Abstract

City Characteristics and Newspaper Coverage of NAFTA:
A Community Structure Approach

Since so many citizens identified the North American Free Trade Agreement (NAFTA) with employment/unemployment issues, this critical event is an ideal opportunity for studying the association between citizen/influential concerns and newspaper coverage. Using a DIALOG national newspaper database, locally generated articles published from September 1, 1993 to December 31, 1993, were analyzed in twenty major newspapers representing a geographic cross-section of cities in the United States. A content analysis technique evaluated both the "amount of attention " an article received and its"direction" to yield a single score for each newspaper. Those scores were compared with a variety of city characteristics to test hypotheses associating several aspects of community structure with reporting variations.

Employing correlation and multiple regression analyses, one key factor was found clearly associated with coverage favoring or opposing NAFTA. Occupational sector, specifically percent employed in manufacturing, outweighs the factors of education, income and occupational status in its association with reporting on NAFTA. Contrary to positions taken by organized labor, the higher the proportion of the labor force employed in manufacturing in a city, the more positive that city's reporting on NAFTA is likely to be. A possible umbrella explanation for the positive association of media coverage and manufacturing is the "guard dog" conception of media as sentries for established groups in social conflict. This perspective, articulated by Gans (1979) and Olien, Donohue, and Tichenor (1995), suggests that media news is primarily about those at or near the top of power hierarchies and those low in the hierarchies who threaten the top. In this case, media coverage was more consistent with the interests of owners and senior managers in manufacturing than with the interests of hourly wage-earners. Mapping coverage of the NAFTA debates demonstrates that archival data comparing newspaper databases and city characteristics can reveal significant variations in reporting on political and economic policies.