Re-Writing the Genetic Code

RE-WRITING THE GENETIC CODE:

AN EXPLORATION OF FRAMING, SOURCES, AND HYPE

IN MEDIA COVERAGE OF CRISPR

Darryn Jayne McCauley

A thesis submitted to the Faculty of the University of Delaware in partial fulfillment of the requirements for the degree of Master of Arts in Communication

Spring 2019

RE-WRITING THE GENETIC CODE:

AN EXPLORATION OF FRAMING, SOURCES, AND HYPE

IN MEDIA COVERAGE OF CRISPR

Darryn Jayne McCauley

Approved: ______Paul R. Brewer, Ph.D. Professor in charge of thesis on behalf of the Advisory Committee

Approved: ______Kami Silk, Ph.D. Chair of the Department of Communication

Approved: ______John Pelesko, Ph.D. Interim Dean of the College of Arts and Sciences

Approved: ______Douglas J. Doren, Ph.D. Interim Vice Provost for Graduate and Professional Education

ACKNOWLEDGMENTS

Writing a thesis is hard work, and I have a great deal of people to thank for the help, patience, and support that has kept me going. First, thank you to Dr. Paul Brewer, the chairman of my committee – this project would not have been possible without your patience, guidance, and thoughtful input. Your knowledge and expertise helped me grow my project from the ground up, and for that I am incredibly grateful. Thank you, also, to the rest of my committee: Dr. James Angelini and Dr. Danielle Catona. Your help, from our very first meeting, nurtured this project and gave me a solid foundation for success; I am grateful for both your time and wisdom. I would also like to thank my family and friends for their constant love and compassion. Your ability to cope with my outbursts of deadline stress, mid-project anxiety, and general senioritis made this project possible, and I am so glad you listened to my stress-driven rants, even though you might not have understood what exactly, I was so stressed about.

iii

TABLE OF CONTENTS

LIST OF TABLES ...... vii LIST OF FIGURES ...... viii ABSTRACT ...... ix

Chapter

1 INTRODUCTION ...... 1

Framing Theory ...... 7 Pressures on Journalists ...... 9 Framing Effects ...... 14 Framing Science ...... 18 Hyped Headlines and Replication Statements ...... 25 Intended Audience and Framing ...... 27

2 METHODS ...... 29

Sample ...... 29 Content Analysis ...... 31 Coders and Coder Training ...... 32

3 RESULTS ...... 34

Research Question 1: Common Biotechnology Frames ...... 34

Social progress ...... 35 Scientific/Technical uncertainty ...... 36 Public accountability and governance ...... 36 Runaway science ...... 38 Morality/ethics ...... 39 Economic development/competitiveness ...... 40 Conflict/strategy ...... 41 Middle way/alternative path ...... 42

Research Question 2: Sources Used by Journalists ...... 43 Research Question 3: Hyped Headlines ...... 45 Research Question 4: Replication Statements and Models ...... 46 Research Question 5: Speed of Clinical Translation ...... 47 Research Question 6: Comparing Newspapers and Science Outlets .... 48

Common biotechnology frames ...... 48 Comparison of sources used by journalists ...... 50

Comparison of hyped headlines and time to clinical translation .... 50

Conclusion ...... 51

4 DISCUSSION ...... 53

Frame Use and Effects in Crispr Coverage ...... 53

Social progress ...... 54 Scientific/technical uncertainty ...... 55 Public accountability/governance ...... 57 Runaway science ...... 58 Morality/ethics ...... 60 Economic development/competitiveness ...... 61 Conflict/strategy ...... 63

Overall Frame Use and Audience Effects ...... 64 Hyped Headlines ...... 68 Replication Statements, Models, and Speed of Clinical Translation ... 70 Limitations of the Study ...... 71 Areas for Further Research ...... 72

REFERENCES ...... 74

Appendix

A NISBET’S TYPOLOGY OF COMMON BIOTECHNOLOGY FRAMES .... 83 B CODING MANUAL ...... 84

Instrument A: Background Information ...... 84 Instrument B: Biotechnology Frames ...... 84 Instrument C: Sources ...... 88 Instrument D: Headlines ...... 89 Instrument E: Replication Statements, Models, and Speed of Clinical Translation ...... 89 Sample Articles ...... 91

C CODING INSTRUMENT ...... 94

Instrument A: Background Information ...... 94 Instrument B: Biotechnology Frames ...... 94 Instrument C: Sources ...... 94 Instrument D: Headlines ...... 95

Instrument E: Replication Statements, Models, and Speed of Clinical Translation ...... 95

D EXAMPLE ARTICLES: CODING RESULTS ...... 96

LIST OF TABLES

Table 1 Timeline of Major Crispr Events and Research Findings ...... 4

Table 2 Summary: Frequencies of Frame Use in Crispr Articles ...... 35

Table 3 Summary Table: Comparing Frame Use in Newspapers and Science Magazines ...... 49

Table 4 Comparing Source Use Between Newspapers and Science Magazines .. 50

vii

LIST OF FIGURES

Figure 1 Newspaper and Science Magazine Coverage of Crispr from 2015- 2018 ...... 6

Figure 2 Frequently Used Words in Hyped Headlines ...... 46

viii

ABSTRACT

In 2015, a small, repetitive section of DNA made international news when it helped scientists modify the genes of a living human zygote. That chunk of DNA, known as Crispr, has become an invaluable tool for genetic editing, with its use by scientists quickly overtaking both other common gene-editing technologies. Innovations using Crispr, from modifying mosquitos to prevent malaria, to creating drought-resistant crops, to the birth of the very first edited babies – twins – in late 2018, have made headlines in news outlets around the world. This study sought to understand how journalists present Crispr in news and science-oriented media in the United States. A content analysis of a three-year census of articles from top newspapers was used to examine the frames, sources, hype, and reliability statements that form the current narrative about Crispr. After analyzing 308 articles, several patterns were uncovered. Social progress is the most common frame used by journalists when writing about Crispr, scientists were the most common sources of information used by journalists, and men were far more likely to serve as sources than women. There were also notable differences in coverage between science magazines and newspaper articles: science magazines were more likely to discuss Crispr in terms of social progress, economic development, and scientific uncertainty frames, while newspapers were more likely to use runaway science frames. Science magazines were also more likely to use female sources than newspapers. In short, this study lays the groundwork for understanding the way media present a biotechnology that will soon be in clinical trials for disease therapies, creating modified plants and animals for us to consume, and potentially changing the genes of our children and humanity as a whole. By understanding the narratives used

ix by mass media to discuss Crispr, future research can examine the effects this presentation has on audience opinion and public policy.

Chapter 1

INTRODUCTION

In 2015, a team of researchers from China made international headlines when they demonstrated that a system called Crispr-Cas9 was capable of changing the DNA inside a living human zygote – the fused pair of human cells, sperm and egg, that allow us to reproduce (Liang et al., 2015). Though Crispr was first popularized as a technological innovation in 2013, its use in a nonviable human zygote in 2015 set it apart, earning the American Academy of Sciences “Breakthrough Award” (Travis, 2015). Since its widespread adoption in 2013, Crispr has been cited more times than TALENs and ZFNs-- two gene-editing techniques that have been in use over forty years--combined (Mariscal & Petropanagos, 2016). To understand the potential applications and implications of Crispr’s use, it is first necessary to understand how it works. “Crispr,” short for “clustered regularly interspersed short palindromic repeats,” refers to both a specific section of DNA discovered in single-celled organisms and the process of editing the genetic information inside a living cell using enzymes produced by the Crispr DNA sequence

(Kozubek, 2016). The process of editing a gene with Crispr begins with the selection of a specific target DNA sequence inside a living cell. Then Cas, an enzyme that functions like a tiny worker bee made out of protein, finds the target DNA and cuts it out. Finally, a new DNA sequence can be added in its place. Compared to previous methods of gene editing, Crispr is much cheaper, works with higher specificity for target DNA sequences, and is highly programmable. Crispr’s ease of use, diversity of

1 application (since it works in all living cells), and small price tag have allowed many scientists to use the technology, with new laboratories, college courses, and entrepreneurial businesses springing up around it. These developments, as well as future innovations using Crispr, may change numerous aspects of contemporary life, particularly in the healthcare and environmental spheres, as scientists use the technology to edit crops, animals, and even humans on the molecular level. Crispr has already been used in efforts to kill off the mosquito-borne parasite that causes malaria (Hammond et al., 2016), to combat muscular dystrophy in beagles (Zou et al., 2018), to increase soybean crop yield (Jacobs et al., 2015), and to augment plant disease resistance (Langner, Kamhoun &

Belhaj, 2018). Future applications of Crispr may help humans suffering from single- gene diseases, such as Huntington’s disease, cystic fibrosis, and muscular dystrophy. Initial trials on animal models have yielded mixed results (Koteliansky et al., 2015; Nelson et al., 2016; Wu et al., 2013; Xue et al., 2014; Yin et al., 2014), but scientists remain hopeful that specialized treatments will soon help patients. Beyond the scientific realm, Crispr has drawn a substantial amount of attention from the media. Like other forms of biotechnology—including stem cell research, cloning, and genetically modified organisms—Crispr has inspired Hollywood, resulting in the film Rampage (Flynn et al., 2018). It made a headlining appearance on Last Week Tonight with John Oliver (Oliver et al., 2018) and was featured on The Late Show with Stephen Colbert (Katsir, Moreschi, & Hoskinson, 2016). In the movie Rampage, Crispr is used by a biotechnology company to create a mutated pathogen that infects three different creatures: a wolf, a silverback gorilla, and an alligator, all of which grow to gargantuan proportions and display aggressive traits

2 (Flynn et al., 2018). The heroes of the story, including a gorilla handler played by Dwayne “The Rock” Johnson, manage to create an “antidote” with Crispr that removes aggressive characteristics, but does not shrink the animals to normal sizes. This film uses three elements of Crispr present in real life: Crispr research, bio- engineered creatures, and biotech companies performing private experiments. However, all three are exaggerated compared to their use in real life, potentially leaving the audience members who contributed to its $99,345,950 box office takings (Box Office Mojo, n.d.) with distorted views of Crispr’s capabilities. Last Week Tonight with John Oliver is a weekly talk show that focuses on a specific social issue or news event and breaks it down in detail over the course of a thirty-minute episode aired on HBO and available on YouTube, where the show has over 6.5 million subscribers. Oliver’s discussion of gene-editing began with soundbites from television spots on Crispr, used an interview with Jennifer Doudna (one of the primary Crispr researchers) to explain Crispr’s mechanism, and laid out the positive and negative aspects of the new technology (Oliver et al., 2018). Oliver’s conversational and humorous tone, coupled with his perspective on the ethical and moral dilemmas of Crispr’s use in human subjects, likely brought gene-editing to the attention of millions of people as a biotechnology with both positive and negative potential outcomes for the human race. On The Late Show with Stephen Colbert, the Harvard geneticist George Church spent 90 seconds attempting to explain Crispr while Colbert quipped, “Can you make me live forever?” and, “The Wolverine claws, I’m up for those, too” (Katsir, Moreschi, & Hoskinson, 2016). Colbert’s treatment of Crispr stands in stark contrast to Oliver’s – it relied on tropes and hyperbole, with Colbert asking for absurd

3 modifications and a means to “live forever.” Though Crispr was mentioned in passing by Church, he was not able to provide a detailed explanation, and Colbert’s hyperbole overshadowed any talk of legitimate research.

Table 1 Timeline of Major Crispr Events and Research Findings Year Month Event 1987 December Discovery of Crispr mechanism 2002 March Term “Crispr-Cas9” first published 2005 January Doudna (UC Berkeley) and Banfield first begin investigating Crispr 2011 March Doudna and Charpentier team up to investigate Crispr 2012 April First commercial use of Crispr by DuPont Corporation May Doudna and Charpentier submit patent application for Crispr mechanism August First Crispr publication breaks from Doudna's lab at UC Berkeley December Zhang (Broad Institute & MIT) submits fast-track patent for Crispr 2013 January Crispr-Cas9 used in human genome editing 2015 March Feng suggests using Crispr for human genome editing US scientists call for voluntary moratorium on human reproductive cell editing April NIH declares no funding for genome editing on human embryos May Huang, Liang, Xu, and Zhang publish report of Crispr use on human embryos October Church proposes modifying pig organs for human transplant with Crispr November US scientists modify mosquitoes to prevent carriage of malaria parasite December International Summit on Human Gene Editing meets to discuss Crispr ethics 2016 January Patent battle between Doudna and Zhang begins February UK scientists are authorized to genetically modify human embryos June NIH gives green light for first clinical trial using Crispr-Cas9 2017 February Zhang (MIT) wins initial patent battle for commercial Crispr rights 2018 January Pre-existing human antibodies target Cas9 proteins, raising possibility of immune responses undermining use of Crispr/Cas9 for use in gene therapy August Crispr-Cas9 clinical trial launched by Vertex Pharmaceuticals and CRISPR Therapeutics September Patent for commercial use of Crispr awarded to Zhang (MIT) in appeal ruling (Hubbard, Kraft & Serpente, 2018; Kozubek, 2016)

Likewise, Church was interviewed by Vanessa Hill (2017) in a YouTube- premiered documentary sponsored by PBS Digital Studios, in which Hill asked, “Could Crispr give us unicorns?” Church’s response exemplifies the disconnect between scientist’s expectations for the technology and the sheer imaginative

4 speculation of the public: “There is a mismatch between expectations and current capabilities.” Last, but not least, Crispr has also garnered a growing amount of news coverage in both science-focused outlets and more general outlets. Crispr coverage has spiked in conjunction with landmark events, defining the technology’s growth. Table 1 lays out a timeline of these events. A census of all Crispr articles published in two top newspapers, the New York Times and the Washington Post, shows that certain events in the Crispr timeline correspond to an increase in news coverage. One of the largest jumps in coverage occurred after Huang, Liang, Xu and Zhang announced their experiment modifying the genes of human embryos in May of 2015, after the National Institute of Health had declared a moratorium on funding for research on human embryos. Crispr coverage in June through December of 2015 was driven by new Crispr findings and debates on the ethics of using human embryos for research, culminating with the international Summit on Human Gene Editing in December. In 2016, coverage of Crispr centered on the patent battle between Doudna and Zhang for the commercial rights to Crispr therapeutic technologies, along with “designer baby” debates and the discovery that Crispr may be able to cure muscular dystrophy. February 2016 also brought the easement of the ban on human embryo testing in the UK, further driving debates about human embryo research and germline modification. In 2017, coverage highlighted the results of the patent battle between Zhang and Doudna, as well as the finding that scientists could edit butterfly genes to change their wing color and patterns. Coverage rose again in 2018 when the results of the

5 patent appeal trial between Doudna and Zhang again awarded the patent rights to Zhang and the Broad Institute/MIT.

Total Number of Crispr Articles 10

July-Sept 2015Oct-Dec 2015 July-Sept 2016Oct-Dec 2016 July-Sept 2017Oct-Dec 2017 July-Sept 2018 Jan-MarchApril-June 2015 2015 Jan-MarchApril-June 2016 2016 Jan-MarchApril-June 2017 2017 Jan-MarchApril-June 2018 2018

New York Times The Washington Post Popular Science The Atlantic Wired

Figure 1 Newspaper and Science Magazine Coverage of Crispr from 2015-2018

Figure 1 illustrates how the number of articles on Crispr increased more rapidly in the Times and the Post when one of the aforementioned events occurred. However, coverage in three science magazines – The Atlantic, Wired, and Popular Science – displayed a somewhat smoother curve in coverage over time. Coverage of Crispr in newspapers and science magazines, in turn, may affect how the public understands and views the technology. Throughout the process of scientific development, application, and regulation, journalism could act as a check, keeping public policy in the hands of those who will benefit from careful deliberation

6 and ethical considerations by the scientists and businesses that will profit. There is much to be gained by those submitting patents on Crispr in an effort to lock down proprietary research. Therapeutic applications may be monopolized, and individuals seeking healthcare for rare diseases may face the choice between exorbitant costs and death. In short, there is much at stake in understanding the media presentation of a nascent technology capable of changing our quality of life. To this end, the proposed research will draw on framing theory in analyzing news coverage of Crispr.

Framing Theory Every news article is framed. That is, it gives one aspect of the story greater attention and other aspects less to organize information into a news narrative (Nisbet, 2009b). Goffman (1974), the sociologist who first developed the concept of framing, posited that primary frameworks act as the means through which we understand reality. Framing provides a narrative structure for a news story, introducing major actors and central themes; it places the journalist and the reader within a broader context, giving them a foothold to understand the narrative. The ways in which journalists frame stories suggest to audience members what is and is not important for understanding the topic at hand. Thus, frames simplify complex issues, allowing audience members acting as “cognitive misers” to interpret them and make judgments (Nisbet, 2016). Put simply, framing sets the boundaries of discussion around an issue in the public arena. Gamson and Modigliani (1987) defined frames as “central organizing idea[s] or story line[s] that provide meaning to an unfolding strip of events, weaving a connection among them” (p. 376). This definition has been widely cited in communication research and was expanded in Gamson and Modigliani’s (1989)

7 research as they argued that journalists look for “pegs,” which are instances or markers that anchor a topic and provide opportunities for elaboration and commentary. In the case of Crispr, each of the major points in its timeline has acted as a “peg” for media coverage (see Figure 1). Furthermore, what is and is not discussed by the journalist becomes part of the frame. Entman (1993) posits that the structure of a news article suggests to the audience what is important to know about an issue, making some aspects more salient than others. The journalist may “…promote a particular problem definition, causal interpretation, moral evaluation, and/or treatment recommendation for the item described” (p. 52). Even when the same information is given, the context provided may alter the perceptions of the audience. This context may come in the form of cues, which are readily available components that signify a frame and are “known to and accepted by a majority of the society as common sense – persuasive and often taken for granted” (Pan & Kosicki, 1993 p. 57). As such, these cues, which signify frames, are used to quickly provide the average “inattentive and marginally informed” reader with an immediate heuristic for understanding what may be a complicated issue (Entman, 1993 p. 57). It follows that an interpretation made more salient, whether through its placement in the headline of text or through repetition, may make a lasting impression on the audience, framing their perceptions as they encounter more news coverage and consider the issue over time. However, frames are not inherently positive or negative. In some cases, the same frame can be used by proponents of both sides of an argument. For example, a frame emphasizing the morality and ethics of stem cell research may be used to argue

8 for the moral obligation to find treatments for those suffering from debilitating diseases, or to argue that using human embryos for research purposes is a moral violation of the right to life (Nisbet et al., 2003). As such, naming the dominant frame in a text does not necessarily denote an argument for or against any given side of an issue on the part of the journalist.

