Department of Journalism
Dr. David Secko
Department of Journalism
Alamir Novin is an MA Student in the Department of Journalism at Concordia University. He is interested in how open source formats can help science journalists.
Dr. David Secko is an associate professor in the Department of Journalism at Concordia University. He currently runs the Concordia Science Journalism Project (www.csjp.ca) and his research covers the analysis of new models of science journalism and the lived experiences of health and science journalists.
Science journalism is often accused of inaccuracy and complying with the production cycles of hype. However, while much of the scholarly literature on science journalism criticizes its problems, few scholars test for solutions. We enter this research gap by testing whether open-source formats can help overcome some critiques of the field. We analyze how a classroom of science journalism students uses a web application called Debate Cited to lay open their research through concept maps. Debate Cited was designed by one of the authors (A. Novin) to help journalists deal with contentious scientific issues. We use Debate Cited with the topic of biofuels. Canada has embraced biofuel production despite complex challenges and heavy criticisms regarding its sustainability. Through a geometrical and textual analysis, we present a first exploration of how science journalism students used Debate Cited.
Science journalism is often accused of being inaccurate, uncritical, and complicit in the production of cycles of hype (Bubela et al., 2009). One particularly thorny issue is the criticism that science journalism can be incomprehensive (Tosse, 2012; Dunwoody, 2008). We define incomprehensiveness as a lack of information required to contextualize a scientific topic for readers. Although scholars have suggested a few solutions to incomprehensiveness (see, for example, Bubela et al., 2009), the literature on methods to implement these solutions is sparse. Just as there is little literature on a suggested method, there is little literature empirically testing any suggested methods.
This paper explores a methodological approach to making science journalism more comprehensive, which is tested in a classroom setting with journalism students. The approach involves an open-source logic system we call Debate Cited. Debate Cited is designed to organize the information accumulated by journalists into a visual representation (i.e. a concept map) that symbolizes the logical relationships between data. The data inputted into Debate Cited includes the variety of premises that construct an article’s logical arguments. Further, a map in Debate Cited allows for the measurement of the validity of certain statements against others. The usefulness of this approach is examined here through a textual analysis with a classroom of science journalism students and their constant feedback. A geometrical analysis of the map-structure will also be conducted via a function of Debate Cited where the shapes of the resulting maps will indicate how the logic behind dialogue is built.
Debate Cited takes advantage of the Internet by using a cloud-based web tool to store data collected by science journalists and then to visually map the relationships between the data. This allowed us to examine correlations between a map in Debate Cited and the methods used by journalism students to construct stories, as well as whether the participants felt the web application aided the structure of their story.
Literature review: A brief account of incomprehensiveness in science journalism
This is not the first scholarly paper to list the issues facing science journalism (for good reviews, see Amend and Secko, 2011; Bubela et al., 2009; Logan, 2001; Schaefer, 2011), but it is one of the few that aims to provide test solutions for improving comprehensiveness. Incomprehensiveness in science journalism is often attributed to factors including an overreliance on press releases, single-sourced articles, the depreciation of the role of science journalists, and now, governmental interference (see, for example, Bubela et al., 2009; Dunwoody, 2008; Kruvand, 2012; Nelkin, 1987; Nelkin, 1987; Pellechia, 1997; Saari et al., 1998). Critics argue that such factors are some of the inputs that lead to a form of science journalism that does not provide a complete scientific story. With the slashing of science journalists from news organizations – including the removal of all science journalists from CNN, for instance – some worry that the public is witnessing a decline in the quality of science journalism (Brainard, 2009). Researchers have found that experience is a greater contributor to quality science journalism than being trained in the sciences (Saari et al., 1998). And yet, the cutting of experienced staff combined with the time required to produce journalism can leave journalists seeking only primary authors as scientific sources for a story. This, in turn, leads to a style of journalism that some scholars have termed single-source journalism, where a researcher’s study is taken at face-value and context is not provided. Some of the problems with this form of journalism is that scientists can provide fringe information, skewed information, or information heavily moderated by PR experts (Bubela et al., 2009; Saari et al., 1998). This is further amplified by the difficulties journalists can face in getting more comprehensive information, such as in the case of the government muzzling of Canadian scientists (O’Hara, 2010; McDonald, 2011).
