CITY ST, 20XX

USE OF FORCE

Use of force occurs when an officer uses or threatens to use their body or an object against a person or in a way that could cause pain, injury, or death.

USES OF FORCE BY RACIAL GROUP [V1/2, standard]

According to LEA NAME data on use of force incidents recorded between 20XX and 20XX:

What does this show?

Each dark blue bar shows the percentage of the total use of force incidents recorded in the assessment period that were against people in one racial group. The orange bars show the percentage of the resident population that are people of the same racial group. Hovering over the blue bars shows the number of incidents that makes up that percentage.



Departments have varying definitions of force and requirements for when officers must report their use of force, so those with more comprehensive reporting requirements may have more recorded incidents than departments with vague or incomplete requirements. For example, some departments do not require officers to report use of hands-on force unless it results in injury or complaint of injury; others do not require officers to report pointing a weapon if it was not discharged. Departments also use different forms to record data on use of force incidents, and forms that encourage more comprehensive or efficient data collection can provide more useful information for analysis than those that solicit fewer, or less specific, incident details.

Our guidance encourages officers to report every incident in which an officer uses or threatens to use their body, a tool, or a weapon against a person, or in any way that could cause pain, injury or death, regardless of the officer’s motivation or whether any injury or complaint results. For more information on the types of incidents officers were required to report during this assessment timeframe, see “What counts as force for these analyses?” at the bottom of this page.

How was this calculated?

We first took the average total recorded incidents per year and calculated the percentage against people of each racial group. Then we compared those percentages to the percentages of the resident population that are of each racial group. See the Data Notes tab for information on how we define racial groups.

We count a use of force incident as a single incident in which any use of force against a person was recorded, regardless of the number of officers, types of force involved, or number of times force was applied. If more than one person had force used against them at the same time, each person who was subjected to force is counted as a separate incident. Our analyses exclude incidents where the only reported type of force is handcuffing, verbal commands, or de-escalation.

Most of our analyses use all use of force data provided by departments, including incomplete years of data. However, certain analyses require complete years of data, so time periods may vary across charts.

We use local demographic data (from the Census Bureau’s American Community Survey 5-year estimates) as the most straightforward and complete representation of the local population. The use of Census data also allows us to perform standardized analyses across law enforcement agencies. We recognize that this measure of demographics may not capture the entire population of individuals with whom police interact. However, the analyses on this page can help shed light on the role that local demographics may play in any observed disparities.

Data required for this analysis:

USES OF FORCE BY RACIAL GROUP [V2/2, homogenous]

insufficient_data

According to LEA NAME data on use of force incidents recorded between 20XX and 20XX:

What does this show?

This chart shows how often police used force on people of each racial group, relative to their share of the residential population of the jurisdiction. If any racial groups have force used on them considerably more often than White people, this may be indicative of racial disparities. 

Departments have varying definitions of force and requirements for when officers must report their use of force, so those with more comprehensive reporting requirements may have more recorded incidents than departments with vague or incomplete requirements. For example, some departments do not require officers to report use of hands-on force unless it results in injury or complaint of injury; others do not require officers to report pointing a weapon if it was not discharged. Departments also use different forms to record data on use of force incidents, and forms that encourage more comprehensive or efficient data collection can provide more useful information for analysis than those that solicit fewer, or less specific, incident details.

Our guidance encourages officers to report every incident in which an officer uses or threatens to use their body, a tool, or a weapon against a person, or in any way that could cause pain, injury or death, regardless of the officer’s motivation or whether any injury or complaint results. For more information on the types of incidents officers were required to report during this assessment timeframe, see “What counts as force for these analyses?” at the bottom of this page.

How was this calculated?

To calculate this, we divided the number of use of force incidents against people of each racial group by the number of the jurisdiciton’s residents of that racial group, and multiplied by 1,000. This gives us the number of use of force incidents per 1,000 residents for each racial group. 

We count a use of force incident as a single incident in which any use of force against a person was recorded, regardless of the number of officers, types of force involved, or number of times force was applied. If more than one person had force used against them at the same time, each person who was subjected to force is counted as a separate incident. Our analyses exclude incidents where the only reported type of force is handcuffing, verbal commands, or de-escalation.

Most of our analyses use all use of force data provided by departments, including incomplete years of data. However, certain analyses require complete years of data, so time periods may vary across charts.

We use local demographic data (from the Census Bureau’s American Community Survey 5-year estimates) as the most straightforward and complete representation of the local population. The use of Census data also allows us to perform standardized analyses across law enforcement agencies. We recognize that this measure of demographics may not capture the entire population of individuals with whom police interact. However, the analyses on this page can help shed light on the role that local demographics may play in any observed disparities.

Data required for this analysis:

USES OF FORCE BY YEAR

The number of use of force incidents recorded each year with complete data varied from a low of XX in 20XX to a high of XX in 20XX.

