Conference Agenda

Scientific networks are key to understanding knowledge creation, and a rich tradition of studies investigates how networks of individual researchers or teams influence their success and creativity. However, less attention has been paid to the question of how networks vary between fields. In this study, we investigate how network processes—such as preferential attachment, homophily, and selection, as well as influence based on similar intellectual orientations—vary across scientific fields for the first time. We introduce a theoretical framework that extends a network ecological perspective to the domain of science and test it by analyzing collaboration and citation networks among ~4.8 million authors and ~12.5 million research articles in over 800 scientific fields derived from the PubMed knowledge graph and OpenAlex. We use relational hyper-event models (RHEMs) to study which contextual moderators—e.g., size, demographic composition, funding type and volume—foster certain network processes. For instance, our analysis allows us to investigate whether preferential attachment in collaboration and citation networks is stronger in fields that rely more on external funding or whether fields with an unbalanced gender composition show more gender homophily in collaborations. Our study makes two primary contributions: first, it provides the first large-scale analysis of network processes in biomedical research; second, it examines how these processes vary across fields. Our results are of interest to scientists and policymakers alike, as they allow us to map which contextual characteristics contribute to detrimental network processes, such as strong gender homophily or an excessive accumulation of resources by scientific elites.

In this comprehensive study, a group of researchers evaluated the reporting practices of relational data across 363 articles published in the REDES journal from 2002 to 2023. The approach involved a meticulous analysis of keywords, themes, and network metrics documented in these articles, followed by an assessment against 18 established reporting recommendations. This method allowed us to identify best practices and potential areas for standardization enhancement in future publications.

The findings reveal a broad range of network metrics and significant diversity in reporting styles. These variations underscore a critical challenge in ensuring comparability and replicability across studies, highlighting the pressing need for standardized reporting formats. A detailed examination shows inconsistent implementation of existing guidelines, pointing towards opportunities for refining these standards to improve their practical application.

The results emphasize the necessity of clear and coherent guidelines that aid in peer review processes, help train emerging researchers, and enhance consistency in scholarly outputs. The proposed recommendations advocate for precise definitions of network boundaries, thorough descriptions of data collection methods, and clear operationalization of network links. Adopting these practices will enhance the transparency and quality of network research, ultimately supporting more robust, replicable, and theoretically sound studies.

The researchers advocate for ongoing revisions of these guidelines to keep pace with methodological advancements in network analysis and evolving academic needs.

The growing complexity of networks has driven the development of various methodological approaches to uncover structural patterns, with applications across multiple fields. In the context of scientific collaboration, these methods offer valuable insights into the pathways of knowledge transfer.

This study introduces a novel approach for analyzing scientific collaboration by applying community detection and core-periphery analysis to a co-authorship network, where nodes represent authors and edges represent co-authored publications. Our analysis examines the collaboration patterns among Italian academic scholars from three distinct fields (statistics, sociology, and business) over a ten-year period (2012–2022). The study utilizes publication data sourced from Scopus and the Ministero dell'Istruzione e del Merito (MUR).

Unlike traditional core-periphery detection methods that focus on the classification of individual actors, our approach identifies core and peripheral communities within the network. To achieve this, we develop an innovative methodological framework building on a stochastic block model and optimizing an objective function that incorporates both inter-community connection density and strength. We further examine how author-specific characteristics, i.e. the academic field, influence community formation.

Our findings provide a foundation for assessing collaborative trends, knowledge diffusion, and the role of cohesive subgroups, contributing to a broader understanding of research networks.

An ambiguous idea rarely remains confined to a specific domain. Some, like free speech, values, and diversity are applied to broad swaths of society, becoming relevant to a wide array of citizens in many different contexts. The wide application of ambiguous ideas creates different emergent and evolutionary pathways for their development and use, resulting in a proliferation of literature and competing or even contradictory terminology. Developing insight into such ideas that diverge or coalesce within a community requires tools capable of untangling their lineage across a wide array of academic disciplines over several decades.

We will present three innovative applications of Social Network Analysis (SNA) methods to explore the emergence and evolution of accountability in higher education across a collection of 450 peer-reviewed articles published from 1974-2017 and their corresponding 12,270 references. First, we integrate qualitative data from articles and references into new interactive joint displays called Narrated Network Diagrams, creating opportunities to more accurately assess themes and meanings in literature by connecting structures in co-citation networks with relevant relational stories. Second, we elevate time in the analysis procedure to capture the dynamism of knowledge formation. Third, underutilized descriptive network statistics are applied to the co-citation network analysis to generate new insights such as different mechanisms for authors gaining influence in a knowledge community. Ultimately, we will present an innovative longitudinal Mixed Methods Social Network Analysis (MMSNA) approach to systematic literature reviews, significantly advancing previous SNA methods integration in this critical research practice.