Unrepresentative Information
The Case of Newspaper Reporting on Campaign Finance
STEPHEN ANSOLABEHERE is a professor of political science at Massachusetts Institute of Technology; ERIK C. SNOWBERG is at the Graduate School of Business, Stanford University; and JAMES M. SNYDER, JR., is a professor of political science and economics at Massachusetts Institute of Technology. Thanks to MIT librarian Anita Perkins who helped us liberate the Los Angeles Times from the grip of Media Conglomeration. The authors also thank the Carnegie Corporation for its support under the Carnegie Scholars program and Phillip Burrowes for his exceptional research assistance. Erik Snowberg would like to thank Dan Arnon and Mathieu Gagné for founding Oryxa, a company where I could earn my livelihood while still pursuing my life. Professor Ansolabehere and Professor Snyder thank the National Science Foundation for its financial support. All three authors thank Chappell Lawson for his helpful comments.
Address correspondence to James M. Snyder, Jr.; e-mail: millet{at}mit.edu.
This article examines evidence of sampling or statistical bias in newspaper reporting on campaign finance. We compile all stories from the five largest circulation newspapers in the United States that mention a dollar amount for campaign expenditures, contributions, or receipts from 1996 to 2000. We compare these figures to those recorded by the Federal Election Commission (FEC). The average figures reported in newspapers exceed the figures from the FEC by as much as eightfold. Press reports also focus excessively on corporate contributions and soft money, rather than on the more common types of donors—individual—and types of contributions—hard money. We further find that these biases are reflected in public perceptions of money in elections. Survey respondents overstate the amount of money raised and the share from different groups by roughly the amount found in newspapers, and better-educated people (those most likely to read newspapers) showed the greatest discrepancy between their beliefs and the facts.