|The literature of environmental communication
February 27, 2001
Andrew Pleasant is a doctoral student in the Department of Communication at Cornell University. James Shanahan is an associate Professor in the Department. Brad Cohen is a Master's student in the Department. Jennifer Good is a doctoral student in the department. Inquiries should be directed to the first author, 314 Kennedy Hall, Ithaca, NY 14853. (email@example.com).
The literature of environmental communication
This paper examines the growth and development of the literature of Òenvironmental communicationÓ research. The paper collects citations of all papers matching specified keywords covering environmental communication topics in the social science journal literature from relevant indices. The indices used thus far are Ñ the ISI Social Sciences Citation Index (Web of Science), the ISI Arts and Humanities Citation Index (Web of Science), Communication Abstracts, PsycINFO, Anthropological Literature, Sociology Abstracts and Periodical Abstracts (Pro-Quest Direct).1
The paper analyzes the collected citations for journal of publication, date and frequency of publication by year, author, and keywords and topics. The paper shows what topics of attention, what fields and journals have been interested in publishing environmental communication research, and some of the authors leading the way. Also, basic topics of interest for environmental communication research are discerned. The literature review shows the need for a more centralized point of publication for environmental communication research.
The literature of environmental communication
The Environmental Communication Commission of the National Communication Association examines Òthe link between communication practices and environmental affairs.Ó In this short statement is carried the most succinct possible definition of the sub-field of communication research known as Òenvironmental communication.Ó In this paper, we attempt to provide a relatively exhaustive compendium of this literature. Because it is difficult to define the field without lapsing into a circular definition, we will define the field by what it has done and focus on the indexed literature accrued in (mainly U.S.) scholarly journals.
Formal environmental communication programs are still relatively rare in US universities, though areas of study and formal programs can be found at universities such as Colorado State University, Cornell University, Northern Arizona University, Ohio State University, SUNY-ESF, the University of Cincinnati, the University of Wisconsin, and a few others.
However, the idea of a specialty focusing on communication issues with respect to the environment has been gradually emerging since the 1960s. The origins of the environmental movement in very visible events such as the furor around Silent Spring focused attention on the importance of the communication media, as well as on the power of individual words and texts to shape and focus opinion.
As scholarly work emerged through the 70s, and especially in the 80s, groups emerged which provided a gathering place for these discussions. Notable, of course, would be the Conference on Communication and the Environment, which began in 1993, started by a small group of interested scholars.2 The Environmental Communication Commission of the NCA was an outgrowth of this. Although the International Communication Association (ICA) does not sustain a particular division devoted to environmental communication, some environmental scholars do present their work there, as well as in the science communication interest group of the Association for Education in Journalism and Mass Communication (AEJMC). As well, there are opportunities for humanists and rhetoricians to present work in an environmental communication context. All in all, given the existence of formalized study programs and conference opportunities, one cannot doubt the vitality of the scholarly enterprise in environmental communication.
However, there is no particular journal that carries the Ògold standardÓ research in communication. For instance, the health communication discipline, though perhaps somewhat larger than environmental communication, is in a similar position within the communication field, covering health topics using the range of theories available to communication scholars. This discipline can look to Health Communication (begun in 1996) to carry some of its most important research, and to provide a recognizable scholarly identity.
No such journal exists for environmental communication, which prompts the question: Does the field of environmental communication need its own journal? This paper attempts to answer that question, primarily by looking at the history of publication in the field.
Several bibliographies of environmental communication are already available, in various channels. For instance, the Center for Environmental Communication Studies' web site contains three bibliographies, one on risk, one on environmental communication, and one of public participation (http://www.uc.edu/cecs/cecs.html). Mark Meisner compiled a bibliography of books dealing with environmental communication themes (http://www.esf.edu/ecn/bibl.htm). There is also a page with a list of environmentally-themed films. The National Cancer Institute maintains a bibliography of risk communication (http://dccps.nci.nih.gov/DECC/riskcommbib/). There is a bibliography on Òenvironmental discourseÓ maintained on a Web-site at Michigan Technological University. William Evans maintains an ongoing bibliography of environmental communication (and science communication) articles in the pages of Public Understanding of Science.