Of the few solutions articulated in the literature, two of the more prominent scholars on science journalism, Nelkin and Dunwoody, have made suggestions that other scholars have echoed since. Dunwoody suggests when the strongest science position on an issue is unclear, science journalists should report (objectively) on as many scientific perspectives as possible with the fallback position that the truth is in there somewhere (Dunwoody, 2008). Nelkin suggests that when different perspectives are presented, the strength of the validity behind those claims should also be presented (Nelkin, 1987). Prior to the growth of the internet in the mid 1990’s, Nelkin made two additional suggestions. The first is that science journalists may help each other resolve disputes by consulting each other. The second is that computers may help science journalists to retrieve relevant information faster (Nelkin, 1987). Nelkin doesn’t tie these three suggestions explicitly together but, given the current era of online networking, we suggest there would be little objection in doing so. The solutions of Nelkin and Dunwoody are taken as our theoretical starting point to develop an online tool to test their usefulness.
Debate Cited maps: From theory to method
Given Dunwoody and Nelkin’s suggestions, we combine several fields of study to explore a methodological approach to the problem of incomprehensive reporting on scientific debates. To improve the practicality of this study, we focus on a contemporary issue that challenges comprehensive science journalism: the coverage of energy (Brainard and Russel, 2009), specifically choosing biofuels as a complex scientific/social issue that highlights many of the challenges studied here (Delshad et al., 2010; McKone et al., 2011; Wright and Reid, 2011). Biofuels, as a subset of energy coverage (Brainard and Russel, 2009), are of growing importance and have reached an important crossroad (Wright and Reid, 2011; Soloman, 2012; Delshad et al., 2010). The coverage of energy issues draws on topics such as climate change, renewable energy, environmental law, energy policy, national security issues, investment banking, material science and genomics. It is a topic that sets traditional energy sectors (coal/oil/gas) against new “green” technologies in a way that is global as well as local, environmental as well as economic, regulatory as well as scientific (Brainard and Russel, 2009). Biofuels showcases a significant challenge for journalism: it is a topic that requires broad expertise to cover effectively (scientific, regulatory, legal, environmental, investment banking, etc.) and therefore a team effort. It is a good topic with which to develop and test innovative approaches to journalism. Thus, this paper focuses on the specifics of how science journalism students may deal with these debates when scientifically accepted positions are unclear.
In developing a web-tool to aid science journalists, we sought a unique yet theoretically informed (i.e., tied to scholarly suggestions in the literature) approach. A heuristic approach led us to develop an online networking mapping technique for science journalists.
Theory: Mapping debates
The history of tree maps can be traced back to the ancient Greek philosopher Porphyry. Porphyrian’s Tree is the earliest example of a tree map (Figure 1). An important element in this map is how he divides categories into two premises that oppose each other (e.g. classifying an object as “animate” or “inanimate” or an animal as “rational” or “irrational”). As we will demonstrate later on, the premises of Debate Cited are divided based on whether their preceding premises agree or disagree with them.
Novak and Gowin’s (1984) “Learning How to Learn” is a seminal work on which the majority of recent concept map literature draws.
Novak and Gowin’s maps have five elements: concepts, propositions, hierarchies, cross links, and examples. Novak calls the six boxes in Figure 2 concepts and the line that links them propositions. For example, “Life” is a concept linked by the proposition “includes” to the concept “Humans.” The “Life” box would be on a different hierarchical stage than the “Humans” box, which, in turn, is on a different hierarchical stage level than the boxes with “Sensitive” and “Rational.” The relationship between “Animals” and “Sensitive” indicated by the arrowed-line is a cross-link. Lastly, Novak calls specific incidents of a concept examples (in Figure 2 that would be “Socrates”).
