Team formation and team impact: The balance between team freshness and repeat collaboration

https://doi.org/10.1016/j.joi.2022.101337Get rights and content

Highlights

  • We extend an indicator, i.e., team freshness, to measure the extent to which a scientific team incorporates new members, by calculating the fraction of new collaboration relations established within the team.

  • Team freshness in scientific teams has been increasing in the past half-century.

  • There exists an inverted-U-shaped association between team freshness and papers’ citations in all the disciplines and in different periods.

  • Team impact is hampered by team freshness in small-sized teams, while medium-sized and large-sized teams can benefit more from team freshness before the fraction of new collaboration reaches its turning point.

Abstract

Incorporating fresh members into teams is considered a pathway to team creativity. However, whether freshness improves team performance or not remains unclear, as well as the optimal involvement of fresh members for team performance. Focusing on team impact, one important dimension of team performance, this study uses a group of authors on the byline of a publication as a proxy for a scientific team and quantifies team impact by citations of a paper authored by this team, i.e., article team impact. We extend an indicator, i.e., article team freshness, to measure the extent to which a scientific team incorporates new members, by calculating the fraction of new collaboration relations established within the team. Based on more than 43 million scientific publications covering more than a half-century of research from Microsoft Academic Graph, this study provides a holistic picture of the current development of article team freshness by outlining the temporal evolution of freshness, and its disciplinary distribution. Subsequently, using a multivariable regression approach, we examine the association between article team freshness and papers’ short-term and long-term citations. The major findings are as follows: (1) article team freshness in scientific teams has been increasing in the past half-century; (2) there exists an inverted-U-shaped association between article team freshness and papers’ citations in all the disciplines and different periods; (3) article team impact is hampered by article team freshness in small-sized teams, while medium-sized and large-sized teams can benefit more from article team freshness before the fraction of new collaboration reaches its turning point. The findings of this study provide implications for the practice of team formation and team management in science.

Introduction

Teamwork is increasingly pervasive and drives innovations in the contemporary scientific landscape (Liu et al., 2020; Milojević, 2014; Hu, 2021; Yang et al., 2021). Compared to singular knowledge production, collaborative work can yield more discoveries and breakthroughs (Liu et al., 2020; Wuchty et al., 2007). Despite considerable efforts to explore the heterogeneity of team members, we know little about how teams are formed to incorporate fresh members who have no prior collaboration with other team members, and incumbents who have repeat collaborative connections with other team members, and the relationship between team freshness and team impact.

The evolution of team freshness might be accompanied by historical changes in science. Contemporary science has witnessed an exponential growth of scientific publications and a rise in the volume of the scientific labor force (Baskaran, 2017; Fortunato et al., 2018; Price, 1963). Scientific research is characterized by a fundamental shift toward team-based research (Rawlings & McFarland, 2011; Wuchty et al., 2007). Due to the increasing specialization in science (Cole & Harriet, 2017; Evans, 2016; Jones, 2009; Moody, 2004), the need to combine diverse and interdisciplinary knowledge and skills to address complex research problems (Katz & Martin, 1997) and the growing costs of scientific facilities and instruments (Shrum et al., 2007), in almost all branches of science, scientists are increasingly involved in teamwork (Wuchty et al., 2007). In addition, advances in information and communication technologies (Binz-Scharf et al., 2015; He & Berry, 2022) and reductions in travel costs (Katz & Martin, 1997), have made constructing new collaboration links easier. These changes have led to the increasing need to involve new members in teams for diverse and distant knowledge, reduced the costs of constructing fresher teams, and might change the development of team freshness. To understand the temporal changes in team freshness from a historical standpoint, we propose RQ1. How has team freshness evolved in the past decades?

