Designing Geoscience Educational Innovations That Propagate

Abstract:

NSF and other funders have supported the development of undergraduate STEM educational innovations over the past decades, only to see many yield limited uptake and impact beyond the grantee institutions. Numerous factors contribute to this complex problem, but one cause is likely insufficient incorporation of the understanding of how innovations propagate into project design. Following J.W. Dearing and colleagues, “dissemination” can be characterized by “push” approaches, which mainly emphasize one-to-many information sharing. In TUES/CCLI proposals, dissemination strategies have commonly taken the form of the “3 Ps” (presenting, publishing and posting) , with overall modest impact. Since the seminal work of Everett Rogers, however, “diffusion” of innovations has been understood as an inherently social process among potential adopters, which interacts with community norms and existing practices. Keys to diffusion include close understanding of the needs and context of the potential-adopter community and the development of “pull” within it, as well as support for implementation of innovations.

Potential approaches to facilitating diffusion of innovations include a) using “lean start-up” methodologies (e.g., NSF’s I-Corps-L program), in which explicit business-model hypotheses are tested through customer-discovery interviews, commonly leading to pivots where initial hypotheses are not confirmed, b) providing a range of potential commitment levels for adopters tailored to levels of support (“reverse Kickstarter model”), c) supporting decentralized communities of practice in which adaptations and tacit knowledge can readily be shared, d) encouraging crowd-sourcing of innovations, with an “architecture of participation” informed by successful open-source projects, and e) integrating innovations with discipline-based educational research, e.g., big-data approaches which allow A/B testing and analysis of clickstream data that reveal behaviors along a novice-to-expert continuum. Such new directions will be facilitated by stronger partnerships with technologists and data scientists, as well as community development of learning objectives and assessment standards that are sufficiently flexible and transparent and allow rapid feedback.