Search

A gubernatorial debate takes place one October evening in Texas. On the local news in San Antonio later that night, reporters state that the debate focused on border issues, and clips are shown of the candidates discussing the need to inspect trucks crossing the border. Hundreds of miles north in Dallas, viewers of the local news learn that the same debate revolved around a recent crisis involving homeowners insurance. Meanwhile, in Michigan, viewers in both Detroit and Grand Rapids are treated to numerous campaign advertisements about various local congressional races. The candidates in these races receive no or very little coverage on the local news, unlike the nearly daily mentions enjoyed by their gubernatorial and senate counterparts. Instead, the congressional candidates must rely almost solely on paid advertising to reach the public - an expensive tactic, but also one that gives candidates a high level of control over the information conveyed to the public. As these anecdotes illustrate, local media coverage of elections embodies a number of dimensions that are not normally picked up in political communication research. In particular, researchers rarely look at both the content and volume of local media coverage, and how those dimensions vary depending on local and regional priorities, and the number, level, and competitiveness of races occurring in each media market. In addition, few studies have presented comprehensive case studies that allow for integrating data about information content and volume across media types. In this paper, we analyze local news broadcasts and campaign advertisements from the 2002 gubernatorial, senatorial and House races in three states: Michigan, Pennsylvania, and Texas. Within each state, we look at either two (in Michigan) or three (in Pennsylvania and Texas) media markets, which vary in terms of population, urbanization, number of races, and race competitiveness. Using the resources offered by the Wisconsin Advertising Project (CMAG) and the Wisconsin NewsLab, we analyzed all candidate advertisements in each market (which account for over 50,000 ad airings) and all highest-rated local news broadcasts from at least two stations in each market (in total, over 700 broadcasts). Our initial analyses, of Michigan and Texas, offer three main conclusions. First, local media is local media. Local news coverage and candidate advertisements for a single race often differ substantially from market to market. In addition, the level of the race and the type of medium matters for how much coverage candidates receive and, in turn, the level of control they have over the information conveyed to the public. Finally, campaign advertisements and news coverage do not exist independently of one another. Instead election information often goes through a cycle in which it breaks in one medium and is then picked up by the others. These findings suggest the value of looking across media types, across races, and across media markets in order to provide a more accurate picture of the information environment. [ABSTRACT FROM AUTHOR]