Digital scholarship has never been more important than for the current generation of undergraduate students. The need to develop one’s technical expertise is not just a concern for those few students aspiring to a career in academia; competency in the use of computer-assisted methods has relevance for the entire student population. Data analytics, knowledge representation, and dissemination techniques are just a few of the many areas with broad professional application to have undergone technology-driven transformation in recent decades. Society’s reliance on technology is such that the digital has permeated our professional lives, transforming the skillsets expected of students upon graduation.

The integration of digital scholarship in the undergraduate curriculum can further the students’ learning experiences and engagement with their core subject matter, but there are numerous obstacles to embedding the skills required into course learning objectives and outcomes. Increasing digital fluencies among undergraduate students in the Arts and Humanities, in particular, presents a number of key challenges: “the technical proficiency of undergraduates and instructors, the timeframe of a single semester or quarter, and the availability of hardware and software”  [Bjork 2012, 101].

As a consequence, institutions of higher education are responsible for exploring a variety of pedagogical approaches to digital scholarship, both within and beyond the confines of the classroom. By “digital scholarship,” we refer to the practice of leveraging digital methods and computer-assisted approaches to research in the broader Arts and Humanities, and indeed, in related disciplines across the Social Sciences, for the purposes of producing new meaning across a multitude of forms.[1] In turn, there is a need to ensure that students are not simply being trained in the use of intuitive tools to produce artifacts of tactical convenience, but rather, that they are developing a deep understanding of the potential for new and supplementary meaning offered by computational methods, as well as an awareness of the digital’s many constraints and the profound repercussions that interdisciplinarity can have for established practices. As Tanya Clement argues:

Until we consider digital humanities undergraduate pedagogy in terms other than training, and rather as a pursuit that enables all students to ask valuable and productive questions that make for “a life worth living,” digital humanities will remain unrelated to and ill defined against the goals of higher education.  [Clement 2012, 372]

In this paper, we present a case study of an engaged scholarship model–which seeks to complement classroom-based learning with out-of-classroom experiences–as a means to explore an alternative pedagogical approach to digital scholarship. Specifically, we consider two evaluative questions: how effective are internships at developing knowledge and skills in digital methods? What are the optimal student learning conditions with respect to structure, guidance, and supervision to nurture the development of such knowledge and skills? To address these questions, we compare two undergraduate internships which proceeded as part of a collaboration in 2015 between the The Pennsylvania State University Libraries and the College of the Liberal Arts, which saw two independent groups working with undergraduate students on research projects with a significant digital component. At Penn State, there has been little distinction between the Digital Humanities work that is housed within the University Libraries, and that which is primarily led at the academic College-level. The structure of the institution’s Digital Humanities effort, and indeed the interdisciplinary nature of the field, is such that it has a range of interdepartmental stakeholders.[2] Both projects hired two paid interns, who were employed in a full-time capacity for a duration of 13 weeks.[3]

One pair of interns worked on a project which availed of computational approaches to text analysis for the purposes of exploring the language utilized by online roleplayers. These students worked primarily under the guidance of the Digital Humanities Research Designer, with support from the University Libraries’ Publishing and Curation Services, as well as the Office of Digital Pedagogy and Scholarship in the College of the Liberal Arts. In this instance, the students worked collaboratively under minimal supervision on the same project, on a research question of their own choosing. Effectively, this internship was an experiment to see the possibility of undergraduate digital scholarly research supported by resources – time, money, and faculty/staff expertise. For the purposes of clarity, these students will be referred to as the “text analysis interns.” The other pairing, which will be addressed as the “geospatial interns,” worked with faculty and staff of the Donald W. Hamer Maps Library, and were tasked with helping the library accomplish the goal of increasing digital access to the Sanborn Fire Insurance map collection. This internship was comparable to a traditional professional internship structurally, but with the goal of developing digital skills (digitization, mapping) that transfer to academic research, along with the freedom within the scope of the internship to pursue independent projects as interns developed skills. While the implementation and scope of the projects varied, both sought to adopt a digital-project-as-pedagogy approach, so that interns developed advanced expertise through direct engagement with applied research. In doing so, the intention was that interns would gain a sense of how to conduct research that is of publishable quality, while seeing what is required in bringing a digital project from conception to fruition. By design, the experience allowed students to develop a number of their broader professional skills, such as time and project management, as well as practice habits of collaboration, all of which would occur in professional environments.