Pressures on Journalists Just as audiences use frames to quickly understand a story, journalists also use frames as a matter of convenience, simplifying complex topics quickly — especially for publication online, where “the norm is to write short” (Witschge & Nygren 2009). As such, members of the journalistic profession face pressures to use repetitive frames that appeal to inattentive individuals (Scheufele & Tewksbury, 2007). Beyond time constraints, multiple pressures – from monetary considerations as newsrooms choose advertisers, to audience analytics and inferences, to journalistic norms and values – inform the way journalists construct stories. News writing is not done in a vacuum. Taken together, these factors lead journalists to emphasize certain aspects of stories, whether out of habit or in an effort to appeal to an already common conception of an issue (Scheufele, 1999). Economic considerations pressure journalists to frame news in a way that is consistent with the wishes of investors and advertisers (Cohen, 2002). In the broadest sense, investors, publishers, networks, and parent corporations exert control over the types of news stories that can be published in a given news outlet. These influences converge in a “profit-motivated struggle between the media firm and corporate, legal, cultural, and technological influences” (p. 534). Cohen (2002) claims that within the newsroom itself, advertisers may influence the “tone, pace, and topics of news” (p.

9 535), using the hanging threat of withdrawn funding and lost advertising deals as a means of exerting control. As Witschge and Nygren (2009) observe, “Financial motives diminish the autonomy of journalists” (p. 48). Just as financial backers may influence what stories are pursued and published, the sources journalists rely on for information may also play a role in frame development and propagation. Davis (2009) defined the relationship between a journalist and a source as reflexive, maintaining that journalists need information and that sources have their own motives: namely, control of information and publicity. Though journalists are charged with maintaining professional autonomy, “sources, whether by fostering information dependency or by more covert means, have regularly gained the upper hand” (p. 206). In the political realm, control of information that is passed to the public acts as a means of influencing the course of democratic decision by changing what aspects of an issue the populace does and does not know about. The source may present journalists with pre-packaged frames, which are then passed along by the journalist to the audience (Davis, 2009). Thus, the dependence of journalists on reliable sources for information may convert journalists into political actors by shaping their choice of story and frame, whether out of convenience or habit. Meanwhile, changes in audience tracking allow journalists to see real-time metrics as news stories are published online. Rather than writing for an audience presumed to be passive and completely dependent on journalistic and editorial choice, web metrics provide information about what the audience wants “at the click of a button” (Tandoc, 2014 p. 785). No longer passive, the online audience can communicate quickly and easily with journalists, as comments on news articles as well as social media feedback and clickthrough data demonstrate audience interests. Use of

10 online metrics allows editors to see what kinds of news stories audiences want to see, “which is then balanced with what readers need” (Vu, 2014 p. 1105). Based on audience metrics, editors may ask journalists to produce content that is similar, expand on existing coverage, or provide additional analysis of existing news stories (Vu, 2014). Web metrics, according to MacGregor’s (2007) research, are seen by senior editors in the field as an objective means of quantifying audience engagement with the content, providing backing for journalistic hunches on what stories to cover. If an audience engages, the story is likely to get a follow up story, expanded coverage, or inspire similar stories. Furthermore, social media platforms act as means of presenting news to a broad audience, though Poell and van Dijck (2014) emphasize that their algorithmic structure does not make them a neutral distributor of information. Since “trending” and “most relevant” algorithms on Twitter and Facebook tend to push posts with high interaction, like comments, likes, and shares, to the top of newsfeeds, users will see popular content more frequently. The feed is not just a linear timeline, where all posts are equally likely to be seen by the user. As such, social media “…algorithmically privilege particular user signals and types of content” (p. 193). Rather than enhancing the ability of the journalist to freely write and disseminate information, the popularity of social media “intensifies the commercial pressures on journalistic independence” (p. 183). Certain kinds of posts may perform better on social media, where only the headline and featured photo are generally seen by the audience. As such, frames used to structure an issue within a headline or within posts on social media may influence audience perceptions (Gabielkov et al., 2016). Online formats may rely more heavily on frames to condense and contextualize information in a short amount of time.

11 Audience metrics also provide insights into the way users interact with news content online, though they may not be accurate predictors of the way individuals who receive the news in print interact with the content. For example, Craft, Vos, and Wolfgang (2015) found that audience members who interact with news content by commenting on the article posted on the original news site tend to see journalism as decreasing in reliability and a commitment to journalistic norms. The commitment to unbiased, objective reporting is cited as the largest “cultural capital” available to journalists that sets them apart from lay people who choose to write news independently (Craft, Vos & Wolfgang, 2015). Commenters, as analyzed by these researchers, tended to critique journalistic objectivity, bias, and partisanship; they saw the news organization as a group of individuals who ought to act as “responsible and morally aware human beings” (p. 11). From within the journalistic field itself, pressure to act in a professional manner may reinforce norms and values related to reporting and the selection of news to cover. In short, journalists are duty-bound to report news, when news is what the population needs to know to self-govern in a democratic society (Craft, Vos & Wolfgang, 2015). Journalists are seen as credible sources, to whom audience members entrust the task of providing objective news. Norms within the field are seen as at odds with what the audience wants: one senior editor at CNN held a low view of audience preferences, stating, “…if I just wanted to chase what people on the Internet wanted to click on, I would do stories about soft porn and football and nothing else” (MacGregor, 2007 p. 291). Additionally, the “balance norm” may influence journalists to frame news in a way that implies an equal representation from both sides of a perceived issue (Gamson

12 & Modigliani, 1989). The balance norm encourages frames that emphasize competition or that unfairly represent an argument that is vastly prevalent as equal to the minority view; journalists often represent themes as frames in dialectic. For example, the Luddite-esque Walden Pond frame may be placed in opposition to a Brave New World stance on technological development (Gamson & Modigliani, 1989). This norm of objectivity, then, may erroneously lead journalists to misrepresent the number of individuals with a minority view by presenting it in balance with the majority opinion. For example, climate change stands as a primary example of false balance in media coverage of a scientific issue. Boykoff and Boykoff

(2007) found that news coverage of climate change frequently relied on portrayals of “dueling scientists” to create a narrative of controversy and a sense of drama, even though climate change was not considered by most scientists to be an issue under serious debate. Using “balance” or “controversy” to frame a news story, then, may “distort unfolding science and policy discourse” when scientific consensus is portrayed not as a consensus, but as a debate (Boykoff & Boykoff, 2007 p. 11). More broadly, the goal of writing objective, informative, and unbiased news that advances what people need to know to act as informed decisionmakers may be at odds with writing quick, clickable internet puff pieces. The norm of journalistic objectivity relies on the implied “empirical validity/facticity” as a means of conveying a journalist’s authority to frame a news item (Pan & Kosicki, 1993). In fact, the labeling of news stories as either “news” or “opinion” pieces implies to the audience that all news articles are inherently factual, free of opinion and selectivity associated with a journalist’s choice of what to write about, whom to quote, and what amount of

13 detail or framing used to present the information (Pan & Kosicki, 1993). As such, journalists may utilize source perspectives, selectively including and excluding quoted information to create frames within stories while still maintaining a front of objectivity. The norm of journalistic objectivity combats the external pressures on journalists, driving them to search for factual news. Awareness of frame theory may, indeed, help journalists use frames as a tool to effectively shape their stories, and potentially shape public opinion. In short, the norms of the profession create yet another pressure guiding journalists as they write and frame news stories.

Framing Effects Just as pressures on journalists combine to influence journalistic choice, particularly in the choice of frame, those choices influence the audience for news articles. Framing effects emerge when a communicator’s “emphasis on a subset of potentially relevant considerations causes individuals to focus on these considerations when forming their opinions” (Druckman, 2001 p. 1042). That is, the impact of framing is to influence audience members to think of particular aspects of a situation when prompted, which may translate into their actions and policy decisions. For example, frames may be used when they are perceived as both important and accessible by audience members, creating a sort of closed loop in which common frames are repeated because they are accessible, and frames are seen as accessible because they are common (Scheufele, 1999). The prevalence of a frame in society in terms of amount of coverage may even influence how important individuals rate that issue (McCombs & Shaw, 1972). That is, the choice of what to cover and how often may, in itself, influence audience perceptions and shape the way individuals view an issue.

14 Framing also affects the way audience members choose to engage with political issues and influences their policy opinions (Bouks et al, 2015; Brossard & Nisbet, 2009; Lecheler, Bos & Vliegenthart, 2015; Maibach et al, 2010). Though one would hope that voters within a democracy make informed, fact-based decisions, individuals in actuality form opinions and make decisions even when they know very little, if anything, about a topic (Scheufele, 2006). Indeed, they often rely on the most readily available information, interpretations, and definitions in mass media as a “convenience sample” to use when making decisions related to policy (Brossard & Nisbet, 2006, p. 26). For instance, international comparisons of news coverage related to climate change showed that the American tendency to present climate change as a hotly- debated issue between scientists shrouded climate change in a frame of uncertainty, rendering national discussion unable to move onto the policy making phase (Brossard, Shanahan & McComas, 2004). Conversely, French reporters do not present climate change as debatable, and Brossard, Shanahan and McComas found that policy action is frequently discussed in the news (2004). French citizens also expressed pessimism and reserve toward innovation and technology, tending to temper policy in a way that created more environmentally friendly outcomes. This differs greatly from the

American perspective, in which policy is not seriously considered and Americans reportedly assert that technology and innovation will improve our quality of life over the next twenty years (Brossard, Shanahan & McCombs, 2004). French frames focused on human causes of climate change may be related to the differing attitudes of the French toward climate change policy development. At the very least, the repetition of frames in news media may reinforce existing public opinion, entrenching

15 individuals and journalists in a repetitive media cycle (Cacciatore, Scheufele & Iyengar, 2016). Media that individuals use to gain information influence their attitude toward science. Heavy television viewers are more likely to distrust scientists (Nisbet et al, 2002), while newspaper and magazine readers exhibited greater understanding and knowledge of science compared to television viewers, even when controlling for education and previous science background (Su et al, 2015). In some cases, simply covering more science topics in television news and print media increases public ratings of importance (Hwang & Southwell, 2009; Takahashi, 2011). The level of attention a consumer gives news has been shown to affect their attitudes. News framing more deeply influences attentive individuals (Scheufele & Tewskbury, 2007). Attentive individuals are more likely to become interested in science topics, then seek out and consume science media, creating a positive feedback loop (Slater, 2007). People who pay attention to science news also hold different attitudes compared to inattentive individuals – the attentive science news consumer is more likely to support biotechnology and innovation (Besley & Shanahan, 2005), and embryonic stem cell research (Ho, Brossard & Scheufele, 2008). They also display more deference to scientific authority and more trust in the decisions made by scientists and scientific individuals compared to the inattentive consumer (Anderson et al, 2011). Furthermore, the scope of the frame used to present an issue affects public perception. In the case of climate change, individuals exposed to thematic frames that represent the issue within a greater context that demonstrates the role of human activity were more supportive of policies that address climate change compared to

16 individuals who viewed episodic frames that did not provide context (Hart, 2011). Likewise, Holton et al (2012) discovered that the episodic approach to framing the measles/mumps/rubella (MMR) and autism relationship was more likely to influence individuals to see MMR as a cause of autism; as such, the newspaper coverage of the vaccine-scare as episodic, using stories from single individuals, likely drove the public to see all vaccines as harmful. In an experiment comparing opinions about emergent technology when exposed to either an entirely factual piece or an article with only framed opinion, it was found that conditions with only facts were no more effective than conditions with only frames (Druckman & Bolsen, 2011). Of course, the combination of facts presented in a frame caused the greatest change in attitude before and after exposure, while individuals who were only exposed to facts then integrated the facts with their prior opinions. The use of thematic frames may also convey that issues deeply affect individuals, as Maibach et al (2010) showed by presenting subjects with climate change news framed as a public health issue. These subjects went on to engage deeply with the novel “public health” frame no matter their prior opinions – even those who were cautious, doubtful, or disengaged with the issue of climate change expressed that climate change was a significant issue with personal significance. As such, issues that are frequently presented in the same frame may engender more engagement and potential policy development through expression via a different frame (Hwang & Southwell, 2009; Maibach, Roser-Renouf & Leisorwitz, 2008). Ultimately, the effects of framing and its relationship to policy development, public attention, and informed decision-making highlight the need for awareness on the part of journalists and the public as news stories are developed. Considering the

17 outcomes related to framing may allow journalists to engage more deeply with the public, increase attention to scientific topics, and pave the way for informed policy development.

Framing Science There is a pervasive misconception among scientists that the facts will out, while in reality the frames through which facts are presented may affect audience members’ perceptions of an issue. As Nisbet (2009b) puts it, “The facts are assumed to speak for themselves and to be interpreted by all citizens in similar ways” (p. 4). If facts are presented to the public but not understood in the same way by all individuals, “the failure in transmission is blamed on journalists, ‘irrational’ beliefs, or both” (p. 5). In short, there is often a disconnect between the way scientists view the world and the way the public views the world. Mariscal and Petropanagos (2016) conducted a qualitative analysis of frames used by scientists in their own academic articles about Crispr. This research reveals the way scientists think the technology should be framed. The authors uncovered three categories of dominant frames, including the "worried and watchful" frame, the "just another biotech" frame, and the "groundbreaking, disruptive, and revolutionary" frame. These frames were gathered from current scientific articles in which Crispr was used as a method for completing gene editing in any number of model systems (Mariscal & Petropanagos, 2016). The “worried and watchful” frame was used by scientists who warned against creating designer babies, implementing eugenics, and wreaking havoc on the environment and natural systems. These scientists heavily cautioned against unregulated or overly inventive use of the technology out of fear of

18 its abuse and potential effects. On the other hand, the “just another biotechnology” frame affords Crispr little in the way of overstatement. That is, these scientists did not see Crispr as different from current gene editing technology (like ZFNs and TALENs) in its potential uses and implications, but as a natural movement forward in scientific progress. Mariscal and Petropanagos (2018) argue that this frame “ignores the potential bad consequences of the technology as well as the ways in which older biotechnologies remain ethically controversial” (p. 2). Of the three, the researchers recommended that journalists use the “groundbreaking” frame, arguing that this frame is one that will provide the best balance between allowing beneficial discovery and regulating adverse experimentation. In their view, framing Crispr as a “revolutionary” new technology allows journalists to elucidate the potential risks, benefits, and need for oversight that may be glossed over if Crispr is portrayed as “just another biotechnology” while avoiding the doomsday perspective inherent in the worried and watchful frame. Mariscal and Petropanagos (2016) highlight the need for further research on how news coverage frames Crispr. Given that the researchers approached the question from a scientist’s perspective, rather than a layperson’s perspective, their frames do not account for the ways in which the public itself is likely to interpret Crispr; nor do these frames align with common ones used by journalists. That is, the “just another biotech” and “worried and watchful” frames are given negative connotations, rather than being seen as a means of constructing a narrative with elements given different salience. Frames are not inherently positive or negative, and Mariscal and Petropanagos (2016) assign value to the frames themselves. Furthermore, the ways scientists discuss the implications of Crispr are likely different than the ways the

19 public understands the issue – as anecdotally exemplified by Hill’s (2017) line of questioning about unicorns in her interview with geneticist George Church. By contrast, prior science communication research provides the basis for understanding how Crispr has actually been presented in the news—and, thus, how audience members may come to view it. In particular, Nisbet (2009b) has developed a typology of eight frames commonly used in science policy debates. These frames include social progress, economic development and competitiveness, morality/ethics, scientific uncertainty, Pandora’s box/runaway science, public accountability, middle way, and conflict/strategy frames (see Appendix A). Though no research to date has examined news framing of Crispr, previous studies have used Nisbet’s typology and similar categorization schemes to analyze news about other scientific topics (Nisbet & Scheufele, 2009; Nisbet, 2016), including a range of biotechnology-related ones such as stem cell research (Nisbet et al., 2003), genetically modified organisms (Nisbet, 2009b), and therapeutic cloning (Jensen, 2012). For example, Nisbet and Scheufele (2009) found that news coverage of GMOs in the United States prominently featured economic competitiveness and social progress frames: the former emphasized how GMOs could provide an economic boon for farmers and the economy alike, while the latter emphasized them as means of feeding a larger proportion of the population worldwide, particularly when drought- resistant crops are made available. Notably, researchers found that the perceptions of GMOs within the U.S. are overwhelmingly favorable, while in the United Kingdom and Europe, the majority of the public holds an unfavorable view (Nisbet & Scheufele, 2009). Abroad, GMOs are discussed in the frame of Pandora’s box, in which using genetically modified crops may lead to any number of unknown environmental

20 consequences; presenting the issue within a different frame has arguably resulted in different policy outcomes and implementations of the same biotechnology in countries that have similar economic and technological capabilities. Likewise, Jensen’s (2012) content analysis of frames used in news articles on therapeutic cloning uncovered a frequent tendency for journalists to use either “hope” or “dystopian” and “Nazi/mad science” narratives in both the U.S. and the United Kingdom. Similar to Nisbet’s social progress frame, “hope” narratives emphasize the positive possibilities for therapeutic cloning to eradicate disease and increase quality of life. On the other hand, the “dystopian” and “Nazi/mad science” narratives function like Pandora’s box/runaway science frames in Nisbet’s typology: they convey the risks of research on cloned embryos, the perils of the unknown, and the potential for science to go awry. Interestingly, though both U.S. and U.K. news coverage included both frames, the U.S. sample demonstrated almost equal representation of “hope” and “dystopian” frames. Jensen (2012) refers to this as “balanced hype,” stating that the two sides are rarely addressed, since U.S. journalists are “leery of appearing to take sides by asserting their own analyses of their sources’ claims” (p. 89). The literature on media framing of biotechnology issues provides a groundwork for understanding news coverage of Crispr. As in the case of GMOs, such coverage could highlight economic opportunity and/or social progress. As in the case of therapeutic cloning, it could also emphasize the Pandora’s box/runaway science frame. Alternatively, coverage of Crispr could depart from previously examined cases. To address these possibilities, the present study asks: RQ1: What frames do news stories use to cover Crispr?

