Modeling Scientists as Agents. How Scientists Cope with the Challenges of the New Public Management of Science

Asso for Info Science & Tech

2019

The impact of university rankings has driven scholars to increase productivity, and expanding collaboration has dramatically changed the publishing patterns in the academic publication system. The aim of this study was to develop a scholar classification scheme from publication patterns in academic science. Classification schemes are ways of describing groups that display different clusters of behaviors, approaches, or perspectives, and useful in the development of typologies. In this research, sample data are selected from three representative universities, considered a leading university, a middle-tier university, and a noncomprehensive university. A final set of 11,427 effective scholars and their 284,128 journal publication records were used to develop the classification scheme via cluster analysis. The results identify six types of scholars, labeled as: singleton (8%), small-team low performer (16%), smallteam high performer (17%), big-team strategist (22%), free-style follower (21%), and life-time warrior (17%). These six scholar types demonstrate different approaches to publishing that can be used to understand both individual and research team performance across different institutional settings. Additionally, possible future work was identified that uses the scholar classification scheme to define the behavior for agents in an agent-based model to simulate the strategicbehavior-driven academic publication system.

Section: Resultsmentioning

confidence: 75%

“…Besides the empirical study on the Bayer classification, Mölders, Fink, and Weyer () proposed a hypothetical scholar typology. In their typology, scholars are categorized as either a Careerist , Orthodox Scientist , or Mass Producer .…”

Section: Introductionmentioning

confidence: 99%

Developing a scholar classification scheme from publication patterns in academic science: A cluster analysis approach

Asso for Info Science & Tech

2019

“…ABMS has been widely used to explore academic activities, with 12 of the 35 studies assigned to this category. These studies involve talent recruitment (Caillou & Sebag, 2009), talent management (Phelan, 2004;Stepanic, Pejić Bach, & Kasac, 2013;Zheng & Lei, 2010), academic collaboration (Fujii & Tanimoto, 2003;Kenna & Berche, 2012), publication (Gilbert, 1997;Mölders, Fink, & Weyer, 2011), peer review (Cabota, Grimaldo, & Squazzoni, 2013;Paolucci & Grimaldo, 2012;Squazzoni & Gandelli, 2012), and academic-level research funding (Hoser, 2013). Caillou and Sebag (2009) implemented an ABMS application to investigate how universities can recruit the best employee candidates with high confidence.…”

Section: What Specific Dynamics Do Existing Abms Applications Cover In He?mentioning

confidence: 99%

“…Similarly, Mölders et al (2011) proposed conceptual model investigates how the scientific publication system works. In the proposed agent-based model, individual academics rationally decide where to publish their papers.…”

Section: What Specific Dynamics Do Existing Abms Applications Cover In He?mentioning

confidence: 99%

A systematic review of agent-based modelling and simulation applications in the higher education domain

Higher Education Research & Development

2015

This paper presents the results of a systematic review of agent-based modelling and simulation applications in the higher education (HE) domain. Agent-based modelling is a 'bottom-up' modelling paradigm in which system level behaviour (macro) is modelled through the behaviour of individual local-level agent interactions (micro). This approach to considering the behaviour of systems of interacting 'agents' has been applied to a wide variety of domains. Of particular interest are the ways that ABMS applications have been used to further understand the dynamics of the HE domain. We conduct a systematic review of literature to analyse publications by year, role of the simulator, development stage of the models, and any associated validation. We also identify areas for future work, which includes an emphasis on validating existing and future models, detailed description of simulations to allow replication and further development, and the use of agent-based models in other contexts within the increasingly complex HE domain.

“…Changes in the journal publishing landscape have created a dynamic system in which individual actors (agents) seek to maximise their performance, usually in terms of standard publication or citation metrics. Modelling approaches such as agent-based modelling may help clarify and even predict the behaviour of actors as well as the performance of the system as a whole (Axelrod, 1997;Bonabeau, 2002;Gu, Blackmore, Cornforth, & Nesbitt, 2015;Macal & North, 2009;Mölders, Fink, & Weyer, 2011). Increasing specialization, the proliferation of journals, a culture of assessment, and greater awareness of journal characteristics and citation metrics have led to heightened competition for space in the top journals as well as competition among journals to capture the most important contributions to the literature (Statzner & Resh, 2010).…”

Section: Introductionmentioning

confidence: 99%

Characterisation of academic journals in the digital age

2017

Scientometrics

Innovations in scholarly publishing have led to new possibilities for academic journals (e.g., open access), and provided scholars with a range of indicators that can be used to evaluate their characteristics and their impact. This study identifies and evaluates the journal characteristics reported in five databases: Ulrich's Periodicals Directory (Ulrichs), Journal Citation Reports (JCR), SCImago Journal & Country Rank (SJR), Google Scholar Metrics (GS), and Cabell's Periodical Directory (Cabells). It describes the 13 indicators (variables) that are available through these databases-scholarly impact, subject category, age, total articles, distribution medium, open access, peer review, acceptance rate, pricing, language, country, status, and issue frequency-and highlights the similarities and differences in the ways these indicators are defined and reported. The study also addresses the ways in which this kind of information can be used to better understand particular journals as well as the scholarly publishing system.