The effect of taxonomic class and habitat salinity on scientific and public media attention

Ross Pearsall, Meghan Haley, Hanna Jackson

Department of Biological Sciences, Simon Fraser University, Burnaby, Canada

*This report is a class project NOT peer-reviewed science

Author Contributions

All authors were heavily involved equally in designing the study, collecting the data, and in the writing of the discussion. Meghan wrote the introduction, Ross wrote the methods, and Hanna analyzed the data, made figures, and wrote the results section. All authors proof-read the entire article multiple times and gave feedback and ideas throughout the entire process of the project.


It is important that science and media are working in synchrony to best inform the public and identifying areas of study where there is a disconnect is useful to fill gaps in this effort. We aimed to identify, amongst five classes of aquatic species, which ones had the disconnect between the amount of scientific attention and media attention. We randomly selected species within each chosen class and determined the number of Web of Science results and Google News results to get a value for scientific attention and media attention respectively, thus allowing us to calculate the science to media attention ratios. Amount of science and media attention on the average species in our five classes were significantly correlated. Chondrichthyes, Mammalia, and Actinopterygii had relatively high science to media ratios, and Anthozoa and Asteroidea had relatively low science to media ratios. On average, science and media report on topics in correlated amounts, however some taxonomic classes such as Anthozoa and Asteroidea do receive less media attention relative to science attention compared to others, and it may be worthwhile for scientists in these areas of study to allocate additional resources to make their work more accessible to the public via the media.


Scientists have expressed concerns over the quality of science communication in the media (Okaka 2009). This concern is likely due to the differences in the factors that affect what science and the mass media are incentivised to report on. In terms of science communication, the goal for the media is to share scientific findings with the public in a way that is understandable and, perhaps more importantly, interesting. This need to spread media that people enjoy throughout popular media platforms affects what the media will report on and the level of complexity of the article.

The consumers of science and media differ in their expectations for communication, legitimacy and even style of writing (Peters et al. 2008). This leads to a disconnect between what topics they cover, and the amount of attention topics receive. Previous research has found that most scientists believe that seeking media coverage for their work is a duty (Peters 2013). It is necessary to communicate scientific findings with the general public to educate as many people as possible on scientific discoveries, and the media is very efficient in doing this. Our study has implications in understanding the efficacy of scientists in achieving this goal.

We aim to identify the categories of species where there is a discrepancy between the relative efforts of the media and the scientific community. This will provide an insight into whether the media is reporting on what scientists are publishing, or rather, if scientists are studying and reporting on things that the general public cares about. If this is the case, we will expect to see differences in the relative volume of media and scientific articles across a couple variables such as taxa, and aquatic habitats. Science to media ratios will help determine if science or the media favour specific taxa, or marine or freshwater species. Our study will answer the question of whether there are discrepancies between what the media reports on and what science reports on. We hypothesize that there will be a significant discrepancy between the taxa that scientific articles report on and those that the media report on. We expect that some taxon will have very few media articles relative to scientific articles while other taxon will have notably more media than scientific articles. We further predict that the media will focus more on marine species than freshwater relative to science.


We chose five classes of marine species for comparison: Anthozoa, Chondrichthyes, Mammalia, Actinopterygii, and Asteroidea. These classes of organisms contained aquatic animals that we thought represented a wide variety of aquatic gaits, size, aquatic ecosystems, and modes of consumption. From each class of organism, we randomly selected 40 species from for a total of n=200 species. We chose these species by using to randomly select a set of numbers that represented the order, then family, genus, and species.

A literature search was then conducted to identify the number of articles each species occurred in, under both The Web of Science and Google News representing science and mainstream media coverage respectively. We also recorded the habitat water salinity and any scientific or common names given to each species. We did not include species where the number of science and media articles they were mentioned in were both zero due to logistical constraints of the study scope. The literature search was conducted over an open time period in order to accurately record the number of results from both literature sites. When species had multiple names, or they had a common and scientific name, both science and media literature searches were conducted using all given names in order to search for all possible articles.