21 Within a news story, actors function as the primary players to move along the narrative and serve as sources of information for journalists. Accordingly, journalists’ choice about whose stories to tell help place audience members within the narrative, allowing them to “register the identity of actors or interests that competed to dominate the text” (Entman, 1993 p. 55). For example, (Nisbet et al., 2003) found that news coverage of stem cell research pitted scientists and clergy members as rivals in conflict/strategy frames and as competing voices on questions of morality and ethics. The use of sources has three purposes: increasing credibility, developing frames, and forming a narrative structure or adding a “human interest” component. Journalists have complicated relationships with their sources. As previously discussed, reliance on sources may have subtle effects on the way journalists frame news articles, often making frames presented in PR packages or from government officials the easiest method for a journalist to structure a story (Davis, 2009). The most frequently used sources for science news include politicians, businessmen, scientists, and laypeople (Listerman, 2010). Other frequent sources, particularly in episodically framed health news, are celebrities, family members (i.e. mothers, children, etc.), health professionals, and nonprofit organizations (Holton et al, 2012). In order to increase perceived credibility, journalists often appeal to authority figures like government officials or scientists at reputable institutions (Berkowitz & Beach, 1993; Nisbet, 2009a). The relationship between scientists and journalists is characterized by two different agendas. Journalists are in need of credible sources who can provide them with breaking news stories, though the journalist needs to frame the news story in a way that draws attention to the issue and is appealing to a broad audience (Nisbet et al, 2003). Journalists may not always understand what their

22 scientific sources are saying, and in those cases may default to “neutral transmitter mode,” in which they do not criticize or object to what scientific sources present, but present information as accurately as possible, without deeper examination (Dunwoody, 2014). Here, the source credibility and accuracy act as a stand in for the journalistic norm of “objectivity.” In short, journalists rely on scientists as sources of credible information and breaking stories. Though the primary driver for journalists is to find credible sources and an entertaining story, scientists look to journalists for media attention with a different agenda. Scientists willingly act as sources for news media to increase their coverage and positive news portrayal, allowing them some level of control over their public image (Dunwoody, 2014). Peters (2013) posits that cooperating with the media is not only a norm, expected within the research realm, but also a means for scientists to garner public support, affirm their legitimacy, gain sponsors for research, and increase their appeal to potential “students, patients, and research development contracts” (p. 14106). These considerations may lead scientists to emphasize the positive potential outcomes of their research; scientists may also use their status as credible sources as a means of presenting framed information that casts their research in a positive light, further reaffirming that research’s relevance and importance (Dunwoody, 2014).

Scientists pass these considerations on to journalists in the form of pre-packaged frames, leaving American coverage of biotechnology “overwhelmingly positive” (Nisbet & Lewenstein, 2002). As Ross (2007) puts it, sources provide “a perspective legitimacy, a corroborating authority of the story to the audience” (p. 466). Furthermore, the use of scientists as sources may play on audience trust of scientists, increasing credibility of

23 the frame, or backfire and produce a boomerang effect based on audience distrust of scientists (Anderson et al, 2011). The use of government organizations as sources may also have unintended consequences: 73% of individuals in Brewer & Ley’s (2013) research indicated that they never trust Washington to do what is right. This distrust may decrease the credibility of a government source, particularly in relation to environmental news. Ultimately, the audience member’s perception of a source as credible is the final marker of whether that source will effectively pitch a frame, altering considerations of what information is salient and important, that will in turn affect individual opinions (Druckman, 2001). The use of elite sources, whether scientists, medical professionals, or government officials, is an appeal to credibility, and “framing effects may be evidence of citizens seeking guidance” from these credible elites (p. 1061). There is also a gender gap in news representation of women as sources: not much more than a decade ago, Ross (2007) concluded that “women rarely feature in news of national or international importance” (p. 454). However, in the case of Crispr, two of the most prominent scientists that discovered Crispr and submitted the first patent applications for Crispr’s adaptation for therapeutic use are women: Jennifer

Doudna and Emmanuelle Charpentier. Zoch and Turk (1999) wrote that women are frequently underrepresented as sources in the media to enforce the belief that women are “virtually without power and thus have no access to information that would be of use to the public” (p. 771). Even so, Doudna and Charpentier’s prominent contribution to Crispr research may mean that women are more frequently represented as sources in Crispr coverage, though further research is necessary to understand this relationship.

24 In short, collecting and categorizing the sources used in Crispr articles may provide a more nuanced perspective of narrative structure and framing used by journalists. Though prior research uses the terms “actors,” “interests,” and “sources” interchangeably (Nisbet et al, 2003; Nisbet & Lewenstein, 2002), this analysis focuses on sources who are quoted within a given news article. These voices represent the journalist’s choices about whom to provide direct representation within coverage. RQ2: Who are the primary sources in Crispr news articles?

Hyped Headlines and Replication Statements The breadth and depth of Crispr’s applications in science and healthcare may inspire the imaginations of scientists and journalists alike. The relatively untested nature of Crispr may also lead the latter to “hype” potential research outcomes and applications that have not yet been tested, replicated, or approved for public use. As MacGregor, Petersen and Parker (2018) observe, hyped claims “inflate expectations of innovations beyond what can realistically be sustained” (p. 77). In particular, “hyped” headlines may present strong causal attributions and draw generalizations (Dumas- Mallet, Smith, Boraud & Gonon, 2018). For example, healthcare coverage may include headlines promising “cures,” “keys,” and “cracked codes.” Molek- Kozakowska (2017) explains that journalists in mainstream news outlets are faced with a dilemma between the need to grab the reader’s attention and the need for conveying accurate news. Against this backdrop, hyped headlines act as one of many “linguistic resources used to emphasize the newsworthiness of science-related news items” (p. 895). This hype can be a kind of misinformation that individuals may use to maximize knowledge gain while minimizing effort. According to Ecker et al. (2014),

25 “Misinformation tends to influence people’s memory and reasoning continuously despite corrections” (p. 28). Coupled with the fact that six in ten articles shared online have not been clicked on by the person who shared the article, the effect of a hyped headline may persist after an individual reads the headline, even with corrective information provided in the body of the article (Gabielkov et al., 2016). Given all this, the present study will examine whether journalists “hype” Crispr in news headlines. RQ3: To what extent, and in what ways, do news headlines hype Crispr? Furthermore, news coverage often presents biomedical research as “cutting- edge” and “breakthrough,” often without mention of reliability or replication (Dumas- Mallet et al., 2018, p. 239). For example, Gonon et al. (2012) found that of the “top

10” most reported ADHD studies that emerged in the 1990s, only “two studies passed the test of the years” (p. 8), indicating a tendency for newspapers to report initial studies but not subsequent scientific articles containing refutations or attenuations. This tendency may lead to a picture of biomedical innovation that is incomplete and misleading to the average newspaper reader unfamiliar with the nature of the scientific process. In the case at hand, healthcare applications of Crispr are only just entering clinical trials, but no Crispr treatment has been approved for mass pharmaceutical production. The gap between the development of a new technology and the actual implementation of therapies may create hope and undue expectation. In reality, Crispr’s applications to human subjects will likely have adverse effects, particularly given that genes are not expressed individually, but in tandem with other genes in a suite of phenotypic reactions (Kozubek, 2016). Representing Crispr as an immediate solution or cure for genetic disease is therefore another potential aspect of hype.

26 Though there is pressure on journalists and press release copywriters to create optimistic narratives about scientific findings, this practice deemphasizes the uncertain nature of technological development. As pressure increases and public funding becomes crucial for scientists to maintain their laboratories and pay their researchers, the temptation to hype findings will continue (Evans et al., 2011). Therefore, the present study will address following research questions: RQ4: Does news coverage of Crispr discuss issues of replication? RQ5: Does news coverage of Crispr discuss the speed of clinical translation related to healthcare?

Intended Audience and Framing A final point to consider is that coverage of biotechnology issues such as Crispr may vary across different types of news outlets. For example, general publications and science-oriented publications may frame Crispr differently. Studies examining media framing of scientific issues often focus on top-circulating newspapers such as the New York Times (Nisbet, 2009b; Nisbet & Lewenstein, 2002) and the Washington Post (Marks et al., 2007). Both of these publications have dedicated sections for science news but also cover a wide array of other topics and subject matter. Meanwhile, publications such as Popular Science and National Geographic focus primarily on science and technology stories. In a census of Crispr coverage, spanning January 2015 to October 2018, six newspapers and six science magazines with the highest number of subscribers were used to count article coverage. Wired published 61 unique stories on Crispr and Popular Science published 49, while the New York Times and Washington Post published 38 and 37 stories, respectively. In spite of their elevated coverage of

27 breaking biotechnology news, previous research on science communication at its interface with the public has paid relatively little attention to science-oriented publications (Brossard, Shanahan & McComas, 2004; Boykoff & Boykoff, 2007; Marks et al, 2007; Nisbet & Lewenstein, 2002; Nisbet et al, 2003; Zimmerman et al, 2001). As such, one goal of this research project is to discover the ways that science- oriented publications may use framing and narrative structures to cover Crispr news, including biotechnology frames, hype, and context, including reliability statements. Thus, the present study will test for differences in coverage across outlet type. RQ6: Do general circulation newspapers and science-oriented publications

differ in how they cover Crispr?

28 Chapter 2

METHODS

For the purposes of this study, a content analysis was conducted to analyze coverage of Crispr since its mainstream debut in 2015. To analyze both the general coverage of Crispr in newspapers and the coverage of Crispr in magazines for science audiences, it was necessary to select publications with wide readership and pertinent content. After selecting publications, I analyzed the common frames used by journalists, recorded the individuals, organizations, and groups cited as sources, determined whether headlines were hyped, whether replication was mentioned, and whether a concrete time to clinical translation was given by journalists. Each of these steps addressed one of the given research questions. The criteria for sample selection and inclusion/exclusion of news articles within those samples are detailed below.

Sample News publications, as sources of information for the general public, were selected by taking the top six newspapers by readership that also included a dedicated science section. Readership was determined using a rank-ordered list of publications (“Top 15 U.S. Newspapers,” 2018). The top six news publications with dedicated science/technology sections were: The New York Times, Washington Post, USA Today, Wall Street Journal, Los Angeles Times, and the Washington Post. Likewise, I determined a set of criteria for selection of relevant, noteworthy science-oriented publications. To be included the publication had to be a magazine with a print version, as print circulation was used as a metric to determine overall readership. Print circulation statistics were gathered from each outlet’s press kit provided online. Furthermore, outlets were selected if they had high print circulation

29 and were also given high marks for “evidence-based reporting” in an expert review of science news sources by the American Council of Science and Health in conjunction with RealClear Science (Ross & Hartsfield, 2017). After combining these two metrics, the following six science-oriented publications were selected: The Atlantic, Wired, National Geographic, Smithsonian, Popular Science, and Discover. I selected these sources as points of comparison to test whether general news outlets and science- oriented publications differ in their framing of Crispr. After selecting publications, I performed an exhaustive online search for relevant articles. Using Google, a search of “Crispr” and the publication’s dedicated website URL was used to find all related articles available online, published between

January 1, 2015 and December 31, 2018. Because Crispr has been slowly gaining popularity in the media, sorting through an exhaustive search of all Crispr related articles was not overwhelming. The broad search was then narrowed down using specific inclusion criteria for individual articles. Articles were included in the sample if Crispr was mentioned by name in the headline, in the article description below the headline, or the first three paragraphs of the article proper. Editorials, opinion pieces, and news articles were included for coding. Press releases, abstracts, and news blurbs were excluded. Articles that were originally published by another source or site were also excluded. All articles were copied and pasted into Word documents to enable easy searching and word count collection. Images and their captions in the body of the articles were excluded from analysis.

30 Content Analysis For each article, the originating news source, headline, author/s, date of publication, and total word count were recorded (see Appendices B & C). The primary coder (the author) then coded for presence or absence of each of the eight common biotechnology frames from Nisbet’s (2009b) typology using their associated keywords and definitions (see Appendix A). Frames were considered present if they appeared at all in the text – even if a frame was briefly mentioned for as little as part of a compound sentence, it was counted as present. Sources were then recorded by examining the name and title given for each individual quoted in the article. The sex was recorded if given in the context of the article as a pronoun. In cases where individual names were mentioned but the journalist did not refer to them by a pronoun indicating sex, Google searches of the source names and organizational affiliations were used to find their sex. Each individual’s profession and/or organizational affiliation was recorded if mentioned. Information given as a quote directly from an organization without citing a specific individual was also recorded. For example, if a journalist mentioned something the University of California, Berkeley had said as a whole, that would count as a source with no individual named, and no sex present. When groups of people (such as scientists or researchers) were mentioned as sources, whether affiliated with an organization or not, they were also recorded. Next, the coder analyzed each article’s headline, coding for presence or absence of hype, and in the presence of hype, collecting keywords. The coding scheme considered hyped words to be those that were emotionally charged, polarizing, or unduly certain. Examples are outlined in the coding manual (Appendix B).

31 The coder then recorded the presence or absence of a replication statement. Any mention of the “initial” nature of a study, or, conversely, the replication of a study or its completion of clinical trials, was considered a replication statement. Words identified by Dumas-Mallet et al. (2018) as indicating a need for replication include “verified,” “repeated,” and “preliminary,” along with phrases such as “if the work stands up,” “tests on larger populations of adults” and “more work is needed to confirm the findings.” If the article was related to healthcare, the amount of time necessary for clinical translation was also recorded, if present. An article was considered to be healthcare related if it mentioned applications of Crispr to human beings as a means of treating diseases or as therapy. Occasionally, Crispr’s human/healthcare applications were mentioned briefly, but were not a focus of the article. As such, those instances did not qualify an article as healthcare related. For example, a journalist might mention Crispr’s potential uses for creating “designer babies” as a mean to illustrate its popularly referenced application, but would primarily discuss the patent battle, or its agricultural applications, in the bulk of the article. Finally, the differences between papers for general consumption and those for scientifically-oriented audiences were analyzed using Pearson’s chi-square. In total,

121 of the articles were from newspapers, and 187 were from science-oriented magazines. The two types of articles were compared across each of the key variables.

Coders and Coder Training Initial sampling provided a means of developing an effective, inclusive, and exhaustive coding scheme. See Appendix B for the coding instrument. A subsample (20%) of the articles was reviewed by a second coder, totaling 61 randomly selected

32 articles from the larger sample of 308 total articles. The second coder was trained using materials from the preliminary sample of articles selected to develop a coding scheme with examples of each frame, source data, hyped headlines, reliability statements. The use of articles from publications not included in the overall sample of Crispr materials provided a thorough representation of language and frames used in current news coverage without using direct material from the broader sample. The author acted as the primary coder, analyzing all articles from both general newspapers and science-oriented magazines from January 2015 through October 2018 (308 in total). The author met with the second coder to discuss the coding manual and data collection. The author then coded an article with the second coder, demonstrating how to keep track of sources and frames within the context of the article. The second coder was then given opportunities to discuss any questions or issues with the coding manual, then asked to code an article independently. The second coder was not able to meet for follow-up training or retraining after the initial coding data was provided. Due to coder drift and the inability of the second coder to meet for re-training and spot-checking during the coding process, the reliability between the primary and second coder’s data was found to be unacceptable for most variables after completing reliability analysis using Cohen’s kappa (Cohen, 1960), as Table 2 reports. For this reason, the primary coder’s data were used for the purposes of this project until further coder training and post-hoc reliability analysis is feasible.

33 Chapter 3

RESULTS

The purpose of this study was to explore how newspapers and science magazines discuss a cutting-edge biotechnology, Crispr, since its mainstream news- debut in 2015, especially in regard to its framing, use of sources, hype, and mentions of uncertainty and the need for replication. The way journalists frame a biotechnology topic, particularly a technology that will soon be in the realm of policy debate and regulation, may affect public perceptions and the outcomes of policy debates. This research includes a census of Crispr-related articles from 12 different newspapers and science magazines, providing a well-balanced and comprehensive picture of Crispr’s media presence. In all, 308 articles were included in the sample, with 121 articles (39.29% of the sample) from news sources and 187 articles (60.71%) from science- oriented publications.

Research Question 1: Common Biotechnology Frames My first research question focused on understanding what common biotechnology frames are used to discuss Crispr. All eight of the common biotechnology frames were used to frame the discussion of Crispr at varying frequencies (see Table 2). Frames in the summary table are ordered from most common to least common. Cohen’s kappa was used as a means of reporting reliability via percent agreement between two coders (Cohen, 1960), and all articles referenced in this chapter may be found cited in Appendix D.

34 Social progress The social progress frame was present in 285 articles (92.5%), with authors referencing the potential for Crispr to solve problems present in society, from genetic disease to pest control and increased agricultural productivity. In one extreme example, a September 8, 2017 USA Today journalist wrote, “To work their healing magic, scientists utilized an incredible genetic engineering tool called CRISPR to modify the ‘broken’ gene in human embryos” (Locwin, 2017). Indeed, the social progress frame was often present in the first line of articles. The first line of a December 3, 2015 New York Times article read, “Biologists in the United States and Europe are developing a revolutionary genetic technique that promises to provide an unprecedented degree of control over insect-borne diseases and crop pests” (Wade, 2015b). Similarly, a September 6, 2016 article in Popular Science opened by saying, “Scientists are already using gene editing enzyme CRISPR to alter human embryos and make antibiotics less resistant” (Ossola, 2016). Offering notable examples of Crispr’s problem-solving capabilities to date was a frequent means of setting the tone for a journalist’s story.

Table 2 Summary: Frequencies of Frame Use in Crispr Articles Frame Number Number Reliability Percent Present Absent (Cohen’s kappa) Agreement Social Progress 285 (92.5%) 23 (7.5%) -.003 63.9 Scientific/Technical Uncertainty 189 (61.4%) 119 (38.6%) .288 60.7 Public Accountability/ Governance 188 (61.0%) 120 (39.0%) .341 65.6 Runaway Science 172 (55.8%) 136 (44.2%) .252 60.7 Morality/Ethics 171 (55.5%) 137 (44.5%) .869 93.4 Economic Development/ Competitiveness 136 (44.2%) 172 (55.8%) .616 80.3 Conflict/Strategy 47 (15.3%) 261 (84.7%) .171 70.5 Middle Way/Alternative Path 20 (6.5%) 288 (93.5%) -.079 85.2

35 Scientific/Technical uncertainty Of the articles examined, 189 used the scientific and technical uncertainty frame (61.4%). These articles frequently referenced uncertainty related to Crispr’s future applications in healthcare. A December 14, 2018 Wall Street Journal article summed up the uncertainty surrounding Crispr neatly: “Bioethicists and many geneticists have raised doubts about applying the gene-editing technology to animals and especially to humans, given the continued uncertainties in both the science and the lab and field results” (Rana & Craymer, 2018). The uncertainty related to Crispr’s use in human subjects was compounded when, in 2017, a letter-to-the-editor published in Nature Methods indicated that Cas9 (the tag-team Crispr enzyme of choice) made off- target, unintended cuts at a rate higher than anyone had anticipated (Molteni, 2017). In a Washington Post op-ed, University of Pittsburgh pathologist and op-ed author Jianhua Luo addressed the “messiness” of Crispr cuts, writing that, “Even if the off- target rate is relatively low, we don’t yet understand the long-term consequences” (2018).

