A new approach for increasing graduate students’ science communication capacity and confidence

Why SciWrite?

Helping science graduate students build more effective writing habits, techniques, and confidence supports their graduate work and prepares them to meet the demands of a career in the sciences. A recent report by the National Science Foundation (NSF) and Council of Graduate Schools recommended the future of graduate education should include training for diverse career opportunities, and this training should include development of broad science communication skills (Linton, 2013). Most of the current emphasis on development of science communication expertise has been directed toward established scientists, both academic and non-academic, and not enough attention has been focused on best practices for training graduate students in broader approaches to effective science communication (Kuehne et al., 2014). The traditional model of graduate student work involves first writing a brief proposal, and then completing milestones of data collection and analysis for research, with the majority of writing relegated to the end of students’ academic programs. This delayed approach has consequences for research practice, for enhancing science competency of the broader public, and for science graduate students’ future endeavors (Kuehne et al., 2014). Early engagement in scientific writing and science communication can contribute to students’ ability to participate in their scientific disciplines (Chinn & Hilgers, 2000) and can improve students’ comprehension of scientific concepts (Keys, 1999; Wallace, Hand & Prain, 2004). There is a demonstrable need for writing programs that engage graduate students in myriad forms of science communication from the outset of their graduate program. Such an approach will help students learn early and often how to best communicate their science through a number of genres to multiple non-specialist and non-academic audiences, as well as to their scientific colleagues.

The SciWrite@URI program (hereafter SciWrite)—initially funded through a NSF Innovations in Graduate Education grant—grew from the idea that barriers to graduate student participation in science communication and science writing include time constraints, lack of knowledge about opportunities, and a perceived lack of support from advisors (Kuehne et al., 2014). In addition, there is often disagreement among science faculty about the most helpful strategies for providing students with writing feedback and support, and many faculty attribute this confusion to a lack of adequate resources and/or inadequate departmental support in developing their own writing and teaching (Pololi, Knight & Dunn, 2004; Reynolds et al., 2009; Kuehne et al., 2014). Graduate students are writing under duress to meet a variety of important deadlines, and yet at many institutions, writing centers and writing classes only exist for undergraduate students. Furthermore, current approaches to writing feedback for graduate students often rely on individual mentors who may or may not have the knowledge necessary to provide useful feedback (Florence & Yore, 2004; Odena & Burgess, 2015) and thus access to writing support becomes an inequitable and unsustainable resource for many graduate students. SciWrite was designed by a community of practice to address many of these constraints by offering students and faculty with opportunities and institutional support for writing and communication. SciWrite responded to the reality that despite the recognized need for improved communication training for scientists, little research has been conducted on the most effective way to implement and assess such programs for STEM graduate students (Heath et al., 2014; Kuehne et al., 2014; Druschke et al., 2018; National Alliance for Broader Impacts, 2018; Harrington et al., 2021). SciWrite faculty members, representing the departments of Writing and Rhetoric, Nutrition and Food Sciences, and Natural Resources Science at the University of Rhode Island (URI), sought to create an innovative, interdisciplinary training model for graduate student writers that was both testable and replicable.

Program objective and foundation

The aim of SciWrite was to create a unique training model that equips science graduate students with the tools necessary to be effective writers for multiple and diverse audiences. To provide students with practical, helpful writing tools, we used a rhetoric-based approach. Though rhetoric is often misunderstood as a term meaning “political spin,” rhetorical studies is in fact a field of research dedicated to better understanding the ways in which we write and communicate (Bitzer, 1968; Selzer, 2004; Druschke & McGreavy, 2016). For instance, rhetoric allows us to understand how writing tasks vary depending on the genre and purpose, as well as the audiences’ needs and expectations—this knowledge is crucial for any developing scientist who wishes to improve their writing and expand the reach of their scientific research (Bazerman, 1988; Druschke & McGreavy, 2016; Druschke et al., 2018).

Rhetorically-based writing instruction provides a number of best practices known to help students improve their writing and ultimately develop autonomy over their own writing and learning processes (North, 1984; Straub, 1996; Coe, 2011; Petersen et al., 2020). Using this rhetorically-based writing instruction framework, SciWrite created tangible and practical learning outcomes that graduate students engaged in during the program and could apply to their own immediate writing practice. At the time the grant was awarded, SciWrite was the first science communication program for graduate students that emerged as a collaborative effort between experts in life sciences and experts in writing and rhetoric to develop specific rhetoric-based learning outcomes, and we currently know of only one such other program (King-Kostelac et al., 2022). Furthermore, we know of very few science communication training programs that incorporate systematic and formal evaluation of student achievement of learning outcomes (but see Baram-Tsabari & Lewenstein, 2016, 2017; Rakedzon & Baram-Tsabari, 2017; Lewenstein & Baram-Tsabari, 2022). We expected that this training would help participants continue to improve their writing even after completion of the program and would serve as a model for other institutions that wanted to do the same.

