View in EDS HTML Full Text PDF Full Text

Using Social Network Analysis to Complete Literature Reviews: A New Systematic Approach for Independent Researchers to Detect and Interpret Prominent Research Programs within Large Collections of Relevant Literature

Saved in:
Bibliographic Details
Title: Using Social Network Analysis to Complete Literature Reviews: A New Systematic Approach for Independent Researchers to Detect and Interpret Prominent Research Programs within Large Collections of Relevant Literature
Language: English
Authors: Cowhitt, Thomas, Butler, Timothy, Wilson, Elaine
Source: International Journal of Social Research Methodology. 2020 23(5):483-496.
Availability: Routledge. Available from: Taylor & Francis, Ltd. 530 Walnut Street Suite 850, Philadelphia, PA 19106. Tel: 800-354-1420; Tel: 215-625-8900; Fax: 215-207-0050; Web site: http://www.tandf.co.uk/journals
Peer Reviewed: Y
Page Count: 14
Publication Date: 2020
Document Type: Journal Articles
Reports - Evaluative
Descriptors: Social Networks, Network Analysis, Literature Reviews, Systems Approach, Computer Software, Authors, Visual Aids
DOI: 10.1080/13645579.2019.1704356
ISSN: 1364-5579
Abstract: Literature reviews are required at early stages of a traditional research progression. Many systematic approaches help researchers identify relevant literature. However, there is far less support for interpreting large collections of references. Understanding the evolution of knowledge within a discipline requires an awareness of the collaborative networks from which significant advances originate. Unfortunately, this relational awareness is only acquired after years of professional experience. Therefore, early-career researchers are at a fundamental disadvantage when conducting literature reviews. Social Network Analysis (SNA) can make the interpretation of large collections of literature more manageable. This paper provides a method for generating a co-authorship network from reference lists. The partition of the co-authorship network into communities provides a reviewer with direction to organise literature into meaningful groups. The interpretation of seemingly disparate and entangled research programs is far more manageable when interpreting smaller collections of highly relevant literature.
Abstractor: As Provided
Entry Date: 2020
Accession Number: EJ1259596
Database: ERIC
Full text is not displayed to guests.
FullText Links:
  – Type: pdflink
    Url: https://content.ebscohost.com/cds/retrieve?content=AQICAHj0k_4E0hTGH8RJwT4gCJyBsGNe_WN95AvKlDbXJGqwxwGUUvxw14bzYW8mCFjrZix4AAAA4zCB4AYJKoZIhvcNAQcGoIHSMIHPAgEAMIHJBgkqhkiG9w0BBwEwHgYJYIZIAWUDBAEuMBEEDGXcWY7Z-a3FHYHvxwIBEICBmzeG81OkichffsmrZ7huwVP-QYblbYnO2f8So1BaYQSW87mv6ydFnJmnOPU-TngO-Gmm_rY2ZWtEaH2wRIKJzeCdxGrNF11eK0fuxNPC412X02xqkBDZUwgaDJfu59T-Yr1r3eCH3huWiK4gvlT2ec5AlPD2WJ5IsctEHD1tZzar931U-2sV6jNkjG_g0D6yU6Gxi1nWrVXeDL6l
Text:
  Availability: 1
  Value: <anid>AN0144473746;9eb01sep.20;2020Jul11.03:29;v2.2.500</anid> <title id="AN0144473746-1">Using social network analysis to complete literature reviews: a new systematic approach for independent researchers to detect and interpret prominent research programs within large collections of relevant literature </title> <p>Literature reviews are required at early stages of a traditional research progression. Many systematic approaches help researchers identify relevant literature. However, there is far less support for interpreting large collections of references. Understanding the evolution of knowledge within a discipline requires an awareness of the collaborative networks from which significant advances originate. Unfortunately, this relational awareness is only acquired after years of professional experience. Therefore, early-career researchers are at a fundamental disadvantage when conducting literature reviews. Social Network Analysis (SNA) can make the interpretation of large collections of literature more manageable. This paper provides a method for generating a co-authorship network from reference lists. The partition of the co-authorship network into communities provides a reviewer with direction to organise literature into meaningful groups. The interpretation of seemingly disparate and entangled research programs is far more manageable when interpreting smaller collections of highly relevant literature.</p> <p>Keywords: Literature review; systematic review; social network analysis; bibliometrics; citation networks; co-authorship networks</p> <hd id="AN0144473746-2">Introduction</hd> <p>A literature review can be understood as an effort to comprehend and evaluate an academic conversation. Researchers start by tracing backwards through the career writings of many authors, some spanning decades. This critical first step in the research process should identify the seminal work in a discipline, allowing the researcher to produce an original contribution after situating their work within the wider body of knowledge.</p> <p>Given its prominent role in the traditional research progression, improving the literature review process has the potential to save time and enrich the overall outcome. This is especially true for new academics that lack familiarity with prominent authors and concepts. Enhancing literature review strategies could well support novice researchers that lack the contextual understanding necessary to avoid overlooking key authors or exploring research questions which have already been answered.</p> <p>Systematic approaches to conducting literature reviews are an attempt at adding rigor and reliability. Being systematic simply means being explicit as you make sense of relevant literature in your discipline (Petticrew & Roberts, [<reflink idref="bib39" id="ref1">39</reflink>]). Reviewers are now encouraged to write reproducible methods for identifying, selecting, and appraising the work of colleagues (Booth, Papaioannou, & Sutton, [<reflink idref="bib6" id="ref2">6</reflink>]; Fink, [<reflink idref="bib16" id="ref3">16</reflink>]). The process of including or omitting a publication in a review is no longer an unchecked editorial privilege as the reviewer is now expected to develop a technical process, which is rational and reproducible (Jesson, Matheson, & Lacey, [<reflink idref="bib24" id="ref4">24</reflink>]).</p> <p>But making sense of the tangled threads of competing and sometimes contradictory research programs is still daunting, especially in areas of research that are multidisciplinary. Adhering to a systematic protocol does not address the fundamental challenge facing reviewers. The literature review is traditionally undertaken at an early stage in the research process. Researchers are therefore tasked with making sense of large quantities of peer-reviewed writing while lacking the necessary background knowledge to support higher order engagement with the literature.</p> <p>This paper asks the reader to first consider the significant variation in systematic rigor that exists in published literature reviews. A wide spectrum of quality stretches from the work of a supervised graduate student incorporating several systematic approaches within a search protocol to the robust systematic review published by an experienced team of researchers, librarians, and research assistants. Many more literature reviews are conducted independently, without funding, and under significant time constraints. So how can the application of quantitative and qualitative methodologies create new or improve existing systematic literature review strategies?</p> <p>This paper first provides a brief explanation of relevant SNA concepts. An initial explanation of SNA is meant to provide background knowledge so the methods presented in this paper are accessible to all readers, even those without formal SNA training. There are reference texts for readers wanting to pursue more advanced analysis of networks (Barabasi, [<reflink idref="bib1" id="ref5">1</reflink>]; Carolan, [<reflink idref="bib11" id="ref6">11</reflink>]; Daly, [<reflink idref="bib15" id="ref7">15</reflink>]; Kadushin, [<reflink idref="bib25" id="ref8">25</reflink>]; Newman, [<reflink idref="bib34" id="ref9">34</reflink>]; Wasserman & Faust, [<reflink idref="bib46" id="ref10">46</reflink>]).