Overall Scope and Coverage:
- Timespan: 2022-2025: This indicates a relatively recent collection, covering research published in the last few years. This is important because it gives us a snapshot of the current state of research within the defined scope.
- Documents: 961: A total of 961 documents suggests a moderate-sized collection. Whether this is “large” or “small” depends heavily on the specific research area and the criteria used to define the collection. You need to compare this number to similar studies or benchmark data in your field to understand its relative size.
- Sources (Journals, Books, etc): 331: The presence of 331 sources indicates the research draws from a fairly diverse set of publications. This is a positive sign, suggesting the research is not overly concentrated in a few niche journals, and the topic benefits from input across a broader range of sources.
- Database: Scopus: Knowing that the data comes from Scopus is crucial. Scopus is a well-regarded, broad database, implying good coverage of scientific literature. However, be mindful of Scopus’s specific indexing policies and potential biases (e.g., towards English-language publications).
Productivity and Trends:
* Annual Growth Rate %: -7.77: A negative annual growth rate of -7.77% is a *significant finding*. It means the number of publications included in this collection *decreased* year-over-year within the 2022-2025 period. Several factors could explain this:
* Shrinking Research Area: The topic may be declining in popularity or funding.
* Changing Research Focus: Research may have shifted to related areas not captured by your search criteria.
* Data Lag/Indexing Delays: Scopus’s indexing may lag, especially towards the end of the period (2024-2025). This could artificially depress the growth rate. It’s important to check if the decrease is consistent across all years or primarily concentrated in the most recent ones.
* Search Query Issues: There might be some limitations in the search strategy to be revised, to capture more information.
*It’s crucial to investigate *why* this negative growth rate is occurring.* This could be a key insight for your research question.
Impact and Influence:
- Average Citations per Doc: 13.96: An average of 13.96 citations per document provides an initial indication of the impact of the publications within the collection. The higher the number of citations, the more influential the articles are. Whether 13.96 citations is “high” or “low” is entirely dependent on the field and the age of the publications. For example, articles in rapidly developing fields (e.g., some areas of computer science or medicine) typically accumulate citations faster than those in more established fields. This statistic needs to be normalized for field and time to be truly informative (i.e., compared to the average citation rate for similar articles published in the same field and year).
- References: 63512: The total number of references (63,512) indicates the depth of scholarship and the extent to which these documents build upon existing knowledge.
Authorship and Collaboration:
- Authors: 2406: A total of 2406 authors contributed to the documents in the collection.
- Authors of single-authored docs: 84: This figure indicates the number of documents within the collection written by a single author.
- Single-authored docs: 89: The single-authored document figure indicates how many documents were written by a single author.
- Co-Authors per Doc: 3.21: An average of 3.21 co-authors per document suggests a reasonably collaborative research environment. This points towards team-based research, which is common in many disciplines.
- International co-authorships %: 35.17: 35.17% of the documents are from international co-authorships, which shows high collaboration among researchers from different countries. It suggests a global interest in the research topic.
Keywords and Focus:
- Keywords Plus (ID): 1817: This metric, unique to Web of Science/Clarivate Analytics, identifies keywords that frequently appear in the cited references of the documents. While you have a Scopus collection, this suggests the data might have been pre-processed in Web of Science and then imported into Biblioshiny. Keywords Plus is useful for uncovering hidden or emergent themes within the research.
- Author’s Keywords (DE): 2587: Author-provided keywords (2587) represent the authors’ own framing of their work. Comparing “Keywords Plus” to “Author’s Keywords” can reveal discrepancies between how authors perceive their work and how it is perceived by the broader research community (as reflected in the cited literature).
Document Types:
- Article: 961: The documents include only articles, so there aren’t any conference proceedings or reviews in this list.
Next Steps and Critical Considerations:
1. Contextualize Citation Impact: Calculate field-normalized citation metrics if possible. Biblioshiny might have functionalities to do this or require you to export the data and use other tools.
2. Investigate the Negative Growth Rate: Dive deeper into *why* the annual growth rate is negative. Analyze publication trends by year within the collection. Check if it is an artifact of your query or a real trend.