Program recruitment and timeline

In 2016, and again in 2017, science graduate students were recruited to be SciWrite Fellows via departmental, college, and university communications (email listservs, newsletters, announcements, etc.), and faculty and staff recommendations. Candidacy required that students were URI graduate students with at least 2 years remaining in their program (M.S. or Ph.D.) and were a science major. We defined “science” using the NSF’s broad definition for STEM disciplines which includes science, technology, engineering, mathematics, and social and economic sciences (O’Meara et al., 2018). We recruited a total of 12 Fellows, 75% of which were female (n = 9) and 25% of which were male (n = 3). 8.3% of Fellows identified as Asian (n = 1), 8.3% as Black or African American (n =1), 66.7% as White (n = 8), and 25% identified as having an ethnicity of Hispanic or Latino origin (n = 3). Nine Fellows were native English speakers, while three fellows spoke English as a second language. Research protocols were approved by the University of Rhode Island Institutional Review Board (IRB # HU1516-009) and all participants completed the informed consent process and signed informed consent forms.

The SciWrite program timeline consisted of two writing-intensive science communication courses and a summer science outreach internship in the first year, followed by a paid writing tutor training course, a multimedia journalism class, and writing tutor employment in the second year (Fig. 1). URI graduate students who were not SciWrite Fellows also enrolled in the courses and attended some of the workshops, but only SciWrite Fellows completed the full suite of courses and workshops and were trained as writing tutors.

Designing courses using best practices for writing instruction

To our knowledge, SciWrite is the first science communication program to systematically and formally assess student achievement of specific learning outcomes based on rhetorical tenets, and best practices were incorporated into the program to help students meet each of the three learning outcomes (Petersen et al., 2020; Table 1). For example, to support students’ ability to meet the learning outcome related to habitual writing (LO1), SciWrite instructors used an approach called scaffolding, in which longer assignments were broken up into simpler, shorter assignments, rather than entire drafts. This helped students slowly get comfortable with smaller chunks of writing and incrementally build up to a full, final draft of each assignment. Such an approach allowed students to take on writing projects that seemed less daunting and worth fewer points than a full writing project, thereby aiding students in developing the habit of writing early and often for their assignments (Coe, 2011; Petersen et al., 2020).

To meet the learning outcome related to multiple genres (LO2), we incorporated the important concept of the “rhetorical situation” into our training (Bitzer, 1968; Druschke & McGreavy, 2016). The rhetorical situation refers to the context within which scientists communicate their research to others. Important parts of this rhetorical situation are the audience they are communicating with, the expectations and needs of that audience, and the purpose for communicating their research. SciWrite Fellows practiced analyzing and engaging in different rhetorical situations by first creating a public engagement project within a SciWrite course, followed by an experiential learning science communication internship. Over the course of 2 years, Fellows practiced writing in multiple genres and engaged with a variety of audiences.

To help students meet the learning outcome related to frequent review (LO3), we incorporated two best practices from writing and rhetoric into our training: a focus on higher-order rather than lower-order concerns, and facilitative feedback rather than directive feedback (Elbow, 1981; North, 1984; Neman, 1995; Straub, 1996). Higher-order concerns deal with matters such as thinking about audience needs and expectations, developing clear arguments, and adhering to genre conventions, rather than sentence-level editing. Facilitative feedback, in contrast to directive feedback and sentence-level editing, is a best practice used by writing centers that encourages writers to consider that there are a variety of choices to make in their revision process (Neman, 1995; Straub, 1996; Reynolds et al., 2009). A more open-ended approach to feedback allows students to have autonomy over their own writing and revision process (Neman, 1995; Straub, 1996). Such writing practices are critical to exercise when science students must determine how best to convey their results in writing (Groffman et al., 2010; Druschke et al., 2018). As peer reviewers, SciWrite Fellows focused on higher-order concerns and providing facilitative feedback in their courses, during their training, and eventually in their work as writing tutors at our newly established Graduate Writing Center. This allowed students to practice giving and receiving peer feedback in a structured, holistic way.

The three required courses (Fig. 1) in the SciWrite program—‘Graduate Writing in the Life Sciences’, ‘Public Engagement with Science’, and ‘Using Multimedia to Communicate Science’ were all carefully designed with the three rhetorically informed program learning outcomes in mind (Box 1; Table 1).

