Scientific discovery increasingly depends on interdisciplinary teams, yet we still lack robust empirical evidence about what makes those teams work well. National reports have called for large-scale, comparable datasets that move beyond case studies and business-sector analogs to examine how scientific teams actually function in real research environments (National Academies of Sciences, Engineering, and Medicine [NASEM], 2025). Advancing team science requires validated measurement frameworks that are methodologically rigorous, validated within science contexts, and scalable across institutions.

The Translational Team Science Initiative (TTSI), established in 2025 at the University of Illinois Urbana-Champaign (Illinois), is developing a multi-site validation platform to address this gap. This effort has two primary aims: (1) to identify team and organizational constructs that reliably predict scientific collaboration outcomes, and (2) to develop consolidated tools capable of supporting both research on science team dynamics and diagnostic uses within research organizations.

TTSI is implementing two pilot testbeds to advance these aims. The first involves evaluating team processes over time during a coaching intervention within a large, multi-site interdisciplinary research initiative (N ≈ 80 members). The second focuses on longitudinal evaluation of early-stage research teams receiving internal seed funding and team science coaching support at Illinois (N = 6 teams with 21 members). Surveys, coaching-session reflections, interviews, and structured feedback allow us to assess whether established team science measures function as expected in active scientific contexts. Because these teams will collaborate over multiple years, the design incorporates longitudinal assessment that supports iterative refinement of measurement strategies while examining how team supports influence collaboration quality, research climate, and related outcomes over time.

Data collection integrates established team science constructs, including transactive memory systems (Lewis, 2003), psychological safety (Edmondson, 1999), team cohesion (Mathieu et al., 2015), task and relational conflict (Jehn & Mannix, 2001), leadership and interpersonal climate (Martinson et al., 2025), and research integrity climate (Martinson et al., 2013; Briskin & Gunsalus, 2025). Quantitative survey data are paired with collaboration network analyses informed by multilevel network research (Barley et al., 2022; Dinh et al., 2024), qualitative interviews, and session-level coaching reflections. This mixed-methods design supports construct validation, triangulation across data sources, and examination of relationships among team processes, institutional context, collaboration networks, and research climate (Stokols et al., 2008; Ruge-Jones et al., 2023, 2024).

Here, we present the evaluation design of the pilot testbeds and outline our current plans for scaling this measurement framework across a broader network of research teams and institutions. Insights from these pilots will guide systematic expansion across interdisciplinary institutes in the Big Ten Academic Alliance, where approximately twenty-five institutes have expressed interest in participating. As we move toward coordinated, cross-institutional implementation, we are actively refining our instruments and protocols to maximize comparability, cumulative knowledge-building, and practical diagnostic value. We view this poster as an opportunity to invite dialogue with members of the SciTS community about how best to scale this effort, design instruments that serve both scholarship and practice, and build a shared empirical foundation for advancing team science.