3. Analyze Keyword Trends: Examine the most frequent “Author’s Keywords” and “Keywords Plus” over time to see how the research focus has evolved.
4. Examine Co-authorship Networks: Visualize the co-authorship network to identify key research groups and collaborations.
5. Compare to Baseline: If possible, compare these statistics to those of a broader collection or a known benchmark in your field. This will provide a more meaningful interpretation of the results.
By thoroughly exploring these aspects of the data, you can develop a much richer and more nuanced understanding of the research landscape captured in your bibliometric collection. Remember that these statistics are just a starting point; the real value comes from interpreting them in the context of your research question and your field. Good luck!
Annual Scientific Production
Three-Field Plot
Overall Structure and Interpretation
The three-field plot visualizes connections between the three metadata categories, with authors in the center, cited references on the left, and keywords on the right. The height of each bar represents the frequency or number of occurrences for each author, cited reference, or keyword. The links (lines) connecting the fields indicate the relationships between them. The thickness of the links often represents the strength or frequency of the association.
Specific Observations and Analysis
1. Central Role of “Business Model Innovation”: You’ll find terms such as ‘business model innovation’ featured across all fields of the graph. It serves as a central concept, connecting prominent authors to frequently cited works and related keywords.
2. Prominent Authors: Analyzing the ‘AU’ column, it’s clear that certain authors have stronger links to both cited references and keywords. For instance:
* *parida v*. and *bashir m* are more connected to ‘business model innovation’
3. Key Cited References: On the left (CR field), some references appear more frequently cited and connected to various authors and keywords. Some of the more prominent cited references include:
* chesbrough h. business model innovation: opportunities and… It seems to be the most important publication in this graph, due to its large amount of connections
* barney j. firm resources and sustained competitive advantage
* clauss t. measuring business model innovation: conceptualiz…
4. Keyword Clusters: The right field (KW_Merged) shows the prominent keywords associated with authors and cited references. Several keywords stand out:
* ‘business model innovation’ is the most frequent keyword.
* ‘business development’
* ‘sustainability’
5. Relationship Patterns:
* The flow of links suggests a general direction from cited references to authors and then to keywords. This implies that the cited references are foundational works that authors build upon, and the keywords represent the themes and concepts explored in their work.
* The links show which authors frequently cite certain references and which keywords are most commonly associated with their work. For example, an author heavily linked to a specific cited reference likely builds their research on that reference.
In summary, this plot offers a high-level view of the intellectual landscape within your Scopus dataset. It highlights the key authors, cited references, and keywords, as well as the relationships between them. This information can be used to identify research trends, influential authors, and potential areas for further investigation.
Further Steps & Questions to Consider:
- Filter by Time: Restrict the analysis to a specific time period to identify emerging trends or shifts in research focus.
- Focus on Specific Authors: Select a particular author to see which references they cite and what keywords are associated with their work.
- Examine Keyword Clusters: Investigate the relationships between different keywords to identify research sub-themes.
- Database-Specific Considerations: Scopus covers a wide array of journals, so your findings may be biased towards the journals indexed more comprehensively in Scopus. Compare with other databases to validate.
By critically examining these relationships, you can gain valuable insights into the structure and dynamics of your research field.
Most Relevant Sources
Sources’ Local Impact
Sources’ Production over Time
“Technological forecasting and social change” started the last and is now the second.
Most Relevant Authors
*Important to watch Vinit Parida
Authors’ Production over Time
Authors’ Local Impact
Most Relevant Affiliations
Affiliations’ Production over Time
Corresponding Author’s Countries
Countries’ Scientific Production
Countries’ Production over Time
Most Cited Countries
Most Global Cited Documents
Most Local Cited Documents
Most Local Cited References
Most Frequent Words
Trend Topics
Clustering by Coupling
Co-occurrence Network
Overall Structure and Interpretation
- Two Distinct Clusters/Communities: The network clearly shows two major clusters, visually distinguished by color (blue and red). This indicates the presence of two dominant, yet interconnected, research themes within your dataset.
- Centrality of “Business Model Innovation” and “Innovation”: The terms “Business Model Innovation” (BMI) and “Innovation” are the most prominent nodes in their respective clusters, indicated by their larger size and bolder font. They act as central connecting points within their clusters, signifying their high frequency and strong co-occurrence with other keywords.
