A. M. Petersen
Structure and dynamics of the U.S. innovation system according to boundary-spanning Academic-Government-Industry research co-production (pdf)
Under review, (2026).
Abstract
Boundary-spanning Academic-Government-Industry (AGI) research co-production is widely appreciated as foundational to national innovation system (NIS) capacity, yet little is known about the structure of the ecosystem of US research institutions and its dynamics across state and sectoral boundaries. We address this gap by analyzing >8.7 million research publications co-produced by >9,000 US institutions from 1980-2020 in order to model the rate of boundary-spanning (cross-state and cross-sectoral) research co-production. Focusing on a balanced panel of 296 prominent institutions, we estimate the relative strength of various factors promoting NIS integration – including institutional (disciplinary alignment, size, pres- tige), spatial (distance, state borders), multi-campus university system (MUS), and cross-sectoral predictors – together with temporal interactions unveiling dynamics across four decades marked by techno-social paradigm shifts. Descriptive results document variation in AGI profiles across states, which fosters co-evolutionary selection environments that condition rates of research co-production across both geographic and sectoral boundaries. Regression estimates show that cross-sectoral interfaces involving government (AG, GI), MUS synergies and prestige hierarchies together contribute positively to NIS integration. In contrast, institutional size disparities and academic-industry (AI) boundaries associate with lower rates of institutional co-production, while persistent negative state-border effects suggest that, despite increasingly global science, regional ecosystems remain fundamental to the NIS. Notably, several of these relationships intensified after the mid-1990s marked by widespread internet emergence. Find- ings provide guidance for policies that strategically invest in spatially-extending cross-sectoral and multi-campus systems that enhance resilience and responsiveness to grand challenges, thereby constituting central mechanisms sustaining NIS cohesion, competitiveness and adaptive capacity.
A. M. Petersen, A. Montaño Ramirez, I. Pavlidis
Intensification of triple-helix research co-production fosters US innovation system competitiveness (pdf)
Under review, (2026).
Abstract
According to the triple-helix model of innovation, the US national innovation system (NIS) is underpinned by synergistic interactions among academia (A), government (G), and industry (I), however the impact of cross-sectoral research co-production on regional competitiveness lacks systematic quantification. To this end, we analyzed $>$11.5 million research publications from 1950-2020, affiliated with $>$9,000 US institutions, and categorized by their sectoral configuration (A, G, I, AG, AI, GI, AGI). Cross-sectoral research rose from 4% in 1970 to 10% by 2020, with complete triple-helix (AGI) configurations featuring the fastest growth. To better understand the impact of the increasing AGE configurational diversity on NIS competitiveness, we model the citation premium attributable to (a) institutional, (b) cross-regional and (c) cross-sectoral dimensions of the US research ecosystem. In particular, we measure a 11.2% citation boost attributable to multi-campus university system synergies, which highlights the multi-scale role of regional integration that supports both local and national AGI interfaces - the analog of ecological ecotones - thereby representing manifold, high-churn selection environments that foster innovation system capacity, competitiveness and resiliency. Moreover, we measure a 14.9% citation attributable to triple-helix research configurations, thereby validating a consequential advantages of cross-sectoral convergence. Notably, the relative strength of citation premiums associated with institutional prestige, national integration and cross-sectoral convergence have grown dramatically post-1990s, likely reflecting the system's response to techno-social and financial paradigm shifts. These insights highlight how boundary-spanning integration enhances NIS capacity, thereby informing policies for government-led moonshots to address grand challenges and sustain global leadership in science & technology.
A. M. Petersen
Institutional prestige, geographic embedding and competitiveness: A comparative analysis of research ecosystems in California and Texas (pdf)
Nature Humanities and Social Sciences Communications 13, 348 (2026). DOI:10.1057/s41599-026-06697-z
Abstract
Among the various institutional layers that shape the structure and function of regional innovation systems (RIS), the network of research universities (RU) stands out as a principal component, constituting the hubs and channels that facilitate knowledge creation, technology transfer and skilled workforce development. Here we analyze the contributions to regional competitiveness associated with (a) local versus cross-border collaboration and (b) prestige signaling associated with a region's flagship institutions of higher education. By applying hierarchical regression models with institutional fixed effects to research profiles constructed from ~800,000 publications affiliated with 28 RU in California (CA) and Texas (TX), in particular including all University of California and University of Texas campuses, we find that research co-produced with a premier regional university that also extends across international borders features a 21% (CA) and 25% (TX) citation premium. We also show that the citation premium attributable to institutional prestige and international collaboration have both increased over time relative to other factors. These findings underscore the importance of balancing investment in regional integration with cross-border engagement, as prestige spillovers can enhance regional competitiveness by attracting elite scientists, fostering impactful collaborations, and integrating distant regions within global research networks. Managing RIS thus calls for strategic investments in regional collaboration networks and flagship institutions to balance local benefits while leveraging the capacity to attract external research expertise and capabilities. Within this context, our study helps explain the growing shift towards multi-institutional consortia and RU satellite campuses as mechanisms for harnessing institutional and cross-regional complementarities at scale.
