes” to three or more statements compared to 17% of White men, 23% of AAPI men, and 19% of Latino men.

Results of multivariate analysis (not shown) reveal that Black-White gaps in the number of agreed-upon statements about trauma symptoms can be explained away by differences in racially motivated harassment. In other words, if Black Bostonians experienced racially motivated harassment at the same rate as White Bostonians, the gaps in the number of trauma statements they agreed with would all but disappear. Number of harassment types is also a strong predictor – the greater the number of harassment types, the greater the number of trauma symptoms reported – although effect size and strength are not as large as that for racially motivated police harassment. Each of these factors, however, explains the difference in Black-White trauma gaps far better than demographic characteristics, arrest status, and neighborhood of residence combined.

We then examined the link between the number of police-involved trauma symptoms and Boston residents’ overall emotional and mental health. To determine this, we asked if residents had an emotional or mental health issue like depression, anxiety, post-traumatic stress disorder (PTSD), or the like. Thirty-eight percent responded affirmatively, with a higher percentage of White residents reporting an emotional health issue than other Boston residents.

For each racial/ethnic group, however, self-reported rates of emotional or mental health vulnerabilities are strongly and positively associated with the number of police-involved trauma statements they agreed to. As the number of police-related trauma statements that Boston’s Black residents agreed to grew, so too did their rates of mental health vulnerabilities. Among Boston’s AAPI community, moving from no reported trauma symptoms to 1-2 such symptoms results in a significant leap in the percentage reporting mental health vulnerabilities – from 25% to 54%. To a lesser extent, this is the case as well among Boston’s Latino residents – from 30% to 43%. Among White residents, the pattern is different. The major distinction appears to be between those who reported two or fewer police-related trauma symptoms and those who reported 3-6 police-related trauma symptoms; for the latter, two-thirds reported emotional and mental health vulnerabilities, a substantially higher rate than the 34-41% who reported two or fewer police-related trauma symptoms.[13] 

5. For some Bostonians, most notably AAPI residents, police harassment and associated distrust and trauma symptoms are linked with chronic health conditions.

A growing body of research links aggressive policing to poor mental and physical health outcomes in communities targeted for such interventions. In fact, in addition to mental health vulnerabilities like depression and PTSD-like symptoms, aggressive policing practices have been linked to high blood pressure, diabetes, and obesity/overweight. Given this, we examined whether the harassment types and associated distrust and trauma symptoms could be linked to high blood pressure, diabetes, obesity/overweight, and heart issues in Boston residents. Because we examined four outcomes – high blood pressure, diabetes, heart issues, and overweight/obesity – across four racial/ethnic groups – Black, Latino, AAPI, and White – by four potential causes – police harassment, racially motivated police harassment, distrust, and trauma symptoms – we have simplified our findings in the table below, highlighting for whom and for what analysis our findings are statistically significant.

In many ways, results are surprising. Although analysis produced several statistically significant findings, they are not always in the direction we would predict, and the strongest associations are not necessarily for the groups we might expect. We find little evidence linking police harassment types, distrust, and police-involved trauma to heart issues. Only among White residents did we find a statistically significant association. Specifically, among White Bostonians, racially motivated police harassment is associated with self-reported heart problems.

For each racial/ethnic group, we find statistically significant associations with diabetes. Among Black residents, diabetes is negatively associated with distrust in law enforcement; people who reported having diabetes also reported lower levels of distrust, on average, a finding of marginal significance. Among Latino residents, however, diabetes is positively associated with police harassment; those reporting police harassment have higher self-reported rates of diabetes than those who reported experiencing no police harassment types. This finding is also of marginal statistical significance. White residents who reported racially motivated police harassment also have higher self-reported rates of diabetes; this, too, is of marginal statistical significance. Among AAPI Bostonians, those who reported at least one police harassment type and racially motivated police harassment are significantly more likely to self-report diabetes.

For Black, Latino, and AAPI residents, we also find statistically significant associations with high blood pressure. Among Black Bostonians, self-reported high blood pressure is negatively associated with both racially motivated police harassment and distrust; a lower percentage of those who reported racially motivated police harassment and distrust also reported having high blood pressure. The opposite is true for AAPI residents, however; self-reported high blood pressure is positively associated not only with police harassment and racially motivated police harassment but also with trauma symptoms. Among Latino residents, distrust in the police is associated with high blood pressure as well.
 

