Preliminary Comparative Results

Project Overview & Interim Results

Scope of the Project & Initial Comparative Results

This section contains information in a poster that was to be presented at the NSF Grantee Conference in September 2010. It is also available in pdf form here.

Abstract
Data Collection Methods
Preliminary Keyword Count Results
Preliminary Article Frame Analysis
Preliminary Actor Statement Network Analysis
Acknowledgments

Abstract

This poster presents the results to date of the current project, with its popular name Comparing Climate Change Policy Networks (acronym COMPON). The purpose of the project is to analyze the effects of social networks and advocacy coalitions upon the responses of different countries or regions (cases) to the problem of global climate change (global warming) and to international negotiations concerned with that problem. These effects proceed in two fields: social relations and discourse. Both fields dovetail in their effects upon policy decisions and changes in social behavior to affect case levels of greenhouse gas emissions (the cause of anthropogenic climate change). The NSF grant funded this research in four country cases (Russia, India, China and the United States) and at the level of international negotiations. In addition, a number of case teams have joined the COMPON project, secured funding (Japan, Taiwan, South Korea, Sweden, Brazil, Lithuania, New Zealand) or are seeking funding (Germany, United Kingdom, Austria, Mexico, Canada), and are producing parallel data (some used in the poster). This poster displays some of this comparative data, as it expands the scope of the project.

Over its 42 month span (now in month 15), the project is collecting, analyzing, and comparing three types of data: 1) news media and other documentary sources, 2) qualitative in-depth interviews with experts and organizational representatives, and 3) a quantitative survey of the networks, policy preferences and resources of (80-130) relevant organizations in each case and internationally. In all the cases, the research teams follow a common protocol to ensure maximally equivalent data for accurate cross-case comparison. The first data-collecting phase of the project (April 1 2009 to March 31, 2010) involves the collection and analysis of the media data: the subject of the present report. The purpose of the media analysis is to compare the attention, interpretation and major actors concerning climate change in the different cases, as portrayed in their main newspapers. The project analyzes the media data on three levels: 1) changing frequency of articles mentioning climate change or global warming (1997-2008); 2) proportions of 6 different frames the media use to interpret the issue (2007-2008); and 3) actors appearing in the media and their policy stances (forming actor-discourse networks) (2007-2008). These three levels represent increasingly labor-intensive coding procedures and have some variance in time span by case. At the present time, most cases have completed Level 1 and made good progress on Level 2. Only the Korean case has coded Level 3 sufficiently to display results. While comparative analysis has only just begun, Level 1 media attention trends display considerable cross-national variation in degree of attention to the issue during the Kyoto Conference in 1997, a long lull in attention from then until 2007, but a much larger and cross-nationally consistent spike in attention to the IPCC Fourth Assessment Report of 2007.

Data Collection Methods

This project involves three types of data collection: media analysis, in-depth interviews, and a quantitative network survey. Data collection and analysis will be a three year process. The first year has been focused on media analysis and refining the survey and interview instruments; this poster describes results from the media analysis portion of the project. The second year of the project will focus on administering the quantitative network survey and qualitative interviews to organizational respondents. During the third year we will analyze individual countries’ data, as well as perform comparative analysis across country cases.

Methods Graphic

Preliminary Keyword Count Results

Figure 2 presents the trend lines in numbers of articles per year mentioning “climate change” and/or “global warming” summed for three newspapers in thirteen national cases. Some teams have not yet completed coding the entire 1997 to 2008 period. The lines do reliably indicate differences in national trends in attention to climate change over time. The lines can also be taken as a very rough indicator of the strength of national attention to the issue. However, some nations have thicker newspapers; they publish more articles in total than the newspapers of another nation. Therefore, the fact that Japan’s trend line is at the top of the graph does not necessarily mean that climate change occupies a greater news share in the Japanese press compared to other nations. We are developing a news share index. At present, this index—absolute number of national articles—remains tentatively indicative but of uncertain validity as a measure of actual differences in strength of national attention to the issue.

