ABSTRACT
Official police records were obtained from the Houston Police Department covering the 4,980 incidents of criminal homicide between 1986 and 1994. These data contained the following variables: victim and offender name, incident location (street address), motive, victim/offender relationship, weapon used, and victim and offender age, gender, and race. These police records were then compared with a newspaper article database consisting of every homicide article published in the Houston Chronicle between 1986 and 1994. Importantly, this database was received directly from the Houston Chronicle in an electronic format, indicating that it is a potentially more reliable and accurate account of all homicide articles than a manual search of newspapers because it is the official list as known by the Houston Chronicle organization. Every homicide incident for the nine-year period was checked with the Houston Chronicle database to determine several important factors concerning news coverage. The following specific questions were addressed: Was there an article covering the homicide? If so, how long was it in both column inches and word length? How many articles were there about the incident? In what section and page of the newspaper was the article placed? Was there a picture with the article? In addition to these questions, other variable were examined, including the mention of race, incident location, motive, victim-offender relationship, weapon, and what stage of the criminal justice process the article covered. Simultaneously, all articles dealing with intimate homicides were separately analyzed by placement into a Word file. This qualitative file was then further divided into two separate files, one containing all articles covered within the first fifteen pages of the newspaper and the second containing all
other intimate homicide articles. Finally, using the incident location provided by the official police data, these homicide incidents were geo-coded for use in spatial analyses within the study.
