Protest events in international press coverage: An empirical critique of cross-national conflict databases

Abstract

The empirical analysis of protest events and other expressions of social conflict is one of the core tasks of the discipline of comparative sociology. The numerous international data sets and empirical country comparisons that rely exclusively on reporting in English-language newspapers such as The New York Times in surveying protest events nevertheless suffer from considerable distortions. Using the example of some 1800 protest events in Argentina, Mexico and Paraguay in the year 2006, the present study shows that there are remarkable differences between national and international (English-language) newspapers when it comes to frequency of reporting. On the one hand, a mere 5.3 percent of all protest events that are reported nationally also attract the attention of the international press. On the other hand, the percentage of international reporting depends considerably and to a statistically significant extent on the country in which the protest takes place. Besides, these country differences persist when additional protest characteristics (e.g. the number of participants, the participation of renowned personalities and the escalation of the protest into rioting) are controlled by means of multivariate logistic regressions. The measurement error that results when surveying protest events on the strength of coverage in the English-language international press is therefore not constant across countries. The frequently used data sets like the World Handbook of Political and Social Indicators or the Cross-National Time Series Data Archive launched by Arthur Banks are thus proving to be highly questionable sources for international comparisons in protest and conflict research.

News and Geolocated Social Media Accurately Measure Protest Size Variation

“…The estimates in newspapers therefore are not as haphazard as their lack of methodological reporting would suggest. Second, CCC generates its data by accepting submissions of protest reports instead of managing a team of researchers; adopting this open-source approach means it includes a broader range of sources than most event datasets (Herkenrath and Knoll 2011;Nam 2006).…”

Terrorism and Corruption

“…The selection of data sources for political violence can be troublesome (LaFree and Dugan 2007;Olzak 2006). Research has shown that databases limited to international media reports do not provide adequate representation across the globe (Herkenrath and Knoll 2011). Fortunately, a more detailed and comprehensive database on incidents of terrorism has recently become available.…”

What Should We Do about Source Selection in Event Data? Challenges, Progress, and Possible Solutions

“…the number of participants) (Barranco and Wisler 1999;McCarthy et al 1996;Mueller 1997a;Oliver and Myers 1999;Fillieule 1998;McCarthy et al 2008;Herkenrath and Knoll 2011); (2) the geographic distance between the event and the media source, especially its reporting market and audience (Barranco and Wisler 1999;Hocke 1998;McCarthy et al 1996;Mueller 1997a;Fillieule 1998;Strawn 2008;Herkenrath and Knoll 2011); (3) Paralleling these general patterns for events, Amenta et al (2009) find that New York Times mentions of social movement families are greater for larger, better organized and disruptive movements that use protest and those with an enforced governmental policy in place. They also find that the power of partisan allies does not influence selection.…”