As seen in...
ABC News
The Economist
Washington Post
The Atlantic
National Police Misconduct Reporting Project

Reporting Project – FAQs

What is the National Police Misconduct Reporting Project (NPMRP)?

The NPMRP is a non-governmental, non-partisan independent project that will attempt to determine the extent of police misconduct in the United States, identify trends affecting police misconduct, and report on issues about police misconduct in order to enhance public awareness on issues regarding police misconduct in the U.S.

What does “” have to do with the NPMRP?

The NPMRP is comprised of several different components and is the general term with which we describe the entirety of the project. One of those components, as it is currently designed, is the site where we present our statistical reports along with other police misconduct related stories and data.

Why do this?

Simply because nobody else does. Only a small fraction of the 17,000 law enforcement agencies actually track their own misconduct in a semi-public manner, and even when they do, the data they provide is generic and does not specify what misconduct occurred, who did it, and what the end result was.

What is the source of information that Cato uses to generate these statistics?

The Cato Institute uses media reports detailing both alleged and confirmed cases of police misconduct from all available media sources in the U.S. All information gathered is manually validated to determine the credibility of the report, whether the report is a duplicate of an existing report, and how each report should be categorized before recording each report to a police misconduct database for use in our statistical analysis.

How do we know the Cato statistics are genuine?

All the data used by the Cato Institute is publicly verifiable by anyone because we limit our findings to media reports of police misconduct, publish each report on a publicly available news feed via Twitter, and then publish a finalized list of reports.

Thus, anyone is free to dispute a report they see on the news feed or call any of the reports listed in our aggregate police misconduct database lists published on this site. We are as open and transparent as possible about our data and statistics because of the contentious nature of police misconduct in the U.S.

Isn’t there a drawback to only using media reports as a source for statistical analysis?

Yes, we accept that there is an indeterminate amount of under-reporting that exists within our statistics since the news media does not report on every complaint of police misconduct nor every lawsuit filed. However, there are problems with any method of data gathering in regards to police misconduct no matter which method is used. At least with this method, our sources of data are public and can be independently verified.

Why not use civil court records as your source of data?

Not all incidents of police misconduct are answered by civil lawsuits. In fact, as far as we can determine, only a fraction of the incidents that occur actually end up in litigation. Despite popular rumor, organizations like the ACLU take only a small number of cases of police misconduct each year and lawyers tend to be very reluctant to take on cases without payment up front. Thus, the margin of under-reporting would be far greater if we relied on civil court filings than our current method.

Why not use criminal court records as your source of information?

Just as with civil courts, very few instances of police misconduct are actually prosecuted. Generally, this is because most police departments are reluctant to charge fellow officers with criminal offense. Prosecutors are reluctant to pursue cases since they rely on the police in their work, and the general public has a strong bias that favors police testimony in court.

Why not use police department misconduct records?

There are over 17,000 law enforcement agencies in the United States. Most states have laws on the books that either prevent those departments from releasing disciplinary and internal investigation records or permit those agencies to keep those records secret. Even if it were feasible to submit public records requests to each law enforcement agency in the U.S. to give us all disciplinary and complaint records for an entire year, it is safe to say that the vast majority of agencies would be unresponsive.

Why not just let the U.S. Department of Justice track statistics?

The last time the Department of Justice generated a statistical report on police misconduct in the United States was in a report based on 2001 statistics that were voluntarily given to it by 5% of the police departments in the United States. That report was not very detailed and did not offer the level of information and geographic data that our reports can.

Why do you hate cops?

This project is about police misconduct. There is a fundamental lack of information about police misconduct in the U.S. and we are simply trying to do what we can to find the truth about how extensive a problem police misconduct really is, what types of misconduct are most prevalent, what factors increase or decrease the likelihood of police misconduct, and what trends might affect police misconduct rates.

The fact is this, without information about police misconduct, it is impossible to say whether police misconduct is a problem in the U.S. or not. We are simply trying to create a ruler with which we can measure police misconduct so that people can determine for themselves if it is really a problem they should care about.

Also, ordinary citizens are not the only people who become victims of police misconduct, sometimes police officers themselves are victims of misconduct and we track those reports as well. Police misconduct affects everyone, the public and police alike. By being open and transparent about police misconduct, this project can help police do their jobs better since gaining the trust of the public helps them gather the information and cooperation they need to do their jobs effectively.

Creative Commons License
This work by Cato Institute is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License.