For each class of aquatic organisms, the total number of articles in each category, science and media, were compared to one another in relative terms by dividing the total number of science articles by the total number of mainstream media articles. Higher values suggest that the aquatic class of interest has more science reports compared to media reports proposing that more attention is paid to these organisms from the science community than media. In trying to capture variation around the science to media ratios, we also analyzed the species disregarding all species that had a value of zero for either science or media, allowing us to create ratios for each species in a group. Ideally each reported article on The Web of Science and Google News represented an article about the organism of interest. Unfortunately, as some organisms in this study had extremely high number of occurrences the relevance of each given article they were mentioned in couldn’t be checked in order to determine if the article was about the actual organism or an unrelated topic.

Data was analysed in base R (version 1.1.442) using linear regression and Tukey HSD tests, as ANOVA tests were unbalanced and therefore unable to give F-values.


Number of science results and number of media results were significantly positively correlated (Figure1, R2=0.33, F1,98=99.4, p<2.2x10-16). Class Mammalia had significantly more media and science results than either Actinopterygii, Anthozoa, Asteroidea, and Chondrichthyes (Figure 2,Tukey HSD Tests, p<0.05) although more data is needed to conduct a full analysis of variance. All other cross-class comparisons were not significantly different. Across classes, Chondrichthyes, Mammalia and Actinopterygii had the most media attention relative to science, and Anthozoa and Asteroidea comparatively had more science attention (Table 1). When only species that had nonzero attention from both science and media were taken into consideration, we see a similar trend across all classes with the exception of Actinopterygii, which received a higher amount of science attention relative to media in this form of analysis (Table 1). There is a significant difference between the science to media ratios from species that had nonzero attention from both science and media between Chondrichthyes and Asteroidea as well as between Mammalia and Asteroidea (Figure 3, Tukey HSD Tests, p<0,05). All other comparisons were insignificant.

When we compared results across the habitat of our aquatic species, we found that there was no significant difference in the number of science or media results (Figure 4). Among the species chosen that had at least one science and at least one media result, there was no significant difference in the average science to media ratio of freshwater species, terrestrial species or species that inhabit both environments (Figure 5).


Figure 1. We sampled n=200 species for the number of Web of Science (science) results and number of Google News (media) results. Number of science results and number of media results are significantly correlated (R2=0.33, F1,98=99.4, p<2.2x10-16)


Figure 2. We sampled the number of Web of Science results (science) and Google News results (media) for species from 5 different classes and analysed the distribution within classes. Class Mammalia had significantly more media and science results than all other classes (Tukey HSD Tests, p<0.05)


Table 1. We sampled the number of Web of Science results (science) and Google News results (media) for species from 5 different classes. We then totaled the science and media results for each class and took the ratio (row 1). We also took species that had nonzero attention from both science and media, took ratios for each species and then averaged the ratios among all species (row 2).


Figure 3. We sampled the number of Web of Science results (science) and Google News results (media) for species from 5 different classes. We then took species that had nonzero attention from both science and media, took ratios for each species and then averaged the ratios among all species. No significant differences were found.


Figure 4. We sampled the number of Web of Science and Google News results of species that live in freshwater and marine habitats, and those that can exist in both. No significant differences were found among these groups.


Figure 5. We sampled the number of Web of Science (science) and Google News (media) results of species that live in freshwater and marine habitats, and those that can exist in both. We then took species that had nonzero attention from both sciences and media and averaged the science to media ratios. No significant difference was found between these groups.


We found that overall, science and media are more synchronous than they are asynchronous, with the number of science and media results being significantly correlated with one another across many sample species (Figure 1). However, there are certainly some classes that get more attention than others. When we took total science and total media reports for each class and turned them into ratios (Table 1). Classes Mammalia, Chondrichthyes and Actinopterygii had relatively low science to media ratios, meaning they got much more attention from the media than science relative to the other classes we tested. Classes Anthozoa and Asteroidea on the other hand, had relatively high science to media ratios, meaning they had more science articles relative to media articles relative to the other three previously mentioned classes (Table 1).