Fresh members enhance team learning and thus spur team creativity by bringing new knowledge and novel perspectives (Perretti & Negro, 2007; Rosenkopf & Almeida, 2003; Skilton & Dooley, 2010). Teams might become less creative over time due to groupthink, homogeneity, and less tendency to disturb the status quo (West & Anderson, 1996). The entry of fresh members could generate conflicts and divergent opinions that can trigger creativity (Badke‐Schaub et al., 2010; Farh et al., 2010; Kane et al., 2005; Santos et al., 2015). In this sense, team freshness could lead to high team performance. A recent study found that team freshness is positively related to a paper's originality and multi-disciplinary impact, which is consistent with this argument (Zeng et al., 2021).

However, too much team freshness could be risky and harmful to team impact due to the high cost of forming new ties (Jackson et al., 1992) and fresh members’ adaption (Chen, 2005), and less trust and familiarity (Van Der Vegt et al., 2010). From the psychological perspective, repeat collaboration entails greater certainty, trust, and reciprocity, and more efficient knowledge transfer, all of which offset the negative consequences of team freshness and thus facilitate team impact (Dahlander & McFarland, 2013; Uzzi, 1997). However, high levels of repeat collaboration could dampen the creation of innovative ideas by reducing collaboration efficiency, homogenizing the pool of knowledge, narrowing the search spaces of teams, and reducing conflicts (Guimera et al., 2005; Porac et al., 2004). Due to the co-existing benefits and disadvantages of team freshness and repeat collaboration, empirical evidence shows that the combination of team freshness and repeat collaboration leads to the best team performance (Guimera et al., 2005; Perretti & Negro, 2007). The mixed arguments about the positive and negative consequences of freshness and repeat collaboration prompt us to explore the “bliss point” between team freshness and repeat collaboration that brings the optimal team impact. Thus, we raise RQ2. What is the relationship between team freshness and team impact?

The relationship between team freshness and team impact might vary with changes in team size. Whether and how team freshness shapes team impact relies on whether the benefits and detriments caused by team freshness outweigh each other, which might be different across team sizes. Fewer and simpler links are embedded in small teams where negative impacts caused by a high level of team freshness could disrupt the whole team (Zhang et al., 2020) and thus worsen team impact, while negative impacts of team freshness might be subtle in large teams and so barely influence team performance. A few empirical studies suggest that stability is more important for small teams’ survival, and that large teams gain benefits from membership dynamics (Palla et al., 2007; Zhang et al., 2020). However, it is still unclear whether and how the relationship between team freshness and team performance is shaped by team size. Thus, we propose RQ3: Is the relationship between team freshness and team impact in small teams different from that in large teams?

To address the three research questions, based on more than 43 million publications between 1950 and 2018 from Microsoft Academic Graph, the research objectives of this paper are threefold: to provide a comprehensive introduction to the evolution of team freshness, to explore how freshness is related to team impact, and to investigate whether and how the relationship between team freshness and team impact depends on team size. This study contributes to the existing literature in multiple dimensions. A better understanding of the balance between freshness and repeat collaboration in teams could improve our knowledge of how the combination of team members’ characteristics relates to team outcomes from the perspective of dynamic team formation. In addition, turnover of members that involves the arrival of new members or the departure of incumbents is increasingly common in scientific teams, which can have profound consequences for team performance by altering the distribution of knowledge and skills within teams, and the relations among team members (Levine et al., 2005). From a practical perspective, the investigation into the balance of freshness and repeat collaboration sheds light on how teams achieve the best performance by maintaining a certain proportion of incumbents and absorbing some fresh members. The remainder of the paper is organized as follows. The related work section reviews the current state of the art, and is followed by section 3, where data and methodology are introduced. Section 4 presents the results. The last section discusses findings and implications for relevant policies.

Section snippets

Related work

In this section, we review three strands of literature concerning the three research questions. We review the literature on team freshness and repeat collaboration, and further provide a discussion on how team size might shape the relationship between team freshness and team impact.