Finally, for each racial/ethnic group, we find statistically significant associations with overweight/obesity. All associations are positive. Among Black, Latino, and AAPI residents, distrust in police is positively associated with being overweight or obese, marginally so for Black and Latino residents and highly significant for AAPI residents. The number of police harassment types and trauma symptoms are also strongly and positively associated with being overweight or obese among Boston’s AAPI residents. Among White residents, racially motivated police harassment is positively associated with being overweight or obese, a finding that is also highly significant.

Across these outcomes, groups, and potential causes, what stands out are the strong bivariate associations between the number of police harassment types and related distrust and trauma symptoms on the one hand and chronic health conditions on the other, especially and specifically for the AAPI community. For all but heart issues, and for half of the relationships studied here, among Boston’s AAPI residents, negative police encounters and related distrust and trauma showed statistically significant associations with poorer physical health.

Conclusion

All things considered, are there any signs to suggest that law enforcement officers treat Black residents of Boston the same as people from other racial and ethnic groups? Based on results of analysis of these survey data, we have little reason to believe that Black Bostonians are treated the same as people from other racial and ethnic groups. Racial disparities in police harassment, including harassment perceived to be racially motivated, are large and consistent with police patterns and practices in Boston described by many in the Black community in the years and decades before George Floyd’s murder, during that year of global protest, and in the years since. It is unclear that reforms responding to Boston’s racial reckoning have done much to alter these very troubling and long-standing patterns.

Further, the consequences for Bostonians’ health and well-being – Black, Latino, AAPI, and White – cannot be overstated. Even though we found no evidence of racial disparities in physical violence perpetrated by the police, the other types of harassment that Boston residents, and especially Black men, experience have major and negative consequences. Indeed, police harassment is not only strongly associated with distrust in the police and a sense that police do not make communities safer, but it is also associated with trauma symptoms and chronic mental and physical health concerns that contribute to poorer quality of life and early mortality. Although we make no claims here that the nature of these relationships is causal, given prior research in other cities, this is a strong possibility that warrants further study and, if borne out, an effective policy response.

Importantly, even samples that are meant to be representative of a population – like the one used for this study – struggle to include the types of people most vulnerable to police harassment and abuse – prime-age Black men who are incarcerated or at risk of incarceration, men who have been deemed by the New York Times to be “missing from everyday life.”  In prior research, their relative absence in study samples has resulted in smaller Black-White gaps in educational attainment, employment, and wages than would have been the case had they been accurately represented in the sample. We assume the same is true here. This means that the racial disparities reported here in police harassment, trust, and trauma are likely underestimated. 

Methodology

A survey of 1,407 Boston residents, including a base sample of 1,000 Boston residents as well as oversamples of an additional 200 Black residents and 200 Latino residents, was organized by a research team at the Harvard Kennedy School’s Program in Criminal Justice (PCJ) and supervised by Sandra Susan Smith, Professor of Criminal Justice, Faculty Chair of the PCJ, and Director of the Malcolm Wiener Center for Social Policy. Data were collected in English and Spanish by the nonpartisan MassINC Polling Group using multiple survey modes – live telephone interviews to landline and cell phones, online survey links sent via text message, and online panel sample. The survey was fielded January 31st-February 14th, 2024, and each respondent was paid $10 for their study participation. Data were weighted to reflect population estimates by age, gender, race/ethnicity, educational attainment, and geography (groupings of neighborhoods within the city). Prior to fielding the survey, MassINC created targets for all parameters, for the base sample and for the two oversamples, and used these to complete interviews. The American Community Survey 5-year estimates for Boston was the source for MassINC targets for age, gender, race, and education. For geography, neighborhood population estimates from Boston Plans were used. Weights were applied to each racial/ethnic group oversample separately. The margin of error for this survey is +/- ­­3.1 percentage points at the 95% confidence level, inclusive of the design effect. A link to the survey questions is provided below.

Click here to access the survey questions.