Public accountability and governance

The public accountability/governance frame was present in 188 articles (61.0%). Indeed, all of the articles discussing the patent fight between the Broad Institute/MIT and the University of California, Berkeley used this frame. As the Los Angeles Times wrote on February 5, 2016, “The 21st century is young, but it may already have its era-defining patent fight” (Hiltzik, 2016). The patent fight was also intertwined with frames of economic development and competitiveness, as seen in a May 11, 2015 New York Times article which references how Jennifer Doudna “…is also fighting for control of what could be hugely lucrative intellectual property rights to the genome

36 editing technique” (Pollack, 2015). In relation to gene editing, journalists often set up this frame using quotes from ethicists. For example, a July 26, 2017 New York Post article used a quote from an NYU School of Medicine medical ethicist who said, “Who should own genetic engineering techniques, and what, if any, requirements will they have to make the taxpayer-funded research that made this possible available and accessible at affordable prices?” (Perez, 2017). Gene editing was not the only Crispr application journalists discussed in terms of control, regulation, and governance. Agricultural and environmental applications, such as the creation of CRISPR’d tomatoes or mosquitos with gene-drive, were also subject to public accountability and governance frames. For example, on May 6, 2018, the Wall Street Journal observed, “Newer gene-editing techniques let scientists make changes without adding any foreign DNA, potentially achieving similar effects with looser regulation, if no outside genes are added” (Bunge & Craymer, 2018). The possibility that Crispr’d plants would not be subject to the same strict regulations and public backlash as GMOs was a recurring theme. In one of the earliest calls for regulation of Crispr in agriculture, an April 18, 2016 Washington Post article wrote, “Whether or not GMOs hold hidden risks, it’s time for U.S. governing agencies to step up and create new, scientifically sound guidelines for studying and regulating CRISPR edited crops” (Feltman, 2016). Similarly, a December 14, 2018 Wall Street Journal article suggested that “many governments work to clear regulatory pathways to bring meat, eggs, and dairy from gene-edited animals to store shelves” (Rana & Craymer, 2014). As for the edited mosquitos, a September 2, 2016 Wall Street Journal article mentioned that, “Field trials and releases of genetically modified mosquitoes require regulatory approval and can take years” (McKay, 2016). The representation of the

37 public accountability/governance frame hinged on the need for guidelines and regulation of Crispr’d plants and animals that may have effects on the environment and the economy.

Runaway science Beyond mentioning unintended consequences as evidence for the need of government intervention in Crispr development, some journalists forewent subtlety in framing Crispr as “runaway” science. The Pandora’s Box/Frankenstein’s Monster/Runaway Science frame was present in 172 (55.8%) of the total articles. When used by journalists, this frame tended to be the most descriptive and speculative means of framing the technology. For example, one November 29, 2018 Washington Post article argued that, “Playing with humanity’s genetic code could open a Pandora’s box” (Thiessen, 2018), while a New York Post article from July 26, 2017 began with the line, “Welcome to the brave new world of gene editing” (Perez, 2017). An interviewee in a June, 2016 Discover Magazine article was quoted as saying, “It’s the tool to modify nature. But when do we stop engineering nature? It’s kind of like Frankenstein” (Weelwright, 2016). Finally, the following excerpt from a National Geographic article published in the August, 2016 issue extensively discussed implications of Crispr as a runaway technology if used to edit the human germline:

In opening the door to one kind of germline modification, we are likely opening it to all kinds. Permitting human germline gene editing for any reason would likely lead to its escape from regulatory limits, to its adoption for enhancement purposes, and to the emergence of a market- based eugenics that would exacerbate already existing discrimination, inequality, and conflict. We need not and should not risk these outcomes. (Darnovsky & Harris, 2017)

38 This slippery slope argument formed the foundation for the runaway science frame. Notably, social progress frames frequently preceded runaway science frames. For example, a December 28, 2018 Wall Street Journal article asserted that, “Crispr holds the promise to correct intractable diseases by rewriting a person’s genetic code or DNA. But it isn’t foolproof and can cause changes in genes other than the ones sought. Unintended consequences could surface years later” (Rana & Fan, 2018). In this case, the runaway frame serves to temper the potential social progress outcomes of the technique, acting as a kind of false balance between progress and potential setbacks. The first sentence emphasizes the problems solved by the gene-editing technique, while the second and third emphasize its dangers and potential unintended consequences.

Morality/ethics Morality/ethics frames were the fifth most common, with a presence in 171 (55.5%) articles. This frame usually invoked ethical lines, boundaries, and moral limits. In a Washington Post opinion article from November 29, 2018, one columnist wrote, “Here is the bottom line: We should not be playing God,” citing a “moral line” that should not be crossed (Thiessen, 2018). Similarly, a July 24, 2017 USA Today op-ed argued, “…keeping a lid on CRISPR will be paramount, as it is just as potent a tool for evil as it is for good” (Feibus, 2017); likewise, in a Popular Science article from November 26, 2018, (following He Jianku’s announcement of the first genetically engineered babies born in China), one scientist discussed the ethicality of He’s actions: “in this case—in secrecy and without oversight--…it’s totally unacceptable” (Eschner, 2018).

39 Journalists often discussed morality in relation to the voluntary moratorium imposed on human genome editing by the scientific community and in conjunction with public accountability and governance frames. In its December 3, 2015 coverage of the first International Summit on Human Genome Editing, a New York Times article pointed out that, “The academies have no regulatory power, but their moral authority on this issue seems very likely to be accepted by scientists in most or all countries” (Wade, 2015a). That is, journalists often appealed to the ability of the scientific community to self-regulate, whether directly and in their own words, or by quoting scientists and bioethicists who were in favor of preventing ethical mis-steps before they occur, particularly in relation to human genome editing. For example, an August

2, 2017 New York Post article quoted Marcy Darnovksy, the director of the Center for Genetics and Society, who commented on the potential human embryo modification abilities of Crispr, saying, “…the idea of using it on the front end to engineer a future generation – we need to draw a bright line there” (Li & Fears, 2017). In short, talk of “lines” drawn, “moral authority,” and “good and evil,” characterized journalist use of the morality/ethics frame to discuss Crispr and the numerous ethical issues it raises.

Economic development/competitiveness The economic development and competitiveness frame was present in 136 articles (44.2%). It frequently referenced the risks and benefits to investment in Crispr technologies for biotechnology startups. In one notable example referencing the economic stakes of the outcome of the patent fight over Crispr, a February 15, 2017 Atlantic piece argued that, “If CRISPR is as widely applicable in biotech as many expect, whoever holds the patent could end up with millions of dollars in licensing fees from companies that want to use the technique to develop new therapies or

40 products” (Zhang, 2017). Alternatively, journalists referenced the potential benefits and risks for investors in biotechnology startups. One WIRED article from January 12, 2018 detailed the goings-on at the J.P. Morgan Healthcare Conference, which took place the same weekend that an initial finding of immunity to Crispr in human cell’s cast doubt on the tech’s future healthcare applications: “All three companies’ stocks took serious hits Monday morning, even as investors crowded into ballrooms to hear Crispr execs speak” (Molteni, 2018a). To investors Crispr represents a technology with the potential to cure big-name diseases and create “…next generation medicines that could do things like erase genetic defects and supercharge the body’s natural defense against cancer. And that mean[s] big money,” as another WIRED article from

April 27, 2018 put it (Molteni, 2018b). In short, economic development and competitiveness focused articles on the potential financial boon of investing or developing Crispr as a lifechanging therapeutic technique or groundbreaking healthcare option.

Conflict/strategy The conflict/strategy frame was present in only 47 articles (15.3%). Journalists frequently used this frame in reference to the patent battle between the University of California, Berkeley/Jennifer Doudna and the Broad Institute/MIT/Feng Zhang. Journalists described the two institutions as “pitted against one another” (Achenbach & Johnson, 2017), as engaged in an “epic patent fight” (Molteni, 2018), and as entangled in an “especially gnarly” case (Greenberg, 2016). The conflict inherent in the patent case was also frequently mentioned in tandem with economic development and competitiveness, as the winner of the case stood to make money on licensing Crispr technology for use by biotechnology companies. As WIRED put it in a

41 February 4, 2016 article, “…investors continue to flood biotech with money. The potential rewards for something like a cure for blindness are too great to ignore” (Greenberg, 2016). There were two interesting exceptions to this pattern of conflict/strategy framing. In the first, a WIRED article from July 10, 2017 described the scientific community as at odds with the government, particularly as genetic editing relates to biohacking and security. Scientists “…fear what might happen to important genetic research if the Pentagon gets too paranoid,” presumably enacting regulations that hinder scientific research in an effort to prevent bioterrorism (Niller, 2017). In a second exception, Popular Science published an article on January 30, 2018 that discussed the “race” between Chinese and American scientists to create innovative healthcare applications for Crispr, pointing out that clinical trials and FDA regulation might mean that the United States falls behind; “Unlike China, the United States has a far more painstaking and demanding system to bring a new drug to market” (Maldarelli, 2018). In both cases, this interpretation of events pits two parties against one another.

Middle way/alternative path The middle way/alternative path frame occurred least frequently of all, appearing in only 20 articles (6.5%). There were two types of instances when this frame was used: 1) in reference to Crispr as a means of genetically modifying plants for food; and 2) when discussing alternatives to human germline editing as a means of combatting disease. In the first instance, an August, 2016 National Geographic article explained that, “…the definitions of the word ‘transgenic’ and the term ‘GMO’ have been based on the practice of combining in the laboratory the DNA of species that

42 could never mate in nature. But scientists hope that using CRISPR to alter DNA could appease the opposition” (Specter, 2016). Since Crispr does not introduce DNA from another species into a plant undergoing gene editing, from a regulatory perspective it is seen as an accelerated means of traditional plant breeding, a la Mendel. The GMO issue is the one where Crispr was most often touted as a moderate solution to blatant gene editing or recombinant DNA. In the second set of instances when the alternative path frame was used, individuals arguing against human germline editing cited pre-implantation genetic diagnosis, or PGD, as a means of screening embryos for disease without modifying their genes. For example, one ethicist wrote in a National Geographic piece from the

August, 2016 issue that, “screening embryos for disease is far safer for resulting children than engineering new traits with germline editing would be. Yet this existing alternative is often omitted from accounts of the controversy about gene editing for reproduction” (Harris & Darnovsky, 2016). PGD as an alternative to human germline editing can only be applied in instances when both parents are not carriers for a genetic disease; Crispr proponents cited this as a shortcoming of the technique (Tucker, 2017; Yong, 2015; Yong, 2017).

Research Question 2: Sources Used by Journalists My second research question focused on identifying the primary sources of information in news articles about Crispr. In the sample of 308 articles, 1708 sources were mentioned (see table 3.2). Of these sources, 658 (38.5%) were organizations or groups rather than individuals. For example, the Associated Press was often quoted in relation to the birth of the Chinese twins whose genes were purportedly Crispr’d by He Jianku in 2018. Of the 1050 individuals cited as sources, 712 (67.8%) were men,

43 241 (23.0%) were women, and 97 (9.2%) were not referred to by a pronoun indicating sex. Of the 97 individuals whose sex was not mentioned, an internet search of their organizational affiliation and name revealed that a further 28 were women, leaving a total of 269 (25.6%) women and 777 (74.0%) men. The sex of four (.4%) individuals could not be determined from the information given. The majority of individuals – 704, or 67% – used as sources were scientists or researchers. Another 74 (7.0%) individuals were legal experts, 65 (6.2%) were ethicists, 109 (10.4%) were businesspeople or representatives of businesses, 64 (6.1%) were journalists or reporters, and 26 (2.5%) were not identified by occupation or fit in none of the other categories. Of the 658 different organizations and groups quoted, 271 (41.2%) represented groups of scientists or scientific organizations, 257 (39.1%) represented businesses or non-profit organizations, 47 (7.1%) represented news organizations quoted in other articles, 31 (4.7%) represented scientific journals quoted as sources, 23 (3.5%) represented ethics-related organizations, 16 (2.4%) represented legal groups, and 13 (2.0%) fit in none of these categories. Some of the most commonly mentioned organizations were research institutions and universities, notably MIT’s Broad Institute and the University of California, Berkeley. Businesses used as sources were primarily involved in biotechnology, like the “Big 3” Crispr companies: Caribou Biosciences, Crispr Therapeutics, and Intellia Therapeutics. Other oft-cited organizations included the National Institutes of Health, the National Academies of Sciences, Engineering, and Medicine, and the United States Department of Agriculture. Journalists also frequently

44 referred to groups of scientists in the broadest sense, often just as “researchers” or “scientists.” Here, journalists made generalizations about the thoughts of a broad group of scientists without naming individuals or universities in particular. Likewise, some journalists also cited other articles published in newspapers and science magazines rather than seeking out original sources for comment. Finally, journalists often discussed science journals, such as the academic, peer-reviewed journal Nature, as sources of information about recent findings. In short, the use of sources by journalists did display a great deal of breadth, but the most common sources were scientists or researchers themselves, with other individuals and organizations used to flesh out the narrative of a news story.

Research Question 3: Hyped Headlines My third research question focused on the ways that news headlines hype Crispr. Of the 308 total articles, 124 (40.3%) had headlines that included hyped words or phrases. Some of the most common words used in hyped headlines are illustrated through the word cloud in Figure 2, where the size of the words indicates how frequently they occurred in newspaper headlines that included hype. Some of the most frequently used emotionally charged words were “battle,” “revolution,” “superpowers,” “terrifying,” “fight,” and “outrage.”

45 Figure 2 Frequently Used Words in Hyped Headlines

Research Question 4: Replication Statements and Models The study’s fourth research question asked whether issues of replication are discussed in articles related to Crispr. Of the 308 total articles, 228 (74.0%) included replication statements. Of these 228 articles, 223 (97.8%) mentioned that the findings were initial rather than replicated. In one example of a replication statement, a May 5,

2017 WIRED stated, “There’s no evidence their trial made any off-target cuts. But they’ll need to run more experiments to make sure that’s absolutely the case, probably using primate models, since their DNA is closer to humans” (Stockton, 2017). Similarly, 217 (70.5%) of the articles included the model used in the experiment. Fully 113 (52.1%) mentioned a human or human cell model, 73 (33.6%) mentioned an animal model, 18 (8.3%) used a plant model, 12 (5.5%) mentioned

46 bacterial models, and a single article (.5%) mentioned a viral model. The high proportion of animal models present in the sample corresponds to the initial, cutting- edge nature of Crispr research.

Research Question 5: Speed of Clinical Translation My fifth research question asked whether the speed of clinical translation was mentioned in articles about Crispr’s health- and medical-related applications. Providing tentative dates audiences can expect a medical technology to be available to the general public gives individuals context for research findings, placing them in the realm of the possible rather than the far-fetched. Of the 308 total articles, 125 (40.6%) had a specific healthcare focus, and of these, 83 (66.4%) mentioned a specific time to clinical translation for a healthcare application. For example, a September 10, 2015 Popular Science article quoted a doctor who said, “he’d be surprised if it took more than five or ten years for CRISPR to be ready, technically speaking, for clinical research” (Yin, 2015). Likewise, a November 5, 2015 Popular Science article mentioned, “In a bid to be the first of at least a dozen companies to use CRISPR in a clinical setting, the company intends to test its treatment on patients before 2017” (Ossola, 2015). Most references to specific clinical translation time used rough estimates of time periods, since approval for clinical trials is a lengthy process run through the FDA in the United States and regulated in Europe. In an instance where clinical trials seemed imminent, WIRED wrote on December 11, 2017 that the French biologist Emmanuelle Charpentier and her team had applied for permission to begin a clinical trial in Europe for a treatment of beta thalassemia (Molteni, 2017). Though no specific date for clinical application was given, the submission of an application implies a

47 concrete, terminal bid to begin a clinical trial. This case exemplifies the difficulty in accurately reporting a time period for individuals to expect a new “cure” or disease treatment – often, the scientists and physicians themselves do not know when clinical trials will be completed and a practice will be deemed safe.

Research Question 6: Comparing Newspapers and Science Outlets My sixth and final research question sought to compare the two types of news coverage of Crispr across all of the measured variables, including frames, sources used, hype, replication statements, and mentions of clinical translation speed. Since these two types of publications have different target audiences and dedicated interests (general news versus science- and technology-centric reporting), comparing them across variables will provide insight into their different approaches for constructing narratives about Crispr.

Common biotechnology frames

To compare frame use in newspapers versus science-oriented publications, chi- square tests for independence were conducted. The results are outlined in Table 3.2. For the social progress frame, a chi-square test indicated a significant difference between newspapers and science-oriented publications, c2 (1, n = 308) = 12.50, p <.01. That is, science-oriented publications were more likely than newspapers to use the social progress frame. For economic development and competitiveness frames, a chi square test indicated a significant difference between the types of publications, c2 (1, n = 308) = 3.92, p = .05. Here, science-oriented publications were more likely than newspapers to talk about Crispr in terms of economic development and

48 Table 3 Summary Table: Comparing Frame Use in Newspapers and Science Magazines

News Science Frame Present Absent Present Absent Chi Signifi- Square cance Social Progress 104 17 181 6 12.495 <.01* (86.0%) (14.0%) (96.8%) (3.2%) Economic Development/ 45 76 91 96 3.922 .05* Competitiveness (37.2%) (62.8%) (48.7%) (51.3%) Morality/Ethics 65 56 106 81 0.262 .61 (53.7%) (46.3%) (56.7%) (43.3%) Scientific/Technical 55 66 134 53 21.275 <.01* Uncertainty (45.5%) (54.5%) (71.7%) (28.3%) Pandora’s 78 43 94 93 6.004 .01* Box/Frankenstein’s (64.5%) (35.5%) (50.3%) (49.7%) Monster/Runaway Public Accountability/ 72 49 116 71 .197 .66 Governance (59.5%) (40.5%) (62.0%) (38.0%) Middle Way/ 4 (3.3%) 117 16 171 3.335 .07 Alternative Path (96.7%) (8.6%) (91.4%) Conflict/Strategy 17 104 30 157 .226 .64 (14.0%) (86.0%) (16.0%) (84.0%) Note: * indicates significant p values lower than .05

competitiveness. A chi-square test comparing the use of the morality and ethics frame was not significant, c2 (1, n = 308) = 0.26, p = .61. Comparing the results for the use of the scientific/technical uncertainty frame, though, did indicate a significant difference, c2 (1, n = 308) = 21.28, p < .01. Here, science-oriented publications were more likely to frame Crispr using mentions of uncertainty. There was also a significant difference in the use of the Pandora’s box/Frankenstein’s monster/runaway frame, c2 (1, n = 308) = 6.00, p = .01. News articles were more likely to use this frame than articles from science-oriented publications. There was no significant difference between the two types of articles for the public accountability/governance frame, c2 (1, n = 308) = 0.20, p = .66, the middle way/alternative path frame, c2 (1, n = 308) = 3.34, p = .07, or the conflict/strategy frame, c2 (1, n = 308) = .23, p = .64.