- Bridge Between Clusters: The grey edges connecting the two clusters indicate relationships (co-occurrence) between the core concepts from each cluster. The thickness of these edges might suggest the relative strength of this interrelation. It appears that there’s a substantial link between “Business Model Innovation” and “Innovation,” acting as the primary bridge.
Cluster-Specific Interpretation:
* Blue Cluster: Traditional Innovation & Sustainable Business: This cluster seems to revolve around more traditional innovation concepts, potentially linked to sustainability. Notable terms include:
* “Innovation” (the core term connecting to the other cluster).
* “Business Models,” “Business Model,” “Business Development”.
* “Sustainable Business Model,” “Sustainable Development,” “Sustainable Business”.
* “Industry 4.0,” “Digitization”.
* “Circular Economy,” “Manufacturing”.
* “Stakeholder,” “Value Proposition”.
*Interpretation:* This cluster represents research focusing on innovation as a driver for business development, potentially within the context of sustainability and circular economy. The presence of “Industry 4.0” and “Digitization” suggests a focus on technological advancements as enablers of these concepts. The presence of “Strategic Approach”, “Value Proposition”, “Stakeholder” suggests a more traditional business and management focus.
* Red Cluster: Digital Business Model Innovation & Strategic Performance: This cluster seems to be centred on the application of business model innovation and strategic management in the digital era. Key terms include:
* “Business Model Innovation” (BMI) as the core term.
* “Dynamic Capabilities,” “Dynamic Capability,” “Digital Technologies”.
* “Digital Transformation,” “Digital Technology,” “Digital Business”.
* “Competitive Advantage,” “Firm Performance”.
* “Open Innovation,” “Strategy,” “Entrepreneurship,” “Case Study,” “COVID-19”.
* “SMEs,” “Commerce”.
*Interpretation:* This cluster represents a research stream concerned with how businesses can innovate their business models, particularly in the context of digital transformation. The emphasis on “Dynamic Capabilities” suggests a focus on the firm’s ability to adapt and evolve. The presence of “COVID-19”, “Strategy” and “Competitive Advantage” implies this research strand is concerned with strategies and performances in the Covid Era. The presence of “Open Innovation,” “Strategy,” and “Entrepreneurship” indicates related interests in the strategic approaches to BMI. Also, the presence of “Case Study” implies there are numerous case studies available on this topic.
Detailed discussion of the parameters:
- Association Normalization: This normalization method highlights terms that co-occur more frequently than expected by chance. This means the network emphasizes strong, meaningful relationships between keywords.
- Walktrap Clustering: The Walktrap algorithm is a community detection method that identifies clusters of nodes that are densely connected internally, but sparsely connected to other clusters. This is very useful to highlight the themes or topics studied in the papers.
- Edge Weight (Edgesize=15, Edges.min=2): The “edgesize” parameter (15) controls the width of the edges, while “edges.min” (2) sets a threshold for the minimum co-occurrence frequency to display an edge. This means the shown relationships are relatively strong (appearing at least twice in your dataset). Thicker edges would represent a higher frequency of co-occurrence.
- Label Parameters (label.n=50, labelsize=3): Only the top 50 most connected terms have labels, with a label size of 3. This focuses the visualization on the most relevant keywords.
Critical Discussion Points & Further Investigation:
- Temporal Trends: This network is a snapshot in time. To understand the evolution of these themes, you should perform the same analysis on subsets of your data (e.g., year by year, or in time periods). This would reveal how the relationships between concepts have changed over time.
- Database Bias: Remember that this is based on SCOPUS data. The results might be different if you used Web of Science or another database, due to different indexing practices and journal coverage.
- Keyword Choice: The quality of the results depends on the keywords used by authors. Consider if there are synonymous terms that are not captured, or if the keywords are too broad/narrow. Consider expanding your keywords to a more comprehensive list.
- Content Analysis: While this network provides a high-level overview, it’s essential to complement it with a qualitative content analysis of the most relevant papers. This will provide a deeper understanding of the nuances of these research themes.