The first course of the SciWrite program, ‘Graduate Writing in the Life Sciences’, was a graduate-level science writing course in the Department of Writing and Rhetoric that served all URI graduate students interested in learning more about scientific writing (Druschke et al., 2018). There were four scaffolded writing projects, each in a unique genre with varied audiences (Table 1). The second SciWrite course (Fig. 1), ‘Public Engagement with Science’, was offered through the Department of Natural Resources Science (Druschke et al., 2018). The course centered around theoretical and practical aspects of public engagement with scientific research, policy, and management, with an emphasis on science communication (Table 1). Students’ final public engagement projects were experiential learning projects based on the individual student’s own research. These projects were created for a variety of different types of public audiences, including K-12 students, professional adults, and members of groups typically underrepresented in the sciences (Druschke et al., 2018). The third SciWrite course, ‘Using Multimedia to Communicate Science’, was a course in the Department of Journalism for upper-level journalism undergraduates and science graduate students (Fig. 1). Students learned about and developed capacity to communicate using a variety of science journalism genres, including news reports, coverage of STEM poster sessions, and public science lectures, interviews, and radio news stories (Druschke et al., 2018). Both SciWrite Fellows and non-SciWrite students were enrolled in all three courses, which were framed around the SciWrite learning outcomes.

Experiential learning: science communication internship

In Year 2, Fellows applied communication skills gained in Year 1 to a science communication internship. With the support of a SciWrite Team Faculty mentor, SciWrite Fellows participated in a writing-intensive summer internship for an external organization related to their own field of research (Fig. 1). The general goal was that SciWrite Fellows would help an organization convey its science to public and specialized audiences. This internship helped students learn how to identify the needs and expectations of different audiences and adjust their approaches accordingly. This internship prompted Fellows to apply SciWrite course and workshop concepts in a practical, hands-on way (Crone et al., 2011; Heath et al., 2014) and also provided Fellows the opportunity to gain professional experience writing in multiple genres for diverse audiences. Fellows identified suitable topics and media outlets with their internship hosts and engaged with their peers on developing and revising each piece (Table 2). Internships, funded by NSF, generally lasted 6 to 8 weeks but the duration was flexible depending on the science communication pieces being developed (all internships were approximately 150 h).

Writing center tutor experience

SciWrite Fellows were trained as writing center tutors for approximately 3 months prior to serving as independent tutors in the Graduate Writing Center. SciWrite Fellows participated in rhetorically-focused writing center tutor training developed around the three program learning outcomes (Box 1; Table 1). The training incorporated a genre theory approach (Troyan, 2014) and writing assignments that required SciWrite Fellows to write multiple, scaffolded drafts as well as provide feedback to other SciWrite tutor trainees on these same writing projects (Table 1). The training incorporated many of the best practices previously mentioned: audience awareness, scaffolding, providing facilitative rather than directive feedback, and focusing on higher-order rather than lower-order concerns (North, 1984; Neman, 1995; Straub, 1996). Fellows were trained using a peer-to-peer framework common to most writing centers (North, 1984; Straub, 1996). SciWrite Fellows then joined other writing center tutors for five hour-long team tutoring sessions at the URI Undergraduate and Graduate Writing Centers. Lastly, SciWrite Fellows were observed by the Graduate Writing Center Trainer and Graduate Writing Center Coordinator and were given advice on how to improve their feedback strategies and how to develop their own personal tutoring philosophy to guide their tutoring approach.

The following year, URI’s Graduate Writing Center was successfully created with tremendous support from the URI Graduate School, Office of the Provost, and Writing and Rhetoric faculty and graduate students, as well as faculty and staff from a variety of other departments and universities. Fellows from our second cohort, as well as one fellow from our first cohort, served as writing tutors alongside our Graduate Writing Center Tutor Trainer and Graduate Writing Center Coordinator.

Program assessment

SciWrite Fellows completed a pre/post survey about achievement of learning outcomes and emergent goals just before the beginning of their program, and immediately after the end of their program (see Supplemental Appendix for surveys). Surveys were developed in collaboration with URI’s Student Learning Outcomes & Assessment team and adopted survey methodologies traditionally used by URI’s Office for the Advancement of Teaching and Learning. These surveys contained identical questions to assess self-perceived achievement of learning outcomes, including changes in writing confidence and writing apprehension. SciWrite Fellows answered most of the questions using Likert scales. For the writing apprehension portion of the survey, students were asked a total of 26 questions about their apprehension with writing, and all Likert scale scores (1, strongly disagree; 5, strongly agree) were added together for a total writing apprehension score (Daly & Miller, 1975; Daly, 1978; Fischer & Meyers, 2017; Güler et al., 2017; for the complete Daly and Miller Writing Apprehension questionnaire, see Supplemental Appendix). All 12 SciWrite Fellows completed the pre-survey; seven SciWrite Fellows completed the post-survey.