Theory to method: Creating Debate Cited
The logic behind Debate Cited’s structure is actively evolving, so for the purpose of this paper, a brief overview will be provided to demonstrate how the software works. Debate Cited maps focus on incorporating debates into a logical format (Figure 3). This means incorporating mapping elements that increase logical structure while discarding mapping elements that decrease communication between members. Therefore, Debate Cited enables a dialogical structure of argument rather than a monological one. Monological arguments tend to be linear syllogisms and Debate Cited’s aim is to incorporate debates to create dialogical arguments (similar to Socratic dialogue or Argumentation-Theory (Walton, 2006)). This, it is hoped, allows debates to be deconstructed logically by dividing premises based on whether they agree or disagree with a preceding point. The bifurcation of premises allows concepts to either affirm or deny each other.
Other messaging systems on the Internet (emails, forums, social networks, etc.) are overwhelmingly linear. The comments are displayed in a top-down method, often chronologically ordered. Chronological hierarchy creates several problems, including (i) not knowing the strength of individual comments, (ii) not knowing which specific points are supported or opposed in relation to others, and (iii) susceptibility to tangential information such as memes, jokes, and spam (Secko et al., 2011). Sometimes rating systems are implemented in commenting systems for other users to rate the value of a comment. The rating feature helps judge a group’s consensus on a comment and Debate Cited incorporates this feature. However, ratings can also be chronologically dependent, in that the first few comments can guide the discourse while subsequent comments become too low on the page to be rated. Furthermore, a comment that attracts ratings early will have more time to accumulate value from them than will an equally valuable, yet late, comment. To address this, Debate Cited adopts a non-linear format in which comments’ placements are dependent on relevance. A user views the relevance of different ideas, comprehends their strength and the logic behind a topic, and opens up the possibility of reflecting on points of argument they may not have considered before.
One of the goals behind Debate Cite was that the program had to be user-friendly for science journalists. This meant that it should be easy to access, understand, and use. Ease of access is enabled by the medium of the Internet itself. Participants did not need to install extra software and could access the application through any computer. Upon opening the application, participants were greeted with a reminder of how Debate Cited maps works in just three steps (Figure 4).
Participants were provided a brief three-step instruction on how to use the software. After the front page, users were presented with a blank canvas to create their maps and a box to login or register an account with DebateCited.com (Figure 5). Alternatively, once created, users who wished to explore Debate Cited maps belonging to others could move their mouse to the left side of the screen to access them. By doing so, a column of maps created by other users would slide out.
Figure 5: Debate Cited’s Blank Canvas
Participants were provided a blank canvas
to begin their maps.
When a topic was clicked, the corresponding map folded out (Figures 6 and 7) to display the root premise and the proceeding premises that agree or disagree with it. The map could be navigated by clicking and dragging the mouse in the direction one reads, or by clicking individually through each concept and auto-scrolling to the next concept.
On any of the premises, users could add related points, facts, citations (in the form of URLs) that agree or disagree with the preceding point. Users simply clicked the Agree/Disagree boxes and were prompted to compose their message. This process allowed both the reader and the author to visualize an in-depth argument based on its general syllogism rather than on the chronology of the message.
Map fold outs may be read by either horizontal or vertical navigation.
With Debate Cited created, we set out to explore the potential of its open source format to help meet some critiques of the field, and, in particular, to help journalism students create more comprehensive science journalism. Students came from the Department of Journalism at Concordia University (Montreal) and were instructed on how to use Debate Cited. Fourteen students participated over four months from January to April, 2011.
Participants were asked to input the angle (main point) or lead (most newsworthy information) of their work as an initial premise for a story. This served as the root premise and allowed the student to lay out the subsequent premises in his or her story as a skeleton for the article. The last step occurred when other participants collaborated on filling out a concept map by contributing information that/ agreed or disagreed with the root premise. Participants were not allowed to edit comments, unless they made a special request, so that we could track all changes. Students were informed that participation was not mandatory and that they may submit their opinions on the application’s usefulness, ideas, and/or bug reports at any time.