When we analyzed only species where attention from both media and science was nonzero, representing the most important species, we were then able to look at the average science to media ratio across many species and understand also the variability, as dividing by zero here is no longer a problem (Table 1, Figure 3). In this scenario, we see that the general trend for overall science to media ratio was similar to when all species were considered, with the exception of class Actinopterygii. For this class, when the zeros were taken out, the science to media ratio increased to a relatively high value, indicating that for those fish species that are at least somewhat important to both science and media, it is science that allocates more resources than media relative to other classes.

We also analyzed the 200 species by their habitat, freshwater, marine, or both, to see what effect this would have on how much science and media report on them. Our results came back insignificant, however due to constraints on how many classes we could include in our study, we cannot claim this result to be indicative of the true relationship, and can only say that for our particular set of data no effect was seen.

Eliminating bias in this study proved to be problematic. When finding a manageable method of species selection, we were limited to randomly selecting species from 5 taxa. This sample also contained an uneven amount of marine and freshwater species, with marine far outweighing freshwater. Therefore, we did not have a representative sample to determine if the media or scientific community reports on marine or freshwater species unequally. Another limitation we encountered came from the inability to create ratios when species had zero science or media search results. This interfered with our methods of analyzing the data. In response, we excluded species with search values of zero when creating figures illustrating the variation in science to media ratios between groups but leaving other figures and results unchanged.

We have shown that connections between scientists and media journalists exist, and are prevalent, yet there is room for improvement as some disconnect remains. Increased effort from the underappreciated branches of science could help bridge the gap. In fact, most scientists believe that seeking media coverage for their work is a duty (Peters 2013). Therefore the difference in journalistic interest on different topics may not always reflect the scientist's effort but instead could be related to variables such as the size of the field of study, size of research team, type of audience and even researcher status (Peters 2013). It is obvious that communication between science and media is important yet tricky due to the differences in the level, kind and depth of knowledge between the two audiences. The source can also have an effect. It has been shown that variation in mainstream media such as hedged language (certain vs. tentative) and the partisanship of the source can lead to differential trust in a source's information (Butterfuss et al. 2020). Trust in a source can be a very important driver of the perception of scientific consensus, an aspect of the gateway belief model, which is when personal judgment of scientific knowledge changes leading to an adjustment in attitude and support for a scientific topic (Sloane and Wiles 2019). It has even been shown that the way in which information is conveyed by a source such as video vs. text prove to be important in increasing the perception of scientific consensus (Goldberg et al. 2019). Therefore, it is not only important for scientists in under-communicated fields, such as those we identified, to reach out to public media, it is important they communicate their knowledge with the appropriate source that will accurately convey their message.

We see this form of analysis as having potential for further and more advanced and species-inclusive analysis, as all data used is easily accessible via the internet. Data for countless species could be collected autonomously by a simple computer program and analyzed. This process would yield insightful results into the relationship between science and the public media, all for a relatively small amount of work given its ability to be automated.

Literature Cited

Butterfuss, R., J. Aubele, and P. Kendeou. 2020. Hedged Language and Partisan Media Influence Belief in Science Claims. Science Communication.

Goldberg, M. H., S. van der Linden, M. T. Ballew, S. A. Rosenthal, A. Gustafson, and A. Leiserowitz. 2019. The Experience of Consensus: Video as an Effective Medium to Communicate Scientific Agreement on Climate Change. Science Communication 41:659–673.

Okaka, W. T. 2009. Science Reporting in the Media: A Scientist’s Perspective.

Peters, H. P. 2013. Gap between science and media revisited: Scientists as public communicators. Proceedings of the National Academy of Sciences of the United States of America 110:14102–14109


Peters, H. P., D. Brossard, S. Dunwoody, and M. Kallfass. 2008. Science-Media Interface; It’s Time to Reconsider:266–276.

Sloane, J. D., and J. R. Wiles. 2019. Communicating the consensus on climate change to college biology majors: The importance of preaching to the choir. Ecology and Evolution 10:594–601.