As we begin to evaluate the internship model as a way to develop digital scholarship skills in students, it is helpful to situate the approach within a pedagogical framework. In this case, both internships were envisioned as project-based learning opportunities to enhance the digital skills and professionalization of undergraduate students. This “learn by doing” approach is common in the classroom, and can be equally effective in the field, given the right conditions. In “Perspectives on Learning in Internships,” David Thornton Moore challenges the notion that academic learning happens exclusively in the classroom, and that only testing or application of that knowledge happens in the field [Moore 1983, 40]. He argues: “Thinking in the real world may indeed supplement and reinforce school-based learning; but it can also do far more to develop valid and important learning in its own right”  [Moore 1983, 42]. Moore puts forth a matrix to evaluate internship experiences through focusing on two dimensions: the ways one uses knowledge, and the ways one relates to others in a particular learning environment. With respect to the mental work of internships, he suggests that we consider how interns are expected to use knowledge: is it fixed and immutable, or are students able to reorganize and transform knowledge? [Moore 1983, 41]. Similarly, we can evaluate the social relationships in particular contexts to see the degree to which interns are relied on and able to participate in the definition and creation of knowledge. This framework provides two spectrums useful for evaluation: is the mental work of the internship more rote and algorithmic, or creative and transformative? And are the social relationships in the environment hierarchical and controlled, or collegial and participatory? [Moore 1983, 44]. What follows is a comparison of the two internship experiences, using Moore’s matrix where appropriate.

In this section, mentor and student intern perspectives give some insight on the approaches and outcomes of the internship experiences. The mentors provide a self-reflective account of the internships, while the student perspectives are based on a thematic analysis of a qualitative survey which was conducted upon completion of their employment.

A similar approach to that utilized in the geospatial internship was pioneered at Bucknell University, where undergraduates worked as research assistants on the Stories of the Susquehanna Valley project.[6] Reporting on the results of their program, Katherine M. Faull and Diane Jakacki conclude:

Extending the classroom outside (both spatially and temporally) allows for the development of rich, deep knowledge in both digital tools and research subject matter. Indeed, extending the faculty-student collaboration to include students from outside traditional humanities departments also reifies the value of interdisciplinary research at an early level and reflects the professional DH research model employed by larger-scale projects.  [Faull and Jakacki 2015]

The experience at Penn State, both in the geospatial and text analysis internships, support these claims. Restating Moore’s spectrums of mental work, knowledge engagement and social relations, undergraduate student internships can be interpreted as serving two distinct but overlapping roles in student preparation. First, they can be seen as a direct extension of classroom work where information literacy and critical thinking–or research–are the focus; application outside of the classroom is the learning environment, and learning outcomes are evaluated in relation to student preparation for further engagement with academics. Alternatively, internships also acknowledge the gaps left by classroom preparation and a response to the reality that most students will not become professional academics, but rather will work in a variety of professional settings. In a student-centric internship model, we encourage advisors and supervisors to be somewhat flexible in adjusting the structure and expectations of their interns according to their stated professional goals rather than preconceived learning objectives or project goals. Although not by design, both internship case studies were successful in this regard, perhaps as a result of careful review and selection of intern candidates. As has already been emphasized, the selection step of an internship plays an important role in setting the stage for internship success and failure from both the perspective of the students and advisors. Any similar undertakings would be best advised to engineer a dynamic which represents the best possible opportunity for success, pairing complementary skillsets and personalities, though these can be difficult to assess, particularly the latter, through a limited application process.

Other institutions looking to implement similar models need to be very clear on their purpose: is the aim to have students develop their ability to negotiate the workplace, or learn how to do advanced research? It is possible to accomplish both, but depending on restrictions on time and resources, it may not be possible to achieve an equilibrium. Nor is it desirable to give students a false sense of a particular dynamic: the reality is that, major interdisciplinary projects excluded, most scholarship is still conducted in a largely isolated manner. This is not to say that we support the status quo, but merely acknowledge that it would be irresponsible to have students believe that a career in scholarship, particularly in the Humanities, will be predominantly occupied by collaborative endeavours–this might change, but most hiring and promotion committees still privilege “traditional” forms of scholarship. Regardless of an internship’s stated purpose, students should develop competencies beyond those cultivated in the classroom, and the supervisory team should stress the importance of transferable skills, so that students can see the utility of any newfound expertise. Internships provide an opportunity to liberate students of the constraints of the classroom, affording them the chance to make and learn from mistakes, to explore potentially fruitless strands of inquiry, and delve deeper into topics of personal interest. When developing a program model in such a context, Moore’s framework is helpful to review, considering the extent to which the work will be rote rather than creative or transformative: will there be enough of an intellectual engagement for the internship to be worth it for the student? How controlled is the situation? Will interns have the ability to make a real difference in the project as equals or is the information very controlled and top-down?