Recommendations for Research:
- Explore the Intersection: Given the strong connection between the two clusters, a fruitful research area might be to explore how traditional innovation theories and practices (Blue cluster) can inform and improve digital business model innovation (Red cluster), and vice versa.
- Focus on Digital Transformation: The prominence of the “Digital Transformation” theme suggests that this is a particularly active and important area of research.
- Investigate Dynamic Capabilities: The presence of “Dynamic Capabilities” in the red cluster indicates that adaptability is a key factor in successful business model innovation. Further research could explore how organizations can develop and leverage dynamic capabilities to drive innovation in a rapidly changing environment.
By carefully interpreting the structure, clusters, and key terms of this network, you can gain valuable insights into the research landscape of your topic. Remember to use this analysis as a starting point for more in-depth investigation, combining it with qualitative methods to develop a nuanced understanding of the field.
Thematic Map
Overall Structure and Interpretation
The strategic diagram you’ve generated is a two-dimensional representation of research themes based on their centrality and density. These metrics provide insights into the importance and development of research areas.
- X-axis (Relevance Degree/Centrality): This axis represents the centrality of a theme, indicating its importance within the network of keywords. Themes positioned further to the right are more central and broadly relevant to the field. Centrality is calculated based on PageRank scores. A higher PageRank for the leading articles in a cluster suggests greater influence and connectivity within the research landscape.
- Y-axis (Development Degree/Density): This axis represents the density of a theme, reflecting its development and maturity. Themes positioned higher up have a greater density of connections, suggesting they are well-developed and actively researched.
Based on these axes, the map is divided into four quadrants, each representing a distinct type of theme:
- Motor Themes (Upper Right): High centrality and high density. These themes are well-developed and central to the field, driving research and innovation.
- Niche Themes (Upper Left): Low centrality but high density. These themes are specialized and well-developed within a specific area, but may not have broad relevance to the overall field.
- Basic Themes (Lower Right): High centrality but low density. These themes are fundamental and widely relevant but may be less actively researched or developed.
- Emerging or Declining Themes (Lower Left): Low centrality and low density. These themes are either newly emerging or declining in importance and may require further investigation to determine their potential.
Cluster Descriptions and Key Articles
Let’s analyze each cluster based on its position on the map and the key articles identified.
1. Business Model Innovation (Motor Theme – Upper Right)
* Position: Located in the upper right quadrant, indicating high centrality and high density.
* Interpretation: This is a “motor theme,” suggesting that business model innovation is a crucial and well-developed area of research within your dataset. It is both central to the field and actively researched.
* Key Articles:
* SJÖDIN D, 2023, TECHNOL FORECAST SOC CHANGE (PageRank: 0.317): This article likely explores future trends and changes in technology related to business model innovation.
* PIZZICHINI L, 2025, TECHNOL SOC (PageRank: 0.298): This article likely investigates the interplay between technology and society in the context of business model innovation, maybe exploring societal impact.
* PALMIÉ M, 2022, TECHNOL FORECAST SOC CHANGE (PageRank: 0.263): Similar to Sjödin, this article probably examines technological forecasting and societal changes impacting business model innovation.
* Overall Impression: The dominance of “Technological Forecasting and Social Change” suggests a forward-looking perspective and a concern with the broader societal implications of business model innovation.
2. Innovation (Basic Theme – Lower Right)
* Position: Located in the lower right quadrant, indicating high centrality and low density.
* Interpretation: This is a “basic theme,” meaning innovation is a fundamental and widely relevant concept but may not be as actively researched or developed *specifically within the context of your dataset*.
* Key Articles:
* BOCKEN N, 2022, TECHNOL FORECAST SOC CHANGE (PageRank: 0.328): Likely addresses the forecasting of technological advancements and their impact on society within the domain of innovation.
* SNIHUR Y, 2022, LONG RANGE PLANN (PageRank: 0.247): Likely focuses on long-term strategic planning in relation to innovation.
* RUITER H, 2022, SUSTAINABILITY (PageRank: 0.241): Likely investigates the intersection of innovation and sustainability.
* Overall Impression: The articles suggest a focus on long-term planning, forecasting, and the integration of sustainability principles within innovation strategies.
