The Research Project
Our research mainly addresses three major policing issues. First, are African Americans and other minority citizens more likely to find themselves victims of police brutality than white citizens? Second, are there certain situations that seem to have a greater likelihood of a violent police-citizen encounter? Third, are police departments taking allegations of violent police misconduct seriously?
Since there are no national statistics concerning police brutality available from any governmental source, we have used data derived from accounts of police misconduct in major newspapers for the period January 1, 1990 to May 31, 1992. We made use of one major Nexis database prepared by Mead Data Central. This database includes fifteen major newspapers updated on a daily basis and made available for computerized searches. These leading newspapers were selected by the Nexis service because of the size of their circulation and their major regional or national importance.
Using the keywords "police brutality," the following major national and regional newspapers were systematically searched for accounts of police malpractice in our time period: The Atlanta Journal and Constitution, The Boston Globe, The Chicago Tribune, The Christian Science Monitor, The Hartford Courant, The Los Angeles Times, The Minneapolis Star and Tribune, The New York Times, Newsday, The San Francisco Chronicle, The Seattle Times, The Wall Street Journal, USA Today, The Washington Post, and The Washington Times.
Out of the database of tens of thousands of articles, a total of 4,770 newspaper stories were retrieved that had at least a brief mention of "police brutality" in them. Due to the general nature of the phrase, not all of these stories dealt with a specific account of police brutality. Some of the articles dealt with "rap" songs that discussed the phenomenon, while others were based on political platforms of reform for candidates running for office. Many of the articles dealt specifically with the Rodney King incident and its aftermath. The King incident prompted many papers to broadly analyze and discuss their own local departments’ practices, while not discussing any specific case of alleged police brutality in any detail. Some of the retrieved articles dealt with the same incident, analyzing the encounter in greater depth or, in some cases, multiple newspapers reported on the same event. Combing through these 4,770 stories we found 130 distinct cases of police violence against civilians with enough detail for analysis.
We must note a few things about these data. For the purposes of this research we have accepted the designation of these incidents reported in these major newspaper accounts as cases of excessive police force, or "police brutality." It should be noted that newspapers use court, police, and community sources and there is a skew in the data toward the larger cities. Incidents of excessive police force were reported for cities in fifteen different states, with the greatest number in the largest states, New York and California. Five newspapers are located in the Midwest, and two are national digest papers that gave modest coverage to the problem of police violence. At the time of the data search, these fifteen media sources were the only newspapers available through the Lexis/Nexis service and no screening or selection process was used by the authors. 
The accounts of the acts of brutality varied in richness and depth. The more serious cases in which the officers were charged and tried were greatly detailed, while some other accounts of single police-citizen incidents presented only some basic details, thereby limiting the number of dimensions we can analyze in this paper.
The possibility of media bias in the major newspaper coverage of police brutality should be noted. In essence, the data is representative of the prevalence of news coverage of claims of police brutality and the level of media interest in these allegations. Why have these incidents found their way into the mass media, while many others have been ignored? In today’s racially charged climate, are violent police-citizen encounters with racial overtones more likely to garner the attention of the media than those with all white participants? Given the nature of the data source, these questions are impossible to answer. Conclusions based on this data should be evaluated with these limitations in mind.
Some white readers of our work have suggested that it is only the black and other minority cases that get into the mass media. However, it seems unlikely that an incident where white citizens were beaten up badly by police officers would not make its way into newspapers, especially if the victims were middle or upper class, as is sometimes the case for minority Americans. The opposite problem seems more likely. Martindale (1986: 133), in a discussion of the media and black Americans, has offered the following:
A vivid illustration of the black community’s concern with this problem was provided in the University of Washington seminar report, which noted that although the meeting was set up to consider media-black relations, a considerable amount of time was spent discussing police-black relations. While the black participants were willing to discuss the media, the report stated, they felt such a strong animosity toward the police that the presence of newsmen was sufficient to unleash a flood of criticism against the police. The participants apparently were eager to inform reporters about their grievances concerning police behavior in the black community because they felt that if journalists became aware of this situation and exposed it to the reading public, some corrective measures might be taken.... The media, however, continue to ignore this problem. They also seem prone to present the police only as upholders of law and order, black victims as possibly deserving of their fate, and black witnesses as probably unreliable.
White newspaper editors may feel pressure to affirm the integrity of the local police department and the desirability of the current societal organization. As phrased by Martindale (1986: 134), "Perhaps the media have failed to inquire deeply into this area because the truths that would emerge might prove uncomfortable to white middle-class society."