2010 Quarterly Q3 Report

Figure 1. Map displaying number of officers involved in reports tracked per county from 01/10-09/10 (clicking on this map will redirect to interactive map at www.targetmap.com)


This is the National Police Misconduct Statistics and Reporting Project (NPMSRP) Police Misconduct Statistical Report for the third quarter of 2010. This report is the result of data captured from January 2010 through September 2010 by the NPMSRP consisting of reports that meet credibility criteria which have been gathered from multiple media sources throughout the United States. For more information about the NPMSRP, the process used to gather data on police misconduct, and other information about our reporting process please visit our FAQ page or About page. You can also review older statistical reports and ancillary reports here.


From January 2010 through September 2010 the National Police Misconduct Statistics and Reporting Project recorded 3,814 unique reports of police misconduct that involved 4,966 sworn law enforcement officers and 5,711 alleged victims.

  • 3,814 – Unique reports of police misconduct tracked
  • 4,966 – Number of sworn law enforcement officers involved (263 were sheriffs or chiefs)
  • 5,711 – Number of alleged victims involved
  • 193 – Number of fatalities associated with tracked reports
  • $213,840,800 – Estimated amount spent on misconduct-related civil judgments and settlements

Misconduct by Type

Of the 4,966 officers involved in reported allegations of misconduct that met NPMSRP criteria for tracking purposes, 1,242 were involved in excessive force reports which were the most prominent type of report at 25% of all reports. This was followed by sexual misconduct complaints at 10.4% of officers reported then drunk driving and drug investigations at 7.9% of all officers reported, (62% of those were DUI allegations). The following chart displays the breakdown of misconduct types by percentage of reports and the number of reports each by type.

Figure 2. Police misconduct by type

Excessive Force

Figure 3. Map displaying number of officers involved in excessive force cases within the first 3 quarters of 2010. (clicking on this map will redirect to interactive map at www.targetmap.com)

Of all 1,242 officers involved in excessive force complaints, 735 (58.4%) were involved in cases of physical use of force complaints which include fist strikes, throws, choke holds, baton strikes, and other physical attacks. 175 officers (14.1%) were involved in firearm-related excessive force complaints, 126 (10.1%) were involved in taser-related cases, and the remaining officers were involved in other cases involving use of police dogs (1.7%), police vehicles (0.2%), chemical weapons (3.1%), or mixed use of force cases that involved a combination of attacks (11.6%).

Figure 4. Excessive Force by Type

There have been 92 fatalities associated with credible excessive force allegations within the first three quarters of 2010, which means 7% of excessive force cases involved fatalities. Of these excessive force fatalities, 68 were caused by firearms, 15 were caused by physical force, 11 by taser, and 3 by other causes. *Note: fatalities listed are only those involved in cases where excessive force or unnecessary force was reported. This does not include all fatalities related to police use of force.

Figure 5. Excessive Force Fatalities by Type

Sexual Misconduct

Officer-involved sexual misconduct describes an entire subset of police misconduct that includes non-criminal complaints such as consensual sexual activity that occurs while an officer is on-duty, sexual harassment, and acts of sexual assault or molestation. Sexual misconduct is the second most common form of misconduct reported through the first three quarters of 2010 with 517 officers involved in sexual misconduct complaints during that period, 297 of which were involved in complaints that involved non-consensual sexual activity such as sexual assault or sexual battery.

Figure 6. Officers involved with sexual misconduct by percentage of incidents involving children or adults.

Of all sexual misconduct complaints, 197 officers were associated with complaints that involved children, 86 were associated with sexual harassment complaints, and 234 officers were involved in more serious sexual misconduct complaints.

Figure 7. Alleged victims of officer-involved sexual assaults by age.

For all sexual misconduct complaints there were 419 victims of misconduct that could be classified as sexual assault, rape, sexual battery, or molestation. Of these 419 victims of more serious types of sexual misconduct, 219 were minors. This would appear to indicate that officers involved with sexual misconduct would appear to be more likely to have multiple victims when minors are involved than adults.

Misconduct Per Capita

Figure 8. Map displaying the Police Misconduct Rate by state. (clicking on this map will redirect to interactive map at www.targetmap.com)

The current US average projected police misconduct rate is an estimated 992.81 officers per 100,000 officers (mean 909.61 per 100k) as calculated using data gathered from the first 9 months of 2010. This is also a slight increase over last year’s estimated average of 980.64 officers per 100k and an increase over last quarter’s midyear reported average of 970.57 per 100,000.

Figure 9. Police misconduct rate by state with corresponding number of officers involved per state.

When current data is filtered to examine only incidents that can be classified as violent crimes as specified per the US FBI/DOJ Uniform Crime Reporting standards and then compared with the 2009 FBI/DOJ UCR Crime in the United States report as a per capita general population and per capita law enforcement basis the results indicate that overall violent crime rates are not too divergent between the two population groups with a difference of only 20.1 per 100k point between the two. However, there appear to be some more significant differences at a more granular level with robbery rates for police far below those reported for the general population but assault and sexual assault rates significantly higher for police when compared to the general population.

Figure 10. Violent Crime Rate comparison between general population UCR data and law enforcement population NPMSRP data.

Geographic Distribution

On a state by state basis, 20 states currently have a police misconduct rate above the US average of 992.81 per 100k and the states showing the lowest misconduct rates include Kansas with a rate of 276.09, Maine with 310.97, and Virginia at 502.10. Here are the 20 states with misconduct rates currently above average:

Figure 11. States with misconduct rates above the US average.