</p> <p>Second, this paper provides a brief review of past applications of networks in literature review methods. Networks have been used most prominently to help researchers systematically chase citations listed in particularly relevant texts to their chosen topic. This paper argues that citation networks are most effectively used when searching for relevant material. However, co-authorship networks are better suited to interpret large collections of references because they are constructed using stronger relational data. This brief review helps differentiate several applications of SNA within different phases of the literature review process.</p> <p>This paper also provides several original contributions. One original contribution is a new R script that allows researchers to transform exported reference lists from databases into separate Excel files that list individual authors and all combinations of co-authorship. Lists of authors and collaboration combinations are the most universal data format for generating networks using networking software.</p> <p>There is a persistent need to better connect popular databases with networking software. Competition between networking software is continuously improving the visualization of networks and making more complex network analysis metrics accessible, even for those without formal training. Easily importing database search results into the many networking software options will make SNA more accessible to reviewers and provide options for the interpretation and presentation of literature.</p> <p>This paper provides a detailed protocol for generating a co-authorship network diagram and partitioning the network into smaller communities. Large reference lists can then be organised into groups of highly related literature based on the community designation of each author. The reviewer can then use many tools to uncover similarities within or differences between the published outputs of co-authorship communities to define prominent research programs.</p> <hd id="AN0144473746-3">A brief explanation of relevant SNA concepts</hd> <p>Social Network Analysis elevates relationships as the primary unit of analysis in research (Borgatti & Ofem, [<reflink idref="bib8" id="ref11">8</reflink>]). Network diagrams can depict individual human actors as small dots, or nodes, and predefined relationships between the human actors as edges, or lines. Network diagrams are therefore useful in visualizing the relational patterns that can influence individual or group behaviour (Newman, [<reflink idref="bib34" id="ref12">34</reflink>]). SNA also provides powerful descriptive and statistical metrics for interpreting networks (Wasserman & Faust, [<reflink idref="bib46" id="ref13">46</reflink>]).</p> <p>When applying SNA to help in the interpretation and presentation of literature, nodes can be used to represent different researchers and edges can connect nodes that co-author a publication. This co-authorship network diagram visualizes patterns of close social relationships between researchers. Understanding this social context is critical when trying to determine prominent lines of enquiry within a larger discipline.</p> <p>By transforming lists of relevant references into network data, the reviewer can apply descriptive and analytical network metrics to differentiate between network actors and also to explore various social mechanisms that propel ideas or authors to greater prominence. For example, each author can be assigned a degree, or a number for how many co-authorship connections they have with other network actors. In a co-authorship network, the total degree of a node represents the number of collaborations that resulted in publication. Comparing the total degree centrality of network actors will yield the names of prolific writers within a larger collection of references.</p> <p>Additionally, SNA allows researchers to partition a whole network into smaller parts, or communities. Network software allows researchers to run many different community detection algorithms with a single click. Once collaborative communities are identified within the co-authorship network, the reviewer can sort long reference lists into more manageable groups based on the community designation of each author. Rigorous analysis of these highly relevant groups of references should provide accurate descriptions of prominent research programs within the larger discipline. This effectively automates the synthesis of literature, allowing the reviewer to more authentically engage with large collections of literature.</p> <hd id="AN0144473746-4">Previous use of networks in literature review methods</hd> <p>Systematic search strategies for identifying relevant literature are well documented. Prominent past research can be identified by systematic search protocols that combine some, if not all of the following methods: searching for previously published reviews, subject searching in multiple databases, manually searching individual volumes of topically relevant journals, chasing citations from particularly compelling literature, and contacting experts. Researchers looking to add systematic approaches to their literature review methods should consult previously published systematic reviews within their discipline. There are also several publications dedicated to explaining systematic approaches for writers in the social sciences (Galvan & Galvan, [<reflink idref="bib17" id="ref14">17</reflink>]; Petticrew & Roberts, [<reflink idref="bib39" id="ref15">39</reflink>]); medicine (Gough, Oliver, & Thomas, [<reflink idref="bib20" id="ref16">20</reflink>]; Khan, Kunz, Kleijnen, & Antes, [<reflink idref="bib26" id="ref17">26</reflink>]); graduate students specifically (Boland, Cherry, & Dickson, [<reflink idref="bib3" id="ref18">3</reflink>]; Galvan & Galvan, [<reflink idref="bib17" id="ref19">17</reflink>]); and those that want a resource that lacks a subject-specific context (Booth, Sutton, & Papaioannou, [<reflink idref="bib7" id="ref20">7</reflink>]).</p> <p>But literature reviews can produce many more insights about a discipline than a summary of findings. A literature review can uncover institutional hubs of activity, significant collaborations between researchers, common methodological approaches, different theoretical frameworks, and exciting frontiers. The literature review is a stand-alone piece of research. Those conducting literature reviews should strive for compelling evidence and consider a variety of methodologies to identify, interpret, and present that evidence.</p> <p>In research, a publication is the most common expression of original findings. Therefore the primary evidence for a literature review is peer-reviewed journal articles, books, and other relevant grey literature. If the reviewer conceptualizes a publication as evidence of originality, then each publication becomes a cultural artefact (Griswold, [<reflink idref="bib21" id="ref21">21</reflink>]) in a constantly evolving knowledge domain (McFarland & Klopfer, [<reflink idref="bib28" id="ref22">28</reflink>]). Conducting a literature review means situating each of these knowledge artefacts (McFarland & Klopfer, [<reflink idref="bib28" id="ref23">28</reflink>]) within the empirical and social contexts that support discovery. This means accessing data both within the text of a publication and data about its publication and consumption.</p> <p>Bibliometrics and Scientometrics are active fields of research dedicated to developing methods and statistics to help researchers take advantage of all types of data a publication has to offer. Networks have long been used to make sense of publication statistics. Most notably, the development of citation indexing (Garfield, [<reflink idref="bib18" id="ref24">18</reflink>]; White, [<reflink idref="bib47" id="ref25">47</reflink>]; Wouters, [<reflink idref="bib48" id="ref26">48</reflink>]) led to the creation of several powerful databases that allow researchers to search for literature that cites specific authors or publications. Web of Science (WoS), Scopus, and Google Scholar all utilize citation indexing to provide users with many advanced search capabilities.