Appendix: Additional Charts

Police Harassment by Gender and Race/Ethnicity
 

Return to Takeaway 1

Police Harassment by Age Categories and Race/Ethnicity

Return to Takeaway 1

Police Harassment by Educational Attainment and Race/Ethnicity

Return to Takeaway 1

Police Harassment by Neighborhood of Residence and Race/Ethnicity

Return to Takeaway 1

Police Harassment by Income and Race/Ethnicity

Return to Takeaway 1

Perceptions of Racially Motivated Police Harassment, by Gender and Race/Ethnicity

Return to Takeaway 1

Perceptions of Racially Motivated Police Harassment, by Age Category and Race/Ethnicity

Return to Takeaway 1

Perceptions of Racially Motivated Police Harassment, by Educational Attainment and Race/Ethnicity

Return to Takeaway 1

Perceptions of Racially Motivated Police Harassment, by Neighborhood and Race/Ethnicity

Return to Takeaway 1

Perceptions of Racially Motivated Police Harassment, by Income Status and Race/Ethnicity

Return to Takeaway 1

Mean Number of Affirmative Responses to Trauma Statements

Return to Takeaway 4 

Trauma Symptoms by Gender and Race/Ethnicity

Return to Takeaway 4 

Trauma Symptoms by Age Categories and Race/Ethnicity

Return to Takeaway 4 

Trauma Symptoms by Educational Attainment and Race/Ethnicity

Return to Takeaway 4 

Trauma Symptoms by Neighborhood of Residence and Race/Ethnicity

Return to Takeaway 4 

Trauma Symptoms by Income and Race/Ethnicity

Return to Takeaway 4 

Emotional or Mental Health Issue(s)

Return to Takeaway 4 

Endnotes

[1] “Other” refers to the following: American Indian/Alaska Native, Middle Eastern/North African, 2 or more races, other, and those who prefer not to say.

[2] Chi-square tests of independence reveal statistically significant associations at p<.001 (2-sided) for being assumed to be a thief, followed for no reason, pulled over for no reason, stopped and searched for no reason, and for being verbally abused. Tests of independence also reveal significant associations at p<.01 (2-sided) for being accused of having or selling drugs, being treated unfairly for how you dress, and being arrested for something you didn’t do.

[3] Chi-square tests of independence reveal a statistically significant association at p<.001 (2-sided). Respondents who reported no contact with law enforcement are included in these figures. If analyses were limited to only those who had contact with law enforcement, rates of harassment would be higher.

[4] One exception to this is the Back Bay, Beacon Hill, South End, and South Boston category of neighborhoods. Here Black, Latino, and AAPI residents each report rates of harassment that far exceed the rate that White residents of these neighborhoods report – 50%, 47%, and 43% relative to White residents’ 20%. Such patterns suggest that law enforcement officers might be engaging in practices to preserve White dominance in neighborhoods that have historically been deemed to be exclusively White by harassing those who are deemed to be “the other.”

[5] Chi-square tests of independence reveal statistically significant associations at p<.001 (2-sided) for each crosstabulation. Respondents who reported no contact with law enforcement are included in these figures. If analyses were limited to only those who had contact with law enforcement, the rates of racially motivated police harassment would be higher.

[6] Chi-square tests of independence reveal statistically significant associations at p<.001 (2-sided) for each crosstabulation.

[7] Trusting responses were given a value of 1 or 2 (agree or strongly agree, respectively), neutral responses (neither agree nor disagree) were given a value of 0, and responses indicative of distrust were given a value of –1 and –2 (disagree or strongly disagree, respectively). To achieve this and to offer ease of interpretation, several items were reverse coded so that agreement corresponded to trust and disagreement corresponded to distrust. Between group differences are statistically significant at p<.001 as determined by one-way ANOVA.

[8] Chi-square tests of independence reveal statistically significant associations at p<.001 (2-sided). Respondents who reported no contact with law enforcement are included in these figures. If analyses were limited to only those who had contact with law enforcement, reports of police harassment would be higher.

[9] Chi-square tests of independence reveal statistically significant associations for White residents (p<.001, 2-sided), Black residents (p=.001), and AAPI residents (p=.050), and marginally significant associations for Latino residents (p=.051). Respondents who reported no contact with law enforcement are included in these figures. If analyses were limited to only those who had contact with law enforcement, reports of police harassment would be higher.