Students were informed that Debate Cited was to serve the purpose of helping them to construct their articles, if they chose to use it, on the topic of biofuels. Students were allowed to approach the topic whichever way they preferred, as long as the result was a journalism article. We focused on the process students used to create a scientific story via: (i) a geometrical analysis of the concept maps used by participants using a point system comparable to Novak’s concept maps (Novak and Gowin, 1984) and (ii) a student questionnaire on functionality of the system. Novak and Gowin’s point system (Novak and Gowin, 1984) requires awarding one point to the first branching of a concept map, followed by additional points to each successive branching (horizontal expansion of a map) and/or each addition to the hierarchy of a map (vertical expansion of a concept map). The student questionnaire centered around whether peer collaboration on Debate Cited helped improve their knowledge on the subject, whether Debate Cited helped their stories overall, and whether the participant would consider using Debate Cited again.
Of the fourteen participants, seven used Debate Cited. In total, twelve Story Constructions were created. Story Constructions are original posts, similar to new topic threads in forums, and are placed at the head of a blank map (see Figures 5 and 6) to set the theme for follow-up posts. Although most students created only one topic, two of the students created more than oneThe nature of the discourse allowed a Debate Cited map to grow vertically and/or horizontally. Table 1 lists the twelve topics created and how they grew.
Max vertical spread refers to how many vertical levels of boxes appear in the associated maps from top to bottom (what Novak would call the hierarchies). Max horizontal spread refers to how many boxes grew horizontally at a single stage. Total number of boxes refers to the number of boxes that were created in the map (what Novak calls propositions). The topics tackled ranged from a comparison between electric vehicles and biofuel cars, the benefits of yeast to biofuel production, tobacco as a biofuel, and corn ethanol.
What was apparent in the discourse that took place in the twelve maps was that almost every point made was well thought out. Unlike comment threads or discussions on forums where certain comments may be superfluous, chatty, inconsequential, or tangential (Secko et al., 2011), the format of Debate Cited restricted the discourse to relevant and meaningful points. The following is an example of a discourse that took place on Topic #11, Whether Nuclear Power Lost the PR Battle (Figure 8).
Fukushima Nuclear Power Plant
The points that were added in Figure 8 are not listed chronologically. For example, even though a point that disagrees with the starting premise was added later in the debate (i.e “as safe as many pro-nuclear…”), the student felt it challenged the root premise and placed it horizontally adjacent in the hierarchy of points so that other users might read it earlier in the list. On the other hand, a premise that agrees with the root premise expands on it by providing a brief point from Wikipedia and creates a vertical line of discourse.
In Topic #2, Biofuels and Wood Residues (Figure 9), the line of discourse continuously expanded vertically and horizontally. Many of the premises carried their own set of controversy that was expanded on. In figure 9 we show an excerpt of the branching of the argument into vertical and horizontal discourse, which would not be visually displayed in the average commenting system online, because most commenting systems follow strictly vertical linearity. If linear commenting systems were sufficient, then one would expect such discourses to only grow vertically or horizontally. However, Figure 9 shows that the students made use of both vertical and horizontal posts, seeing a need and different purpose for each type. Furthermore, there is very little noise to signal in this system. Each of the premises that appear in this discussion provides the signal of agreement/disagreement with a corresponding justification and a frequent citation.