</p> <p>Citation networks have been used to search for influential, or highly cited ideas (Colicchia, Creazza, & Strozzi, [<reflink idref="bib12" id="ref27">12</reflink>]) and journals (Wang & Bowers, [<reflink idref="bib45" id="ref28">45</reflink>]) within a discipline. Citation networks have successfully traced the cognitive development of new ideas and provided evidence of collaborative relationships between researchers (Mullins, Hargens, Hecht, & Kick, [<reflink idref="bib31" id="ref29">31</reflink>]). Reviewers have also used citation indexing to identify researchers that interact with similar literature and can therefore be considered members of 'invisible colleges,' or scientific specialities (Persson & Beckmann, [<reflink idref="bib38" id="ref30">38</reflink>]). Citation analysis has been used to identify individuals that instigate scientific change (Small & Griffith, [<reflink idref="bib41" id="ref31">41</reflink>]) and to identify publications that promote the development of specific theories (Hummon & Dereian, [<reflink idref="bib23" id="ref32">23</reflink>]). Citation indexing is also being used to develop new impact measures for publications (Bollen, Sompel, De, Hagberg, Chute, & Mailund, [<reflink idref="bib4" id="ref33">4</reflink>]; Bollen, Van de Sompel, Smith, & Luce, [<reflink idref="bib5" id="ref34">5</reflink>]).</p> <p>Citation networks can provide powerful insights about the development of knowledge in a discipline. But it is difficult to establish a strong relational link between researchers simply because one cites the work of another. Alternatively, the distribution of keyword co-occurrence in education journals has been used to study the evolution of research trends (Huang et al., [<reflink idref="bib22" id="ref35">22</reflink>]). Reviewers have also been able to access readership statistics of particular articles to construct networks for how information in a discipline is consumed (Carolan, [<reflink idref="bib10" id="ref36">10</reflink>]). But if the goal of a reviewer is to understand the collaborative nature of knowledge formation, the use of similar keywords or reading the work of someone conducting similar research, does not provide strong evidence of meaningful collaboration (Liu, Bollen, Nelson, & Van de Sompel, [<reflink idref="bib27" id="ref37">27</reflink>]; Newman, [<reflink idref="bib33" id="ref38">33</reflink>]).</p> <p>SNA is a methodology that relies on strong relational data. A more substantial relationship tends to exist between researchers that write together compared to those that cite each other's work, use similar key words, or attract similar audiences. Co-authorship networks have been used to explore patterns of international collaboration (Melin & Persson, [<reflink idref="bib29" id="ref39">29</reflink>]; Wagner & Leydesdorff, [<reflink idref="bib44" id="ref40">44</reflink>]). They have also been used to explore the geographic and gender distribution of collaborators (Cunningham & Dillon, [<reflink idref="bib14" id="ref41">14</reflink>]). Also, like citation networks, co-authorship networks are effective in identifying prominent authors in a discipline (González-Teruel, González-Alcaide, Barrios, & Abad-García, [<reflink idref="bib19" id="ref42">19</reflink>]; Moody, [<reflink idref="bib30" id="ref43">30</reflink>]; Newman, [<reflink idref="bib32" id="ref44">32</reflink>]; Otte & Rousseau, [<reflink idref="bib36" id="ref45">36</reflink>]).</p> <p>All reviewers must figure out a method for organizing large collections of literature into coherent descriptions of prominent research efforts. This involves a two-step process where a reviewer first groups literature that is somehow related, and then infers meaning from the various similarities between publications in each group. These two related steps are generally known as <emph>synthesis</emph> and <emph>analysis</emph>. Co-authorship networks can help reviewers organise publications into meaningful groups, effectively automating the synthesis process in a literature review.</p> <p>When a co-authorship network is partitioned into communities, these communities represent strong groups of collaborators. Reviewers can then group publications based on the community membership of each author. Because the network is formed using strong relational data, the resulting groups of publications should contain meaningful similarities that can support the identification of prominent research efforts within a discipline.</p> <p>The reviewer is then left with the singular task of making inferences from groups of highly related publications. While citation, common keyword, or readership networks can help in identifying relevant literature, networks constructed from these types of relational data do not ensure meaningful communities of collaboration. The relational strength of edges in co-authorship networks is therefore more effective when sorting publications for the purpose of defining prominent research programs. Identifying communities in co-authorship networks effectively automates the time-consuming process of synthesis. This allows reviewers to more authentically engage with large collections of literature, leading to greater quality in the interpretive results of reviews.</p> <hd id="AN0144473746-5">Methods</hd> <p></p> <hd id="AN0144473746-6">Using co-authorship networks for the synthesis of literature</hd> <p>The following protocol was developed to help reviewers organise large collections of literature into meaningful groups, making the identification of prominent research programs more manageable (Figure 1).</p> <p>PHOTO (COLOR): Figure 1. Protocol for generating a co-authorship network.</p> <p></p> <ulist> <item> Compile a collection of literature into a.CSV file using systematic approaches for searching digital and print resources.</item> <p></p> <item> Select network software and transform the.CSV file into a network template that is compatible with the chosen platform.</item> <p></p> <item> Generate a co-authorship network diagram and partition the network into communities.</item> </ulist> <hd id="AN0144473746-7">Compile a collection of literature into a.CSV file using systematic approaches for searching...</hd> <p>A comprehensive explanation of systematic approaches for finding relevant literature is beyond the scope of this article. But it is worth highlighting that reviewers demonstrate thoroughness by using multiple systematic search approaches in literature review methods. Productive strategies can include searching for previously published reviews, subject searching in multiple databases, manually searching individual volumes of topically relevant journals, chasing citations from particularly compelling literature, and contacting experts.</p> <p>Regardless of which search strategies are used, the reviewer must export reference lists from each search as a.CSV file. Exporting search results as a.CSV file allows a reviewer to clean and manipulate the reference data in Excel. This flexibility to interact with reference data will be important in later steps because the provided R script requires an author list in a particular format.</p> <p>Sometimes export capabilities for a database are obscured. For example, some databases only allow the export of search results from personal libraries or profile folders. This requires the reviewer to create a user account and navigate to a separate page to export results. For example, EBSCO Information Services requires users to create an account and move search results into a personal folder to export reference lists. Web of Science only allows a user to export search results after moving references into a separate marked list. Google Scholar also requires users to individually click on a star to save references to a personal library. This can be a tedious task for a systematic reviewer and might encourage some less experienced reviewers to make decisions about the inclusion of specific references while scrolling through search results.