Sample Design
Twenty-six media outlets—7 newspapers, 18 broadcasts and 2 internet sites—were monitored for five weeks of coverage between February 2000 and June 2000. The first period began on February 7 and continued through February 13. The second period began March 13 and continued through March 19. The third period began April 24 and ran through April 30. The fourth period began May 15 and continued through May 21. The final monitoring period began June 5 and continued through June 11. This monitoring period provided a steady sampling of coverage from the heat of the primaries up through the eve of the convention kickoffs.
Newspapers were selected to develop a sampling of coverage by both national and regional publications. We studied a total of 7 newspapers: Washington Post, New York Times, Atlanta Journal & Constitution, Boston Globe, Indianapolis Star, San Francisco Chronicle and Seattle Times.
Television and radio programs, selected to provide the widest range of coverage, included the following:
Inclusion and Screening
The selected outlets were monitored first through the use of Lexis-Nexis. The search criteria was designed to cast the widest possible net so that we could capture all news stories having to do with George Bush or Al Gore and the presidential campaign. This provided a total of 2,004 news stories and 400 television and radio broadcasts or cablecasts.
Next we read each story individually to search for mentions of the six themes in our study: Bush coasts on family connections, Bush lacks intelligence, Bush is a different kind of Republican, Gore is scandal tainted, Gore exaggerates and Gore is competent and experienced. Rather than code the story as a whole, we wanted to examine each individual assertion of a theme. Therefore, every time a relevant statement was made, it was highlighted for inclusion.
To be included, a statement needed to assert or refute one the six character themes. There was a wide range of acceptable language for each theme. For example the wording of a statement about Bush coasting on family connections could be that he "has relied on his Dad's name to get him where he is" or "Without his family ties, Bush would not be never make it in national politics" or something else. In general, the statement needed to somehow convey or refute one of the six themes.
Some stories had five or six assertions while other had just one. Stories with no assertions were discarded. The resulting project sample consisted of 1330 specific statements—344 about Gore and scandal, 320 about Bush as a different Republican, 269 about Gore's exaggerations, 176 about Bush's intelligence, 132 about Gore's competence and 89 about Bush's family ties.
Coding Process
Each of the 1330 assertions was coded individually. Researchers worked with a detailed, standardized coding scheme. All assertions were first coded for basic inventory variables: date, outlet and story type. Next assertions were coded for content variables: source, evidence, context and whether it was a rebuttal of the theme.
Intercoder Reliability
Intercoder reliability measures the extent to which individual coders, operations independently of one another, reach the same coding decision. Tests were performed throughout the project: no systematic errors were found. When necessary, the coding supervisor reviewed decisions on content variables to be sure coders were in agreement.