49 Comparison of sources used by journalists To compare the use of sources by the two different types of publications, two separate chi square analyses were performed (see Table 4). Looking at the sex of the source as mentioned in the article, there was no significant difference between the use of male and female sources by news publications versus science-oriented publications, c2 (2, n = 1050) = 3.85, p = .15. However, once the sex of the sources whose pronouns were not used in the context of the article was determined, there was a significant difference in the use of male versus female sources in science-oriented articles compared to news articles, c2 (1, n = 1046) = 5.34, p = .02. Specifically, science- oriented articles were more likely to cite female sources compared to news publications, though male sources were still cited a majority of the time.

Table 4 Comparing Source Use Between Newspapers and Science Magazines

News Science Male Female Not Male Female Not Chi Signifi- Mentioned Mentioned Square cance Sex of source 287 80 37 425 161 60 3.85 .15 mentioned (71.0%) (19.8%) (9.2%) (65.8%) (24.9%) (9.3%) Sex of source 313 87 n/a 464 182 n/a 5.34 .02* (78.3%) (21.8%) (71.8%) (28.2%) Note: * indicates significant p values lower than .05

Comparison of hyped headlines and time to clinical translation

As for hyped headlines, a chi square analysis discovered no significant difference in the use of hype when comparing news and science-oriented publications, c2 (1, n = 308) = 1.26, p = .26. That is, neither type of outlet was more likely to use hyped words or phrases in their headlines. There was also no significant difference in whether replication statements were present, c2 (1, n = 308) = .14, p = .70. In cases where a replication statement was present, there were not any differences in whether

50 journalists mentioned if results were initial or replicated, c2 (1, n = 228) = .00, p = .99. That is, neither newspapers nor science-oriented outlets was more likely to report initial findings. Likewise, there was no significant difference between the two types of articles in whether they mentioned the model used for the research study, c2 (1, n = 308) = .20, p = .66. Furthermore, there was no significant difference in whether science or news publication articles were more likely to be related to healthcare applications, c2 (1, n = 308) = .55, p = .15. However, there was a significant difference in whether articles that were related to healthcare applications mentioned the time to clinical application, c2 (1, n = 125) = 8.57, p < .01. Here, newspapers were more likely than science-oriented outlets to mention a discrete time to clinical translation.

Conclusion A census of 308 articles from 12 different newspapers and science magazines indicated that Crispr is, indeed, a topic of discussion in several spheres: agriculture, healthcare and medicine, and business development. Sources used to discuss Crispr findings varied but were primarily scientists and researchers. Men were more likely to be used as sources than women, but science magazines were more likely than newspapers to use female sources. Headlines were hyped less than half of the time, indicating a bit of tempering, rather than overwhelming inflation, of expectations about the technology’s capabilities. Most articles indicated that findings by scientists were initial and in need of replication, and the model used was usually human or human cells, though animal models were a close second. Finally, the time to clinical application was mentioned in the majority of articles that had a healthcare focus.

51 Comparing science magazines and news articles indicated some differences in the use of frames. Science magazines were more likely than newspapers to frame Crispr in terms of social progress, economic development and competitiveness, and scientific/technical uncertainty. Meanwhile, newspapers were more likely to frame Crispr as a runaway technology, while neither type of outlet was more likely to use hyped headlines. Newspapers were also more likely to indicate a discrete time to clinical translation than science magazines. In short, a content analysis of a census of news articles from six different newspapers and science magazines has uncovered differences in the way that outlets geared toward science-minded and general audiences discuss Crispr. This powerful technology will likely have widespread effects on our way of life, from healthcare applications to agricultural developments and environmental changes, and news framing of the issue may affect audiences differently based on individual media consumption. Understanding how Crispr is discussed in media provides deep insight into the way the technology is perceived by journalists. The narrative created by journalists may then influence public perceptions of the technology and later affect public policy outcomes and technological development. As this research covers only the preliminary developments in this nascent biotechnology, however, discussion of

Crispr may change as it moves from basic, initial research to clinically tested, widely available and developed applications.

52 Chapter 4

DISCUSSION

This study contributed to existing science communication literature by examining, in-depth, how journalists are framing and discussing an up-and-coming biotechnology with applications in agriculture, medicine, and environmental ecology. By analyzing a census of Crispr articles from six top newspapers and six top science magazines, this study collected data on frame presence, sources used, hyped headlines, replication statements of scientific findings, and mentions of the speed of clinical application. Given recent developments and continued coverage of Crispr (for example, the March 2019 cover of WIRED featured the tech), these findings add depth to our understanding of its presentation in the media, and potentially its effects on audiences.

Frame Use and Effects in Crispr Coverage By examining common frames journalists used to discuss Crispr, my aims were twofold: first, to understand which frames are used across news outlets; and second, to examine how newspapers and science-oriented magazines differ in their use of frames. The study’s census, drawing from the newspapers and science magazines with the largest number of subscribers in the United States, gives a glimpse into the ways journalists construct narratives about Crispr. Moving frame by frame, I will discuss which frames were commonly used in articles, how newspapers and science magazines used frames differently, and why journalists may have used frames in certain ways. Furthermore, I will speculate about the potential effects of these frames on audiences who consume news media.

53 Social progress The social progress frame was present in an overwhelming 92.5% of all articles. Journalists likely used social progress as a means to make a complex technology relevant to the audience, who may or may not be familiar with gene editing. By leading with a catchy line about the potential innovations and problem- solving capabilities of a complicated biotechnology, journalists give readers something to latch onto and create a focal point for the article. The problem-solving emphasis placed on Crispr may serve as a cue, giving readers a frame of reference for the technology as an innovation. The positive spin and ubiquitous presence of the social progress frames is consistent with Nisbet and Lewenstein’s (2002) finding that American coverage of biotechnology is overwhelmingly positive – even in articles critiquing the potential for Crispr to be used as a bioweapon or to make designer babies, journalists still mentioned its positive outcomes for agriculture, medicine, and therapies. The analysis comparing the use of frames across outlets showed that science magazines were significantly more likely than newspapers to use the social progress frame – indeed, it was absent from only 6 science magazine articles. Further research could elucidate where this difference arises. It may be that science magazines are more attuned to the ways the technology could be applied and help solve problems in areas as diverse as agriculture and healthcare. Conversely, news coverage may tend to present the technology as more dangerous and negatively hyped. However, one must remember that the vast majority of articles from both outlets enumerated the potential problem-solving capabilities of Crispr. The overwhelming presence of the social progress frame may also have effects on audiences, particularly as Crispr moves from the development stage into clinical

54 applications. Given that Crispr is so frequently discussed as a technology capable of solving problems, there may be a positive relationship between the use of the social progress frame by journalists and positive public perceptions. The emphasis on problem-solving and progress in the media may also affect upcoming legislation in a more direct way. As Brossard and Nisbet (2006) wrote, mass media functions as a “convenience sample” for policy makers when making policy decisions – that is, legislators often look to media presentation of an issue to gauge public support (p. 26). Crispr’s positive presentation in news media may encourage legislators to view the tech favorably, as well.

Scientific/technical uncertainty The second most common frame, scientific/technical uncertainty, was present in 61.4% of all articles coded. This finding may be heartening, because it shows that most journalists understand – or are at least conscious of – the issues with claiming a technology as fully developed and completely certain. In reference to Crispr, uncertainty was generally framed as a hurdle to overcome – that is, it was still mentioned as a positive means of solving problems related to DNA, but with more research, development, and testing. Notably, the scientific/technical uncertainty frame is often used to cast doubt on the (definitive) existence of global climate change (Nisbet, 2009b). Here, the frame’s use by journalists as a means to increase credibility stands as a counterpoint, indicating the importance of valence in frame use. This frame was used in one of two ways: journalists either discussed “what is known/unknown” or portrayed complicated issues as “a matter of expert understanding” (Nisbet, 2009b). Though it may sound counterproductive to mention what scientists still do not know, developing a technology requires a lot of trial and

55 error; by mentioning that even scientists are not sure about all the ways Crispr works or how it can be used gives the public an accurate presentation of the scientific process and all its flaws. For example, an August 26, 105 Popular Science article observed that, “CRISPR is a powerful tool that scientists don’t fully understand” (Ossola, 2015b). When comparing newspapers and science magazines, the data indicated that science magazines were more likely to discuss Crispr via scientific/technical uncertainty frames. This could be related to both the knowledge of science journalists writing for specialized magazines and the amount of time and attention they have to dedicate to writing science articles. One of the major pressures facing journalists is creating content as quickly as possible, and it could be that journalists for monthly publications with online components are not pushed to create content as quickly as individuals working for daily newspapers, which includes all six of the surveyed news outlets: the New York Times, Washington Post, LA Times, New York Post, USA Today, and Wall Street Journal. Though presenting a technology as uncertain may seem, at first, likely to increase public distrust of both the tech and scientists, research by Fiske and Dupree (2014) demonstrated that public trust in scientists as communicators relies on perceptions of their truthfulness. That is, “Audiences judge the communicator’s apparent trustworthiness as the communicator’s inferred motivation to be truthful” (p. 13594). As such, scientists who do not shy away from communicating their uncertainty related to Crispr may be seen as more trustworthy than individuals who hype Crispr’s potential or claim that it is perfectly understood, since any attentive reader would realize that complete faith in a brand-new technology would not be a

56 truthful stance from a scientist at this point in the research development. Given that scientists are used as sources for much of the information related to scientific/technical uncertainty framing, one can infer that this heuristic of trust applies to Crispr articles. However, further research would be required to see whether trust in scientists in the face of uncertainty applies to scientific/technical uncertainty framing used by journalists.

Public accountability/governance The public accountability/governance frame was used in 61% of the collected articles. Journalists frequently introduced this frame through comments from ethicists and scientists. Using a credible source as a means of calling for regulation helps preclude any potential bias from the shoulders of the journalist writing the piece -- that is, by using poignant quotes from trusted sources, like Crispr co-inventor Jennifer Doudna, or prominent bioethicist Alta Charo, journalists can present the call for regulation as a trusted source’s opinion. In reality, calling for regulation is a political move, even via a source’s quoted words. This use of sources as framing mechanisms was present in several notable cases. For example, one anecdote that preceded calls for regulation was Jennifer Doudna’s dream about explaining Crispr to Hitler. This dream featured in the Washington Post, where Doudna is described as having, “…a particularly awful dream…in which her colleague wanted her to explain CRISPR to a funny little man with a mustache – a man she suddenly realized was Adolf Hitler” (Achenbach, 2016). Dr. Doudna’s dream was also recounted in the LA Times (Hiltzik, 2017) and the Wall Street Journal, though the Journal mentioned that, “…it was Hitler, but with the face of a pig” (Peterson, 2017). The rest of the Washington Post article discussed, in detail,

57 how scientists are attempting to self-regulate the technology via talks at the First International Summit on Human Genome Editing (2016). This particular example illustrates the (sometimes dramatic) means journalists used to introduce the need for governance of Crispr. Calls for public accountability (on the part of scientists) and governance (both internally by scientists and externally via government departments such as the USDA, FDA, and NIH) often referenced the potential negative outcomes of unregulated Crispr use. Take the veiled eugenics concerns in Doudna’s stress dream about Hitler, among other references to class disparities in access to Crispr therapies, potential changes to the human genome through germline editing, and total environmental disruption through organismal editing and gene drive. However, when comparing newspaper and science magazine articles, there was no significant difference in their use of public accountability/governance frames. This could be a product of the unilateral concern about regulation related to Crispr, from journalists and scientists alike.

Runaway science The Pandora’s Box/Frankenstein’s Monster/Runaway Science frame was used by journalists in 55.8% of articles analyzed. This frame was typified by “a call for precaution in face of possible impacts or catastrophe” or any time Crispr was discussed as a technology that would soon be out of control (Nisbet, 2009b). The runaway science frame often followed on the heels of the social progress frame, usually in the first paragraph or two. The use of this frame to temper the unbridled enthusiasm for Crispr embodied in the social progress narrative seemed to be an attempt by journalists to be objective. The norm of objectivity in journalism has, in the

58 past, led journalists to pit scientific sources against clergy members, and global warming proponents against skeptics in an effort to “balance” the content of the article, resulting in “balanced hype” (Jensen, 2012). The pressure on journalists to be objective results in narratives that do not reflect the actual ratio of support or dissent present in a population, whether in political or ethical debates. The runaway science frame also featured some colorful imagery on the part of journalists; it appealed to archetypes such as Frankenstein and Pandora, and painted scenes of slippery slopes. For example, a December 2, 2015 Atlantic article proclaimed, “With this power at hand, old questions about playing God, making designer babies, and ushering in dystopian Brave New Worlds of genetic haves and have-nots, take on fresh urgency” (Yong, 2015). These evocative images point one’s thoughts immediately toward the possible outcomes of unchecked scientific experimentation with gene editing via Crispr. The results indicated that newspapers were more likely than science magazines to use the runaway science frame. This could be an effort to incite audience connections with the complicated topic by instilling a sense of urgency, or could, again, be an attempt to be objective, resulting in “balanced hype” (Jensen, 2012). Further research into the prevalence of social progress frames followed by runaway science frames could lend more weight to the hypothesis that newspapers are more likely than science magazines to use balanced hype. Furthermore, balanced hype may affect the audience by reminding them of the potential negative outcomes of Crispr technology, in spite of its benefits. Understanding how “coupled” frames influence audiences could more clearly demonstrate the relationship between these two frames and their effects on the audience.

59 Morality/ethics The morality/ethics frame was present in 55.5% of the coded articles. Moral and ethical judgments about Crispr’s use were not usually found in a journalist’s own words. Rather, journalists often pitched this frame through quotations from sources. Using a scientist or bioethicist’s words to convey moral/ethical judgments may be an effort by the journalist either to appear as an objective reporter, or to appeal to authority. To process information efficiently, individuals rely on heuristics. Sundar (2008) argues that, “the authority heuristic is likely to be operational whenever a topic expert or official authority is identified as the source of content” (p. 84). Identifying bioethicists and scientists as information sources may lend greater credibility to a moral or ethical framework relating to Crispr. By using direct quotes, journalists appeal to authority and avoid perceived bias in their original writings. Some articles were even constructed as pseudo-conversations between bioethicists and scientists with opposing views on the morality of Crispr use. For example, an August 6, 2015 Popular Science article first quoted an interview with psychologist Steven Pinker, who said, “CRISPR is key to reducing human suffering, …and bioethicists should strive to ‘get out of the way’ of progress” (Ossola, 2015a). The journalist then used a combination of direct quotes from bioethicists and generalized statements from bioethicists as a whole to indicate their thoughts. Unfortunately, the article became less of an argument about the morality of Crispr use, and more of a justification for bioethicist participation in scientific conversation. The morality/ethics frame was also present in articles focused entirely on moral issues. Some headlines illustrate this emphasis: for example, the New York Times article “Ethical Divide for China and West” (Tatlow, 2015); the USA Today article “Profound ethical questions as it appears eugenics is upon us: Get ready for

60 CRISPR babies” (Tilley, 2018); the WIRED article “Crispr is getting better. Now it’s time to ask the hard ethical questions” (Zhang, 2015); and the Popular Science article “The gene-edited Chinese twins represent a multi-generational ethical quandary” (Eschner, 2018). Interestingly, there was no significant difference in how often science magazines and news outlets used the morality/ethics frame. Since moral and ethical considerations are inherently opinion-based, the journalistic norm of objectivity may be keeping journalists from providing hard-and-fast judgments about the morality of Crispr’s use. Furthermore, the overwhelming use of scientists as sources of information may be swaying journalists to present Crispr in terms of social progress and scientific uncertainty. Nisbet and Lewenstein (2002) found that scientists framed the rDNA debates of the 1970s in reference to technical risks, without mention of broader ethical and social implications. Scientists may be working to present their research in the best possible light while also staying away from moral judgments or ethical imperatives. In short, the presence of this frame across outlets indicates a general preoccupation with the ethics of gene-editing, from science-savvy journalists to classic news reporters. Using morality and ethics in headlines may make the frame more influential on audiences. Perhaps the repeated exposure to the idea of ethical issues with Crispr, using the cue of the word “ethics” or “ethical,” may influence audiences to view the technology as morally ambiguous. However, further research is needed to understand how this emphasis affects audiences.

Economic development/competitiveness The economic development/competitiveness frame was present in 44.2% of articles. This frame focused on financial gains for either individuals or countries and

61 may have been a product of the heavy use of scientists as sources. There are pressures on scientists to make their research relevant and interesting to the public, given that public support for science translates to increased funding for scientists (Caulfield & Condit, 2012). The pressure on scientists to present their research in a positive light may then translate into media framing of Crispr – if scientists mention the potential benefits Crispr may have for economic development, that frame may then become a part of the Crispr narrative presented by journalists. It follows, then, that the results of this study indicated that scientific/technical uncertainty frames often referred to the developing biotechnology and therapeutic industry using Crispr as a means to create healthcare interventions.

Interestingly, this frame was more likely to be used by journalists writing for science magazines than by newspaper reporters. This could be for a variety of reasons. Journalists for science magazines are writing for an audience that is likely more interested and invested in research outcomes. Giving details about the impact research findings will have on the economy, development of new medicines and therapies, and on research itself may be of more interest to this specialized audience. Since science magazines were more likely to use the social progress and scientific and technical uncertainty frames, too, it could be that science journalists are working to provide the broadest view of the technology, from its testing to its applications and eventual economic benefits. Another aspect of economic development/competitiveness frame use is the potential monetary ties influencing journalists to mention economic gain. Further research would be required to see who supports each magazine and whether they are

62 financially invested in the therapeutic developers who stand to gain the most if their interventions become widespread after clinical testing.