Following the creation of the maps, we did an analysis of how the maps compared to the final stories produced. Student articles were on the same topics as their maps and did not seem to change their mind on the topic. However, there were slight changes in how students approached the controversy behind their topic. For example, in Figure 9, one of the student’s premises was “there is no such thing as dead wood, according to Nicolas Mainville, biologist for Greenpeace.” On the map, the statement was disagreed with by another study in California, which had specified exactly what type of deadwood exists. In the article, the premise was retracted and instead replaced with a paragraph on both sides of the issue on what constitutes “deadwood.” When controversial points on Debate Cited’s map appeared, students presented the information objectively in the article. One thing was certain: Mistaken statements in the maps that were pointed out were not repeated in the final articles. The questionnaire explored whether Debate Cited had caused the students to become aware and how much students credited Debate Cited’s usefulness.
Of the students who used Debate Cited, six provided detailed feedback. While there were suggestions for improvement, which will be taken into consideration for future studies, most were satisfied with the results. When students were asked if they “feel that communicating with other science journalists on the website improved [their] knowledge on the subject [they] were covering” four of the six participants gave a clear “Yes,” while the remaining two gave positive results with slight reservations. On whether Debate Cited helped or hindered results, two of the six students responded “Yes” and one reported that it “moderately helped.” The remaining two others gave views that were still positive but slightly reserved (Table 2). The least positive was one student who wished more students in the class participated with Debate Cited as this student believed it held potential to be “very helpful” to their stories. When asked if the students would consider using Debate Cited again, five of the six students responded positively. The remaining student said that he or she would consider using it again as long as there were enough participants to use Debate Cited at its full potential. The most frequent complaint by students was that not enough participants were using Debate Cited. However, none of the students provided strongly negative feedback.
Controversial scientific topics require comprehensive journalism. Yet current science journalism has drawn criticism for being incomprehensive. To counter this, we have begun to explore how to combine the work of science journalists (interviews, facts, and counter-factuals) so that different sides of a scientific debate have a greater chance of being presented alongside the context of their scientific validity. This project built on software developed from previous code created by Novin and his brother, which was advanced here to be more user-friendly, relatively bug-free, secure, and useful to journalists. This initial work found that the science journalism students responded favourably to this format and that they felt it helped their overall articles.
The common method of aggregating user input is through a linear format. Whether through comments, emails, or message boards, the method of input most often resembles a single line of discussion. However, this study required a mapping scheme that could represent the complexities of the topic. Debate Cited contains a mapping element that is not present in the more common structure of concept maps yet is critical to journalism. Science journalism often deals with concepts that are new and uncommon. Therefore, the importance of the possibility of negation, counter-factuals, and disagreements is stressed in Debate Cited because these elements are important to scientific theory, results, and debate. This is also made quantifiable in that negations are clearly labeled and placed as separate concepts. In this study, students made use of the possibility of direct negation to premises to further the discourse.
The students involved in this study indicated in the questionnaire that Debate Cited helped improve their level of knowledge. This echoes scholarly work on group concept-mapping schemes in science education, which show it to be an effective learning tool (Ryve, 2004). It has also been recently shown that the combination of concept mapping and web-based learning can help students understand scientific topics (Liu and Wang, 2010). However, these studies are often aimed at educating children. Our purpose is to represent newsworthy scientific debates where the science is not always clear and other fields may need to be drawn from to provide comprehensive view of the issue. In science journalism, many forms of information need to be shared, and Debate Cited’s challenge is to fit them into a visual diagram. Students in this study were able to fit a maximum of 11 propositions into a concept map that was used to produce a science journalism story.
While concept maps were created prior to the Internet and were meant to be created on paper, recent research has found that web-based concept maps are more advantageous than their non-web counterpart (Novak and Canas, 2008). Two of the advantages of Debate Cited is that it can be accessed by participants from any computer with access to the Internet (due to its cloud system) and that ultimate/future participation on the web site is not intended to be performed in the confines of a classroom (like many traditional experiments with concept maps) but at a student’s own leisure.