</p> <p>However, systematic literature reviews require the use of explicit criteria to screen potential references for inclusion in a review. Do not allow the design of a database to undermine your systematic approaches for identifying relevant literature. Export all search results as.CSV files and apply predetermined inclusion or exclusion criteria consistently to each reference list produced by every systematic search.</p> <p>After individual searches are completed, reviewers should create a master.CSV file to serve as a final repository for relevant references. In order to take advantage of the R script provided by this paper, the final collection of relevant references must be compiled in a.CSV file. The first column of the.CSV file needs to be titled 'authors' and each cell in the column should include a list of all authors for every reference the reviewer wants to include in the co-authorship network.</p> <hd id="AN0144473746-8">Select network software and transform the.CSV file into a network template that is compatible...</hd> <p>Databases provide different forms of relational data with references. All databases list the authors of each reference. But some databases go further and provide the institutional affiliation of the authors or sources of funding that supported the research. This paper uses authorship data to create networks because co-authoring an article implies a strong relationship between researchers. However, a reasonable connection might be inferred if two researchers work at the same institution or receive funding from the same source.</p> <p>Whatever type of relational data a reviewer wants to use, it must be transformed into a format that is compatible with their preferred network software. This is because databases currently lack the ability to construct networks and apply descriptive and statistical metrics of SNA using reference lists. An R script[<reflink idref="bib1" id="ref46">1</reflink>] is provided to automate the conversion of database.CSV reference files into network data templates. Specifically, the R script uses author lists to generate separate nodes and edges lists.</p> <p>To run the R script, the user must first create a new folder on their desktop titled, 'input_files'. Deposit multiple.CSV reference files from separate searches or a master.CSV reference file into the input_files folder. The R script will generate a new folder on the desktop titled, 'combination_out'.</p> <p>Two new.CSV files will be available in the combination_out folder. The first file will contain a single-column list of each individual author listed in every author cell of every reference. This is known as a nodes list by SNA researchers. The second file will contain a two-column list of every co-authorship combination based on the contents within each individual author cell of every reference. This is known as an edges list.</p> <p>All network researchers develop preferences for particular network software based on ease of use. The visual and analytical features of different networking software can also vary significantly. The examples in this paper were created using Polinode (Pitts, [<reflink idref="bib40" id="ref47">40</reflink>]). Polinode was selected because the free features of Polinode allow a user to generate a co-authorship network and partition the network into communities.</p> <p>Furthermore, Polinode was selected over other networking options (NodeXL, Gephi, and UCINet) because the visual controls are impressive and easy to navigate. Most visual characteristics of the network diagram are adjustable using a toolbar. The network statistics are also easily accessible to those without formal SNA training. Each network statistic that Polinode offers includes a written description for the user if you hover your cursor over the command.</p> <p>VOSviewer is another network software worth exploring for those interested in creating co-authorship networks (van Eck & Waltman, [<reflink idref="bib43" id="ref48">43</reflink>]). VOSviewer has two clear advantages over the network protocol presented in this paper. VOSviewer is a program that can be downloaded onto your computer. This means you can generate networks without signing up for a user account or uploading network data to a website.</p> <p>VOSviewer can also generate a network from several forms of bibliographic data including database export files, reference manager files, and APIs. VOSviewer is a good option for readers not familiar with RStudio or for those that want to generate a network from a collection of references compiled in a specific reference manager like EndNote or Zotero.</p> <p>However, better connections between databases and networking software is needed. VOSviewer only allows users to import search results from several databases, mostly citation-indexing databases like Web of Science and Scopus. Although this is a powerful capability, researchers conducting systematic reviews use many databases when compiling relevant references. Therefore, a researcher cannot use VOSviewer to generate a co-authorship diagram for their entire reference collection if they exported references from a database that isn't compatible with VOSviewer.</p> <p>The R script provided by this paper was designed to interact with all database reference list formats and generates more universal network data outputs. The R script can manipulate author lists that separate individual authors with all forms of punctuations. It then produces universal nodes and edges lists. This gives reviewers the greatest database coverage and the greatest flexibility in their networking software selection. Connecting databases and networking software is critical if SNA is to become an accessible option for literature synthesis and analysis. Providing greater flexibility also forces competitors to retain users by improving visual and statistical capabilities rather than individualizing data formats.</p> <hd id="AN0144473746-9">Generate a co-authorship network diagram and partition the network into communities</hd> <p>To generate a co-authorship network in Polinode, create a two-sheet Excel file. The first sheet should contain a list of all individual authors you want represented as nodes in your co-authorship network. This sheet should be titled, 'Nodes'. The first column (A1 cell) in the Nodes sheet needs to be titled, 'Name'. The second sheet in the.CSV file should be titled, 'Edges'. The A1 cell needs to be labelled, 'Source' and the B1 cell needs to be labelled, 'Target'.</p> <p>Polinode provides this network data template. The template can be downloaded when creating a new network in the Polinode platform. Furthermore, if there are formatting errors in your network data file, Polinode will identify each error and list them when the user attempts to upload the network dataset.</p> <p>An example co-authorship network is provided in Figure 2. The example network was created from a single-column author list, referred to as a nodes list. The example co-authorship network also requires a duel-column co-authorship table, known as an edges list. The nodes and edges lists were transferred into a network data template compatible with Polinode. Figure 2 therefore depicts every author from the original reference list as a node and each authorship collaboration as an edge.</p> <p>PHOTO (COLOR): Figure 2. Example co-authorship network diagram.</p> <p>Several visual settings in Polinode make the co-authorship network easier to read. A 'light' template can be selected so the network is presented on a white background. The layout of the network can also be displayed as 'lens' rather than 'force directed' so that all nodes are visible and so there is less overlap when all node labels are visible. The node labels in Figure 2 were nonetheless filtered based on the total degree of each node. A total degree threshold was selected so that only the highest total degree nodes would be labelled. This means labelled nodes have a rich collaborative history, making them potential experts in the depicted discipline.</p> <p>This initial co-authorship network diagram must then be partitioned into communities. There is much debate within SNA about how to best identify communities within a larger network (Newman & Girvan, [<reflink idref="bib35" id="ref49">35</reflink>]). And even the often-utilized Louvain community detection algorithm (Blondel, Guillaume, Lambiotte, & Lefebvre, [<reflink idref="bib2" id="ref50">2</reflink>]) used by Polinode now has critics (Traag, Waltman, & Eck, [<reflink idref="bib42" id="ref51">42</reflink>]).