Library and Information Science (LIS) has been undergoing a radical change since the 1980s when some universities closed their traditional library schools (Wiggins & Sawyer, 2010) as the iSchool movement began (Shu & Mongeon, 2016). Then LIS has gradually become an interdisciplinary field (Tang, 2004) ingesting the library science, information science, computer science and other fields (Bruce, 2011). As an original contribution to the advancement of knowledge (Johnson, 2009; O'Connor & Park, 2001), the doctoral research topics has been used to investigate the LIS disciplinary identify (Sugimoto, Li, Russell, Finlay, & Ding, 2011) and its interdisciplinary relations (Shu, Larivière, Mongeon, Julien, & Piper, 2016); but LIS doctoral research co-authorship network has never been investigated. The purpose of this study is to investigate the evolution of the network of LIS doctoral research collaboration, which reflects the expanding LIS research universe.
Literature Review
Scholars with shared research interests collaborate with each other and form communities (Girvan & Newman, 2002) that play important roles in knowledge creation (Lambiotte & Panzarasa, 2009). Co-authorship networks provide a copious and meticulously documented record of the social and professional networks of authors (Newman, 2004); they can therefore be used to understand the research landscape within or between disciplines (Biscaro, Giupponi, & Ouzounis, 2014).
An increase in the interdisciplinarity in LIS research is well documented.by Tang (2004) and Shu et al. (2016). Chang and Huang (2012) report an increase in collaborations between LIS doctoral students and researchers affiliated with non-LIS institutes, in which LIS PhDs could benefit from the collaborations and improve their publication productivity (Kamler, 2008; Lariviere, 2012). However, no study has investigated the evolution of the LIS doctoral co-authorship network.
Methodology
First, a manually validated list of doctoral students who graduated between 1960 and 2013 and their advisors was compiled using the MPACT database (MPACT, 2010), which stores all LIS doctoral graduates from 1930 to 2009. Second, LIS doctoral students who graduated on or after 2010 and their advisors were identified and added to the list by searching the ProQuest Thesis and Dissertation Database and corresponding university websites. This process produced a list of 3,561 LIS doctoral graduates and their 928 doctoral advisors.
The papers published by the identified graduates during their supervised doctoral studies, defined as between six years before and two years after graduation, were retrieved from the Web of Science (WoS). Based on the journals in which the papers were published, all publications were categorized into 144 disciplines (LIS is one of 114 disciplines) according to the NSF classification system, which assigns each journal to a single discipline. All authors and the affiliated institutions listed on the papers were identified.
Findings
From 1960 to 2013, 3,561 doctoral students graduated from 44 LIS programs. The number of LIS doctoral graduates has increased from 18 in 1960 to 114 in 2013, peaking at 116 in 2010. Excluding128 students whose advisors were not identified, 3,433 LIS doctoral students were supervised by 928 advisors. 469 advisors (50.5%) obtained a doctoral degree in LIS supervised 2,097 LIS doctoral students (61.1%), and the remaining 459 advisors (49.5%) graduated from non-LIS fields and supervised 1,336 students (38.9%).
Only 26.1% (930/3,561) of LIS doctoral graduates published at least one paper indexed by the WoS during their doctoral studies. The percentage of published students increased from 3.5% in the 1960s to 42.8% in the 2010s. These 930 LIS doctoral graduates contributed 1,804 papers of which 75.2% (1,357/1,804) are published in a LIS journal; they also published papers in journals in Computers (8.0%), Law (2.6%), Management (2.4%), Communication (2.1%) and 36 other disciplines. The percentage of papers published in LIS journals has been decreasing from 90.0% in the 1960s to 59.7% in the 2010s.
1,218 of these 1,804 papers are co-authored papers, including 616 papers showing collaborations within the same institution and 602 papers between different institutions. 593 of 984 (60%) external collaborators are affiliated with non-LIS institutes in co-authorship between different institutions. Wisconsin-Madison is the largest source institution in terms of the number of LIS collaborators while Penn State is the largest non-LIS contributor. A visual mapping (see Appendix) presents the LIS doctoral co-authorship network from the 1970s to 2010s. The co-authorship network is shown as 9 separated small clusters in the 1970s while a big cluster and 5 other small clusters appear in the 1980s. The meaningful co-authorship network emerges in the 1990s; the number of collaborators from a LIS institution (red nodes) and from a non-LIS institution (red nodes) are the same in the 1990s but the percentage of non-LIS collaborators increased from 50% in the 1990s to 66% in the 2010s.
In addition, LIS doctoral students collaborated with more non-LIS collaborators (79%, 232 out of 294) when publishing the paper in non-LIS journals; but the ratio of non-LIS collaborators is only 52% (361 out of 690) when the co-authored papers were published in a LIS paper. The impact of advisors’ disciplinary background on students’ collaborators’ background is not significant. LIS doctoral students supervised by non-LIS advisors collaborated with more non-LIS collaborators compared with those supervised by LIS advisors (LIS supervision: 59%; non-LIS supervision: 62%).
Conclusion
This study presented an analysis of LIS doctoral co-authorship network since the 1970s, which showed a trend in collaboration with researchers affiliated with non-LIS institutes. Both the journals in which their papers are published and the advisors’ disciplinary background have impact on LIS students’ collaborators’ background. The evolution of LIS doctoral co-authorship network reflects the expansion of LIS research as more and more external collaboration with researchers from non-LIS instutions.
Reference
Biscaro, C., Giupponi, C., & Ouzounis, C. A. (2014). Co-Authorship and Bibliographic Coupling Network Effects on Citations. PLoS ONE PLOS ONE, 9 (6), e99502.
Bruce, H. (2011). The Audacious Vision of Information Schools. Journal of Library and Information Science (Taipei), 37 (1), 4-10.
Chang, Y.-W., & Huang, M.-H. (2012). A study of the evolution of interdisciplinarity in library and information science: Using three bibliometric methods. Journal of the American Society for Information Science and Technology, 63 (1), 22-33.
Girvan, M., & Newman, M. E. (2002). Community structure in social and biological networks. Proceedings of the National Academy of Sciences of the United States of America, 99 (12), 7821-7826.
Johnson, I. (2009). Education for Librarianship and Information Studies: fit for purpose? Information Development, 25 (3), 175-177.
Kamler, B. (2008). Rethinking Doctoral Publication Practices: Writing from and beyond the Thesis. Studies in Higher Education, 33 (3), 283-294.
Lambiotte, R., & Panzarasa, P. (2009). Communities, knowledge creation, and information diffusion. JOI Journal of Informetrics, 3 (3), 180-190.
Lariviere, V. (2012). On the Shoulders of Students? The Contribution of PhD Students to the Advancement of Knowledge. Scientometrics, 90 (2), 463-481.
MPACT. (2010). The MPACT Project. http://www.ibiblio.org/mpact/
Newman, M. E. (2004). Coauthorship networks and patterns of scientific collaboration. Proceedings of the National Academy of Sciences of the United States of America, 101 , 5200-5205.
O'Connor, D., & Park, S. (2001). Crisis in LIS research capacity. Library and Information Science Research, 23 (2), 103-106.
Shu, F., Larivière, V., Mongeon, P., Julien, C.-A., & Piper, A. (2016). On the Evolution of Library and Information Sci