Defining suitable barometers for the measurement of success also presents a challenge. The primary purpose of the text analysis internship was to give students a sense of what is required in the completion of a digital project, a condition which, from assessment of their final research outputs, was a success. However, it cannot be said with confidence that they could replicate some of the methodologies that were utilized without assistance, and a future iteration of the initiative would do well to ensure that the project’s dataset can be gathered quickly, allowing more time for students to gain familiarity with some of the more technical aspects. This experience was shared by the geospatial interns, who spent a considerable amount of time having to digitize physical materials. When defining those objectives which will determine the success of a project, it is essential to allow for the time necessary to investigate challenges and problems, both technical and intellectual, that arise throughout the research process. Supervisors must also account for seemingly obvious yet often overlooked social realities, like students wanting to take holidays during the summer semester, which was when these internships took place.

The danger in such programs is that mentors will impose their own value systems on their participants, expecting them to adhere to scholarly principles and standards that undergraduates cannot reasonably be expected to have attained. Even faculty with extensive teaching experience might find that they are demanding too much of participants who are embarking on what is likely their first major research undertaking, and expectations must be continuously revisited before the project begins so as to ensure intended outcomes are realistic. While it is important to have a clear set of expectations before the project begins, flexibility is crucial, as it affords interns the scope necessary to pursue unforeseen developments, thus emphasizing learning outcomes over tangible deliverables. Students must not be afraid to “fail,” for any such fear will restrict critical and creative exploration. Equally, considering the value of the opportunity, the demands placed on interns should be as challenging as they are reasonable, and mentors need to be comfortable reprimanding any participant whose performance might not be meeting expectations–this is difficult, in that you may well be dealing with inexperienced individuals with whom you have started to build a rapport atypical of what is usually established between faculty and undergraduates. Striking a balance between authority and understanding is a leadership quality that is not easily attained.

Institutional limitations must also be considered–the reality of engaged scholarship is such that it is not always feasible. Penn State, and other institutions like Bucknell, where similar programs are already in place, has a network of faculty and staff whose remit is to support digital scholarship and undertakings of this nature. At other institutions, where faculty may be working without appropriate support from suitably-qualified peers, implementing such an approach might prove to be far more challenging. Having undergone significant investment in personnel whose mandate was to develop the institution’s capacity for digital scholarship, Penn State is in a position to pursue such initiatives. The culture of the departments involved is such that the inherent power differentials in staff and faculty-like employee classifications did not influence the internships–the projects operated as a nearly flat hierarchy, with all faculty, staff, and students contributing as important stakeholders. In a different context, wherein such an initiative might be contained within the hierarchical structures embedded within an institution, one could envision problematic scenarios where student time is prioritised in terms of labour, and dichotomies within employee classifications are reinforced in the minds of emerging scholars. The danger in these models, which involve both intellectual and practical components, is that interns might develop false scholar versus technician personas, based on the perceived roles of contributors. It is imperative to the future of digital scholarship, which has considerable issues around the division and acknowledgement of labour, that artificial power-structures are not reinforced in the minds of the next generation. This extends to both the students [Di Pressi et al. n.d.] and their perceptions of the roles played by all those staff enabling and contributing to the internship. At institutions where faculty-staff classifications exist, measures must be taken to ensure that each partner is encouraged to make a direct contribution to the scholarship’s intellectual vision, and that any technical effort is recognized for its inherent value.

Administrative support must also be in place, particularly if course credit is going to be one of the motivating factors for students. The geospatial interns had the option of obtaining credit through their respective departments, an option which one of them pursued. While a commitment to alternative modes of learning is present in particular schools and departments at Penn State, one cannot assume that cultural differences do not exist across disciplines, and that similar administrative support for engaged scholarship would be present across the entire institution. Cost is also a major consideration, particularly here, where the Bednar program allowed us to pay students for their participation in the internships.[7] The likelihood is that most institutions would not be in a position to pay interns, nor would they have the capacity to provide the technology necessary for students to pursue digital modes of scholarship. Unpaid internships might only serve to further the field’s current issues with diversity, in that they would be exclusive to those students in the privileged position of not having to consider remuneration when pursuing extracurricular placements.

Engaged approaches to teaching digital scholarship provide a mechanism to explore areas of interest in a collaborative and multidisciplinary manner. A digital project requires contact with primary and secondary sources through the lens of digital technologies. By creating opportunities to take an inquiry and explore it in a project environment, interns are learning the importance of understanding how these sources can be negotiated and manipulated through the digital, and what the deep and significant repercussions of this act might be. Given the resources, appropriate planning, and clear objectives and success metrics, we conclude that the internship model can be a highly effective learning experience for students, both to develop their digital literacies and professional skills. With the right guidance and meaningful work, the digital-project-as-pedagogy can be a powerful teaching approach in the digital humanities.

The conveners of this initiative would like to thank Nicki Hendrix, Joe Fennewald, Barbara Dewey, and Christopher P. Long for their support. Funding for this project was generously provided by library endowments made possible through the charitable contributions of Donald W. Hamer and Marie Bednar.