The Collaboration Patterns of Institution during the COVID-19 Pandemic

Chenbo Fu, Xuejiao Liang, Yong Ki Min, Qi Xuan, Shanqing YuSustainability 2023 · DOI

TL;DR

This paper analyzes how scientific institutions collaborated before and during the COVID-19 pandemic using a novel weighted network approach and "Institution Cooperation Freshness" (ICF). They found that while the pandemic generally fostered both new and stable collaborations, institutions became more cautious in partner selection, especially those involved in COVID-19 research, suggesting a strategic response to uncertainty.

What Problem Does This Paper Address?

The COVID-19 pandemic highlighted the critical role of scientific collaboration in addressing global crises and ensuring the sustainability of research. However, the pandemic also disrupted traditional research workflows and introduced new challenges. This paper aims to understand how the pandemic specifically altered the patterns of collaboration among scientific institutions. It seeks to answer: How did institutions adapt their collaboration strategies in response to the pandemic? Did the crisis lead to more open or more conservative collaboration? And how did these patterns differ across fields and for institutions involved in pandemic-specific research?

Methods

The authors developed a novel approach to analyze institutional collaboration:

  • Weighted Cooperation Network: Instead of simply counting co-authored papers, they introduced a weighted network that accounts for the attention an institution gives to a collaboration. This is crucial because a paper with many authors from one institution and few from another implies different levels of engagement and influence. They proposed a method to calculate this weight based on author distribution within papers.
  • Institution Cooperation Freshness (ICF): This metric quantifies how "fresh" an institution's collaborations are. It's defined as the proportion of new collaborators an institution engages with in a given period, relative to its total collaborators. An ICF of 1 means all collaborators are new, while 0 means all have collaborated before. This helps understand whether institutions are seeking novel partnerships or sticking with established ones.
  • Causal Inference (GPSM): To understand the causal relationship between collaboration strategies (ICF) and an institution's influence (measured by PageRank), they employed Generalized Propensity Score Matching (GPSM). This method allows for analyzing the effect of a continuous treatment variable (ICF) on an outcome variable (change in influence).

Datasets

The study utilized the Microsoft Academic Graph (MAG) database, a large collection of academic documents. Specifically, they analyzed:

  • 2,252,016 journal papers published between 2019 and 2020.
  • This dataset was used to construct institution cooperation networks and analyze collaboration patterns.
  • Papers published in 2020 were further categorized into "COVID-19 related" (identified by keywords) and "non-COVID-19 related" for comparative analysis.

Key Insights and Ideas

  • Pandemic as a Catalyst for Both Stability and Novelty: The crisis didn't just push institutions to collaborate more; it fostered both the strengthening of existing ties and the formation of new ones. This suggests a dual strategy of leveraging established trust while seeking fresh perspectives.
  • Increased Caution in Partner Selection: Despite the push for collaboration, institutions, particularly those involved in COVID-19 research, became more selective. This "cautious strategy" likely stems from the inherent risks and uncertainties of a global crisis, where resource allocation and research direction are critical.
  • Influence Dictates Strategy: Highly influential institutions (top 25% PageRank) displayed a wider range of ICF values, indicating they employ diverse strategies – some conservative, some aggressive. Less influential institutions tended to be more conservative, sticking to known collaborators.
  • Field-Specific Collaboration Norms: STEM fields (biology, computer science, engineering, medicine) showed a bimodal ICF distribution, suggesting a willingness to engage in both familiar and novel collaborations. Humanities fields, however, leaned towards more conservative, less "fresh" collaborations.
  • COVID-19 Research Demands Novelty (but with caution): Institutions working on COVID-19 research showed a higher propensity for new collaborators (higher ICF) compared to non-COVID-19 research. This might be driven by the urgent need for diverse expertise and innovative solutions, even if it means navigating less familiar partnerships.

Key Results

  • The number of cooperative relationships increased significantly (19.4%) from 2019 to 2020, indicating a surge in collaboration during the pandemic.
  • Institutions with higher influence (PageRank) showed greater dispersion in their ICF values, suggesting varied collaboration strategies.
  • Institutions involved in COVID-19 research had a higher average ICF compared to those in non-COVID-19 research, with a statistically significant difference of 0.052 (Table 3).
  • The GPSM analysis indicated that institutions with higher ICF (more new collaborators) experienced a larger fluctuation in their influence (PageRank), suggesting a higher probability of facing both opportunities and risks associated with new partnerships.

Limitations and Future Work

  • Short Time Window: The study's duration is limited to two years (2019-2020). Future research should explore longer timeframes to understand the long-term impact of the pandemic on collaboration.
  • Journal Paper Focus: The analysis is based solely on journal papers, potentially missing collaborations documented in other publication types.
  • Unaccounted Factors: The study did not consider external factors like differing national collaboration policies, the lag in scientific research cooperation, talent crises, economic recessions, or broader societal contexts, which could influence collaboration patterns.
  • Future Directions: The authors suggest incorporating more diverse data sources, longer time windows, and a broader range of influencing factors to gain a more nuanced understanding of research collaboration dynamics in the post-pandemic era. They also emphasize the value of this research for informing policy during future global crises.

Summary updated

Abstract

During the COVID-19 pandemic, scientific collaboration has profoundly influenced the sustainability of human science. The pandemic has exposed the fragility of the global health system, forcing various research institutions to cooperate more actively and giving rise to a pressing inquiry regarding its implications on the patterns of scientific institution collaboration. With this in mind, we investigate 2,252,016 scientific papers between 2019 and 2020 to study this issue from an institution cooperation view. Specifically, by proposing a novel weighted cooperation network and utilizing institution cooperation freshness, we investigate different collaboration patterns and strategies when institutions respond to the pandemic. The results show that the crisis generally promotes both fresh and stable cooperation relationships among institutions, and different fields (e.g., STEM and Humanities) demonstrate different collaboration patterns. During the pandemic, institutions with a strong influence have a variety of cooperation strategies, while institutions with a low influence prefer a conservative strategy. Moreover, institutions participating in COVID-19 research are very cautious in choosing partners, which may originate from the risk of cooperation, e.g., the fluctuation of influence, and the causal inference results also verify this conclusion. Overall, our study reveals the impact of the COVID-19 pandemic on scientific research work, which may pave the way for designing policy when other global crises occur again.
BibTeX
@inproceedings{fu2023collaboration,
  author = {Chenbo Fu and Xuejiao Liang and Yong Ki Min and Qi Xuan and Shanqing Yu},
  title = {{The Collaboration Patterns of Institution during the COVID-19 Pandemic}},
  booktitle = {Sustainability},
  year = {2023},
  doi = {10.3390/su152416915},
}
Figure