(Note: This chart includes our “Transparency Index” which is a method under development to rank agencies or states according to how transparent misconduct reporting appears to be in order to determine if data reported is under-reported or closer to actual rates. The current average index is 1.32 and 0.0 is the most transparent.

At the agency level the NPMSRP splits groupings by agency size in order to reduce the effect that small sample sets in agencies with fewer officers might have on the resultant rankings. Currently this split divides agencies into four different groups with the first including agencies with 1,000 sworn law enforcement officers or more. The second grouping includes agencies with between 500 and 999 officers, then 100 to 499, and finally 50 to 99 officers. We do not rank agencies that have fewer than 50 officers since the sample sets for those agencies are too small for reliable comparative statistical analysis.

1000+ Officer Agency Rates

The following chart displays the 20 agencies with 1,000 or more sworn law enforcement officers with the highest misconduct rates for that group of agencies:

500 – 999 Officer Agency Rates

The following chart displays the 20 agencies with between 500 to 999 sworn law enforcement officers with the highest misconduct rates for that group of agencies:

100 – 499 Officer Agency Rates

The following chart displays the 20 agencies with between 100 to 499 sworn law enforcement officers with the highest misconduct rates for that group of agencies:

50 – 99 Officer Agency Rates

The following chart displays the 20 agencies with between 50 to 99 sworn law enforcement officers with the highest misconduct rates for that group of agencies:

Police Misconduct Trending

While the overall US average police misconduct rate appears to be climbing in comparison to both last year’s rate and the previously reported rate 3 months ago it is difficult to see a clear causative factor for the increase and it isn’t clear what type of misconduct is increasing to cause this trend though the number of officers involved in excessive force reports appear to be demonstrating an overall trend increase since the beginning of 2010.

Figure 12. Officers involved in misconduct reports tracked per month.

Figure 13. Officers involved in excessive force reports per month

Figure 14. Officers involved in sexual misconduct reports per month

While overall misconduct appears to be trending higher, disciplinary actions against officers and the number of convictions on criminal charges appear to be relatively flat overall. Also, while conviction rates do not show any correlation with the number of reported officers, internal disciplinary rates do appear to show a very slight matching trend.

Figure 15. Trends for officers reported compared officers disciplined and officers convicted

When examining the trending data on a state by state basis we see that states that are trending upward outnumber states with a downward trend for misconduct by 28 to 24. The largest differentials between 2009 and 2010 have been in Washington DC with a decline from a 2009 PMR of 2313.33 to a 2010 PMR of 685.60 and Oklahoma with an increase from 1027.41 to 2013.90 per 100k.

Figure 16. Misconduct Rate comparisons by state from 2009 to 2010


With general misconduct rates showing an increase over last year and over the previous reporting cycles this year as well, it is difficult to pinpoint why the rates are climbing slightly, especially since excessive force rates don’t appear to be climbing at the same rate, though excessive force rates are climbing as well. While some states have exhibited large fluctuations between 2009 and 2010, most states showed a deviation of under +/-25% and 7 showed a deviation of under +/-10%. 8 out of 10 States showing the largest deviation between 2009 and 2010 had law enforcement populations under 10,000 which is as expected since smaller sample sizes tend to be most susceptible to individual incident influences. Since the NPMSRP has only been operational since April of 2009, trending data is limited in scope and long-term trending data is unavailable at this time, hopefully we will be able to generate more meaningful and useful trending information as the project continues.

General responses to police misconduct on a judicial/criminal justice level appear unchanged with no corresponding fluctuation to the increase in general misconduct rates which demonstrates a possible bias built into the justice system which continues to limit prosecutorial effectiveness against law enforcement officers charged with criminal wrongdoing. This appears to correspond to a previous study performed by the NPMSRP showing a large disparity in conviction and incarceration rates between law enforcement officers and the general public.

Accordingly, we’ll perform another comparison between conviction and incarceration rates in our end-of-year statistical report in order to examine the issue further and look to trend that data as well. However, on potential positive was to see an apparent correlation between report rates and internal disciplinary rates.

One of the persistent problems the NPMSRP faces is determining whether reporting rates are abnormally low for any given state or agency based on how effective that agency or that state’s laws are at keeping misconduct information hidden from the public. The NPMSRP is currently in the process of implementing a “Transparency Index” that might be used to determine if laws or actions meant to hide misconduct information from the public are affecting rates for a given agency or state. While currently in a beta phase, we hope to have that index fully functional for our 2011 reporting cycle.


The National Police Misconduct Statistics and Reporting Project began in April of 2009 in order to address the lack of statistical data concerning police misconduct in the United States. Despite becoming a more prominent issue in landscape of American public opinion, police misconduct is still a largely unstudied issue and no other sources of current statistical and trending data exist with which we could use to analyze the nature, persistence, and prominence of police misconduct in America. The NPMSRP has been created to address this gap and, in doing so, hopefully help address the causative factors of police misconduct in the process.

The NPMSRP utilizes the only consistent source of data available for police misconduct information since most states currently have laws that prevent the examination of police misconduct information recorded by individual agencies themselves by the public and no other agency tracks police misconduct data in any publicly available way. Therefore the NPMSRP must rely on media reports of police misconduct in order to gather data for statistical and trending analysis.

Reports of police misconduct are recorded in an internal database where these reports are analyzed at the end of each quarter in order to filter out duplicate reports and adjust for status changes for previously recorded incidents. This filtered data is then used to generate our quarterly and yearly reports which are also tied to a public release of the underlying data for public review. In order to maintain credibility the NPMSRP does not record all reported police misconduct allegations but uses a set of criteria in order to limit recorded reports to only those reports which appear to be credible and which exclude minor internal matters such as tardiness or other minor policy infractions.