Conflict/strategy The conflict/strategy frame, as defined by Nisbet (2009b), creates a narrative involving “a game among elites; who’s ahead or behind in a winning debate; a battle of personalities.” This frame was used in 15.3% of articles, and there was no significant difference in its use by science magazines and newspapers. Journalists used the frame most often in reference to the patent case between Feng Zhang and Jennifer Doudna for control of rights to Crispr’s use in mammalian models. This case did have high stakes – the winner controls who gets to use the specific enzyme that works well in human DNA. However, covering the details of a patent case is less than riveting bedtime reading without a strong narrative structure. Creating a conflict frame to pit Doudna and Zhang gives the story a bit more zip. In other cases, when journalists referred to conflict between the government and scientists (Niller, 2017) or as a race between Chinese scientists and Americans for the next big innovation (Maldarelli, 2018), tension between two groups creates a story where, perhaps, no tension really exists. Due to its limited presence and clearly defined focus among articles on Crispr, it is difficult to anticipate how this frame may affect audience views of the technology. Perhaps it would encourage some to support one individual over the other based on journalist portrayals of actors involved. However, further research into the way conflict/strategy frames influences audiences would increase our understanding of its potential effects.

63 Overall Frame Use and Audience Effects As Crispr moves from its adolescence, filled with exploration and wild speculation, into its adulthood as a full-fledged, commonplace biotechnology, framing within news coverage provides a window into both media and population-level views of its importance and viability. Here, the results indicate that framing is, indeed, varied based on intended audience, and that one frame in particular is common to most narratives of Crispr: Social progress. Social progress emphasizes Crispr’s capacity to solve problems, from treating disease to creating better crops and banishing pests. This frame is common in the media, and Scheufele (1999) demonstrated that the most common frames are often seen as the most accessible to both journalists and audiences – they provide the easiest cues to create a familiar narrative that draws in the consumer. This accessibility may lead audience members and journalists to latch on to the frames in a kind of closed loop, where the most common frames become seen as the most accessible, and the most accessible become the most common (Scheufele, 1999). In the case of the social progress frame, its ubiquity throughout Crispr coverage over three years-worth of news samples shows its staying power. This frame has likely become the dominant way audiences view Crispr – as a kind of semi-magical means of editing genes, as easy as cutting and pasting in a word document. Unfortunately, this heuristic does not convey Crispr’s nuance or difficulties. However, the well-documented presence of the scientific/technical uncertainty frame in this sample could indicate one of two things: either journalists are doing their best to prevent the setbacks of unbridled optimism, or they are undermining the technology’s potential and progress by referencing its current problems and the lack of scientist understanding of the human genome itself. Being open about the uncertainty involved in developing technology prevents the

64 public from expecting too much, too soon, though it may also convey that scientists do not fully understand the inner workings of Crispr as a technology. The interaction of these two frames on audience opinion is a wide-open area of research. For science-oriented outlets, the combination of social progress and scientific/technical uncertainty frames bears a striking resemblance to the proposed “groundbreaking” frame recommended by Mariscal and Petropanagos (2018) as the best way for scientists to frame Crispr in their own academic articles. The “groundbreaking” frame highlights the benefits of Crispr while still recommending a watchful gaze is applied as the technology is developed and regulated. The findings in this study indicated that science magazines were more likely to frame Crispr in terms of social progress, economic development and competitiveness, and scientific/technical uncertainty. These three frames, taken together, almost perfectly encompass the ideas presented by Mariscal and Petropanagos. Understanding why these three frames are more likely to be used by science magazine journalists than newspaper journalists requires some speculation. Surveying journalists and logging their credentials may indicate that more science magazine writers have scientific training or education. It could be that journalists covering biotechnology beats for science magazines have a deeper understanding of the nuance involved in scientific discovery, from its potential to solve myriad problems, to its economic boons, to its need for replication and testing, testing, testing. Conversely, it may also be that science magazines are less likely to take a critical stance toward scientists due to their close and constant interaction. Dudo (2015) cites press embargos (allowing certain journalists to view scientific articles before their publication), press conferences, and press releases as means for scientists to control what information

65 journalists publish. Doing so may allow scientists to increase their visibility, potentially increasing their sources of funding (Nelkin, 1995). Journalists may take the views presented by scientists at face value, thus conveying the information scientists present without further examination. That science-oriented magazines so closely aligned with framing recommended for scientists, by scientists, is a finding that requires further research to understand; however, discovering what is at the root of the framing alignment between journalists and scientists would help elucidate their complicated give-and-take relationship. Understanding who is chosen by journalists to speak about Crispr on the platform created by top newspapers and science magazines provides a glimpse into which voices journalists see as important. Journalists select sources that increase their perceived credibility – that is, the credentials of the source then increase both the derived credibility of the journalist and the credibility of the source information (Berkowitz & Beach, 1993; Nisbet, 2009a). For sources, on the other hand, gaining a platform to present one’s work in one’s own words is a valuable form of control over public image and perception (Dunwoody, 2014). As such, there are conflicting motivations between journalists, who seek sources in order to meet the norm of journalistic objectivity and develop credibility, while sources themselves are thinking of ways to present their work in the best possible light. Here, data indicated that scientists were far and away the most quoted sources by journalists covering Crispr. Of all individuals quoted, 67% were scientists or researchers, and the second most represented group was legal experts, with only 7% of individuals quoted. For an up-and-coming technology, the way it is perceived by the broader public has massive implications for scientists and their future work. As

66 Dunwoody (2014) put it, scientists seek to cast their research in a favorable light in order to increase their credibility and reaffirm their importance. Upcoming regulation in regard to Crispr could potentially leave scientists in the cold; if basic research using human embryos is suspended, or clinical trials using human models is not approved by regulatory bodies, all Crispr research related to human health and therapeutics could be stopped – and with it, the massive influx of money to biotechnology startups. Even groups used as sources by journalists were primarily groups of scientists or scientific organizations (41.2% of the sample). Furthermore, Nisbet and Lewenstein (2002) posited that scientists use pre- packaged frames that present their research in the best positive light, and that these frames influence overall coverage of biotechnology. This was evident in the findings of this study – not only are scientists the primary sources of information on Crispr, but the most prominent frame, social progress, also had an overwhelmingly positive slant. As for the sex of sources mentioned, 67.8% were male and 23.0% were female, with 9.2% referred to without a pronoun indicating sex. This ratio of male to female pronouns is actually more representative of women compared to other studies. In one report examining the ratio of pronouns He and She in a sample of 80,000 Reuters messages indicated that He pronouns are used 9 times more than She pronouns

(Sendén, Sikström & Lindholm, 2014). This heightened (though by no means equal) representation of female sources could be due to the active role Jennifer Doudna has chosen to take in spearheading discussions about the moral and ethical implications of Crispr, never mind being one of the three primary discoverers of the technology. Doudna alone was quoted as a direct source in 55 separate articles, making up about a fifth of the total female sources. Doudna’s active role in Crispr outreach and her

67 appointment of a female CEO for her biotechnology company, Caribou Biosciences, may be a big step forward for representation of women, both in the news and in the sciences. When comparing science magazine and newspaper articles, there was no significant difference in the use of male and female sources by sex. However, once sources whose sex was not mentioned were added to the sample, there was a significant difference between the two types of outlets: science magazines were more likely to use female sources. This may indicate a willingness to quote female scientists more often in science magazines. However, there may be another reason that begs further investigation. It could be that there are more female science journalists working beats in science magazines than working as newspaper reporters. Examining whether female journalists quote female sources more often would be a worthy further study. It would also be enlightening to see whether audiences view female scientists as more or less credible than their male counterparts.

Hyped Headlines Journalists use hyped headlines as a means of gaining audience attention; these headlines may draw broad generalizations and create causal attributions where none may, in fact, exist (Dumas-Mallet, Smith, Boraud & Gonon, 2018). There is pressure on journalists to grab attention while, at the same time, conveying accurate news. This pressure is seen in the data: 40.3% of headlines from the sample were classified as hyped. These headlines included emotionally charged words and/or overt generalizations of Crispr’s potential as a cure for disease. Interestingly, many of the hyped headlines had to do with the patent battle between Feng Zhang (of the Broad Institute and MIT) and Jennifer Doudna (of the

68 University of California, Berkeley) for rights to Crispr’s use for human interventions and therapeutics. These headlines often used words such as “battle,” “fight,” and “outrage.” Notably, the conflict/strategy frame was almost exclusively used in reference to the patent battle, and perhaps journalists felt pressure to create a strong narrative about the patent case (a dry subject without high stakes hype). Hype was also used in reference to Crispr’s capabilities as a gene editing technique. Words such as “terrifying,” “superpowers,” “revolution,” “eugenics,” “cure” and “promise” indicate the tendency for journalists to use emotional appeals to create interest in a broader story. Since prior research has indicated that six in ten shared articles on social media have not been clicked on (and presumably, read) by the person who shared them, the persistence of hype to frame science news is concerning (Gabielkov et al., 2016). Furthermore, many headlines were not classified as hyped due to the presence of one simple word: “may.” Take WIRED’s June 8, 2017 article titled “Crispr may cure all genetic disease – one day” (Molteni, 2017), or the Los Angeles Times article from December 7, 2017 titled “Safer DNA editing may fix human diseases, Salk scientists report” (Fikes, 2017). Using this modifier to dampen expectations and remove total generalization is a clever way to attract attention without overt hype.

There was no significant difference between science magazines and newspapers in tendency to use hype. This demonstrates the pressure on journalists across the board to come up with catchy headlines and appeals for audiences to read their work.

69 Replication Statements, Models, and Speed of Clinical Translation Replication is an important part of the scientific process: it allows scientists to see, over time, whether their assumptions hold true across a variety of situations, samples, and variables. However, research by Gonon et al. (2012) found that newspapers have a tendency to report initial findings – those exciting breakthroughs that have the potential to revolutionize an area of science or medicine – but not the altogether less-exciting details about whether those research endeavors hold up over time. Here, the data indicated that 74% of journalists writing about Crispr included a replication statement in the body of the article. That is, they clarified whether this research finding is initial or has been replicated over time. As was expected, 97.8% of those articles indicated that research findings were initial. Given Crispr’s youth as a technology, this finding is consistent with the state of its development. Finding that roughly three-fourths of all articles about Crispr mention its need for further and refined research may work as something of a counter to hype of its potential to solve problems or cure diseases. There was also no difference between science and news articles in the tendency to provide replication statements, nor any difference in mention of findings as initial or replicated. Overall, this indicates an attention to detail on the part of journalists: most are including statements about the need for further research to refine this technology, and that may have positive effects on audiences by tempering hype related to the tech. In articles specifically discussing a research finding, 70.5% included the model used to test Crispr. This is another helpful practice: journalists are letting audiences know about the state of the research based on its use either in bacterial, animal, plant, or human cells or organisms. Letting the audience know that these findings were

70 tested in, say, mice or pigs rather than human cells, journalists are adding another layer of dampened expectation to the cautious and attentive reader. The third component of complete and thorough reporting investigated in this study was inclusion of the speed of clinical translation for Crispr’s healthcare applications. The purpose of including this variable was to see whether journalists are giving audiences a tentative time frame for the availability of healthcare interventions, so expectations about Crispr’s capabilities can be contained. This can be seen in a December 2, 2015 Washington Post article detailing geneticist George Church’s quest to “cure aging” (Achenbach, 2015). Achenbach wrote, “I asked him if he was on track to reversing the aging process in the next five years or so. He said yes – and that it’s already happening in mice in the laboratory.” In two sentences, Achenbach has given readers an overview of the technological capability and an expectation for clinical translation – though this, of course, relies on accurate information from the scientists themselves, who are likely feeling pressure to churn out clinically tested research. Understanding just how much these statements effect audience attitude and opinion toward technological development would require further research.

Limitations of the Study Though the goal of this study was to take a comprehensive census of Crispr articles from top newspapers and science magazines, it does have some limitations. The first is its scope. Conducting a full census of all articles published on Crispr might yield even more robust results, particularly regarding framing and source use by journalists. There were also issues with the second coder and replication (see Methods section).

71 Another limitation of this study is that it considers only the final output by journalists – there is no way of knowing, without further research, what pressures journalists were facing as they developed stories, nor whether they attempted to reach more diverse sources. There is also no examination of the financial pressures placed on journalists, nor the nature of those pressures. The same can be said for scientists. Though it is in a scientist’s best interest to cooperate with the press and generate positive attitudes toward research, it could be that some stories were not told because of an unwillingness for scientists to risk getting “scooped” and losing their claims to research findings. It can be advantageous to keep research ideas to oneself in a climate that promotes monetizing scientific findings and creating biotechnology companies to capitalize on those findings.

Areas for Further Research As an exploratory study, this work provides a foundation for further research into the ways Crispr is presented by news media, how its presentation effects audiences who read news media, and how scientists are presented by journalists from two different kinds of media outlets. One of the obvious routes for further research is conducting experiments that present audiences with Crispr framed in different ways and examine their attitudes toward the technology. It would also be enlightening to see how frames work in tandem, and whether frames presented together have different effects than frames presented alone. For example, a future study could show how the pairing of social progress and runaway frames influences audience attitudes and understanding of Crispr. Another useful group of frames to examine in tandem would be the combination of social progress, scientific/technical uncertainty, and economic

72 development/competitiveness frames – the three more likely to be used by science journalists compared to news reporters. As Crispr moves from its initial stages into clinical trials and eventually widespread interventions, continued research on the way journalists use frames, audiences are affected by those frames, and how policy is developed related to both frames and audience effects will be a fruitful area of research. Furthermore, analyzing how frame use has changed over time, even within the scope of the current study, may yield further insights into how frames have increased, decreased, or stayed the same in frequency as the technology has been covered by mainstream media. Another line of research on framing and biotechnology that may increase our understanding of journalistic practices is exploring the educational background of journalists covering biotech, as well as their sex. Relationships between individual journalists and their decisions about framing and source selection related to biotech coverage could demonstrate connections. For example, it could be that female journalists are more likely to use female sources, or that journalists with a background in science and technology cover stories using different frames, and thus tell different narratives compared to those without a science background. In short, this study has brought a deeper understanding to the methods used by journalists to develop a narrative about Crispr. These results lay the foundation for understanding how different sources, frames, and devices like hype affect audience members. As Crispr moves into the regulatory stage of its development, and even as clinical trials progress and medical interventions reach the market, the results of this study will help researchers understand the effects of journalist framing on audience understanding of a complex, yet pervasive and public, biotechnology.

73 REFERENCES

Achenbach, J. (2015, December 2). A Harvard professor says he can cure aging, but is that a good idea? Washington Post.

Achenbach, J. (2016, May 3). Pondering 'what it means to be human' on the frontier of gene editing. Washington Post, Health and Science

Anderson, A. A., Scheufele, D. A., Brossard, D., & Corley, E. A. (2011). The role of media and deference to scientific authority in cultivating trust in sources of information about emerging technologies. International Journal of Public Opinion Research, 24(2), 225–237. https://doi.org/10.1093/ijpor/edr032

Besley, J. C., & Shanahan, J. (2005). Media attention and exposure in relation to support for agricultural biotechnology. Science Communication, 26(4), 347– 367. https://doi.org/10.1177/1075547005275443

Berkowitz, B. D., & Beach, D. W. (1993). News Sources and News Context: The Effect of Routine News, Conflict and Proximity. Journalism Quarterly, 70(1), 4–12.

Boukes, M., Boomgaarden, H. G., Moorman, M., & De Vreese, C. H. (2015). Political News with a Personal Touch. Journalism and Mass Communication Quarterly, 92(1), 121–141. https://doi.org/10.1177/1077699014558554

Box Office Mojo. (n.d.). Rampage (2018). Retrieved April 3, 2019, from Box Office Mojo website: https://www.boxofficemojo.com/movies/?id=newlinetentpole 2018.htm

Boykoff, M. T., & Boykoff, J. M. (2004). Balance as bias: global warming and the US prestige press. Global environmental change, 14(2), 125-136.

Brewer, P. R., & Ley, B. L. (2013). Whose Science Do You Believe? Explaining Trust in Sources of Scientific Information About the Environment. Science Communication, 35(1), 115–137. https://doi.org/10.1177/1075547012441691

Brossard, D., & Nisbet, M. C. (2006). Deference to Scientific Authority Among a Low Information Public: Understanding U.S. Opinion on Agricultural Biotechnology. International Journal of Public Opinion Research, 19(1), 24– 52.

74 Brossard, D., Shanahan, J., & McComas, K. (2004). Are Issue-Cycles Culturally Constructed? A Comparison of French and American Coverage of Global Climate Change. Mass Communication & Society, 7(3), 359–377. https://doi.org/10.1207/s15327825mcs0703

Cacciatore, M. A., Scheufele, D. A., & Iyengar, S. (2016). The End of Framing as we Know it … and the Future of Media Effects. Mass Communication and Society, 19(1), 7–23. https://doi.org/10.1080/15205436.2015.1068811

Caulfield, T., & Condit, C. (2012). Science and the sources of hype. Public Health Genomics, 15(3-4), 209-217.

Cohen, E. L. (2002). Online Journalism as Market-Driven Journalism. Journal of Broadcasting & Electronic Media, 46(4), 532–548. https://doi.org/10.1207/s15506878jobem4604

Cohen, J. (1960). A coefficient of agreement for nominal scales. Educational and Psychological Measurement, 20(1), 37-46.

Craft, S., Vos, T. P., & David Wolfgang, J. (2016). Reader comments as press criticism: Implications for the journalistic field. Journalism, 17(6), 677–693. https://doi.org/10.1177/1464884915579332

Davis, A. (2009). Journalist–source relations, mediated reflexivity and the politics of politics. Journalism Studies, 10(2), 204–219. https://doi.org/10.1080/14616700802580540

Druckman, J. N. (2001). On the Limits of Framing Effects: Who Can Frame? The Journal of Politics, 63(4), 1041–1066.

Druckman, J. N., & Bolsen, T. (2011). Framing, motivated reasoning, and opinions about emergent technologies. Journal of Communication, 61(4), 659–688. https://doi.org/10.1111/j.1460-2466.2011.01562.x

Dudo, A. (2015). Scientists, the Media, and the Public Communication of Science. Sociology Compass, 9(9), 761–775. https://doi.org/10.1111/soc4.12298

Dumas-Mallet, E., Smith, A., Boraud, T., & Gonon, F. (2018). Scientific Uncertainty in the Press: How Newspapers Describe Initial Biomedical Findings. Science Communication, 40(1), 124–141. https://doi.org/10.1177/1075547017752166

Dunwoody, S. (2014). Science journalism: prospects in the digital age. In Routledge handbook of public communication of science and technology (pp. 43-55). Routledge.

75 Ecker, U. K. H., Lewandowsky, S., Chang, E. P., & Pillai, R. (2014). The effects of subtle misinformation in news headlines. Journal of Experimental Psychology: Applied, 20(4), 323–335. https://doi.org/10.1037/xap0000028

Entman, R. (1993). Framing : Toward Clarification of A Fractured Paradigm. Journal of Communication, 43(January), 51–58. https://doi.org/10.1111/j.1460- 2466.1993.tb01304.x

Eschner, K. (2018, December 6). The gene-edited Chinese twins represent a multi- generational ethical quandary. Popular Science.