Our purpose of this paper is to build on these bibliographies, but to do so in a systematic way. Using computer retrieval techniques, with a specified list of keywords, from known indices, we can at least begin the process of identifying the 'literatureÓ of the field of environmental communication, perhaps even identifying canonical moments in this literature.
We wanted to be not only systematic, but relatively exhaustive as well. Because the question of journal publication was what ultimately interested us, we decided to exclude all other forms of publication from our analysis. That means we decided not to look at books, chapters, conference proceedings, choosing to focus on the peer-reviewed literature, even though many important pieces can be found in those formats.
This means we purposefully exclude a number of the intellectual contributions to the field of environmental communication, although the reader will soon see that even our limited task of reviewing journal publications was nearly unmanageable. We caution that the data base resulting from our efforts is not the entire field but a specific sample; we advise that the importance of the many interesting and important books in the field (which, again, are quite ably reviewed on the ECN web site) should not be ignored.
There are certainly virtues to our approach. By focusing on peer-reviewed literature, there is a bias for stronger and more empirically rooted research to emerge. Essays and thought pieces, more likely to be reserved for book chapters and other channels, will be less important in this review. By focusing on articles published in indexed journals, there is also a bias toward increasing the quality of the pieces reviewed. Though some good journals are not indexed in major indices, it helps us to have an outside authority in making determinations about what sources should be included in our analysis.
All of this leads toward our final goal, which is to arrive at a relatively comprehensive and systematic account of the literature of environmental communication. We will probably err on the side of systematicity; by making our method transparent others can praise it, criticize it, or build upon it. As for being comprehensive, we admit that our methods will miss certain important segments (see below).
To gather citations, we used the electronic databases ISI Social Sciences Citation Index (Web of Science), ISI Arts and Humanities Citation Index (Web of Science), Communication Abstracts, PsycINFO, Anthropological Literature, Sociology Abstracts and Periodical Abstracts (Pro-Quest Direct).
Search word combinations used in each database were the combinations of ÒenvironmentÓ and Òcommunication,Ó ÒenvironmentalÓ and Òcommunication,Ó ÒenvironmentÓ and Òmedia,Ó ÒenvironmentalÓ and Òmedia,Ó ÒenvironmentÓ and Òrhetoric,Ó environmentalÓ and Òrhetoric,Ó ÒriskÓ and Òcommunication,Ó ÒriskÓ and Òmedia,Ó ÒriskÓ and Òrhetoric,Ó ÒscienceÓ and Òcommunication,Ó ÒscienceÓ and Òmedia,Ó ÒscienceÓ and Òrhetoric,Ó ÒnatureÓ and Òcommunication,Ó ÒnatureÓ and Òmedia,Ó ÒnatureÓ and Òmedia,Ó ÒecologyÓ and Òcommunication,Ó ÒecologyÓ and Òmedia,Ó ÒecologyÓ and Òrhetoric.Ó When databases allowed the use of wild cards such as enviro* or scien*, they were employed. Searches were limited to keywords, titles and abstracts.
Only published journal articles are included in the final database used for analysis Ñ specifically excluding book reviews, editorials, conference proceedings, dissertations, books, and book chapters.
Initial search results were compiled in EndNote libraries. At that stage, we removed citations that responded to the search terms but were plainly not related to environmental communication or were not journal articles. (Examples include articles dealing with communication across media in both an artificial intelligence and a chemistry sense.) Tabulation and analysis of results utilized EndNote, File Maker Pro, Excel, and CATPAC.
A base rate
In order to say anything valid about any growth or decline in environmental communication research, we first had to determine a base-line growth rate for journal article publication patterns in the social sciences. Obviously, there has been an overall increase in the frequency of journal publication in the last fifty years; this basic ÒinflationÓ rate might obscure whether environmental communication research has grown because of its increasing importance, or merely because research on all topics has become more frequent. Any deviation from that rate will indicate either growth or decline in interest given to environmental communication beyond that occurring in the broader academic discipline.
