How We Analyzed Our Police Officer Data

A used chemical agent grenade found on the streets in downtown Oakland following Occupy demonstrations in 2011. Photo by Eric K Arnold.

We began this project by examining 1,368 lawsuits and complaints filed against the Oakland Police Department between Jan. 1, 1990 and Jan. 1, 2013 that the City of Oakland settled out of court. We chose to focus on settled cases because the city has settled at least 75% of the cases filed against the OPD. The Oakland City Attorney supplied us with the dataset of lawsuits.

Using public records requests we obtained many of the reports that the City Attorney sent to the city council about each settlement. Using that information and state and federal court documents, we determined that in the last 22 years, 417 of the settled cases involved alleged civil rights violations. That dataset of 417 cases is the foundation for the stories we’ve published.

Data Sources

  • Oakland City Attorney: individual lawsuit settlement reports sent to the City Council; all case settlement data, including the city’s internal identification number for the case, court case number, dollar amount awarded to the plaintiff, court filing date, plaintiff name or names, plaintiff lawyer or lawyers, and case summary
  • Alameda County District Attorney: officer involved shooting reports
  • Oakland Police Department: officer involved shooting data, officer awards and commendations, officer names
  • US District Court, Northern District of California; Superior Court of California, Alameda County: case summaries
  • Federal Bureau of Investigation: national and regional crime statistics, which are compiled from data supplied by individual law enforcement agencies

The information in our databases comes directly from official court and city documents. Because of that, it may contain errors made by record keepers in the federal and state courts, the City Attorney’s office, and the police department. Please contact Editor Abraham Hyatt if you feel there is a problem.


City’s problems with data: When the City Attorney supplied us with the settlements dataset we found seven civil rights-related cases where plaintiff names from one case had been transposed with another. After we caught those mistakes, the city gave us new data. It contained misplaced plaintiff names as well.

Rather than try a third time, we used the case reports the City Attorney sends to the City Council to identify the correct plaintiffs. The city confirmed that the names we identified were correct. We encourage other researchers and journalists who obtain the full 1990-2013 settlement database from the city of Oakland to be extremely cautious when reporting on plaintiff names.

Case summary: While the cases in our database did not go to trial, we included the allegations because it’s necessary to show one of the main factors that influence why the City Attorney chose to settle each case. The summaries in our database were taken directly from court and/or City Attorney records.

OPD chiefs: We did not include OPD chiefs named in lawsuits in our database. Under the federal civil rights act, police supervisors can be held liable if department policies — including officer training policies — result in the violation of someone’s civil rights. Because of that, chiefs are regularly named in suits even when they did not have direct involvement in an incident. In those cases we chose to focus on individual officers who present at the time of the alleged misconduct.

Officer names: The City Attorney rarely mentioned officers by name in the reports for the City Council. We used PACER (the federal court system’s online portal) and both physical and digital documents from the Superior Court of Alameda County to identify officers named in the civil rights-related cases.

We only included officers who were explicitly named in court documents. We did not include officers who were removed from a case prior to it being settled. We made every effort to not include officers who were named in civil rights lawsuits but who were accused of non-civil rights violations. When there was a discrepancy in how officer names were spelled between our various sources of data, we used the spelling found on payroll records and employment lists provided by the OPD.

Most Decorated Officers Methodology

Our definition of most decorated included both the importance of the award an officer earned as well as the number of awards he or she received. Only officers who had received a medal — the highest recognition the department gives out — were included.

We began by asking the OPD for a list of all awards given between 2013 and 1990. The department gave us data going back to 2007 and informed us that they didn’t keep records of individual officer awards prior to that date. The data they gave us included 558 officers who’d earned a total of 2762 medals, awards and commendations.

We only included awards that were given for outstanding or notable performance or behavior, and that were given by a person inside the department: Advanced POST Bar, Good Conduct Bar, Outstanding Performance Evaluation Bar, Perfect Attendance Bar, Perfect Driving Bar, Recognition of Distinguished Service, Tactical Operations Team Bar, Unit Citation Award, Medal of Merit, Medal of Valor, Silver Star Medal.

We gave each type of award a numerical value based on their importance: Medal of Valor: 12, Medal of Merit: 11, Silver Star Medal: 10; other awards and bars, 3. We added up each officer’s score. An officer who, for instance, received two awards and a medal — i.e. Silver Star Medal, Perfect Driving Bar and Good Conduct Bar — would get a 16.

We ranked each officer by their score and then narrowed the list to the officers who had received three or more non-medal awards, which yielded a list of the top 35 officers.

We used the same ranking method to create the list of 100 officers who had won awards but not medals: We only included awards that were given for outstanding or notable performance or behavior; we gave each award a value (3) and then added up each officer’s score.

Limitations: We’re contrasting five years of awards with more that 20 years of lawsuits and officer involved shooting data. If we were able to include the full 20 years of awards it would potentially result in more or fewer officers who were involved in lawsuits and shootings near the top of the awards ranking.

However, this smaller data set still creates a contemporary snapshot of the department. While limited, the five years of awards still reflect the department’s current attitude towards officers who have been accused of using excessive force.

If you have questions about our data or methodology, please contact us anonymously us via our contact page or email Editor Abraham Hyatt directly.