Data

In this study, we use Microsoft Academic Graph (MAG), a heterogeneous graph comprised of 173.67 million scientific papers published between the years 1800 and 2018. MAG provides related information about the publications, such as institutions, venues, the field of study and citation relationships. Due to its comprehensive coverage, MAG has become an important resource for scholarly communication studies in recent years (Cui et al., 2022; Huang et al., 2022; Kanakia et al., 2019; Wang et al.,

RQ1: The temporal evolution of article team freshness in the past half-century

Generally, article team freshness increased over time. Article team freshness has grown over years from 1970 to 2000, remained stable in the 2000s, and increased again after 2010 (see Fig. 2(a)). The average article team freshness reached 0.443 in 2018, which means that nearly 44% of author pairs in an article team have no prior collaboration, on average. To investigate the extreme cases of article team freshness, we calculate the proportion of papers with freshness of value 0 and that of value

Discussions and conclusion

Newcomers and old members bring freshness and experience to teams, respectively. Based on more than 43 million research articles published between 1950 and 2018 from the MAG dataset, we extend a measurement that quantifies freshness in teams, investigate the temporal evolution of freshness, and explore its correlation to papers’ citations and how this association changes with the variations in team size.

On average, we observe that article team freshness is growing over time,3

Supplementary Information

Supplementary Information (SI) is available for this paper: https://zenodo.org/record/7124156#.YzVyA3ZBzGJ.

CRediT authorship contribution statement

Meijun Liu: Conceptualization, Writing – original draft, Data curation, Investigation. Ajay Jaiswal: Data curation, Writing – original draft. Yi Bu: Conceptualization, Supervision, Investigation, Writing – review & editing. Chao Min: Data curation. Sijie Yang: Data curation. Zhibo Liu: Data curation. Daniel Acuña: Conceptualization. Ying Ding: Conceptualization, Supervision, Writing – review & editing.

Acknowledgments

This work is supported by the Youth Program of National Natural Science Foundation in China (No: 72104054 and 72104007), the Youth Project of Humanities and Social Sciences of the Ministry of Education (MOE) of China (No: 21YJC870001), Shanghai Pujiang Program (21PJC026) and “ISTIC-Taylor & Francis Joint Lab for Academic Frontier Observation Open Fund”. The authors deeply appreciate the constructive comments from the reviewers.

References (133)

  • Y.-N. Lee et al.

    Creativity in scientific teams: Unpacking novelty and impact

    Research policy

    (2015)
  • M. Liu et al.

    Will collaborators make scientists move? A Generalized Propensity Score analysis

    Journal of Informetrics

    (2021)
  • M. Liu et al.

    Further divided gender gaps in research productivity and collaboration during the COVID-19 pandemic: Evidence from coronavirus-related literature

    Journal of Informetrics

    (2022)
  • J.F. Porac et al.

    Human capital heterogeneity, collaborative relationships, and publication patterns in a multidisciplinary scientific alliance: a comparative case study of two scientific teams

    Research Policy

    (2004)
  • C.M. Rawlings et al.

    Influence flows in the academy: Using affiliation networks to assess peer effects among researchers

    Social Science Research

    (2011)
  • D.W. Aksnes et al.

    Citations, citation indicators, and research quality: An overview of basic concepts and theories

    Sage Open

    (2019)
  • H.E. Aldrich et al.

    Environments of organizations

    Annual Review of Sociology

    (1976)
  • H. Arrow et al.

    Membership matters: How member change and continuity affect small group structure, process, and performance

    Small Group Research

    (1993)
  • R. Axelrod et al.

    The evolution of cooperation

    Science

    (1981)
  • U. Backes-Gellner et al.

    Effort provision in entrepreneurial teams: effects of team size, free-riding and peer pressure

    Journal of Business Economics

    (2015)
  • K. Badar et al.

    Knowledge network centrality, formal rank and research performance: evidence for curvilinear and interaction effects

    Scientometrics

    (2015)
  • P. Badke-Schaub et al.

    How does cognitive conflict in design teams support the development of creative ideas?