In Figure 2, the trend lines of most countries increase overall from 1997 to 2009, with a small peak in 2001 and a much larger peak in 2007. Only Japan shows a major peak in 1997, likely due to its hosting the Kyoto Conference (COP3). The small peak in 2001 is probably related to the release of the IPCC Third Assessment Report in that year. The dramatic increase in articles published in 2007 is likely related to the release of the IPCC Fourth Assessment Report then. After 2007, with the exception of Japan, Korea, and Russia, the number of articles decreases, sometimes quite dramatically. For China, after declining from 2007 to 2008, the number of articles rises to 2009. To the extent that raw article numbers validly indicate national strength of attention to climate change, they reveal a clear hierarchy with Japan and the UK at the top and China, Russia and the US at the bottom.

What social principles explain these outcomes? Previous studies of media coverage of climate change mostly concern individual countries, mainly the US, UK and Germany (Ungar, 1992; Mazur
and Lee, 1993; Trumbo, 1996; Liu et al., 2006). The present study is the first multi-country cross-national study, thereby widening the scope of analysis. Looking within a single country (the US), Downs predicts an issue attention cycle with an initial rise and then fall of media coverage with issue fatigue (1972). Other scholars predict a news share or issue prominence effect (Oliver and Maney 2000). In the latter, new issues can fill the pages and crowd out old but continuing issues (for a time, at least). If the trend lines of the 13 nations in Figure 2 varied independently, it would signal the effect of autonomous national-level factors. The fact that all 13 lines show a similar curve slowly rising to a peak in 2007 with most then declining indicates, to the contrary, a common global cause. The evident factors are the 2001 and 2007 Assessment Reports from the Intergovernmental Panel on Climate Change (IPCC) stating the anthropogenic causes and likely disastrous effects of climate change with increasing certainty. This common trend indicates a pan-national media focus on a common issue defined by scientific analysis, not immediate substantive events. This supports a model of global diffusion of acceptance of scientific predictive rationality defining environmental effects (Meyer, et al. 1997). However this pan-national trend follows a serial issue-attention cycle around concentrated points of scientific intervention (the IPCC reports). The overall upward trend may be due to the increasing use of this scientific frame on weather disasters. The post-2007 decline may also be affected by the sudden issue prominence of the global financial breakdown.

Preliminary Keyword Count Results

Preliminary Article Frame Analysis

This step of the research aims to discern how the news media represent or frame the issue of climate change in each case. One type of framing concerns the main topic or general frame that the news media use when covering climate change. The project uses a typology of six general frames drawn from previous studies: Civil Society, Culture, Ecology/Meteorology, Economic and Energy Interests, Policy-making, and Science and Technology (Boykoff 2008). Each article was coded according to its primary (and if present, secondary) general frame. At minimum, each team coded climate change articles from 2007 and 2008, with some teams coding additional years. As noted in the methods flow chart, this level of data collection also includes other variables, not presented here, such as scale, source, and coding domestic public debates and public policies about climate change into a set of (inductive) content categories. As above, the figures show comparative distribution of general frames in the coded articles, not the intensity or news share of climate change coverage.

The frame figures to the left show how national news media frame climate change and how that framing changes over time. While in each case the distribution of frame types varies from year to year, there are some general trends. “Policy-making” is the predominant frame across nearly all years in all cases except India, where it is more equal to other frames. Ecological/meteorological aspects are a large focus in India and China, but are fairly minor elsewhere. One interpretation would be that, compared to highly industrialized cases, China and India, with vast agrarian populations, are more vulnerable to ecological changes. Additionally, that economic and energy interests make up a larger portion of media coverage in Russia may be due to that country’s heavy dependence on the export of fossil fuels.