Each law enforcement agency’s policy manual requires that officers record use of force, and defines the behaviors that count as “force” which must be reported. For this reason, the number of incidents recorded each year may fluctuate due to changes in officers’ use of force practices or departments’ reporting policies. Departments should investigate whether variations in use of force may be related to policy changes that govern officers’ use of force or define the types of behavior that count as force which officers are required to report.

What does this show?

This chart shows the number of use of force incidents recorded during each year of the assessment period for all racial groups combined.

Departments have varying definitions of force and requirements for when officers must report their use of force, so those with more comprehensive reporting requirements may have more recorded incidents than departments with vague or incomplete requirements.  Disruptions related to the COVID 19 pandemic may have affected the volume of police-community interactions recorded in 2020 and 2021, although the extent of this impact likely varies by jurisdiction according to the local policies and restrictions implemented in response to the outbreak. It is worth noting whether or the extent to which racial disparities persisted over this time period, despite the reduction in overall police activity.

How was this calculated?

We count a use of force incident as a single incident in which any use of force against a person was recorded, regardless of the number of officers, types of force involved, or number of times force was applied. If more than one person had force used against them at the same time, each person who was subjected to force is counted as a separate incident. Our analyses exclude incidents where the only reported type of force is handcuffing, verbal commands, or de-escalation.

Data required for this analysis:

COMPARING USE OF FORCE INCIDENT RATES

After using a statistical technique called regression analysis to account for the influence of different crime rates, poverty levels, and percent of Black residents in neighborhoods:

We take into account the share of Black residents, crime rates, and poverty levels in a neighborhood because these factors affect the likelihood that a person of any racial group in a neighborhood will have police contact. For details on CPE’s regression analysis, see “More Information” below.

What does this show?

This infographic displays findings from CPE’s regression analysis, a statistical technique that allows CPE to investigate differences in use of force rates by race, taking into account other socioeconomic factors that may affect policing strategies and deployment. Specifically, this regression tests how much more or less likely each racial group is than White people to have force used on them in a neighborhood with an average poverty rate, crime rate, and percentage of Black residents – three factors commonly associated with increased police contact. The results of this analysis show the size of racial disparities in use of force that remain even when the influence of poverty levels, crime rates, and the percentage of Black residents across neighborhoods are removed from the equation.

We take into account the share of Black residents, crime rates, and poverty levels in a neighborhood because these factors affect the likelihood that a person of any racial group in a neighborhood will have police contact. This relationship between police presence and the percent of Black residents in a neighborhood is, in part, a result of systemic racism and structural disadvantage (for example, a lack of community services can lead to more calls for police service). But police-driven factors, such as departmental policy or officer behavior, also contribute to increased police activity in neighborhoods with more Black residents, crime, and poverty.

How was this calculated?

To represent neighborhoods, we use Census tracts — small geographic areas of approximately 4,000 residents — as defined by the Census Bureau. We use publicly available Census data to measure the percentage of Black residents in each neighborhood

To measure serious crime rates, we count crimes in each neighborhood that are recorded by the department. Specifically, we count reports of Part 1 offenses. The FBI’s Uniform Crime Reporting Statistics defines Part 1 offenses as: murder and non-negligent homicide, rape , robbery, aggravated assault, burglary, motor vehicle theft, larceny, and arson. Racial groups that made up less than 2% of all use of force incidents, or which had fewer than 40 total incidents, were excluded from this analysis (see the Data Notes tab for information on how we define racial groups).

Data required for this analysis:

PERCENTAGE OF USE OF FORCE INCIDENT FREQUENCY EXPLAINED BY NEIGHBORHOOD FACTORS

This chart displays findings from CPE’s regression analysis, a statistical technique that tests how much neighborhood poverty levels, crime rates, and share of Black residents—three common explanations for increased police contact—contribute to how often all recorded use of force incidents occur.

Statistical analysis showed that neighborhood crime rates, poverty, and share of Black residents explained XX% of the frequency of use of force, while XX% was not explained by these factors.

For details on CPE’s regression analysis, see “More Information” below.

What does this show?

This infographic displays findings from CPE’s regression analysisa statistical technique that investigates how certain factors contribute to how often all use of force incidents occur. Specifically, it shows the results of testing how much neighborhood poverty levels, crime rates, and share of Black residents—three common explanations for increased police contact—are contributing to the frequency of use of force overall. The results of this analysis show that the frequency of use of force is largely not explained by (or predicted by) these external factors. It is likely that factors within the control of the department, such as departmental policy and practice or officer behavior, play a significant part in determining when, where, and who is subjected to force.

Our model cannot precisely distinguish the extent to which incident frequency is due to reasons within or outside of a department’s control, but we are able to use statistical tools to account for three of the most commonly proposed, quantifiable factors related to frequency of police contact. Thus, our model controls for the predictive effect of crime rates, poverty levels, and racial demographics on use of force frequency, but there are many additional factors that may have an impact on whether a person is likely to come into contact with law enforcement and subsequently be subjected to force.

How was this calculated?

