Keywords

6.1 Introduction

This chapter presents the results of using the analytical framework for popularization discourse to analyze a corpus of professional science journalisms texts. The analysis is quantitative and instrumental. It investigates how often each strategy is used in the texts, aiming to provide insights into the broader genre of science journalism (instead of at the level of the individual text, such as in Chap. 5), to answer questions about how professional science journalists communicate research findings to lay audiences. The goal of the chapter is to show what kinds of results are generated when analyzing a corpus of texts using the framework in this manner. The chapter also includes examples from the professional texts for each strategy.

6.2 Corpus Construction

This corpus consists of 38 science journalism texts. These texts are written by professional science journalists and published online. The texts cover a range of topics, from immunotherapy in llamas and bioengineered catnip to digital dementia and COVID-19 vaccines, and a range of disciplinary fields, such as biology, archeology, musicology, and technology. They also present a range of popularization subtypes: some articles presented insights from one academic publication, while others took multiple studies into consideration, showed an overview of current knowledge, or presented science news. This corpus was constructed for the validation phase of the development of our framework, using Berezow’s (Berezow, 2017) infographic on the quality of science reporting (see Chap. 4). This infographic arranges science reporting on the axes of evidence-based reporting and compellingness of content. For this corpus, texts were used from outlets from all represented quadrants except the quadrant that scores poor on both criteria.

We visited the websites of each of the outlets and downloaded the most recent article from the science section, or if there was none, the general website. In some cases, the most recent article was not an article about research findings, but, for example, an interview. In that case, we downloaded the five most recent articles and chose the one that was most compliant with the genre of science journalism. All texts in the corpus are written by science journalists, bar one text about holograms, which is written by the researchers themselves and therefore technically seen as a form of science communication (see Dahiya, 2021).

6.3 Analysis

We analyzed this corpus using our analytical framework. For each text, we performed an analysis such as the one that was shown in Chap. 5. We then tallied scores for each of the 34 strategies and for each text. In Fig. 6.1, we present a quantitative overview of the number of strategies that are used in each text. The figure shows that although all texts used a variety of strategies, none use (near) all of them.

Fig. 6.1
A bar graph of the amount of texts versus the number of strategies used from 1 to 34. The highest bar is at strategy 21 with the amount of texts being 8 approximately.

Number of strategies (out of 34 strategies) used per text in a corpus of professional science journalism texts

Full size image

In Table 6.1, for all 34 strategies, we present the percentage of texts using each strategy. We also show, for all five themes, an aggregated percentage for the number of strategies that are used per theme. In Table 6.2 we present one or multiple examples for the use of each strategy. We furthermore included an overview of additional strategies that are also used in the example shown (since, as discussed in Chap. 5, multiple strategies can be used within one phrase or sentence), to illustrate the complexity of the texts in the corpus and, therefore, the complexity of their analysis. If context is added for the example to make sense, the strategy under discussion is underlined. Hyperlinks are displayed in blue.

Table 6.1 Quantitative results from the corpus analysis of professional science journalism texts
Full size table
Table 6.2 Qualitative results from the corpus analysis of professional science journalism texts
Full size table

6.4 Interpretation

After using the framework to analyze these 38 science journalism texts, the following insights can be gathered about the corpus. The average number of strategies that was used is 18.7 strategies per text, and the median is 20 strategies with the lowest score of 10 strategies and the highest of 27 strategies. This also means that no single text uses all available strategies. Several strategies appear in every single text in the corpus (100%): presenting results/conclusions, hyperlinks, visuals, stance markers, and titles/subheadings. Other strategies are used in almost all texts (75–99%): contextualization, announcing the new finding or new contribution to the discipline, and imagery. In most texts (50–75%), the following strategies are used: lede, describing the method, applied implications, explanations, giving the researcher an active voice, lexical mention of the original research, additional sources, and inclusive pronouns. A few strategies are only used in some texts (25–50%): novelty, examples from daily life, academic implications, mentioning more research is necessary/next step in research, contribution to research, mention of statistics, link to the academic publication, opinion, references to the reader, features of conversational discourse, and questions. And lastly, some strategies are hardly used at all (0–25%): direct quote from the academic publication, in-text specification of a source, references to popular lore and beliefs, and popular culture, self-disclosure of the author’s public or personal life, giving the non-researcher an active voice, humor, and explicit self-reference.

Looking at the use of strategies per theme, the themes Subject Matter (75.9%), Tailoring Information to the Reader (80.3%), and Stance (81.6%) are used most. Or, to explain it differently, the theme Subject Matter contains six strategies that could be used across 38 texts, which means that the corpus enables 228 ‘options’ to use Subject Matter strategies, of which 173, or 75.9%, are in fact used. The aggregated scores for Credibility (42.8%) and Engagement (36.6%) are much lower. These aggregated results per theme of course do gloss over differences on the level of strategies. To give an example, for the theme Stance, stance markers are used in 100% of the texts but opinions are used in ‘only’ 63.2% of the texts—which then leads to an aggregated score of 81.6%. Especially in the themes Credibility and Engagement, there are big differences in the use of strategies, with some strategies being used across all texts (titles) while others are used hardly to not at all (self-disclosure of the author’s public or personal life).

Although we did not undertake this analysis to contrast professional journalism writing with student science journalism writing, anecdotally we can share that the use of strategies by professionals is much denser than is the case in the corpus that liberal education students wrote, which is discussed in Chap. 7. This might imply that professional writers have a better grasp on language use and are better equipped to utilize the options provided in the form of different strategies. The use of strategies also shows much more overlap, that is, multiple strategies being used at the same time. A visual representation of this overlap in strategies and richness in the use of strategies can be seen in the text that is discussed in Chap. 5, which was also part of the corpus discussed in this chapter.

When comparing the results of the analysis with Berezow’s (2017) infographic on quality of science reporting, we see that those texts that are part of the quadrant from ‘evidence-based reporting’/‘almost always compelling science content’ are more academic in register—especially the texts from the websites of Nature and Science—which is particularly visible through generally lower scores on the theme Engagement. Furthermore, many differences are visible between individual texts such as the length of texts, the number of strategies used within them, and the themes that are used the most.

A next step, which was not performed for this corpus, would be to analyze the specific use of each strategy. In other words, if a writer uses, for example, the strategy reference to the reader, what does that look like in the text? This step is performed for the corpus of student writing that is discussed in Chap. 7.

What You Have Learned in This Chapter

  • This chapter presented the analysis of a corpus of science journalism texts written by professional science journalists.

  • Science journalism articles from a whole range of science journalism outlets (based on a scale of being evidence-based and showing compelling science content) and from a spectrum of different topics/research fields were used.

  • The analysis was quantitative and instrumental, meaning the focus was on how often each strategy is used in the text.

  • Information is given about the overall number of strategies that were used in texts, on the percentage of use of each individual strategy, and about aggregated scores per theme.

  • Aggregated scores for the themes Subject Matter, Tailoring Information to the Reader, and Stance are generally high, with the themes Credibility and Engagement scoring much lower.