The Causes and Consequences of Differential Press Coverage of Legislators

Princeton University, Princeton NJ

Investigators

Abstract

This project explores how local newspapers covered members of the House of Representatives during a two-year period. The project involves searching all editions of the chosen newspapers for any articles, editorials, columns, or letters that mention the selected representatives, coding the items for their content, and analyzing the coded data. The project is designed to determine whether newspapers report the kinds of information that citizens need to hold representatives accountable for their actions in office. The assumption is that unbiased and readily available information enhances the prospects for accountability, while incomplete or biased information thwarts it. The study explores the regularity with which local newspapers report roll-call votes, the extent to which newspapers cover representatives' other activities, including introducing bills, holding hearings, and building coalitions, and the types of policy issues that attract journalists' attention. The project involves collecting four data sets to answer questions about the volume, content, causes, and consequences of newspaper coverage. It includes: (1) a systematic account of how 25 newspapers covered 25 representatives between January 1, 1993, and November 8, 1994, in 8,003 articles, editorials, columns, and letters; (2) a comparative analysis of how six pairs of newspapers covered six representatives from Boston, Chicago, San Francisco, Seattle, Tucson, and Washington during 1993 and 1994; (3) an analysis of the factors that affected how extensively 67 local newspapers covered 187 representatives during a two-year period, with a data set of 61 ,084 citations; and (4) an analysis of how the volume of coverage in local newspapers affected how much citizens knew about their local representatives, achieved by linking the data set of 61 ,084 citations with the 1994 survey conducted by the National Election Studies. The current proposal is for a project to clean and document these four data sets and prepare them for permanent archiving and public use. The project is the fIrst systematic study of how newspapers cover individual legislators outside the campaign period. It is also the first study to examine press coverage for a period longer than a few weeks. Most previous studies estimated the extent of press coverage for representatives without analyzing the content of coverage. The project demonstrates the utility of computers for identifying and retrieving articles about representatives from newspaper archives. All previous studies of how the press covered legislators have required that researchers browse through newspapers searching for articles that mention individual legislators. This study employs computers to do the browsing. Computers have the advantage of being more efficient and less error-prone than ordinary labor. Thus, they permit a vast increase in sample size. A better understanding of how the press covers representatives' actions should provide the foundation for a better understanding of citizens' electoral decisions, representatives' policy decisions, and the nature of political accountability in a large polity with separated powers and weak political parties.