To represent neighborhoods, we use Census tracts, which are small geographic areas of approximately 4,000 residents each, defined by the Census Bureau. We use publicly available Census data to measure neighborhood poverty levels and the percentage of Black residents.

To measure serious crime rates, we count crimes in each neighborhood that are recorded by the department. Specifically, we count reports of Part 1 offenses. The FBI’s Uniform Crime Reporting Statistics defines Part 1 offenses as: murder and non-negligent homicide, rape (legacy and revised), robbery, aggravated assault, burglary, motor vehicle theft, larceny theft, and arson. Racial groups which made up less than 2% of all incidents, or which had fewer than 40 total incidents, were excluded from this analysis (see the Data Notes tab for information on how we define racial groups).

For an explanation of why we measure the share of Black residents as a potential factor influencing use of force, see “More information” under the analysis above.

Data required for this analysis:

TOTAL USES OF EACH FORCE TYPE

The 2 most common force types recorded overall were “Holds” and “Takedown.”

What does this show?

This chart shows the total number of uses of each particular force type recorded for all racial groups combined.

How was this calculated?

We took the total recorded uses of force and separated them into force type. We count each distinct type of force that was recorded against any person as one force type, regardless of the number of other force types recorded against a given person, the number of officers who applied it, or the number of times it was applied. We combine categories for easier interpretation. See the Data Notes tab for details on how we sort force types provided by the department into these categories, and how we define racial groups.

Data required for this analysis:

FORCE TYPES USED, BY RACIAL GROUPS

According to LEA Name data on the types of force that deputies recorded using:

Identifying officers’ most commonly used force types can help serve as a starting point for addressing racial disparities and decreasing use of force overall. If the results show that use of force is more frequent or racially disparate in incidents involving a particular force type, departments should investigate their policies and practices relating to that force type. If disparities are consistent among uses of all or most types of force, departments should explore broad interventions that reduce the use of all types of force, such as changes to departmental policy and organizational culture.

Note: Force types used less than 5 times are not included in the finding text for this analysis.

What does this show?

Each colored bar shows the percentage of uses of each force type recorded against people of each racial group. Hovering over a colored bar shows the number of uses of each force type that makes up that percentage. Any force type that was recorded in high numbers, or in which racial disparities were large, will influence overall racial disparities in use of force.

How was this calculated?

We took the total recorded uses of force and separated them into force type. Then we calculated the percentage of each type that was against people of each racial group.

We count each distinct type of force that was recorded against any person as one force type, regardless of the number of other force types recorded against a given person, the number of officers who applied it, or the number of times it was applied. We combine categories for easier interpretation. See the Data Notes tab for details on how we sort force types provided by the department into these categories, and how we define racial groups.

Data required for this analysis:

USE OF FORCE INCIDENTS BY WORK UNIT AND RACIAL GROUP

Incident Totals For Each Work Unit, Separated by Racial Group

insufficient_data

What does this show?

“Work unit” describes the work groups in a department. It can refer to the assignment of the officer who applied force (e.g. Detective Unit, Narcotics, Traffic, etc.), or the geographic areas where incidents occurred (e.g. precincts, districts, zones, etc.).

Each colored bar shows the percentage of use of force incidents recorded by each work unit of people of each racial group. The Multiple Work Units category, if used, represents incidents involving officers from two or more work units. Hovering over a colored bar shows the number of incidents that make up that percentage. The grey bars on the right show the total number of incidents recorded by each work unit. Any work unit that records a large number of use of force incidents or records large racial disparities will influence overall racial disparities in use of force. If disparities are present among most or all work units, the different racial makeup of various neighborhoods is likely not the whole explanation for the observed disparity.

How was this calculated?

We took the total recorded incidents of force and first separated them by the work unit that recorded the incident. We then calculated what percentage of incidents was recorded for people of each racial group.

The “Other Work Units” category, if used, combines the work units recording less than 2% of incidents. See the Data Notes tab for information on how we define racial groups.

Data required for this analysis:

WHAT COUNTS AS FORCE FOR THESE ANALYSES?

We reviewed LEA NAME’s policies that defined the behaviors deputies were required to report as force during the assessment timeframe. Understanding what types of incidents were required to be recorded, as well as gaps in what was required to be recorded, can help determine whether any incidents might have been missing from the provided dataset and reveal opportunities for improving data collection policies and practices.

We encourage departments to adopt use of force policies that clearly define what counts as reportable “force” and explicitly include every type of behavior that officers are required to report as a use of force incident. A comprehensive use of force reporting policy would require officers to record every incident in which they use their body, a tool, or a weapon to overcome resistance, secure compliance, or in any way that could cause pain, injury or death, regardless of the officer’s motivation or whether any injury or complaint results. CPE’s resources provide further details on how law enforcement agencies can effectively collect use of force data and communities can advocate for comprehensive use of force reporting.

In this section, we list reportable uses of force according to the policy provided by the department.

INSERT DEFINITION 

Scroll to Top