Evans, J. P., Meslin, E. M., Marteau, T. M., & Caulfi, T. (2011). Deflating the Genomic Bubble “. Science, 331(February), 18–19. https://doi.org/10.1126/science.1198039

Fikes, B. J. (2017, December 7). Safer DNA editing method may fix human diseases, Salk scientists report. Los Angeles Times.

Fiske, S. T., & Dupree, C. (2014). Gaining trust as well as respect in communicating to motivated audiences about science topics. Proceedings of the National Academy of Sciences, 111, 13593–13597. https://doi.org/10.1073/pnas. 1317505111

Flynn, B., Rickard, J., & Garcia, H. (Producers), & Peyton, B. (Director). (2018). Rampage [Motion picture]. USA: Warner Bros. Pictures.

Gabielkov, M., Ramachandran, A., Chaintreau, A., & Legout, A. (2016). Social Clicks: What and Who Gets Read on Twitter? HAL.https://doi.org/10.1145/2896377.2901462

Gamson, W. A., & Modigliani, A. (1987). The changing culture of affirmative action. In R. G. Braungart & M. M. Braungart (Eds.), Research in political sociology (Vol. 3, pp. 137–177). Greenwich, CT: JAI Press.

Gamson, W. A., & Modigliani, A. (1989). Media discourse and public opinion on nuclear power: A constructionist approach. American Journal of Sociology, 95, 1–37.

Goffman, E. (1974). Frame analysis: An essay on the organization of experience. Cambridge, MA, US: Harvard University Press.

Goffman, E. (1986). Framing analysis. Boston: Northeastern University.

76 Gonon, F., Bezard, E., & Boraud, T. (2011). Misrepresentation of neuroscience data might give rise to misleading conclusions in the media: The case of attention deficit hyperactivity disorder. PLoS ONE, 6(1). https://doi.org/10.1371/journal.pone.0014618

Hammond, A., Galizi, R., Kyrou, K., Simoni, A., Siniscalchi, C., Katsanos, D., … Nolan, T. (2016). A CRISPR-Cas9 gene drive system targeting female reproduction in the malaria mosquito vector Anopheles gambiae. Nature Biotechnology, 34(1), 78–83. https://doi.org/10.1038/nbt.3439.A

Hart, P. S. (2011). One or many? The influence of episodic and thematic climate change frames on policy preferences and individual behavior change. Science Communication, 33(1), 28–51. https://doi.org/10.1177/1075547010366400

Hill, V. (2017, May 10). MUTANT MENU | The Ethics of Gene Editing [Video file]. Retrieved from https://www.youtube.com/watch?v=NrDM6Ic2xMM&t =0s&list=PLQwg0PxpUPlrFWJX2OBb1Gkbjn02k5UjT&index=2

Ho, S. S., Brossard, D., & Scheufele, D. A. (2008). Effects of Value Predispositions, Mass Media Use, and Knowledge on Public Attitudes Toward Embryonic Stem Cell Research. International Journal of Public Opinion Research, 20(2), 171–192.

Holton, A., Weberling, B., Clarke, C. E., & Smith, M. J. (2012). The Blame Frame: Media Attribution of Culpability About the MMR-Autism Vaccination Scare. Health Communication, 27(7), 690–701. https://doi.org/10.1080/10410236.2011.633158

Hubbard, G., Kraft, A., & Serpente, N. (2018, August). Crispr-Cas9: Timeline of Key Events (L. Marks, Ed.). Retrieved November 16, 2018, from http://www.whatisbiotechnology.org/index.php/timeline/science/CRISPR- Cas9/

Hwang, Y., & Southwell, B. G. (2009). Science TV News Exposure Predicts Science Beliefs. Communication Research, 36(5), 724–742. https://doi.org/10.1177/0093650209338912

Jacobs, T. B., LaFayette, P. R., Schmitz, R. J., & Parrott, W. A. (2015). Targeted genome modifications in soybean with CRISPR/Cas9. BMC Biotechnology, 15(1), 1–10. https://doi.org/10.1186/s12896-015-0131-2

Jensen, E. (2012). Scientific sensationalism in American and British press coverage of therapeutic cloning. Journalism & Mass Communication Quarterly, 89(1), 40- 54.

77 Katsir, J., & Moreschi, O. (Writers), & Hoskinson, J. (Director). (2016, January 5). Episode 64, season 2 [Television episode]. In S. Colbert, J. Stewart, C. Licht, T. Purcell, & B. Julien (Producer), The Late Show with Stephen Colbert. Manhattan, NY: CBS Television Studios.

Koteliansky, V., Sharp, P. A., Jacks, T., & Anderson, D. G. (2015). Genome editing with Cas9 in adult mice corrects a disease mutation and phenotype. Nature Biotechnology, 32(6), 551–553. https://doi.org/10.1038/nbt.2884.Genome

Kozubek, J. (2016). Modern Prometheus: Editing the Human Genome with Crispr- Cas 9. Cambridge, UK: Cambridge University Press.

Langner, T., Kamoun, S., & Belhaj, K. (2018). CRISPR Crops: Plant Genome Editing Toward Disease Resistance. Annual Review of Phytopathology, 56(1), annurev-phyto-080417-050158. https://doi.org/10.1146/annurev-phyto- 080417-050158

Lecheler, S., Bos, L., & Vliegenthart, R. (2015). The mediating role of emotions: News framing effects on opinions about immigration. Journalism and Mass Communication Quarterly, 92(4), 812–838. https://doi.org/10.1177/1077699015596338

Listerman, T. (2010). Framing of science issues in opinion-leading news: International comparison of biotechnology issue coverage. Public Understanding of Science, 19(1), 5–15. https://doi.org/10.1177/0963662505089539

Liang, P., Xu, Y., Zhang, X., Ding, C., Huang, R., Zhang, Z., … Huang, J. (2015). CRISPR/Cas9-mediated gene editing in human tripronuclear zygotes. Protein and Cell, 6(5), 363–372. https://doi.org/10.1007/s13238-015-0153-5

MacGregor, P. (2007). Tracking the online audience: Metric data start a subtle revolution. Journalism Studies, 8(2), 280–298. https://doi.org/10.1080/14616700601148879

MacGregor, C., Petersen, A., & Parker, C. (2018). Promoting a healthier, younger you: The media marketing of anti-ageing superfoods. Journal of Consumer Culture. https://doi.org/10.1177/1469540518773825

Maibach, E. W., Nisbet, M., Baldwin, P., Akerlof, K., & Diao, G. (2010). Reframing climate change as a public health issue: an exploratory study of public relations. BMC Public Health, 10(299), 2010.

78 Maibach, E. W., Roser-Renouf, C., & Leiserowitz, A. (2008). Communication and Marketing As Climate Change-Intervention Assets. A Public Health Perspective. American Journal of Preventive Medicine, 35(5), 488–500. https://doi.org/10.1016/j.amepre.2008.08.016

Mariscal, C., & Petropanagos, A. (2016). CRISPR as a driving force: the Model T of biotechnology. Monash Bioethics Review, 34(2), 101–116. https://doi.org/10.1007/s40592-016-0062-2

Marks, L. A., Kalaitzandonakes, N., Wilkins, L., & Zakharova, L. (2007). Mass media framing of biotechnology news. Public Understanding of Science, 16(2), 183- 203.

McCombs, M. E., & Shaw, D. L. (1972). The agenda-setting function of mass media. Public opinion quarterly, 36(2), 176-187.

Molek-Kozakowska, K. (2014). Hybrid Styles in Popular Reporting on Science: A Study of New Scientist’s Headlines. Hybridity and the News Hybrid Forms of Journalism in the 21st Century, (December), 135–159.

Molteni, M. (2017, June 8). Crispr may cure all genetic disease -- one day. WIRED.

Nelkin, D. (1995). Selling science: How the press covers science and technology (2nd ed.). New York, NY: W.H. Freeman.

Nelson, C. E., Hakim, C. H., Ousterout, D. G., Thakore, P. I., Moreb, E. A., Rivera, R. M. C., … Gersbach, C. A. (2017). In vivo genome editing improves muscle function in a mouse model of Duchenne muscular dystrophy. Science, 351(6271), 403–407. https://doi.org/10.1126/science.aad5143.In

Nisbet, M. C. (2009a). Communicating climate change: Why frames matter to public engagement. Environment, 51, 514-518.

Nisbet, M. C. (2009b). Framing science: A new paradigm in public engagement. In Communicating science (pp. 54-81). Routledge.

Nisbet, M. C. (2016). The ethics of framing science. In Communicating Biological Sciences (pp. 67-90). Routledge.

Nisbet, M. C., Brossard, D., & Kroepsch, A. (2003). Framing science: The stem cell controversy in an age of press/politics. Harvard International Journal of Press/Politics, 8(2), 36-70.

79 Nisbet, M. C., & Lewenstein, B. V. (2002). Biotechnology and the American media: The policy process and the elite press, 1970 to 1999. Science Communication, 23(4), 359-391.

Nisbet, M. C., & Scheufele, D. A. (2009). What's next for science communication? Promising directions and lingering distractions. American Journal of Botany, 96(10), 1767-1778.

Nisbet, M. C., Scheufele, D. A., Shanahan, J., Moy, P., Brossard, D., & Lewenstein, B. V. (2002). Knowledge, reservations, or promise? A media effects model for public perceptions of science and technology. Communication Research, 29(5), 584–608+609. https://doi.org/10.1177/009365002236196

Oliver, J. (Writer), & Perota, J., Werner, C., Hoskinson, J., Pennolino, P., & Leddy, B. (Directors). (2018, July 1). Gene Editing [Television episode]. In J. Oliver, T. Carvell, J. Taylor, J. Thoday, & L. Stanton (Producer), Last Week Tonight with John Oliver. New York, NY: CBS Broadcast Center.

Ossola, A. (2015a, August 6). Should bioethicists "Get out of the way" of CRISPR research? Popular Science.

Ossola, A. (2015b, August 26). Biohackers are now using CRISPR. Popular Science, Science.

Pan, Zhongdang, Kosicki, G. (1993). Framing Analysis: An Approach to News Discourse Framing Analysis: An Approach to News Discourse. Political Communication, 10, 55–75. https://doi.org/10.1080/10584609.1993.9962963

Peters, H. P. (2013). Gap between science and media revisited: Scientists as public communicators. Proceedings of the National Academy of Sciences, 110(Supplement 3), 14102-14109.

Peterson, K. (2017, July 7). The gene editors are only getting started. Wall Street Journal, Opinion.

Poell, T., & van Dijck, J. (2014). Social media and journalistic independence. In J. Bennett & N. Strange (Eds.), Media Independence: Working with Freedom or Working for Free? (pp. 182–201). London: Routledge.

Ross, K. (2007). The journalist, the housewife, the citizen and the press: Women and men as sources in local news narratives. Journalism, 8(4), 449–473. https://doi.org/10.1177/1464884907078659

80 Ross, P., & Hartsfield, T. (2017, March 6). Ranked: The Best and Worst Science News Sites. Retrieved November 15, 2018, RealClear Science: https://www.realclearscience.com/blog/2017/03/06/ranked_the_best__worst_s cience_news_sites.html

Sendén, M. G., Sikström, S., & Lindholm, T. (2014). “ She ” and “ He ” in News Media Messages : Pronoun Use Reflects Gender Biases in Semantic Contexts. Sex Roles, 72, 40–49. https://doi.org/10.1007/s11199-014-0437-x

Scheufele, D. A. (1999). Framing as a theory of media effects. Journal of Communication, (August 1996), 103–122. https://doi.org/10.1111/j.1460- 2466.1999.tb02784.x

Scheufele, D. A. (2006). Five lessons in nano outreach. Materials Today, 9,64.

Scheufele, D. A., & Tewksbury, D. (2007). Framing, Agenda Setting, and Priming: The Evolution of Three Media Effects Models. Journal of Communication, 57, 9–20. https://doi.org/10.1111/j.1460-2466.2006.00326.x

Slater, M. D. (2007). Reinforcing spirals: The mutual influence of media selectivity and media effects and their impact on individual behavior and social identity. Communication Theory, 17(3), 281–303. https://doi.org/10.1111/j.1468- 2885.2007.00296.x

Su, L. Y. F., Akin, H., Brossard, D., Scheufele, D. A., & Xenos, M. A. (2015). Science News Consumption Patterns and Their Implications for Public Understanding of Science. Journalism and Mass Communication Quarterly, 92(3), 597–616. https://doi.org/10.1177/1077699015586415

Sundar S (2008) The MAIN model: A heuristic approach to understanding technology effects on credibility. In: Metzger M and Flanagin A (eds) Digital Media, Youth, and Credibility. Cambridge, MA: The MIT Press.

Takahashi, B. (2011). Framing and sources: A study of mass media coverage of climate change in Peru during the V ALCUE. Public Understanding of Science, 20(4), 543–557. https://doi.org/10.1177/0963662509356502

Tandoc, E. C. (2015). Why Web Analytics Click: Factors affecting the ways journalists use audience metrics. Journalism Studies, 16(6), 782–799. https://doi.org/10.1080/1461670X.2014.946309

Tatlow, D. K. (2015, June 30). Ethical divide for China and West. New York Times, Science Desk.

81 Tilley, S. (2018, November 29). Profound ethical questions as it appears eugenics is upon us: Get ready for CRISPR babies. USA Today, Technology.

Top 15 U.S. Newspapers by Circulation [Fact sheet]. (2018, July). Retrieved September 18, 2018, Agility PR Solutions: https://www.agilitypr.com/resources/top-media-outlets/top-15-daily-american- newspapers/

Travis, J. (2015, December 18). Making the cut: CRISPR genome-editing technology shows its power. Science, 350(6267), 1456-1457.

Vu, H. T. (2014). The online audience as gatekeeper: The influence of reader metrics on news editorial selection. Journalism, 15(8), 1094–1110. https://doi.org/10.1177/1464884913504259

Witschge, T., & Nygren, G. (2009). Journalism: A Profession Under Pressure? Journal of Media Business Studies, 6(1), 37–59. https://doi.org/10.1177/1748048512461762

Wu, Y., Liang, D., Wang, Y., Bai, M., Tang, W., Bao, S., … Li, J. (2013). Correction of a Genetic Disease in Mouse via Use of CRISPR-Cas9. Cell Stem Cell, 13(6), 659–662. https://doi.org/10.1016/j.stem.2013.10.016

Xue, W., Chen, S., Yin, H., Tammela, T., Papagiannakopoulos, T., Joshi, N. S., … Jacks, T. (2015). CRISPR-mediated direct mutation of cancer genes in the mouse liver. Nature, 514(7522), 380–384. https://doi.org/10.1038/nature13589.CRISPR-mediated

Yin, H., Xue, W., Chen, S., Bogorad, R., Benedetti, E., Grompe, M., … Anderson, D. G. (2015). Genome editing with Cas9 in adult mice corrects a disease mutation and phenotype. Nature Biotechnology, 32(6), 551–553. https://doi.org/10.1038/nbt.2884.Genome

Zoch, L. M., & Turk, J. V. (1998). Women Making News: Gender as a Variable in Source Selection and Use. Journalism & Mass Communication Quarterly, 75(4), 762–775.

Zou, Q., Wang, X., Liu, Y., Ouyang, Z., Long, H., Wei, S., … He, J. (2015). Letter to the Editor Generation of gene-target dogs using CRISPR / Cas 9 system. Journal of Molecular Cell Biology, 7(6), 580–583.

82 Appendix A

NISBET’S TYPOLOGY OF COMMON BIOTECHNOLOGY FRAMES

Frames that Consistently Appear Across Policy Debates (Nisbet, 2009b) Frame Defines Science-Related Issue As… Social progress …improving quality of life, or solution to problems. Alternative interpretation as harmony with nature instead of mastery, “sustainability.” Economic development/competitiveness …economic investment, market benefits or risks; local, national, or global competitiveness. Morality/ethics …in terms of right or wrong; respecting or crossing limits, thresholds or boundaries. Scientific/technical uncertainty …a matter of expert understanding; what is known versus unknown; either invokes or undermines expert consensus, calls on the authority of “sound science,” falsifiability, or peer- review. Pandora’s box / Frankenstein’s monster / …call for precaution in face of possible runaway science impacts or catastrophe. Out-of-control, a Frankenstein’s monster, or as fatalism, i.e. action is futile, path is chosen, no turning back. Public accountability/governance …research in the public good or serving private interests; a matter of ownership, control, and/or patenting of research, or responsible use or abuse of science in decision-making, “politicization.” Middle way/alternative path …around finding a possible compromise position, or a third way between conflicting/polarized views or opinions. Conflict/strategy …as a game among elites; who’s ahead or behind in a winning debate; battle of personalities; or groups; (usually journalist-driven interpretation).

83 Appendix B

CODING MANUAL

Instrument A: Background Information

1. Name of Media Outlet:

• Record the name of the newspaper/magazine.

2. Article Headline:

• Record the article’s headline in its entirety.

3. Date of Publication:

• Record the article’s publication date.

4. Author:

• Record the author(s) listed in the byline.

5. Word Count:

• Using Microsoft Word, record the word count from the article, not including the headline, byline, publication date, or photograph titles/captions.

Instrument B: Biotechnology Frames

• Record whether each of the following frames was present or absent in the article using the keywords from Nisbet’s (2009b) typology (See Appendix A).

1. Social Progress: P/A?

• Ex: “Over the next decade, gene editing could help humanity overcome some of the biggest and most persistent challenges in global health and development. The technology is making it much easier for scientists to discover better diagnostics, treatments, and other tools to fight diseases that still kill and disable millions of people every year, primarily the poor” (Smith, 2018).

84 • Ex: “For the first time, scientists in the US have successfully edited the DNA of viable human embryos using the powerful gene-editing tool CRISPR. Gaining the ability to edit human DNA is the first step toward one day allowing scientists to prevent babies from being born with incurable diseases or disabilities” (Potenza, 2017).

2. Economic Development/Competitiveness: P/A?

• Ex: “Crispr promises to create a large amount of economic value. Beyond simply editing out the genes for cancer, heart disease, or arthritis, people might be able to use the technique to improve both their happiness and their productivity…Considering the anti-anxiety and anti- depression drug markets are already worth tens of billions of dollars, more effective genetic treatments could become an even bigger industry” (Smith, 2018).

• Ex: “As these and other unprecedented discoveries continue to occur, expect the value of CRISPR companies to grow as fast as the potential practical applications of this technology” (Glatter, 2018).

