Statement of Method
Updated June 2019
We monitor 96 sources (across newspapers, radio and TV) in 43 countries in seven different regions around the world. We assemble the data by accessing archives through the Lexis Nexis, Proquest and Factiva databases via the University of Colorado libraries. These sources are selected through a decision processes involving weighting of three main factors:
- geographical diversity (favoring a greater geographical range)
- circulation (favoring higher circulating publications)
- reliable access to archives over time (favoring those accessible consistently for longer periods of time)
Through Lexis Nexis, we search these sources (listed alphabetically): The Age (Australia), The Australian (Australia), Business Day (South Africa), the Courier-Mail (Australia), Daily Mail (Mail on Sunday) (United Kingdom), the Daily Telegraph & Sunday Telegraph (Australia), Dominion Post (New Zealand), the Express and Express on Sunday (United Kingdom), the Ghanaian Chronicle (Ghana), the Globe and Mail (Canada), the Guardian and Observer (United Kingdom), Gulf Daily News (Bahrain), the Herald (Zimbabwe), the Irish Times (Ireland), the Jerusalem Post (Israel), the Korea Times (South Korea), the Los Angeles Times (United States), the Daily Mirror (and Sunday Mirror) (United Kingdom), the Nation (Thailand), National Post (Canada), the New Straits Times (Malaysia), the New York Times (United States), New Zealand Herald (New Zealand), Philippines Daily Inquirer (Republic of the Philippines), the Prague Post (Czech Republic), The Press (New Zealand), The Scotsman (and Scotland on Sunday) (United Kingdom), the South China Morning Post (China), the Straits Times (Singapore), The Sun and The Sun on Sunday (United Kingdom), Sydney Morning Herald (Australia), the Telegraph and Telegraph on Sunday (United Kingdom), the Times and Times on Sunday (United Kingdom), the Toronto Star (Canada), USA Today (United States), the Wall Street Journal (United States), and the Washington Post (United States). We use the Boolean string ‘climate change or global warming’ and the date ranges as well as individual sources selected through the ‘advanced options’ function. For all searches through the Lexis Nexis database, the default option for duplicates was chosen.
Through Factiva, we search these sources (listed alphabetically): the Bangkok Post (Thailand), Daily Star (Lebanon), Die Tageszeitung (Germany), El Pais (Spain), El Mercurio (Chile), the Hindu (India), Hindustan Times (India), the Indian Express (India), the Jakarta Post (Indonesia), Japan News (Japan), La Nacion (Argentina), Manila Bulletin (Philippines), the Nation (Pakistan), O Globo (Brazil), The Saigon Times Daily (Vietnam), Süddeutsche Zeitung (Germany), and Times of India (India). We use the Boolean string ‘climate change or global warming’ and the date ranges as well as individual sources selected for all English-languages searches. We conduct individual searches for ‘calentamiento global’ as well as ‘cambio climático’ and then manually search for (and eliminate duplicates. For all English-language searches through the Factiva database, the default option for duplicates was chosen.
National-level counts in Japan are undertaken by Midori Aoyagi and Shoko Yamaguchi, from the National Institute for Environmental Sciences (NIES) in Japan. They use the Fujitsu Limited "G-search" database to monitor coverage in Asahi Shimbun, Yomiuri Shimbun and Mainichi Shimbun through the search terms ‘climate change’, ’global warming’, or ‘greenhouse effect, from January 2000 through the present. National-level counts in Spain and Latin America are undertaken by Rogelio Fernández Reyes, member of the research group GREHCCO of the Universidad de Sevilla, and member of the research group MDCS of the Universidad Complutense de Madrid. Rogelio uses My News database (a referent provider of journalistic documentation in Spain) to monitor coverage in El Mundo, La Vanguardia and Expansión through the search terms ‘‘calentamiento global’ or ‘cambio climático’.
International and national-level counts in Australia, Canada, India, New Zealand, the United Kingdom and the United States are undertaken by Maxwell Boykoff, Andrew Benham, Meaghan Daly, Rogelio Fernández-Reyes, Lucy McAllister, Marisa McNatt, Ami Nacu-Schmidt, David Oonk, and Olivia Pearman. National-level counts in each of these countries begin in January 2000 with the exception of the following: Dominion Post (July 2002). International counts of the fifty-two noted sources begin in January 2004 with the exception of the following: El Nacional (December 2007); Indian Express (December 2007); The Nation (Pakistan) (June 2004); Gulf Daily News (Bahrain) (May 2008); Philippine Daily Inquirer (September 2004); Ghanaian Chronicle (January 2010); The Herald (Zimbabwe) (January 2010).
To arrive at comparable assessments of numbers of articles per month per source, we take the total number of articles each month in each region and divide by the number of sources searched during that month. For the archives that become available after the starting date of January 2004, this denominator changes over time. As some regions contain more sources than others, this 'normalization' process then provides a useful metric for comparisons across regions.