3. Circular Economy (Motor Theme – Upper Center/Left)
* Position: Located in the upper center/left quadrant, indicating high density but less centrality compared to “Business Model Innovation”.
* Interpretation: “Circular Economy” appears to be a well-developed theme that is significant to the field, and closely related to topics such as Industry 4.0 and reviews of the literature. It is not quite as *central* as “Business Model Innovation,” but has a strong research presence.
* Key Articles:
* OTTERBACH N, 2024, RESOUR CONSERV RECYCL (PageRank: 0.293): This article focuses on resource conservation and recycling within the context of the circular economy.
* CHAUHAN C, 2022, TECHNOL FORECAST SOC CHANGE (PageRank: 0.231): Similar to previous entries from this journal, it likely deals with technological forecasting and societal changes affecting the circular economy.
* MAIS F, 2024, SUSTAINABILITY-a (PageRank: 0.229): This article explores sustainability aspects within the circular economy.
* Overall Impression: The cluster’s key articles suggest a strong emphasis on resource efficiency, recycling, sustainability, and the role of technological advancements in enabling circular economy models.
4. 3D Printing (Niche Theme – Upper Left)
* Position: Located in the upper left quadrant, indicating low centrality and high density.
* Interpretation: “3D Printing” is a niche theme, meaning it’s a specialized area that is well-developed and actively researched but may not be as broadly relevant to the overall field covered by your dataset.
* Key Articles:
* LIM JJ, 2024, PROD PLANN CONTROL (PageRank: 0.181): Likely focuses on production planning and control aspects of 3D printing.
* DAI Y, 2025, J MANUF TECHNOL MANAGE (PageRank: 0.127): Likely addresses manufacturing technology and management related to 3D printing.
* JIN Y, 2022, EUR J INNOV MANAGE (PageRank: 0.093): Likely explores innovation management within the context of 3D printing.
* Overall Impression: The articles point to a focus on the practical applications of 3D printing in manufacturing, including production planning, technology management, and innovation processes.
Critical Discussion Points and Further Investigation
- Journal Representation: Notice the frequent appearance of “Technological Forecasting and Social Change.” This suggests that the research area is strongly influenced by studies considering the future impacts of technology on society. Consider whether this journal is over-represented in your dataset and how that might skew the results.
- Temporal Trends: The dates of the articles (mostly 2022-2025) suggest a relatively recent focus on these topics. Consider exploring how these themes have evolved over time (e.g., by filtering your analysis to different time periods).
- Inter-cluster Relationships: Consider how the clusters might interact. For example, how might 3D printing contribute to circular economy initiatives or enable new business models? You could perform additional analyses (e.g., co-occurrence analysis) to explore these relationships more directly.
- Parameter Sensitivity: Be aware that the position of clusters on the map is sensitive to the parameters used in the analysis (e.g., `minfreq`, `n.labels`, clustering algorithm). Experiment with different parameters to see how the map changes and assess the robustness of your findings.
- SCOPUS Bias: Remember that your analysis is based on data from SCOPUS. SCOPUS has its own biases in terms of journal coverage. Compare your findings to analyses using other databases (e.g., Web of Science) to assess the generalizability of your results.
- Community Detection Algorithm: The “walktrap” algorithm is a community detection method. Investigate if other community detection algorithms could lead to different cluster configurations and potentially highlight alternative relationships in your data.
In summary, this strategic map provides a valuable overview of the research landscape related to innovation, business models, circular economy, and 3D printing. By carefully interpreting the position of each cluster and considering the key articles within them, you can gain insights into the most important and actively researched areas within your field.
Factorial Analysis
Okay, let’s break down this factorial map derived from your bibliometric analysis of the SCOPUS dataset. This map uses Multiple Correspondence Analysis (MCA) on the merged keywords (“KW_Merged”) to visualize relationships between research themes. The parameters indicate that you’re focusing on single-word keywords (ngrams=1) and have filtered for terms appearing with a minimum degree of 14 (minDegree=14). The clustering (clust=1, k.max=8) suggests an attempt to identify up to 8 thematic clusters within the data, though the graph itself doesn’t explicitly show cluster boundaries. Stemming is disabled, and the labelsize is small (5) for readability. The fact that ‘graph=FALSE’ is specified suggests this is a static representation and not an interactive network visualization.