Debate Cited retained many of the criteria from Novak’s notion of concept map (Novak and Gowin, 1984) except for what Novak called “cross links.” Cross links were abandoned for this particular study to provide a more straightforward map for users that would be easier to navigate–keeping in mind that user friendliness is important to Debate Cited since the concept-map was not to be used in a live classroom setting but by individuals when they were away from an instructor’s direct aid. Furthermore, a straightforward flowchart would be easier to transform into a news article than one with excessive relationships. Propositions, hierarchies, and examples correspond tightly with a common form of making an argument in a newspaper article. A point is made (proposition), a hierarchy of rational statements provides information, and real-world examples are provided.
Finally, one value of Debate Cited only touched on here is the rhetorical concept of providing two or more sides to a story. As mentioned, there is debate on when and how journalists should provide the ‘other side’ of a science story. One of the intentions of Debate Cited is to provide an interface that allows journalists to evaluate the different sides of science topics more efficiently. To measure whether propositions are receiving more than one hierarchical relation, the horizontal growth that occurs within a stage is measured. If a proposition receives several responses on Debate Cited, then it is more likely to be receiving more than one side to the proposition. The maps produced in this study (Figure 9) did begin to show the potential of horizontal growth, and indeed, the hierarchies on Debate Cited are more apparent than on a linear format such as a comment system.
This study had several limitations. First, while concept maps have been shown to improve the comprehension skills of students studying science, the study only involved university students studying science journalism. The students may have used Debate Cited in ways they wouldn’t outside a class and may have led to responses on usefulness that are overly positive. Thus, although the study had positive results for inexperienced science journalists, this may not translate directly to professional science journalism or other environments. Second, the sample size of this study is small and therefore only presents an initial exploration of how Debate Cited may function. Further work is needed to expand this sample in terms of numbers and diversity. Third, although Debate Cited is an original concept design, the authors of this paper wanted to adopt point systems that have already been used for prior concept maps. This was problematic for a number of reasons: i) Many point systems already exist; ii) There is still an ongoing debate on the value of point systems; and iii) as a result, point systems are still continuously being developed (West et al., 2002). While a point system may provide some indication that a certain map may be more robust than relative maps created by the same participants, it does not logically follow that this will always be the case. Several point systems were considered. Most of them gave one point per element studied (i.e. links, concepts, and hierarchies). Therefore, the point system used only served as an indication of how useful the map itself was. More specifically, it only indicated whether linking premises through hierarchy and bifurcation was used over standard linear style arguments.
Debate Cited is a work in progress. It forms part of a research direction that is situated at the crossroads of science journalism, science education, and new media. Debate Cited aims to add to the ongoing scholarly debate on how to help improve comprehensiveness in science journalism production. However, it does this not by describing or listing a problem, but by seeking to develop a method to address current limitations. Debate Cited is one attempt that is evolving through testing and is based on how scientific debates can be mapped prior to science journalism production. Signs of improvement presented here are based on whether the participants themselves reported satisfaction with the use of Debate Cited and found it to be a useful tool. Overall, the authors discovered that Debate Cited shows promise. Future studies will aim to see whether the reported satisfaction carries on to science journalists with professional experience.
This work was supported by Genome Canada and Genome Quebec as part of the GE3LS component of the Genozymes for Bioproducts and Bioprocesses Development project. We would like to thank the students of Jour 389G for their participation in the study.
Amend, E. and Secko, D. M. (2011). In the face of critique: A metasynthesis of the experiences of journalists covering health and science. Science Communication, 34(2):241-282.
Brainard, C. (2009). Science journalism’s hope and despair. Columbia Journalism Review. http://www.cjr.org/the_observatory/science_journalisms_hope_and_d.php?page=all
Brainard, C. and Russel, C. (2009). The new energy beat. Columbia Journalism Review. http://www.cjr.org/feature/the new energy beat.php?page=all.
Bubela, T., Nisbet, M., Borchelt, R., Brunger, F., Critchley, C., Einsiedel, E., Geller, G., Gupta, A., Hampel, J., and Hyde-Lay, R. (2009). Science communication reconsidered. Nature biotechnology, 27(6):514-518.