As is so often the case, developing an indicator for a complex and dynamic system is fraught with difficulty and necessarily incorporates uncertainty. We acknowledge that no existing index is a complete and adequate representation of what is inherently a cross-discipline enterprise such as environmental communication research. That position is reinforced by our use of several electronic indexes to develop our database. For our purposes, we are using the growth in the number of journals included in the ISI Social Sciences Index (Web of Science) and the ISI Arts and Humanities Index (Web of Science) as a baseline for comparison purposes.
The social science and arts & humanities indices make up two-thirds of the Institute for Scientific InformationÕs (ISI) World of Science electronic database (http://www.webofscience.com). According to ISI, the social sciences citation index includes references from 1,700 social science journals from 50 disciplines and selected items from over 5,700 science and technology journals with back issues going to 1956. The arts and humanities index includes references from 1,400 arts and humanities journals and items from over 7,000 science and technology journals back to 1975.
We developed a basic growth trend for both databases based upon information in the comparative statistical summaries provided by ISI for both indexes. The arts and humanities and social sciences index have each grown at an average annual rate of 2.7 percent.
Growth through time
The number of articles in the database (as of 2/27/01) is 9633. These articles range in time of publication from 1945 to the current year. Graph A depicts the growth apparent in the database of environmental communication articles we developed in comparison to the growth in the Social Sciences and the Arts and Humanities indexes. For comparison purposes, we chose 1990 as a base year (setting that equal to 100).
The rate of growth in the environmental communication research, in comparison to the ISI databases, is phenomenal. The number of articles per year in our database grows at an annual average growth rate of 25.5 percent from 1970 to 2000, the years where at least one article appears each year. As is apparent from graph A, the growth really began to take off in 1985 when the number of articles doubled from the previous year.
Between the years of 1985 and 1990, the average annual growth rate was 44 percent while between 1985 and 1995 the rate was 32 percent. Those years center on the heart of the explosion that seems apparent in the growth of environmental communication research published in journals. It also corresponds to the enormous growth in public and media attention directed toward the issue (McComas & Shanahan, 1999).
All of that growth, however, occurred in a very diffuse number of academic journals. There are 499 journals represented in the database, 59.5 percent of which (297) published only one article that matched our search parameters. By far, the strongest showing in terms of concentration of articles per journal occurred in the aspect of environmental communication concerning risk. The journal Risk Analysis topped the list of articles per journal with a total of 48 articles, followed by another risk related journal, International Journal of Mass Emergencies and Disasters which has 41 articles in the database.
Number of Articles 1 2 3 4 5
Number of Journals 297 79 72 16 13
Number of Articles 6 7 8 9 10
Number of Journals 3 2 3 3 2
A second area of emphasis within the broader arena of environmental communication research that seems to exhibit some clustering of articles per journal is that of science and the communication of science. Journals represented in that area include Science Communication with 14 articles, Public Understanding of Science with eight articles and Science magazine with five articles.4 The average is 29 articles per journal with a standard error of 17.7. The median is two articles per journal.
Journal Number of Articles
Risk Analysis 48
International Journal of Mass Emergencies and Disasters 41
Journal of Communication 17
Science Communication 14
Australian Journal of Communication 13
Environment and Behavior 13
Journal of Environmental Education 12
Journalism & Mass Communication Quarterly 11
Media Culture & Society 11
Human Organization 10
Science Technology & Human Values 9
Socijalna Ekologija 9
Communication Research 8
Forschungsjournal Neue Soziale Bewegungen 8
Public Understanding of Science 8
Just as a vast majority of the journals in the database are represented by only one article, a majority of articles have only one author. In fact, 62 percent (598) of the articles have one author. Ten authors are the most for any article in the database Ñ only one article has that number. Articles with two authors make up 24 percent (233) of the database, while articles with three authors make 9 percent (83) of the database.5
A cast of frequently cited names appears when authors are analyzed in terms of their frequency. (See Table B) Topping the list are Paul Slovic, Peter Sandman and Robert Griffin with eight appearances.6 Reflective of the large number of articles in the risk-related journals, those top three authors Ñ Slovic, Sandman, and Griffin Ñ generally work in that dimension of environmental communication research.
