No News Is Good News: Human Rights Coverage in the American Print Media, 1980–2000

Abstract

Several studies have analyzed media representations of human rights practices, but none has systematically examined whether those representations lead to skewed perceptions of repression or abuse. This phenomenon, referred to in the literature as the “information paradox,” poses serious problems for scholars attempting to gauge the extent of human rights violations or to evaluate measures taken to remediate them. This article considers whether increased awareness of and attention to human rights practices in fact distorts perceptions of abuse with an analysis of human rights coverage in two major American newspapers, the New York Times and the Washington Post, between 1980 and 2000. Multivariate regression analyses are employed to identify the characteristics of countries that are mentioned in conjunction with human rights in these newspapers. Results show that human rights coverage tends to be negative, with repressive countries garnering the most attention. However, human rights reporting appears not to be tainted by an information paradox, inasmuch as coverage does not vary systematically by a country's respect for press freedoms, linkage to international civil society, or the extensiveness of its communication infrastructure.

Wade Cole received his PhD in sociology from Stanford University and is currently an assistant professor of sociology at Montana State University in Bozeman. He researches the expansion and impact of global human rights norms, focusing especially on the characteristics of countries that join human rights treaties and the effect of those treaties on state practices. A forthcoming book also examines the emergence and spread of postsecondary institutions for indigenous peoples around the world and includes an analysis of indigenous sovereignty claims under international and domestic law. I thank John Meyer and Francisco Ramirez for their support of this project, which was partially funded by grants to them from the National Science Foundation, Sociology Program (NSF 0214168) and the Bechtel Initiative on Global Growth and Change, Institute for International Studies, Stanford University. I also thank David C. W. Parker for his comments on a previous draft, as well as Howard Ramos and James Ron for generously making their data on human rights coverage in The Economist and Newsweek available. None of these individuals or institutions, of course, bears responsibility for the conclusions reported herein.

Notes

1. See: http://www.accessabc.com/index.html (accessed December 8, 2008).

2. Nevertheless, it is noteworthy that Keck and Sikkink devoted less than a page to the information paradox in a book of more than 200 pages.

3. Coding the dependent variables in this manner may underestimate media coverage of human rights practices. In the above example, several clearly repressive countries were not coded using the within-10-words protocol. Thus, the variable is underinclusive. The alternative, coding mentions of country names and keywords if they appear anywhere in the same article, would likely be overinclusive. Moreover, the coding scheme does not capture articles in which the practices of countries, whether positive or negative, are discussed in terms other than “human rights.”

4. See: http://www.freedomhouse.org/template.cfm?page = 350&ana_page = 348&year = 2008 (accessed August 24, 2009). Prior to 1989, Freedom House assigned separate ratings to the print and broadcast media in each country. I assigned countries the lowest rating for these years. More detailed numerical scores on press freedoms, ranging from 0 (best) to 100 (worst), are available only from 1994 onward.

5. These estimates come from the New York Times and Washington Post models labeled “All” in Table 2 and are calculated as (exp[β]−1) x 100.

6. These findings were robust to the exclusion of GDP per capita, which correlates highly with the communications index (r = .79).

7. I thank an anonymous reviewer for suggesting this possibility.

8. An alternative measure of communication infrastructure is Internet connectivity. I conducted a cross-sectional analysis for the effect of Internet hosts and users per capita, from Banks (2008), on the number of human rights mentions in 2000 and found that connectivity tends to decrease the number of human rights mentions, but only slightly.

9. As a check on the results obtained from the random-effects models in Table 3, I estimated parallel fixed-effects regression models and found that only one interaction effect—between repressors and democracy for the NYT—was statistically significant. Its effect, however, was negative, which contradicts the information paradox thesis.

10. The communications index and civil war indicator correlated more strongly with the five-category human rights measure than with the binary repressor variable, which obscured their independent effects on the number of human rights mentions in Table 2.

11. Results obtained from analyses of the Ramos et al. data were substantively identical for random-effects models using a five-year panel design.

12. Results for each media outlet independently were broadly similar to those reported here.

13. These figures are based on keyword searches using the advanced search feature of the online New York Times archive (www.nytimes.com).

14. The United States, for example, has ratified the International Covenant on Civil and Political Rights but not the corresponding covenant on socioeconomic rights. The United States was also the only Western country to oppose the creation of a formal treaty body charged with monitoring compliance under the International Covenant on Economic, Social and Cultural Rights (Whelan and Donnelly 2007: 937).

15. According to Hafner-Burton (2008: 698), “the U.S. edition of Newsweek distributes the news to nearly 20 million people every year; fewer than half are college graduates and their median personal income is around $40,000. The Economist has a readership of just fewer than one million; 95 percent are college educated and their median personal income is around $150,000.”