</p> <p>However, for the sake of readers less familiar to SNA, this methods section will not dwell on discussions about which method for identifying communities within larger networks is best. It is sufficient to know that Polinode uses a standard community detection algorithm and that researchers should always verify that their preferred networking software uses the Louvain method or a well-tested alternative. Figure 3 depicts the same co-authorship network from Figure 2, but each node is coloured based on the author's community membership.</p> <p>PHOTO (COLOR): Figure 3. Co-authorship network partitioned into communities.</p> <hd id="AN0144473746-10">Using SNA metrics to analyse groups of literature</hd> <p>A reviewer can create more manageable groups of highly relevant literature by sorting references based on the community designation of an author. Because these communities are based on co-authorship relationships, there should be significant similarities in the published output of each community. Reviewers can use these similarities to differentiate between research programs within the larger discipline. But it is still difficult to develop understanding from large collections of publications. Several SNA metrics and concepts are useful to reviewers attempting to interpret publications from larger co-authorship communities.</p> <hd id="AN0144473746-11">Prioritising the analysis of publications based on different centrality measures</hd> <p>The publications of central authors will be more influential when defining the collective work of a co-authorship community. Total degree centrality has been used to identify influential authors within a network (González-Teruel et al., [<reflink idref="bib19" id="ref52">19</reflink>]; Moody, [<reflink idref="bib30" id="ref53">30</reflink>]; Otte & Rousseau, [<reflink idref="bib36" id="ref54">36</reflink>]). However, there are more accurate centrality measures to help with the prioritization of publications in a co-authorship community.</p> <p>It is more efficient for a reviewer to first analyse the publications of authors that serve as collaborative links between otherwise separate communities. For example, Figure 4 highlights an author that serves as a collaborative link holding community 13 together. Community 13 only exists because of the co-authorship edges of J. Little. If J. Little were removed from the network, Community 13 would split into two smaller co-authorship communities. Even if J. Little doesn't have the highest total degree centrality, these publications need to play a central role in defining the research output of the community.</p> <p>PHOTO (COLOR): Figure 4. Identification of a central author within community 13.</p> <p>Betweenness centrality (Wasserman & Faust, [<reflink idref="bib46" id="ref55">46</reflink>]) can explain the significance of authors that might not have the highest total degree centrality but still serve as collaborative links within communities. Nodes have greater betweenness centrality measures when they occupy network positions that connect otherwise isolated areas of the network. Betweenness centrality should be used as an additional indicator of significance when prioritizing the analysis of publications from large co-authorship communities.</p> <hd id="AN0144473746-12">Using network density and structural holes to identify new frontiers</hd> <p>Network density can be used by a reviewer to make assertions about the developmental stage of a discipline. Network density refers to the percentage of existing edges compared to possible edges in a network (Wasserman & Faust, [<reflink idref="bib46" id="ref56">46</reflink>]). Unless a discipline is actively decaying, more time should result in a thicker lattice of co-authorship connections.</p> <p>New collaborations can develop through a number of different relational mechanisms. A new collaboration between researchers might develop after meeting at a conference. Collaboration might also begin when one researcher hires the graduate student of another. Productive collaborations can also happen by chance meetings (Cornelissen, [<reflink idref="bib13" id="ref57">13</reflink>]). In a co-authorship network diagram, greater density will result in a growing central component of connected researchers. However, it is important to note that just because two authors publish together does not mean they will be members of the same co-authorship community. This is an important distinction from the previously mentioned use of betweenness centrality to prioritize the publications of authors within the same co-authorship community.</p> <p>Authors that collaborate with researchers from different communities have greater exposure to diverse ideas. These authors fill special positions in networks known as structural holes (Burt, [<reflink idref="bib9" id="ref58">9</reflink>]). Therefore, nodes occupying structural holes have greater opportunity to synthesise information in innovative ways. Publications by nodes filling structural holes should be searched for innovative theoretical frameworks and methods. Therefore, authors filling structural holes can help reviewers efficiently identify the frontiers in the discipline rather than help define a specific co-authorship community.</p> <hd id="AN0144473746-13">Adding node attributes</hd> <p>A reviewer can add attribute characteristics to individual nodes in a network diagram. Adding institutional affiliations or nationality might lead to interesting insights into a co-authorship community of researchers. Scopus includes author affiliations as an export option when downloading reference lists. Google Scholar allows reviewers the opportunity to click on author names to see verified email addresses and author affiliations.</p> <p>These attribute characteristics can provide interesting insights when searching for similarities within groups of community publications. Common nationality of authors within a community could indicate the influence of national legislation. If a co-authorship research community shares the same nationality, the reviewer can easily search the collection of community publications for references to legislation or national education policies. This can be done using text analysis software like NVivo or Atlas.ti. Simple keyword searching of a PDF can produce the same result.</p> <p>Conversely, different nationalities of authors within a community would signify international collaboration. This might justify a reviewer contacting central authors within the community to determine the origins of such collaborations. Furthermore, individual attribute data might uncover skewed reference lists that only account for one national context or output from a select group of institutions. Reviewers should look to add as much attribute data about authors to their network dataset. However, this data will be difficult to access and will likely be incomplete as various databases include different types of relational data.</p> <hd id="AN0144473746-14">Discussion</hd> <p></p> <hd id="AN0144473746-15">Using the co-authorship network to design additional searches</hd> <p>Multiple systematic approaches are needed to thoroughly search literature. As mentioned previously, reviewers often combine systematic approaches when developing a search protocol to ensure thoroughness. This paper primarily advocated for the application of SNA to help reviewers synthesise and analyse large collections of references. However, a co-authorship network can be used to design additional productive searches.</p> <p>As discussed earlier in this paper, total degree centrality and betweenness centrality can be used to identify prominent authors in a co-authorship network. Conducting author searches can expand collections of relevant literature. Many databases allow users to search for all publications by an individual author. Cross-referencing the complete writings of a central author with the original reference collection can help validate the initial systematic search protocol and specifically the search queries used in various databases.