First name Middle Last name # articles
Paul Slovic 8
Peter M. Sandman 8
Robert J. Griffin 8
Sharon Dunwoody 7
Robert E. O'Connor 6
Michael Greenberg 6
Richard J. Bord 6
Kandice L. Salomone 5
E. L. Quarantelli 5
K. A. McComas 5
Allan Mazur 5
Bruna De Marchi 5
Ann Marie Major 5
C. M. Harrison 5
William R. Freudenburg 5
Ann Fisher 5
J. Burgess 5
Lee Wilkins 4
K. R. Stamm 4
David B. Sachsman 4
Dennis S. Mileti 4
D. B. McCallum 4
M. J. Killingsworth 4
Colleen Fitzpatrick 4
B. Fischhoff 4
Caron Chess 4
James G. Cantrill 4
JoAnn M. Valenti 3
James Shanahan 3
Donald E. Riechard 3
H. P. Peters 3
Philip Patterson 3
Mark Meister 3
Timothy L. McDaniels 3
Tom Jagtenberg 3
J. M. Gutteling 3
David Griffith 3
Cynthia-Lou Coleman 3
Ognjen Caldarovic 3
Jacob Bendix 3
Lawrence J. Axelrod 3
We analyzed titles, keywords, and abstracts for basic word frequencies. Remembering that we chose certain words as the keywords, guaranteeing that they will appear frequently in our analysis, the relative frequency of these words and their relation to other keywords that may emerge is of interest. Not to anyoneÕs surprise, the most frequent word was Òenvironmental,Ó at 16.4 percent. The second most frequent word in keyword sections was ÒinteractionsÓ (10.7 percent) as many sociology journals use the keyword pair Òenvironmental interactionsÓ a great deal of the time. In the titles, the word ÒriskÓ was the second most frequent (10.6 percent), while in abstracts the word ÒcommunicationÓ filled that spot (7.4 percent). On at least a very basic level, that confirms the validity of our methods. Table C contains the 25 most frequent appearing words for the titles, abstracts, and keyword sections.
In titles and abstracts, combinations of the words Òenvironmental,Ó Òcommunication,Ó Òrisk,Ó Òanalysis,Ó and ÒpublicÓ create the largest clusters. In keywords, the words Òanalysis,Ó Òissues,Ó Òattitudes,Ó Òcommunication,Ó ÒenvironmentalÓ are in the most frequent cluster. A second smaller cluster occurs with the words Òscience,Ó Òperception,Ó ÒhealthÓ and Òmass.Ó (Wards method).
Word Frequency Lists
Keywords Titles Abstracts
WORD FREQ PCNT WORD FREQ PCNT WORD FREQ PCNT
ENVIRONMENTAL 582 18.5 ENVIRONMENTAL 278 17.9 ENVIRONMENTAL 1152 12.9
INTERACTIONS 337 10.7 RISK 164 10.6 COMMUNICATION 656 7.4
COMMUNICATION 245 7.8 COMMUNICATION 145 9.3 RISK 649 7.3
MASS 171 5.4 SOCIAL 93 6 MEDIA 550 6.2
SOCIAL 168 5.3 MEDIA 86 5.5 PUBLIC 527 5.9
RISK 157 5 PUBLIC 76 4.9 SOCIAL 510 5.7
SOCIOLOGY 155 4.9 ENVIRONMENT 72 4.6 INFORMATION 422 4.7
STUDIES 150 4.8 SCIENCE 54 3.5 REFERENCES 356 4
DISASTER 129 4.1 DEVELOPMENT 46 3 DEVELOPMENT 319 3.6
PHENOMENA 112 3.6 CASE 41 2.6 ENVIRONMENT 284 3.2
MEDIA 98 3.1 NATURE 41 2.6 SOURCE 284 3.2
PUBLIC 83 2.6 INFORMATION 39 2.5 RESEARCH 266 3
ATTITUDES 82 2.6 NEWS 37 2.4 ANALYSIS 254 2.8
POLICY 78 2.5 COMMUNITY 36 2.3 DOCUMENT 254 2.8
HEALTH 70 2.2 ANALYSIS 34 2.2 ISSUES 253 2.8
ENVIRONMENT 69 2.2 RHETORIC 33 2.1 ADAPTED 251 2.8
SCIENCE 64 2 COVERAGE 32 2.1 PROBLEMS 240 2.7
DEVELOPMENT 62 2 ECOLOGICAL 32 2.1 POLITICAL 230 2.6
PROTECTION 56 1.8 POLICY 32 2.1 SCIENCE 226 2.5
COVERAGE 51 1.6 STUDY 32 2.1 POLICY 218 2.4
NEWS 49 1.6 HEALTH 31 2 COMMUNITY 216 2.4
ANALYSIS 48 1.5 MANAGEMENT 31 2 HEALTH 211 2.4
ISSUES 46 1.5 GLOBAL 30 1.9 STUDY 211 2.4
PERCEPTION 46 1.5 MASS 29 1.9 NATURE 192 2.2
POLITICAL 41 1.3 ISSUES 28 1.8 SCIENTIFIC 192 2.2
The word frequencies seem to support the earlier supposition based upon journal clustering that, after moving past the dominance of environmental and communication again risk is a strong element on its own. In this mode of analysis, science is present throughout but not at the same level of frequency as risk.