Preliminary Article Frame Analysis

Preliminary Actor Statement Network Analysis

Discourse Network Analysis (DNA) (Liefeld 2010) reveals the relationship between types of organizations and the positions they take on climate change issues in national newspaper discourse. The DNA coder categorizes all actors cited in the national newspaper database articles and also develops categories for the climate change issues they raise and their stances on those issues. The organizational types and issue categories for South Korea are listed below. After determining a limited set of types and categories, network analysis identifies their relationships, clustering organizational types by the similarity of their stances on the main issues. Comparative analysis of these findings will show the main organizations and ideas dominating the national media discourse. At present, the South Korean team is most advanced in this analysis. The network figures show patterns of relations between types of organizational actors (546 actors coded into 10 categories) and categories of statements (from 316 articles coded into 10 categories) for South Korea in 2007 and 2008. The network placement of organizational types and statement categories as well as the width of the lines between them indicates the distribution and intensity of their association. These figures provide evidence bearing upon project hypotheses about the roles of different types of actors and their probable effects upon emissions outcomes. As comparative figures accumulate, they will permit more reliable testing and broader generalizations.

The Discourse Network (Issues and Actors) figure shows that domestic government strongly acknowledged Korea’s ecological changes but defended the appropriateness of domestic countermeasures and supported nuclear power. In contrast, the strongest supporters of Korea adopting binding emissions targets were foreign governments, with the domestic government a weak second. Absent from the Korean media discourse field were topics such as alternative energy, cap-and-trade, and carbon tax. The Discourse Network (Actors’ Position) figure shows whether an actor agrees or disagrees with a category, with the size of the category (red ball) indicating the number of mentions. Here, the economic opportunities of climate change (indicated by the largest red ball) dominated the field of discourse and won the support of almost all organizational types. The broad organizational support in Korea for framing climate change as economic opportunity, so in contrast to the discourse in the United States, resembles the alacrity with which its economy has penetrated other market sectors such as automobiles. Of course, this network is filtered by the national newspapers, so the actual organizational survey in the next phase may reveal divergent patterns of advocacy clusters.

Preliminary Actor Statement Network Analysis
Preliminary Actor Statement Network Analysis

Acknowledgments

Thank you to the following people for their contributions to this poster: Brazil: Myanna Lahsen, International Geosphere-Biosphere Project | China: Jun Jin, Tsing Hua University | Germany: Volker Schneider, University of Konstanz; Katrin Vogt, University of Konstanz | Greece: Moses Boudourides, University of Patras; Iosef Botetzagias, University of the Agean; Giouzepas Georgios, University of the Aegean | India: Sony Pellissery, Institute of Rural Management; Debaprashad Chatterjee, Maulana Azad College; Atanu Sarkar, Queen's University (Canada); Suman Ranjan Sensarma, The Louis Berger Group Inc. | Japan: Koichi Hasegawa, Tohoku University; Jeffrey Broadbent, University of Minnesota (USA); Chika Shinohara, Momoyama University; Susumu Kitagawa, Yamanashi University; Tomomi Shinada, Tokyo Institute of Technology; Hiroshi Noda, Kyoto Prefectural University; Keiko Hirao, Sophia University; Syunsuke Kishi, Keio University; Kazuhiro Ikeda, Tokyo Institute of Technology; Keiichi Satoh, Hitotsubashi University; Tomoyuki Tatsumi, Hitotsubashi University; Takashi Nakazawa; Hitotsubashi University; Yasuhiko Sugiyama, Hitotsubashi University; Akiko Inoue, Hitotsubashi University | Korea: Dowan Ku, Environment and Society Research Institute; Sun-Jin Yun, Seoul National University | New Zealand: Philip Vaughter, University of Minnesota | Russia: Irina Shmeleva, Saint Petersburg State University; Alyona Rydannykh, Saint Petersburg State University | Sweden: Christopher Edling, Jacobs University; Marcus Carson, University of Stockholm; Johannes Ackva, University of Gronigen (Netherlands); Emily Akerblom, University of Stockholm | UK: Clare Saunders, University of Southampton; Chris Rootes, University of Kent; Stephan Price, University of Kent | US: Dana R. Fisher, Columbia University | Comparative: Sarah Burridge, University of Minnesota; Anne Kaduk, University of Minnesota

See also: Comparing Climate Change Policy Networks (COMPON) - presentation delivered in March 2010 at the "Comparing National Responses to Climate Change" panel at Institut du developpement durable et des relations internationales (IDDRI) in Paris, France.