</p> <p>Furthermore, the publications of central authors can be used to develop citation searches. Web of Science and Scopus are citation-indexing databases. Therefore, these databases provide users with search capabilities for identifying publications that cite specific writings. Researchers that cite the publications of central authors can produce additional references for the reviewer.</p> <p>Finally, contacting prominent authors in a discipline is a suggested systematic approach to identifying relevant literature. As discussed previously, network diagrams can help reviewers identify prominent authors. However, a reviewer does not need to construct a co-authorship network to differentiate between authors. Scanning the author column of a reference list will give any skilled reviewer a good idea as to which authors are prominent.</p> <p>However, constructing a co-authorship diagram before contacting prominent authors provides two advantages. First, a network diagram can help solicit a response from a prominent researcher. Although many researchers are willing to engage by email with graduate students and early-career researchers, providing a network diagram can spark greater interest and might result in a more authentic conversation and potentially an interview with the researcher.</p> <p>Second, a co-authorship network diagram provides the reviewer direction when writing interview questions. Asking prominent authors to explain specific research relationships will result in richer contextual narratives to explain significant collaborations in the cognitive development of the discipline. The resulting transcript is valuable data for the literature review. It also represents an opportunity for the reviewer to network with more senior researchers in their discipline.</p> <hd id="AN0144473746-16">Increasing the level of systematic rigor in literature reviews</hd> <p>Consider a team of researchers, librarians, and hired assistants conducting a systematic literature review. This team collectively has years of experience within a discipline and already possesses important background knowledge about many of the researchers and publications they will encounter. This is important contextual knowledge that allows for more meaningful analysis of the literature.</p> <p>However, these skilled working groups are the exception, rather than the norm. Many more literature reviews are conducted by individual researchers under significant constraint. The application of qualitative and quantitative methods can improve existing and create new systematic methods to support individual researchers in conducting literature reviews.</p> <p>Researchers are expected to master systematic approaches for searching literature. This requires researchers to learn many different techniques and use many different resources to access an ever-expanding body of literature. But this phase of the literature review is supported by information professionals and powerful databases. It is the synthesis and analysis of large collections of relevant references that is far more challenging. Many reviewers fail to meaningfully engage with literature because simultaneously synthesising and analysing large collections of publications can quickly become overwhelming.</p> <p>The application of SNA to these later processes in the literature review progression creates much needed support. An independent researcher is not able to match the level of rigor of a full-funded systematic review team. However, the application of different qualitative and quantitative methods provides opportunities to accelerate tedious tasks in the literature review process. This allows an individual reviewer to spend far more time interacting with the literature.</p> <hd id="AN0144473746-17">Conclusion</hd> <p>Social Network Analysis is based on the simple but powerful premise that relationships matter (Wasserman & Faust, [<reflink idref="bib46" id="ref59">46</reflink>]). A comprehensive understanding of an academic field requires a researcher to pay attention to more than just key findings in publications. A researcher must understand the collaborations that yield new knowledge (Paavola, Lipponen, & Hakkarainen, [<reflink idref="bib37" id="ref60">37</reflink>]). Data to help uncover this relational context is available to reviewers that are willing to dig into the many different search tools being developed by databases.</p> <p>Publications contain more significant data than findings of research. There are many publication statistics that provide information on how the publication was written and how the publication is consumed. The application of new tools into the literature review process allows authors to take advantage of this powerful data surrounding these knowledge artefacts.</p> <p>There exists such diversity in methods for data collection, especially in the social sciences. We betray this beautiful plurality of methods when conducting the literature review by defaulting to a mind-set that solely focuses on completion rather than creative expression. Network analysis has long been applied to literature searching. However, strong relational data now exists within publication statistics. This allows for the use of SNA in the interpretive phases of the literature review.</p> <p>Like any standalone research, those conducting literature reviews must decide what is compelling evidence. The reviewer should then consider the application of any method that will help identify, interpret, and present that evidence.</p> <hd id="AN0144473746-18">Disclosure statement</hd> <p>No potential conflict of interest was reported by the authors.</p> <ref id="AN0144473746-19"> <title> Note </title> <blist> <bibl id="bib1" idref="ref5" type="bt">1</bibl> <bibtext> The R script can be downloaded from GitHub at https://github.com/EdNetResearch/Co-authorship-Network.</bibtext> </blist> </ref> <ref id="AN0144473746-20"> <title> References </title> <blist> <bibtext> Barabasi, A. L. (2016). Network Science. Cambridge, UK: Cambridge University Press.</bibtext> </blist> <blist> <bibl id="bib2" idref="ref50" type="bt">2</bibl> <bibtext> Blondel, V. D., Guillaume, J. L., Lambiotte, R., & Lefebvre, E. (2008). Fast unfolding of communities in large networks. Journal of Statistical Mechanics: Theory and Experiment, 2008 (10), P10008.</bibtext> </blist> <blist> <bibl id="bib3" idref="ref18" type="bt">3</bibl> <bibtext> Boland, A., Cherry, M. G., & Dickson, R. (2013). Doing a systematic review: A student's guide (2nd ed.). London, UK: Sage Publications Ltd.</bibtext> </blist> <blist> <bibl id="bib4" idref="ref33" type="bt">4</bibl> <bibtext> Bollen, J., Van de Sompel, H., Hagberg, A., Chute, R., & Mailund, T. (2009). A principal component analysis of 39 scientific impact measures. PloS One, 4 (6), e6022.</bibtext> </blist> <blist> <bibl id="bib5" idref="ref34" type="bt">5</bibl> <bibtext> Bollen, J., Van de Sompel, H., Smith, J. A., & Luce, R. (2005). Toward alternative metrics of journal impact: A comparison of download and citation data. Information Processing & Management, 41 (6), 1419 – 1440.</bibtext> </blist> <blist> <bibl id="bib6" idref="ref2" type="bt">6</bibl> <bibtext> Booth, A., Papaioannou, D., & Sutton, A. (2012). Systematic approaches to a successful literature review. London, UK: Sage Publications Ltd.</bibtext> </blist> <blist> <bibl id="bib7" idref="ref20" type="bt">7</bibl> <bibtext> Booth, A., Sutton, A., & Papaioannou, D. (2016). Systematic approaches to a successful literature review (2nd ed.). London, UK: SAGE Publications Ltd.</bibtext> </blist> <blist> <bibl id="bib8" idref="ref11" type="bt">8</bibl> <bibtext> Borgatti, S. P., & Ofem, B. (2010). Overview: Social network theory and analysis. In Social network theory and educational change (pp. 17 – 30). Cambridge, MA: Harvard Education Press.</bibtext> </blist> <blist> <bibl id="bib9" idref="ref58" type="bt">9</bibl> <bibtext> Burt, R. S. (2004). Structural holes and good ideas. American Journal of Sociology, 110 (2), 349 – 399.