Overall Structure and Interpretation of Dimensions:
- Dimensions 1 & 2 Explained Variance: Dimension 1 explains 37.87% of the variance, and Dimension 2 explains 20.66%. This means that Dimension 1 is the primary axis differentiating the keywords in your dataset, followed by Dimension 2.
* What do the Dimensions Represent? Without deeper topic knowledge, it’s impossible to *definitively* label these axes. However, we can infer potential meanings based on the distribution of keywords:
* Dimension 1 (Horizontal Axis): The left side seems to be related to *Innovation Models in the Digital Era*, ‘Open Innovation’, ‘Business Model Innovation’,’Covid-19′ and ‘Digital Transformation’, while the right side appears to be linked to *Sustainability and Enterprise Resource Management*, including ‘Sustainable Business Model’, ‘Environmental Economics’ and ‘Enterprise Resource Management’. This dimension may capture the shift from traditional business processes to the integration of more contemporary digital, environmental, and sustainable practices.
* Dimension 2 (Vertical Axis): The upper area might represent *Strategic and Macro-Level Business Development*, encompassing ‘Strategic Approach’, ‘Literature Review’, and ‘Business Development’. The lower section concentrates on *Digital and Dynamic Capabilities*, including ‘Servitization’, ‘Digital Business’, and ‘Dynamics Capability’. Dimension 2 may represent the distinction between broad strategic approaches and specific, technologically-driven capabilities.
Cluster Identification (Inferred – No Explicit Clusters):
While the parameters suggest a clustering attempt, the visual output doesn’t show defined clusters. We can infer potential clusters based on proximity:
1. Cluster 1 (Upper Left): “Innovation Models in the Digital Era”: Keywords like “open innovation,” “business model,” “strategy,” and “COVID-19.” This suggests research focused on innovative business models, strategies, and how the COVID-19 pandemic has affected these.
2. Cluster 2 (Center-Left): “Digital Transformation and Innovation”: “Digital Transformation,” “Artificial Intelligence,” and “Business Model Innovation” are grouped together. This likely represents a body of research exploring the impact and processes of digital transformation.
3. Cluster 3 (Center): “Industry 4.0 and Related Technologies”: Terms like “Industry 4.0”, “Manufacturing”, “Business Models”, “Digital Technologies” and “Competition”. Focuses on technologies and applications that drive business transformation and competitiveness.
4. Cluster 4 (Center-Right): “Sustainability and Business Development”: “Sustainability”, “Innovation”, “Business Development,” “Sustainable Business Model”, “Environmental Economics”, and “Technological Development” indicate the convergence of these themes in the research.
5. Cluster 5 (Bottom): “Dynamic Capabilities and Servitization”: This cluster contains terms like “Dynamics Capability,” “Servitization,” “Digital Business”, “Digital Servitization”, and “Enterprise Resource Management.” This group could represent research on the dynamic capabilities necessary for organizations to adapt in a changing environment, particularly in the context of servitization and digital business models.
Relevance of Contributing Terms:
The position of a term on the map reflects its contribution to the dimensions. Terms further from the origin (0,0) have a stronger influence on defining the dimensions.
- Most Contributing Terms to Dimension 1: ‘Environmental Economics’, ‘Enterprise Resource Management’, ‘Open Innovation’, ‘Covid-19’, ‘Sustainable Business Model’
- Most Contributing Terms to Dimension 2: ‘Enterprise Resource Management’ and ‘Dynamics Capability’ (negative side); ‘Strategic Approach’ and ‘Environmental Economics’ (positive side).
Critical Discussion Points & Further Investigation:
- Limited Variance Explained: The two dimensions only explain about 58.53% of the total variance. This means there are other important factors (dimensions) not captured in this two-dimensional representation. Consider exploring additional dimensions in the MCA if possible.
- Missing Cluster Boundaries: While the ‘clust’ parameter was set, the graph lacks explicit cluster boundaries. You might need to explore the clustering results separately (e.g., using cluster membership tables) to understand the composition and characteristics of each cluster fully.