Creswell, J. (2009). Research Design – Qualitative, Quantitative, and Mixed Methods Approach, volume 3. Sage.
Delshad, A. B., Raymond, L., Sawicki, V., and Wegener, D. T. (2010). Public attitudes toward political and technological options for biofuels. Energy Policy, 38(7):3414-3425.
Dunwoody, S. (2008). Science journalism. Handbook of public communication of science and technology, Chapter 2, pp.15-26. http://www.bpatc.org.bd/elibrary/files/12713227600415386179.pdf
Kruvand, M. (2012). “Dr. soundbite”: The making of an expert source in science and medical stories. Science Communication. March 22, DOI: 2012 1075547011434991.
Liu, M. and Wang, J. (2010). Investigating knowledge integration in web-based thematic learning using concept mapping assessment. Educational Technology & Society, 13(2):25-39.
Logan, R. A. (2001). Science mass communication: Its conceptual history. Science Communication, 23(2):135-163.
Markham, K., Mintzes, J., and Jones, M. (1994). The concept map as a research and evaluation tool: Further evidence of validity. Journal of Research in Science Teaching, 31(1):91-101.
McDonald, B. (2011). Big hole in ozone layer, bigger hole in government policy – quirks and quarks. http://www.cbc.ca/news/technology/quirks-quarksblog/2011/10/big-hole-in-ozone-layer-bigger-hole-in-government-policy.html.
McKone, T. E., Nazaro_, W. W., Berck, P., Au_hammer, M., Lipman, T., Torn, M. S., Masanet, E., Lobscheid, A., Santero, N., Mishra, U., Barrett, A., Bomberg, M., Fingerman, K., Scown, C., Strogen, B., and Horvath, A. (2011). Grand challenges for life-cycle assessment of biofuels. Environ. Sci. Technol., 45(5):1751-1756.
Nelkin, D. (1987). Selling Science – How the Press Covers Science and Technology. New York: W.H. Freeman and Company.
Novak, J. and Canas, A. (2008). The theory underlying concept maps and how to construct and use them. Florida Institute for Human and Machine Cognition, 2008.
Novak, J. and Gowin, D. (1984). Learning how to learn. London: Cambridge University Press.
O’Hara, K. (2010). Canada must free scientists to talk to journalists. Nature, 467(7315):51.
Pellechia, M. (1997). Trends in science coverage: a content analysis of three US newspapers. Public Understanding of Science, 6(1):49-68.
Ryve, A. (2004). Can collaborative concept mapping create mathematically productive discourses? Educational Studies in Mathematics, 56:157-177.
Saari, M. A., Gibson, C., and Osler, A. (1998). Endangered species: science writers in the Canadian daily press. Public Understanding of Science, 7(1):61-81.
Schaefer, M. S. (2011). Sources, characteristics and effects of mass media communication on science: A review of the literature, current trends and areas for future research. Sociology Compass, 5(6):399-412.
Secko, D. M., Tlalka, S., Dunlop, M., Kingdon, A., and Amend, E. (2011). The unfinished science story: Journalist-audience interactions from the Globe and Mail’s online health and science sections. Journalism, 12:814-831.
Soloman, E. (2012). Power and politics with Evan Solomon – politics – CBC player. http://www.cbc.ca/player/News/Politics/ID/1883264868/.
Tosse, S. E. (2012). Aiming for social or political robustness? Media strategies among climate scientists. Science Communication. April 29, 2012. DOI: 1075547012438465.
Walton, D. (2006). Examination dialogue: An argumentation framework for critically questioning an expert opinion. Journal of Pragmatics, 38(5):745-777.
West, D. C., Park, J. K., Pomeroy, J. R., and Sandoval, J. (2002). Concept mapping assessment in medical education: a comparison of two scoring systems. Medical Education, 36(9):820-826.
Wright, W. and Reid, T. (2011). Green dreams or pipe dreams?: Media framing of the U.S. biofuels movement. Biomass and Bioenergy, 35(4):1390-1399.