</bibtext> </blist> <blist> <bibtext> Carolan, B. V. (2008). The structure of educational research: The role of multivocality in promoting cohesion in an article interlock network. Social Networks, 30 (1), 69 – 82.</bibtext> </blist> <blist> <bibtext> Carolan, B. V. (2014). Social network analysis and education: Theory, methods and applications. London, UK: SAGE Publications Ltd.</bibtext> </blist> <blist> <bibtext> Colicchia, C., Creazza, A., & Strozzi, F. (2018). Citation network analysis for supporting continuous improvement in higher education. Studies in Higher Education, 43 (9), 1637 – 1653.</bibtext> </blist> <blist> <bibtext> Cornelissen, F. (2015). Getting your knowledge "out there": Finding your knowledge ambassadors, or how they find you. In Surviving and thriving with teacher action research: Reflections and advice from the field (pp. 191 – 196). New York, NY: Peter Lang Publishing Inc.</bibtext> </blist> <blist> <bibtext> Cunningham, S. J., & Dillon, S. M. (1997). Authorship patterns in information systems. Scientometrics, 39 (1), 19 – 27.</bibtext> </blist> <blist> <bibtext> Daly, A. J. (2010). Social network theory. Cambridge, MA: Harvard Education Press.</bibtext> </blist> <blist> <bibtext> Fink, A. (2010). Conducting research literature reviews: From internet to paper (3rd ed.). London, UK: Sage Publications Ltd.</bibtext> </blist> <blist> <bibtext> Galvan, J. L., & Galvan, M. C. (2017). Writing literature reviews: A guide for students of the social and behavioral sciences (7th ed.). New York, NY: Routledge.</bibtext> </blist> <blist> <bibtext> Garfield, E. (1955). Citation indexes for science. Science, 122, 108 – 111.</bibtext> </blist> <blist> <bibtext> González-Teruel, A., González-Alcaide, G., Barrios, M., & Abad-García, M. F. (2015). Mapping recent information behavior research: An analysis of co-authorship and co-citation networks. Scientometrics, 103 (2), 687 – 705.</bibtext> </blist> <blist> <bibtext> Gough, D., Oliver, S., & Thomas, J. (2017). An introduction to systematic reviews (2nd ed.). London, UK: Sage Publications Ltd.</bibtext> </blist> <blist> <bibtext> Griswold, W. (1987). A methodological framework for the sociology of culture. Sociological Methodology, 17, 1 – 35.</bibtext> </blist> <blist> <bibtext> Huang, C., Yang, C., Wang, S., Wu, W., Su, J., & Liang, C. (2019). Evolution of topics in education research: A systematic review using bibliometric analysis. Educational Review, 1 – 17. doi: 10.1080/00131911.2019.1566212</bibtext> </blist> <blist> <bibtext> Hummon, N. P., & Dereian, P. (1989). Connectivity in a citation network: The development of DNA theory. Social Networks, 11 (1), 39 – 63.</bibtext> </blist> <blist> <bibtext> Jesson, J. K., Matheson, L., & Lacey, F. M. (2011). Doing your literature review: Traditional and systematic techniques. London, UK: SagePublications Ltd.</bibtext> </blist> <blist> <bibtext> Kadushin, C. (2012). Understanding social networks: Theories, concepts, and findings. New York, NY: Oxford University Press.</bibtext> </blist> <blist> <bibtext> Khan, K., Kunz, R., Kleijnen, J., & Antes, G. (2011). Systematic reviews to support evidence-based medicine (2nd ed.). Boca Raton, FL: CRC Press, Taylor and Francis Group.</bibtext> </blist> <blist> <bibtext> Liu, X., Bollen, J., Nelson, M. L., & Van de Sompel, H. (2005). Co-authorship networks in the digital library research community. Information Processing & Management, 41 (6), 1462 – 1480.</bibtext> </blist> <blist> <bibtext> McFarland, D., & Klopfer, E. (2010). Network search: A new way of seeing the education knowledge domain. Teachers College Record, 112 (10), 2664 – 2702.</bibtext> </blist> <blist> <bibtext> Melin, G., & Persson, O. (2005). Studying research collaboration using co-authorships. Scientometrics. doi: 10.1007/BF02129600</bibtext> </blist> <blist> <bibtext> Moody, J. (2004). The structure of a social science collaboration network: Disciplinary cohesion from 1963 to 1999. American Sociological Review; Washington, 69 (2), 213 – 238.</bibtext> </blist> <blist> <bibtext> Mullins, N. C., Hargens, L. L., Hecht, P. K., & Kick, E. L. (1977). The group structure of cocitation clusters: A comparative study. American Sociological Review, 42 (4), 552 – 562.</bibtext> </blist> <blist> <bibtext> Newman, M. E. J. (2001a). Scientific collaboration networks. II. Shortest paths, weighted networks, and centrality. Physical Review E, 64 (1). doi: 10.1103/PhysRevE.64.016132</bibtext> </blist> <blist> <bibtext> Newman, M. E. J. (2001b). The structure of scientific collaboration networks. Proceedings of the National Academy of Sciences, 98 (2), 404 – 409.</bibtext> </blist> <blist> <bibtext> Newman, M. E. J. (2010). Networks: An introduction. New York, NY: Oxford University Press.</bibtext> </blist> <blist> <bibtext> Newman, M. E. J., & Girvan, M. (2004). Finding and evaluating community structure in networks. Physical Review E, 69 (2). doi: 10.1103/PhysRevE.69.026113</bibtext> </blist> <blist> <bibtext> Otte, E., & Rousseau, R. (2002). Social network analysis: A powerful strategy, also for the information sciences. Journal of Information Science, 28 (6), 441 – 453.</bibtext> </blist> <blist> <bibtext> Paavola, S., Lipponen, L., & Hakkarainen, K. (2004). Models of innovative knowledge communities and three metaphors of learning. Review of Educational Research, 74 (4), 557 – 576.</bibtext> </blist> <blist> <bibtext> Persson, O., & Beckmann, M. (1995). Locating the network of interacting authors in scientific specialties. Scientometrics, 33 (3), 351 – 366.</bibtext> </blist> <blist> <bibtext> Petticrew, M., & Roberts, H. (2006). Systematic reviews in the social sciences: A practical guide. Oxford, UK: Blackwell Publishing Ltd.</bibtext> </blist> <blist> <bibtext> Pitts, A. (2016). Polinode: A web application for the collection and analysis of network data. 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM) (pp. 1422 – 1425). San Francisco, CA. doi: 10.1109/ASONAM.2016.7752435</bibtext> </blist> <blist> <bibtext> Small, H., & Griffith, B. C. (1974). The structure of scientific literatures I: Identifying and graphing specialties. Science Studies, 4 (1), 17 – 40.</bibtext> </blist> <blist> <bibtext> Traag, V. A., Waltman, L., & Eck, N. J. V. (2019). From Louvain to Leiden: Guaranteeing well-connected communities. Scientific Reports, 9 (1), 1 – 12.</bibtext> </blist> <blist> <bibtext> van Eck, N. J., & Waltman, L. (2019). VOSviewer (Version 1.6.13).</bibtext> </blist> <blist> <bibtext> Wagner, C. S., & Leydesdorff, L. (2003). Mapping Global Science Using International Co-authorships: A Comparison of 1990 and 2000.</bibtext> </blist> <blist> <bibtext> Wang, Y., & Bowers, A. J. (2016). Mapping the field of educational administration research: A journal citation network analysis. Journal of Educational Administration; Armidale, 54 (3), 242 – 269.</bibtext> </blist> <blist> <bibtext> Wasserman, S., & Faust, K. (2009). Social network analysis: Methods and applications. Cambridge, UK: Cambridge University Press.</bibtext> </blist> <blist> <bibtext> White, H. D. (2000). Toward ego-centered citation analysis. In The web of knowledge: A festschrift in honor of eugene garfield (pp. 475 – 498). Medford, NJ: Information Today, Inc.</bibtext> </blist> <blist> <bibtext> Wouters, P. (2017). Eugene garfield (1925–2017). Nature, 543, 492–492. doi: 10.1038/543492a</bibtext> </blist> </ref> <aug> <p>By Thomas Cowhitt; Timothy Butler and Elaine Wilson</p> <p>Reported by Author; Author; Author</p> <p></p> <p>Thomas Cowhitt is a former secondary teacher and current PhD candidate with the Faculty of Education at the University of Cambridge. His research interests include using Social Network Analysis to track the expression of innovation and new ideas in school organizations. Thomas is also interested in the creative application of SNA to improve important research processes.</p> <p>Timothy Butler is a postdoctoral fellow in the Cancer, Ageing and Somatic Mutation Programme at the Wellcome Trust Sanger Institute. His research interests involve using sequencing of pre-cancerous and normal human tissues to understand the earliest steps of tumour development. Timothy is particularly focused on understanding the impact of smoking on mutational acquisition of the lung, and how the mutational burden differs in the breast tissue of women at varying risks of developing breast cancer.