We compared changes in learning outcomes, writing confidence, and writing apprehension over time for all Fellows in the SciWrite program. Because all 12 participants completed the pre-survey, but only seven completed the post-survey, we compared population means (rather than individual scores) for each metric. We used t-tests to compare means between data that were normally distributed and non-parametric Kruskal-Wallis tests for data that were not normally distributed.

Study limitations

Two limitations to our study were small sample size and lack of a “control” group of non-SciWrite Fellows. However, in a separately published manuscript, we report on an additional portion of our study in which we also evaluated non-SciWrite Fellows. Harrington et al. (2021). In that study, we found that written work from SciWrite Fellows scored higher than those from non-SciWrite Fellows in all three genres, and most notably thesis/dissertation proposals were higher quality. Unfortunately, due to lack of funding and resources, our sample size was limited to 12 students. Despite these limitations, our major goal (program development), was performed robustly and our evaluation shows that students responded positively to the program that was designed to produce these outcomes. We believe this study provides the groundwork for future science communication programs to develop larger research projects that address additional research questions.

Lessons learned and recommendations for future programs

SciWrite was a highly collaborative, interdisciplinary endeavor, and we believe that it is because of this collaboration that the legacy of the SciWrite program has continued for several years now past the period of the NSF-IGE grant. The research team included staff and faculty collaborations such as the URI Writing Center staff, faculty from the Department of Writing and Rhetoric, the College of the Environment and Life Sciences, and the College of Health Sciences, and the Director of Writing Across URI, a writing across the curriculum program. These interdisciplinary collaborations, coupled with the founding of the Graduate Writing Center, the development and launch of a Graduate Certificate in Science Writing and Rhetoric, and the creation of three new courses in science communication situated in several departments, resulted in SciWrite creating self-sustaining communities of writing practice for our graduate students and an infrastructure that remains and represents the legacy of SciWrite. Importantly, a final lasting effect of SciWrite is that science graduate students can now receive training very similar to that provided through the SciWrite program through enrollment in the Graduate Certificate in Science Writing and Rhetoric at URI, which includes completion of a set of science writing and communication courses, a science communication internship off campus, and the potential of completing tutor training and working in the Graduate Writing Center. Previous research has found such communities of practice are crucial for graduate student development (Kriner et al., 2015; Coffman et al., 2016; Grainger et al., 2017; O’Meara et al., 2018). Communities of practice are also particularly important when considering issues associated with a lack of systematic instruction for graduate students, and thus disparity in individual support and success (Florence & Yore, 2004; Odena & Burgess, 2015). Creating equitable infrastructures of support are crucial when developing writing/communication programs for graduate students, because many graduate students do not have access to the writing help they need and are solely relying on support from mentors who may or may not have the skills necessary to provide useful feedback (Florence & Yore, 2004). This inequity is of particular significance when considering research has found science publication to be a male and privileged space (Macaluso et al., 2016). SciWrite—thanks to support from the National Science Foundation and the faculty and key administrative leaders of our university—fundamentally and visibly changed the landscape of writing at URI. We recommend that future science communication programs develop similar strategies to provide infrastructure that will maintain interdisciplinary communities of practice to ensure the program has lasting impacts for students.

We believe that the Graduate Certificate in Science Writing and Rhetoric program that URI developed in collaboration with SciWrite can be used as a model for other science graduate departments lacking the funding and resources to develop a Science Communication training program. The Graduate Certificate in Science Writing and Rhetoric program adopted already existing courses in our English Department as well as various science departments, and worked with instructors to slightly modify the curriculum to incorporate our learning outcomes and best practices. For example, some best practices that are relatively simple for instructors to incorporate into already existing courses are scaffolding and peer review (Table 1). Courses in the Graduate Certificate in Science Writing and Rhetoric program adopted an approach called scaffolding, in which longer assignments were broken up into simpler, shorter assignments, rather than entire drafts. This helps students slowly get comfortable with smaller chunks of writing and incrementally build up to a full, final draft of each assignment. Such an approach allows students to take on writing projects that seem less daunting and are worth fewer points than a full writing project, thereby aiding students in developing the habit of writing early and often for their assignments (Coe, 2011; Petersen et al., 2020). Courses in the Graduate Certificate in Science Writing and Rhetoric program also adopted a teaching tool known as “peer review” in which students practice giving and receiving peer feedback in a structured, holistic way. Science graduate departments looking to provide students with additional science communication training opportunities that don’t require additional funding and resources can consult our Supplemental Materials and the Graduate Certificate in Science Writing and Rhetoric program’s website (https://web.uri.edu/cels/academics/graduate-programs/certificate-in-science-writing-and-rhetoric/) for additional guidance.