Covering the Epidemic:
AIDS in the News Media, 1985-1996
A Content Analysis
METHODOLOGY
National newspapers were selected on the basis of circulation, distribution, and audience; geographic distribution and diversity of parent corporation were factors in regional selections. News stories were downloaded in their entirety from the NEXIS on-line data base.
Broadcast news was provided by the Vanderbilt University Television News Archives, taped in Nashville, Tennessee according to published televisions and radio timetables. Thus, the television archives are subject to preemption by the local affiliates, a common occurrence on weekends.
| NATIONAL NEWSPAPERS
| | | |
| The New York Times | | | daily publication |
| The Washington Post | | | daily publication |
| USA Today | | | M-F publication |
| | | |
| REGIONAL NEWSPAPERS | | | |
| The San Francisco Chronicle | | | daily publication |
| The St. Louis Post-Dispatch | | | daily publication |
| | | |
| BROADCAST NEWS | | | |
| ABC World News Tonight | | | daily broadcast |
| CBS Evening News | | | daily broadcast |
| NBC Nightly News | | | daily broadcast |
Screening and Inclusion
All Typical News Weeks and Major News Event Weeks were searched in their entirety, on a Sunday-Saturday system. (Exception: Magic Johnson Week I, which began on a Thursday and continued for 11 days; however, all totals for that week have been adjusted to reflect only the first seven days of coverage.) If the terms HIV or AIDS appeared in the news story, it was included in this study, with some exceptions:
If the story merely mentioned HIV or AIDS but was truly focused on another issue, it was catalogued, but not coded for all tone and policy-related variables.
For broadcast, all references were coded, but anchor lead-ins of less than 35 seconds were considered part of the upcoming report unless it was clear that the lead-in was intended to stand alone.
Intercoder Reliability
Intercoder reliability measures the extent to which coders, operating autonomously, code or classify the same story the same way. Intercoder reliability tests were performed throughout this study, and no significant differences were found to exist on a recurring basis.