</p> <p>Elaine Wilson is a senior lecturer in Education at the University of Cambridge and a Fellow of Homerton College. Elaine currently supervises multiple PhD students, leads the Chemistry PGCE new teachers cohort, and directs the EdD programme. She has received a Cambridge Pilkington Teaching Prize and a National Teaching Fellowship in recognition of excellence in teacher education leadership. Her research interests include Implementation and Improvement Science, Education Reform, Teacher Education, and Digital Technology.</p> </aug> <nolink nlid="nl1" bibid="bib39" firstref="ref1"></nolink> <nolink nlid="nl2" bibid="bib16" firstref="ref3"></nolink> <nolink nlid="nl3" bibid="bib24" firstref="ref4"></nolink> <nolink nlid="nl4" bibid="bib11" firstref="ref6"></nolink> <nolink nlid="nl5" bibid="bib15" firstref="ref7"></nolink> <nolink nlid="nl6" bibid="bib25" firstref="ref8"></nolink> <nolink nlid="nl7" bibid="bib34" firstref="ref9"></nolink> <nolink nlid="nl8" bibid="bib46" firstref="ref10"></nolink> <nolink nlid="nl9" bibid="bib17" firstref="ref14"></nolink> <nolink nlid="nl10" bibid="bib20" firstref="ref16"></nolink> <nolink nlid="nl11" bibid="bib26" firstref="ref17"></nolink> <nolink nlid="nl12" bibid="bib21" firstref="ref21"></nolink> <nolink nlid="nl13" bibid="bib28" firstref="ref22"></nolink> <nolink nlid="nl14" bibid="bib18" firstref="ref24"></nolink> <nolink nlid="nl15" bibid="bib47" firstref="ref25"></nolink> <nolink nlid="nl16" bibid="bib48" firstref="ref26"></nolink> <nolink nlid="nl17" bibid="bib12" firstref="ref27"></nolink> <nolink nlid="nl18" bibid="bib45" firstref="ref28"></nolink> <nolink nlid="nl19" bibid="bib31" firstref="ref29"></nolink> <nolink nlid="nl20" bibid="bib38" firstref="ref30"></nolink> <nolink nlid="nl21" bibid="bib41" firstref="ref31"></nolink> <nolink nlid="nl22" bibid="bib23" firstref="ref32"></nolink> <nolink nlid="nl23" bibid="bib22" firstref="ref35"></nolink> <nolink nlid="nl24" bibid="bib10" firstref="ref36"></nolink> <nolink nlid="nl25" bibid="bib27" firstref="ref37"></nolink> <nolink nlid="nl26" bibid="bib33" firstref="ref38"></nolink> <nolink nlid="nl27" bibid="bib29" firstref="ref39"></nolink> <nolink nlid="nl28" bibid="bib44" firstref="ref40"></nolink> <nolink nlid="nl29" bibid="bib14" firstref="ref41"></nolink> <nolink nlid="nl30" bibid="bib19" firstref="ref42"></nolink> <nolink nlid="nl31" bibid="bib30" firstref="ref43"></nolink> <nolink nlid="nl32" bibid="bib32" firstref="ref44"></nolink> <nolink nlid="nl33" bibid="bib36" firstref="ref45"></nolink> <nolink nlid="nl34" bibid="bib40" firstref="ref47"></nolink> <nolink nlid="nl35" bibid="bib43" firstref="ref48"></nolink> <nolink nlid="nl36" bibid="bib35" firstref="ref49"></nolink> <nolink nlid="nl37" bibid="bib42" firstref="ref51"></nolink> <nolink nlid="nl38" bibid="bib13" firstref="ref57"></nolink> <nolink nlid="nl39" bibid="bib37" firstref="ref60"></nolink>
Header DbId: eric
DbLabel: ERIC
An: EJ1259596
AccessLevel: 3
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Using Social Network Analysis to Complete Literature Reviews: A New Systematic Approach for Independent Researchers to Detect and Interpret Prominent Research Programs within Large Collections of Relevant Literature
– Name: Language
  Label: Language
  Group: Lang
  Data: English
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Cowhitt%2C+Thomas%22">Cowhitt, Thomas</searchLink><br /><searchLink fieldCode="AR" term="%22Butler%2C+Timothy%22">Butler, Timothy</searchLink><br /><searchLink fieldCode="AR" term="%22Wilson%2C+Elaine%22">Wilson, Elaine</searchLink>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="SO" term="%22International+Journal+of+Social+Research+Methodology%22"><i>International Journal of Social Research Methodology</i></searchLink>. 2020 23(5):483-496.
– Name: Avail
  Label: Availability
  Group: Avail
  Data: Routledge. Available from: Taylor & Francis, Ltd. 530 Walnut Street Suite 850, Philadelphia, PA 19106. Tel: 800-354-1420; Tel: 215-625-8900; Fax: 215-207-0050; Web site: http://www.tandf.co.uk/journals
– Name: PeerReviewed
  Label: Peer Reviewed
  Group: SrcInfo
  Data: Y
– Name: Pages
  Label: Page Count
  Group: Src
  Data: 14
– Name: DatePubCY
  Label: Publication Date
  Group: Date
  Data: 2020
– Name: TypeDocument
  Label: Document Type
  Group: TypDoc
  Data: Journal Articles<br />Reports - Evaluative
– Name: Subject
  Label: Descriptors
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Social+Networks%22">Social Networks</searchLink><br /><searchLink fieldCode="DE" term="%22Network+Analysis%22">Network Analysis</searchLink><br /><searchLink fieldCode="DE" term="%22Literature+Reviews%22">Literature Reviews</searchLink><br /><searchLink fieldCode="DE" term="%22Systems+Approach%22">Systems Approach</searchLink><br /><searchLink fieldCode="DE" term="%22Computer+Software%22">Computer Software</searchLink><br /><searchLink fieldCode="DE" term="%22Authors%22">Authors</searchLink><br /><searchLink fieldCode="DE" term="%22Visual+Aids%22">Visual Aids</searchLink>
– Name: DOI
  Label: DOI
  Group: ID
  Data: 10.1080/13645579.2019.1704356
– Name: ISSN
  Label: ISSN
  Group: ISSN
  Data: 1364-5579
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Literature reviews are required at early stages of a traditional research progression. Many systematic approaches help researchers identify relevant literature. However, there is far less support for interpreting large collections of references. Understanding the evolution of knowledge within a discipline requires an awareness of the collaborative networks from which significant advances originate. Unfortunately, this relational awareness is only acquired after years of professional experience. Therefore, early-career researchers are at a fundamental disadvantage when conducting literature reviews. Social Network Analysis (SNA) can make the interpretation of large collections of literature more manageable. This paper provides a method for generating a co-authorship network from reference lists. The partition of the co-authorship network into communities provides a reviewer with direction to organise literature into meaningful groups. The interpretation of seemingly disparate and entangled research programs is far more manageable when interpreting smaller collections of highly relevant literature.
– Name: AbstractInfo
  Label: Abstractor
  Group: Ab
  Data: As Provided
– Name: DateEntry
  Label: Entry Date
  Group: Date
  Data: 2020
– Name: AN
  Label: Accession Number
  Group: ID
  Data: EJ1259596
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=eric&AN=EJ1259596
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1080/13645579.2019.1704356
    Languages:
      – Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 14
        StartPage: 483
    Subjects:
      – SubjectFull: Social Networks
        Type: general
      – SubjectFull: Network Analysis
        Type: general
      – SubjectFull: Literature Reviews
        Type: general
      – SubjectFull: Systems Approach
        Type: general
      – SubjectFull: Computer Software
        Type: general
      – SubjectFull: Authors
        Type: general
      – SubjectFull: Visual Aids
        Type: general
    Titles:
      – TitleFull: Using Social Network Analysis to Complete Literature Reviews: A New Systematic Approach for Independent Researchers to Detect and Interpret Prominent Research Programs within Large Collections of Relevant Literature
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Cowhitt, Thomas
      – PersonEntity:
          Name:
            NameFull: Butler, Timothy
      – PersonEntity:
          Name:
            NameFull: Wilson, Elaine
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 01
              Type: published
              Y: 2020
          Identifiers:
            – Type: issn-print
              Value: 1364-5579
          Numbering:
            – Type: volume
              Value: 23
            – Type: issue
              Value: 5
          Titles:
            – TitleFull: International Journal of Social Research Methodology
              Type: main
ResultId 1