A. M. Petersen, A. Montaño Ramirez
Research university assortativity conditions the integration of regional innovation systems (pdf)
PLOS Complex Systems 3, e0000088 (2026). DOI:10.1371/journal.pcsy.0000088
Abstract
Substantial policy efforts to develop regional innovation systems (RIS) highlight the importance of understanding the institutional factors that promote integration and synergy at the regional scale. To this end, we analyzed historical patterns of research co-production within and across California (CA) and Texas (TX), two US regions that account for >5% of global research publication. This predominance is largely attributed to the University of California and the University of Texas, two multi-campus university systems (MUS) that feature distinct configurations of institutional specialization. We exploit these differences to analyze four institutional assortativity channels that foster RIS synergy: institu- tional proximity, prestige, homophily, and specialization. Descriptive analysis reveals that institutional co-publication rates differ within and across RIS and are influenced by external socio-economic shocks, such as the 2007-08 financial crisis, which intensified institutional clustering within these research university ecosystems. We also develop institutional specialization profiles for exploring the structure and role of institutional alignment within RIS. Results indicate that regional integration is mediated by the alignment of institutional specialization and moderated by institutional homophily. These findings underscore the critical role of the MUS backbone that supports RIS integration and generates resiliency to socio-economic shocks. Moreover, MUS provide institutional redundancy and variation that generates a broad combinatorial space fostering multi-university research synergies. All together our framework can offer guidance for resource-allocating administrators facing the innovator’s dilemma of whether to exploit institution-specific capabilities or to strategically identify and invest in novel multi-institutional synergies that leverage the complex configurational space of institutional specializations that uniquely characterize each RIS.
- Presented at ICCSI 2026: Lightning talk
- Dashboard (link): Research co-production (co-occurrence) matrix - across 28 research universities in California and Texas, 1970-2020.
A. M. Petersen
University digital media co-occurrence networks reveal structure and dynamics of brand visibility in the attention economy (pdf)
Nature Humanities and Social Sciences Communications 12, 117 (2025). DOI:10.1057/s41599-025-04419-5
Abstract
As universities compete for visibility to attract student enrollment and build scientific reputation, the management of institution of higher education (IHE) brand has emerged as an important strategic endeavor. Yet our understanding of the factors that condition the structure and dynamics of brand stratification within regional IHE ecosystems is limited. Instead, our best approximation for brand equity derives from widespread IHE rankings, which lack contextual and relational features for understanding the patterns of engagement in the fast-moving attention economy, and in particular how institutional partnerships can generate brand equity by leveraging ecosystem network effects. To this end, here we develop a framework for measuring two dimensions of brand equity, namely visibility and association, according to the frequency of digital media articles featuring a university’s official name. We demonstrate this approach for 29 universities in California and Texas based upon 2 million media articles published between 2000-2020, and validate our approach by correlating university digital media visibility with ARWU Shanghai rankings and freshman enrollment growth. As roughly 10% of the article sample features > 1 university, longitudinal analysis of institutional co-occurrence reveals the extent to which brand association stratifies according to regional proximity, institutional specialization and prestige. Interestingly, despite the shared value generated from media co-visibility, the frequency of multi-university media articles has declined over time, which we attribute to paradigm shifts in media content production following the 2007-08 financial crisis and the COVID-19 infodemic, in addition to increased competitiveness of the attention economy. Topic classification of media article titles shows how specialized institutions may strategically manage their brand equity by aligning content production with dominant media topics to reinforce brand visibility with broader social, technological and environmental narratives.
- Dashboard (link): Digital media co-visibility (co-occurrence) matrix - across 29 research universities in California and Texas, 2000-2020.
A. M. Petersen
University ecosystem analytics: Case study of regional integration and competitiveness in California and Texas (weblink)
Dryad repository dataset, (2025). DOI: 10.5061/dryad.2rbnzs7zc
Abstract
Despite substantial policy efforts aimed at developing regional innovation systems (RIS), our understanding of institutional factors that promote synergy and integration at the regional scale is limited. To address this gap, we constructed 2 representations of research university ecosystems in California (CA) and Texas (TX) that identify institutional co-occurrences in research and news media, within and across these regions. The selection of these regions is attributed to the University of California and the University of Texas, two multi-campus university systems (MUS) that feature distinct configurations of institutional specialization. As such, we exploit these differences to analyze four institutional assortativity channels that foster system-level synergies: institutional proximity, prestige, homophily, and specialization. The first representation we constructed is based upon ~3 million publications collected from Clarivate Analytics Web of Science Core Collection (WOS) that are affiliated with at least one of the 28 institutions in our sample, which together represent >5% of publications indexed by WOS over the sample period 1970-2020. The 28 institutions consist of 10 institutions belonging to the University of California (UC) system and 12 institutions belonging to the University of Texas (UT) system; we complement these two public multi-campus university systems (MUS) by including six prominent private universities, which represent a non-MUS comparison group. As universities increasingly compete for visibility to attract student enrollment and build scientific reputation, the management of institution of higher education (IHE) brand has emerged as an important strategic endeavor. Hence, the second representation we constructed is based upon ~2 million digital news media articles published between 2000-2020 that specifically mention at least one of these universities. Similar to the first representation, mapping the rates of digital media co-visibility among IHE facilitates a systems-level understanding of the factors that condition the structure and dynamics of brand stratification within research university ecosystems, and fosters the development of novel measures for two dimensions of brand equity – namely, visibility and association.