3. Morality/Ethics: P/A?

• Ex: “The scientific outcry has been so swift because He’s purported work, conducted in secret, bulldozes past ethical guidance on so called ‘germline editing,’ in which alterations to an embryo’s DNA will be passed down to subsequent generations” (Molteni, 2018).

• Ex: Headline from The Conversation: “Should we edit the genomes of human embryos? A geneticist and social scientist discuss” (Boardman & O’Neill, 2018).

4. Scientific/Technical Uncertainty: P/A?

• Ex: “The report was met with instant concern and skepticism by the scientific community. He’s experiment altered the genomes of embryos produced through IVF; their genetic changes will therefore be passed on to any future generations. What’s more, most experts in CRISPR are not convinced that the technology is ready – or safe – for treating humans” (Park, 2018).

85 • Ex: “However, a study just published in Nature Biotechnology has found that when CRISPR-Cas9 is used to edit genomes, off-target DNA damage is more common than had previously been appreciated…This raises the possibility that non-target genes or regulatory sequences could be affected by the editing process, a discovery which comes in the wake of other recent work which raised concerns that CRISPR-Cas9 gene-editing might trigger cancers” (“The Safety of Crispr-Cas9 Gene Editing Is Being Debated,” 2018).

5. Pandora’s Box/Frankenstein’s Monster/Runaway Science: P/A?

• Ex: “While the two baby girls were born healthy, the future potential risk to humans if the procedure was continued was immeasurable, the letter said. The scientists called for a swift investigation by regulators. ‘The Pandora's box has been opened, and we may have a chance to close it before it is irreparable,’ they said” (Needham, 2018).

• Ex: “Gene drives have similar dystopian potential. In theory, a single lab could alter the entire planet. And the technology has arrived far quicker than our ability to grapple with its staggering implications” (Jacobsen, 2018).

• Ex: Headline: “China’s ‘Dr. Frankenstein’ Faces Down His Critics” (Tsoi, 2018).

6. Public Accountability/Governance: P/A?

• Ex: “But earlier this month, the U.S. Patent and Trademark Office accepted a challenge to the Broad Institute group’s patent from the University of California, which is backing a rival group of scientists. Without licensing from the eventual winner of the patent fight, Editas and other companies investing in commercial CRISPR research could be blocked from marketing products based on the intellectual property in question, putting at risk hundreds of millions of dollars in investment” (Chideya, 2016). *Note that this example could also be categorized as an “Economic Development/Competitiveness” frame.

• Ex: “It is the latest round of litigation of an ongoing patent dispute. The Broad Institute holds patents on the gene editing

86 technology known as CRISPR, and attorneys for UC Berkeley have been challenging the claim for more than a year” (Daley, 2018).

• Ex: “What’s unclear is who will reap the potential windfall on royalties for commercial applications in medicines, health treatments, and improved foods. ‘I’m a company and I want to practice CRISPR, who do I license, who do I pay?’” (Decker & Cortez, 2018).

7. Middle Way/Alternative Path: P/A?

• Ex: “Increasing crop yields through conventional plant breeding is inefficient – the outcomes are unpredictable and it can take years to decades to create a new strain. On the other hand, powerful genetically modified plant technologies can quickly yield new plant varieties, but their adoption has been controversial…but now a new genome editing technology known as CRISPR may offer a good alternative” (Li, 2018).

• Ex: “The idea of using gene-editing technology to tweak a baby’s DNA before birth has been the topic of fierce debate for years. But now, most Americans say using this technology on embryos would be acceptable under certain conditions” (Rettner, 2018). Here, editing embryos with diseases is the middle way between no embryonic editing and complete freedom.

8. Conflict/Strategy: P/A?

• Ex: From the article “Three Crispr Scientists Win Prestigious Award, Fanning Controversy Over Credit:” “It was only the latest verdict on the controversial question of who deserves credit for turning a bacterial immune system worth billions of dollars race to turn CRISPR into a human therapeutic, everyone…is clashing over who did what when and how important their contribution was…By then seemingly everyone had filed for patents, and Zhang, though he filed after Doudna and Charpentier, was awarded foundational CRISPR patents starting in 2014. Those are now the subject of a bitter, years-long…legal dispute pitting financially strapped UC against the mega-endowment Broad” (Begley, 2018).

87 • Ex: An article with the headline, “With prestigious prize, an overshadowed CRISPR researcher wins the spotlight,” the following was stated: “For Šikšnys, who works at Vilnius University’s Institute of Biotechnology, the recognition was double sweet because his part in the discovery of CRISPR often has been overlooked. Šikšnys will share the $1 million award with two researchers who have received far more attention, Jennifer Doudna…and her collaborator, Emmanuelle Charpentier…Conspicuously absent from the award was another researcher who has enjoyed the CRISPR spotlight, chemist Feng Zhang” (Cohen, 2018).

Instrument C: Sources

• For each distinct source quoted within the body of an article, record the following:

1. Name (if given):

• Record the name of the individual, group, or organization used as a source of information by the journalist.

2. Title (if given):

• Record the title given for the individual quoted (i.e. scientist, researcher, representative, spokesperson).

3. Organization/Affiliation (if given):

• Record the organization or affiliation of the individual quoted.

4. Sex (if given):

• If a pronoun is used by the journalist to refer to a source, record the corresponding sex of the source.

5. Sex (if not given):

• If no pronoun is used by the journalist, conduct a google search of the individual and his/her corresponding affiliation to determine the individual’s sex.

88 Instrument D: Headlines

• For each article headline, assess the following:

1. Hype: P/A?

• Record whether hype is present in the headline by assessing whether the headline makes unsubstantiated promises or overt generalizations (i.e. mentioning “cures,” “keys,” or “cracked codes” for a disease).

2. Keywords/Phrases

• If hype is present, record any hyped key words or phrases.

1. Ex: How CRISPR Gene Editing is Revolutionizing Medicine and the Companies Who Invest in It” (Glatter, 2018)

a. Hype: P

b. Keyword: “Revolutionizing”

2. Ex: “Cure or Carcinogen? CRISPR-Cas9 May Cause Cancer” (Treur, 2018)

a. Hype: P

b. Keywords: “Cure” and “Carcinogen”

3. Ex: “Scientists Wiped Out a Mosquito Population by Hacking Their DNA With CRISPR” (Houser, 2018)

a. Hype: P

b. Keywords: “Wiped Out” and “Hacking”

Instrument E: Replication Statements, Models, and Speed of Clinical Translation

1. Replication Statement: P/A?

• Record whether or not the journalist discusses the amount of testing completed with a given Crispr method in the article.

89 • If a replication statement is present, record whether the findings were described as “initial” or “replicated.”

1. Ex: Replication Statement Present, Initial: “The perspective also calls for more research on local gene drives” (Borel, 2017).

2. Ex: Replication Statement Present, Initial: “’The two papers present preliminary results,” biochemist Bernhard Schmierer of the Karolinska, co-leader of its study, told STAT” (Begley, 2018a).

3. Ex: Replication Statement Present, Initial: “The Scientists Who Sparked CRISPR Panic Couldn’t Reproduce Their Study Results” (Brown, 2018).

2. Model: P/A?

• Record whether or not a model was mentioned.

• If mentioned, record the type of model as one of the following:

1. Human (including human cells and embryos)

2. Animal

3. Plant

4. Bacteria

5. Other – list:

a. Ex: Model P, Animal: “Olson is encouraged by the results, even if they came from just a few dogs” (Park, 2018a).

b. Ex: Model P, Animal: “The researchers treated four dogs…with the CRISPR-Cas9 system, which snipped out a short stretch of the animals’ DNA” (“Canine CRISPR trial raises hopes for humans with deadly disease,” 2018).

90 c. Ex: Model P, Animal: “The technique has been tested in live mice, as well as dogs, and appears to be working” (Potenza, 2018).

3. Healthcare Related: P/A?

• If the article has a primary focus on the healthcare applications of CRISPR and discusses a specific therapy or intervention, code as present.

• If healthcare related, record the presence of a statement about the speed of clinical translation.

1. Ex: Healthcare P, Clinical Translation P: “The biotech firm CRISPR Therapeutics is using a similar approach in a clinical trial for beta thalassemia that’s slated to begin this year in Europe” (Mullin, 2018).

2. Ex: Healthcare P, Clinical Translation P: “If studies continue to go smoothly, with promising results, clinical trials could start within a couple of years” (Potenza, 2018).

Sample Articles

Begley, S. (2018a, May 31). Three CRISPR Scientists Win Prestigious Award, Fanning Controversy Over Credit. Scientific American.

Boardman, F., & O'Neill, H. (2018, September 24). Should we edit the genomes of human embryos? A geneticist and social scientist discuss. Retrieved November 28, 2018, from The Conversation website: https://theconversation.com/should- we-edit-the-genomes-of-human-embryos-a-geneticist-and-social-scientist- discuss-100355

Borel, B. (2017, November 16). New Model Warns About CRISPR Gene Drives in the Wild. Quanta Magazine.

Brown, K. V. (2018, March 29). The Scientists Who Sparked CRISPR Panic Couldn't Reproduce Their Study Results. Retrieved November 28, 2018, from Gizmodo website: https://gizmodo.com/the-scientists-who-sparked-crispr-panic-couldnt- reprodu-1824181703

Canine CRISPR trial raises hopes for humans with deadly disease. (2018). Nature.

91 Chideya, F. (2016, January 25). The Battle Over CRISPR Could Make Or Break Some Biotech Companies. Retrieved November 27, 2018, from https://fivethirtyeight.com/features/ the-battle-over-crispr-could-make-or- break-some-biotech-companies/

Cohen, J. (2018, June 4). With prestigious prize, an overshadowed CRISPR researcher wins the spotlight. Science.

Daley, J. (2018, May 1). CRISPR Patent Dispute Heard in Federal Court. TheScientist.

Decker, S., & Cortez, M. (2018, April 29). This Court Battle Will Decide Who Will Make a Fortune From Gene-Editing Tech. Bloomberg.

Gates, B. (2018, May). Gene Editing for Good: How Crispr Could Transform Global Development. Foreign Affairs.

Glatter, R. (2018, June 25). How CRISPR Gene Editing Is Revolutionizing Medicine And The Companies Who Invest In It. Forbes, Med Tech.

Houser, K. (2018, September 25). Scientists Wiped Out a Mosquito Population by Hacking Their DNA. Retrieved November 28, 2018, from https://futurism.com/the-byte/gene-drive-mosquitos-crispr

Jacobsen, R. (2018, June 20). Deleting a species. Pacific Standard.

Li, Y. (2018, May 22). These CRISPR-modified crops don't count as GMOs. Retrieved November 28, 2018, from Phys.Org website: https://phys.org/news/2018-05-crispr-modified-crops-dont-gmos.html

Molteni, M. (2018, November 27). Scientist who Crispr'd babies bucked his own ethics policy. Wired.

Mullin, E. (2018, April 11). CRISPR trials are about to begin in people—but we still don’t know how well it works in monkeys. MIT Technology Review.

Needham, K. (2018, November 27). 'Pandora's box has been opened: Scientist's baby gene-editing claim has fueled backlash". Sydney Morning Herald, Science.

Park, A. (2018a, August 30). CRISPR Gene Editing Fixes Muscular Dystrophy in Dogs. Are Humans Next? Time.

Park, A. (2018b, November 27). Chinese University 'Shocked' By Researcher’s Report of First CRISPR Gene-Edited Human Babies. Time, Health.

92 Potenza, A. (2017, August 2). Human embryos edited for first time in the US using CRISPR. Retrieved November 28, 2018, from The Verge website: https://www.theverge.com/2017/7/27/16049340/human-embryos-dna-crispr- gene-editin-us

Potenza, A. (2018, January 31). New gene therapy could cure a deadly disease that stops the heart from beating. Retrieved November 28, 2018, from The Verge website: https://www.theverge.com/2018/1/31/16954678/gene-editing- duchenne-muscular-dystrophy-cure-crispr-cas9-dystrophin-protein

Rettner, R. (2018, July 26). Most Americans Support Gene Editing for Babies to Treat Diseases, Poll Finds. Retrieved November 28, 2018, from Live Science website: https://www.livescience.com/63168-gene-editing-babies-pew- poll.html

Smith, N. (2018, January 25). Gene Editing Needs to Be Available to Everyone. Retrieved November 28, 2018, from Bloomberg website: https://www.bloomberg.com/opinion/articles/2018-01-25/ gene-editing-needs- to-be-available-to-everyone

The safety of Crispr-Cas9 gene editing is being debated. (2018, July 19). The Economist.

Treur, T. (2018, June 12). Cure or Carcinogen? CRISPR-Cas9 May Cause Cancer. Retrieved November 28, 2018, from Nova Next website: https://www.pbs.org/wgbh/nova/article/cure-or-carcinogen-crispr-cas9-may- cause-cancer/

Tsoi, G. (2018, November 28). ‘China’s Dr Frankenstein’ faces down his critics. Retrieved November 28, 2018, from Inkstone website: https://www.inkstonenews.com/science/scientist-he-jiankui-defends-gene- editing-babies-leaves-questions-unanswered/article/2175418

93 Appendix C

CODING INSTRUMENT

Instrument A: Background Information

1. Name of Media Outlet: ______

2. Article Headline: ______

3. Date of Publication: ______

4. Author(s): ______

5. Word Count: ______

Instrument B: Biotechnology Frames

• Circle whether each of the following frames was present (P) or absent (A):

1. Social Progress: P A

2. Economic Development/Competitiveness: P A

3. Morality/Ethics: P A

4. Scientific/Technical Uncertainty: P A

5. Pandora’s Box/Frankenstein’s Monster/Runaway Science: P A

6. Public Accountability/Governance: P A

7. Middle Way/Alternative Path: P A

8. Conflict/Strategy: P A

Instrument C: Sources

• Record the following for each distinct source used in the article (if not mentioned, write “NM”):

1. Name: ______

94 2. Title: ______

3. Organization/Affiliation: ______

4. Sex (if given): ______

5. Sex (if not mentioned): ______

Instrument D: Headlines

• Record the presence/absence of headline hype, and if present, key words/phrases.

1. Hype: P A

2. If present, key words/phrases: ______

Instrument E: Replication Statements, Models, and Speed of Clinical Translation

• Record whether each of the following items is present in the article. If present, provide the given details.

1. Replication Statement: P A

• If present, was it replicated or initial? ______

2. Model: P A

• If present, circle the type:

1. Human (including human cells and embryos)

2. Animal

3. Plant

4. Bacteria

5. Other – list: ______

3. Healthcare Related: P A

• If present, speed of clinical translation statement: P A

95 Appendix D

EXAMPLE ARTICLES: CODING RESULTS

Achenbach, J., & Johnson, C. Y. (2017, February 15). Broad Institute scientist prevails in epic patent fight over CRISPR. The Washington Post, Science.

Berezow, A., & Locwin, B. (2017, September 8). 'Designer babies' won't be a fad. It's too hard to create them. USA Today, Opinion.

Bunge, J., & Craymer, L. (2018, May 6). Scientists in China race to edit crop genes, sowing unease in U.S. The Wall Street Journal, Business.

Eschner, K. (2018, November 26). Scientists 'went rogue' and genetically engineered two human babies--or at least claimed to. Popular Science, Health.

Feibus, M. (2017, July 24). CRISPR gene editing tool: Are we ready to play God? USA Today.

Feltman, R. (2016, April 18). Why this genetically modified mushroom gets to skip USDA oversight. The Washington Post, Science.

Greenberg, J. (2016, February 4). Crispr gene-editing upstart Editas goes public as patent battle rages. WIRED.

Harris, J., & Darnovsky, M. (2016, August). Pro and con: Should gene editing be performed on human embryos? National Geographic.

Hiltzik, M. (2016, February 5). The billion-dollar CRISPR patent battle: A case of big money shaping science. Los Angeles Times, Business.

Li, D. K., & Fears, D. (2017, August 2). Inside researchers' amazing – and terrifying – gene editing discovery. New York Post, News.

Luo, J. (2018, September 3). Here's what we know about CRISPR safety -- and reports of 'genome vandalism'. The Washington Post, Health & Science.

Maldarelli, C. (2018, January 30). China might be winning the CRISPR race, but we have the FDA. Popular Science, Health.

McKay, B. (2016, September 2). Mosquitoes are deadly, so why not kill them all? The Wall Street Journal, General News.

Molteni, M. (2017, December 11). Crispr Therapeutics plans its first clinical trial for genetic disease. WIRED.

96 Molteni, M. (2018a, January 12). Clashes over the future of gene therapy at US' biggest biotech meeting. WIRED, Science.

Molteni, M. (2018b, April 27). The WIRED guide to Crispr. WIRED.

Niller, E. (2017, July 10). The Pentagon ponders the threat of synthetic bioweapons. WIRED.

Ossola, A. (2015, November 5). Biotech company plans to use CRISPR on people in next two years. Popular Science.

Ossola, A. (2016, September 6). CRISPR-modified corn may soon be ready for market. Popular Science.

Perez, C. (2017, July 26). Scientists genetically modify a human embryo for the first time. New York Post, Tech.

Pollack, A. (2015, May 11). Jennifer Doudna, a pioneer who helped simplify genome editing. The New York Times, Science.

Rana, P., & Craymer, L. (2018, December 14). Big tongues and extra vertebrae: The unintended consequences of animal gene editing. The Wall Street Journal, Ideas.

Rana, P., & Fan, W. (2018, December 28). Chinese gene-editing experiment loses track of patients, alarming technology’s inventors. Wall Street Journal, World.

Specter, M. (2016, August). How the DNA revolution is changing us. National Geographic.

Stockton, N. (2017, May 5). How Crispr could snip away some of humanity's worst diseases. WIRED.

Thiessen, M. (2018, November 29). Gene editing is here. It's an enormous threat. The Washington Post, Opinions.

Tucker, R. (2017, March 19). The terrifying DNA discoveries that are making science- fiction fact. New York Post, Tech.

Wade, N. (2015a, December 3). Scientists seek moratorium on edits to human genome that could be inherited. The New York Times, Science.

Wade, N. (2015b, December 21). Gene drives offer new hope against diseases and crop pests. New York Times, Science.

97 Weelwright, J. (2016, June). The revolution will be edited. Discover Magazine.

Yin, S. (2015, September 10). Stem cell experts support using CRISPR in human embryos. Popular Science.

Yong, E. (2015, December 2). What can you actually do with your fancy gene-editing technology? The Atlantic.

Yong, E. (2017, August 2). The designer baby era is not upon us. The Atlantic.

Zhang, S. (2017, February 15). What the CRISPR Patent Decision Means for Gene Editing. The Atlantic, Science.