Moving past those two apparent areas of academic concentration (science and risk), a rough depiction of other areas of concentration begins to emerge when we combine results of word frequency from titles, keywords, and abstracts. That analysis is partially presented in Table D. This look at the data mirrors the journal types to some extent, as there are a vast number of sociology related journals with a small number of articles each in the database, but we can begin to see a rough definition of the central concepts of the academic concentration emerge. For example, a focus upon the
WORD PERCENT WORD PERCENT
SOCIAL 5.9 ATTITUDES 2.6
MEDIA 4.9 POLITICAL 2.6
SOCIOLOGY 4.9 ANALYSIS 2.5
PUBLIC 4.5 NATURE 2.4
INFORMATION 3.6 NEWS 2.4w
PHENOMENA 3.6 COMMUNITY 2.4
SOURCE 3.2 POLICY 2.3
RESEARCH 3 HEALTH 2.2
DEVELOPMENT 2.9 COVERAGE 2.1
ADAPTED 2.8 ECOLOGICAL 2.1
ISSUES 2.8 RHETORIC 2.1
PROBLEMS 2.7 MANAGEMENT 2
media appears (media at 4.9 percent, news at 2.4 percent, coverage at 2.1 percent) as well as a concept of environmental communication being public (public, 4.5 percent and community at 2.4 percent). Pushing the analysis a bit further, tentative ideas about an applied nature to environmental communication research can be derived from the showings of words like development (2.9 percent), problems (2.7 percent), adapted (2.8 percent), issues (2.8 percent), political (2.6 percent), policy (2.3 percent), and management (2 percent).
Summarizing our conclusions:
¥ The literature on environmental communication has grown phenomenally, and seems poised to continue this growth;
¥ The literature is spread across many journals; while no single journal predominates, a few in risk and science, as well as some in communication, are notable;
¥ Authors in risk appear most frequently;
¥ Analysis of keywords shows the same broad clustering as the journal listings.
This bibliography shows a lot of diversity, with articles on topics as disparate as risk communication, science communication, disasters, interpersonal communication, rhetoric, and much more. We ask the question, would the existence of a journal titled Environmental Communication improve the growth, development and dissemination of environmental communication research? The answer, inevitably, seems to be yes. How will a risk communication scholar become familiar with the work of the rhetorician? How will those interested in science communication learn about media effects research on the environment? Although any scholar interested in cross-disciplinary work will have to voyage around the indices to some extent, an environmental communication oriented journal does seem to us to be a missing piece.
1. This paper is an ongoing project; more indices are being added at time of writing.
2. For the story of this see Cantrill's account of the emergence of COCE as a bi-annual conference (http://www.esf.edu/ecn/hist.htm).
3. Appendix A is a printout of the entire database.
4. Again, it is worth emphasizing that our database is just a sample. While we have confidence in the relative proportions as they developed via the database and keyword combinations used, we readily admit that in general our numbers are lower Ñ as is the case with all samples Ñ than have actually occurred.
5. This analysis does not lend any weight to position in the list of authors in any way (e.g., first authorship counts the same as fifth).
6. Variations in the formatting of names create a problematic factor to this listing. An attempt was made to combine reasonable variations, but there certainly may be authors with more articles in the database.