    Creativity and Innovation Management

    (2010)
  • F. Barjak et al.

    International collaboration, mobility and team diversity in the life sciences: impact on research performance

    Social geography

    (2008)
  • A. Baskaran

    UNESCO science report: Towards 2030

    Institutions and Economies

    (2017)
  • M.C. Binz-Scharf et al.

    Making science: New generations of collaborative knowledge production

    American Behavioral Scientist

    (2015)
  • M. Bordons et al.

    Advantages and limitations in the use of impact factor measures for the assessment of research performance

    Scientometrics

    (2002)
  • Y. Bu et al.

    Understanding persistent scientific collaboration

    Journal of the Association for Information Science and Technology

    (2018)
  • C. Catalini et al.

    How do travel costs shape collaboration?

    Management Science

    (2020)
  • G. Chen

    Newcomer adaptation in teams: Multilevel antecedents and outcomes

    Academy of Management Journal

    (2005)
  • J.R. Cole et al.

    The emergence of a scientific specialty: The self-exemplifying case of the sociology of science

    The idea of social structure

    (2017)
  • T. Cole et al.

    Non-zero-sum collaboration, reciprocity, and the preference for similarity: Developing an adaptive model of close relational functioning

    Personal Relationships

    (2004)
  • D. Cooper et al.

    Integrator or gremlin? Identity partnerships and team newcomer socialization

    Academy of Management Review

    (2021)
  • O.C.A. COUNCIL

    Informing Research Choices: Indicators and Judment

    The Expert Panel on Science Performance and Research Funding

    (2012)
  • Cui, H., Wu, L., & Evans, J. A. (2022). Aging Scientists and Slowed Advance. arXiv preprint...
  • J.N. Cummings et al.

    Group heterogeneity increases the risks of large group size: A longitudinal study of productivity in research groups

    Psychological Science

    (2013)
  • L. Dahlander et al.

    Ties that last: Tie formation and persistence in research collaborations over time

    Administrative Science Quarterly

    (2013)
  • N. De Bellis

    Bibliometrics and citation analysis: from the science citation index to cybermetrics

    (2009)
  • V.U. Druskat et al.

    Group emotional intelligence and its influence on group effectiveness

    The emotionally intelligent workplace: How to select for, measure, and improve emotional intelligence in individuals, groups and organizations

    (2001)
  • E.D. Evans

    Measuring interdisciplinarity using text

    Socius

    (2016)
  • M. Färber et al.

    The Microsoft Academic Knowledge Graph enhanced: Author name disambiguation, publication classification, and embeddings

    Quantitative Science Studies

    (2022)
  • J.-L. Farh et al.

    Task conflict and team creativity: a question of how much and when

    Journal of Applied Psychology

    (2010)
  • S. Fortunato et al.

    Science of science

    Science

    (2018)
  • R.B. Freeman et al.

    Why and wherefore of increased scientific collaboration

    The changing frontier: Rethinking science and innovation policy

    (2014)
  • Funk, R. J., & Owen-Smith, J. (2012). A dynamic network approach to breakthrough innovation. arXiv preprint...
  • E. Garfield

    Journal impact factor: a brief review

    Cmaj

    (1999)
  • W. Glänzel et al.

    Journal impact measures in bibliometric research

    Scientometrics

    (2002)
  • M. Granovetter

    Economic action and social structure: The problem of embeddedness

    American journal of sociology

    (1985)
  • R. Guimera et al.

    Team assembly mechanisms determine collaboration network structure and team performance

    Science

    (2005)
  • R.A. Guzzo et al.

    Teams in organizations: Recent research on performance and effectiveness and Marcus W. Dickson

    Annual Review of Psychology

    (1996)
  • R.F. Haans et al.

    Thinking about U: Theorizing and testing U-and inverted U-shaped relationships in strategy research

    Strategic Management Journal

    (2016)
  • Cited by (6)

    View full text