- Keyword Interpretation: Be cautious about over-interpreting individual keyword positions. The relationships are relative, and context is crucial.
- Database Bias: Remember this analysis is based on SCOPUS data. Results might differ with other databases (Web of Science, etc.).
- Temporal Trends: This is a static snapshot. Consider analyzing trends over time to see how these themes have evolved.
Recommendations for Further Analysis:
1. Examine Cluster Details: Investigate the specific documents associated with each inferred cluster to gain deeper insights into the research being conducted in each area.
2. Explore Additional Dimensions: Analyze additional dimensions of the MCA to capture more variance in the data.
3. Compare with Other Analyses: Compare these findings with other bibliometric techniques (e.g., co-citation analysis, bibliographic coupling) to triangulate the results and gain a more comprehensive understanding of the field.
4. Qualitative Review: Conduct a qualitative review of a subset of the most relevant papers to validate and enrich the interpretations derived from the quantitative analysis.
By considering these points, you can move beyond a simple description of the map and develop a more nuanced and insightful interpretation of the research landscape represented in your SCOPUS dataset. Remember to always relate your findings back to the specific research question you are trying to address.
Co-citation Network
Overall Network Structure:
- Core and Periphery: The network exhibits a clear core-periphery structure. A dense cluster, predominantly colored in red, sits at the center, indicating a group of highly co-cited publications. This central cluster likely represents the most influential and foundational works in your research area. Publications farther from the core, represented by smaller, more isolated nodes (and different colors), represent works that are less frequently co-cited or possibly address more niche or emerging aspects of the field.
- Communities/Clusters: The application of the walktrap community detection algorithm reveals distinct communities within the network. This suggests that the research field is not monolithic but is composed of several sub-disciplines or schools of thought. Each community (represented by a different color) groups together publications that are frequently cited together, indicating shared intellectual lineages or common research agendas.
Community-Specific Observations (Based on Available Information):
- The Red Cluster: The largest, densest (red) cluster appears to be centered around a paper by “foss n.j. 2017-1”. The high co-citation of this paper, and the other papers in that cluster, indicates that it is a cornerstone of the area investigated. Papers like “Massa I. 2015” and “Ghezzi A. 2020” are also part of this large community.
- The Green Cluster: Another community, indicated by the green nodes, includes authors like “Miles M.B. 1994” and “Eisenhardt K.M. 1989”. This cluster likely represents a different, but related, strand of research, perhaps focused on qualitative research or different context.
- Other Clusters: The presence of smaller clusters in other colors (e.g., blue, pink, orange, grey) suggests other, possibly less central or more specialized, research streams. For example the blue community has “Osterwalder A. 2010” and “Teece D.J.”
Interpretation and Critical Discussion Points:
1. Central Themes and Foundational Works:
* The dominance of “foss n.j. 2017-1” in the co-citation network implies that this paper is highly influential. You should investigate what is the theme of this paper and whether it is still relevant in the field.
* The presence of other key authors/papers within the central cluster highlights other critical contributions. Identify the main arguments and research questions addressed by these core publications to understand the fundamental themes driving the field.
2. Community-Specific Knowledge:
* Analyze the key publications within each community to understand the specific research focus. What are the core concepts, methodologies, and empirical settings associated with each community? Are there any key theoretical debates or methodological differences between them?
* The “walktrap” algorithm is useful but sensitive to parameters. Consider experimenting with different clustering algorithms (e.g., Louvain, Leiden) and parameter settings to assess the robustness of your community structure.
3. Peripheral Nodes and Emerging Trends:
* Pay attention to the isolated nodes or smaller clusters on the periphery of the network. These may represent newer research areas or interdisciplinary connections that are gaining traction.
* Identify any recent publications (e.g., those from 2020 or 2021) in the peripheral areas, as these might indicate emerging trends or shifts in the research landscape.
4. Limitations and Caveats:
* Remember that co-citation analysis captures intellectual influence as perceived through citation patterns. It doesn’t necessarily reflect the true impact or quality of a publication.
* The choice of bibliographic database (SCOPUS) and search terms can influence the composition of your dataset and, consequently, the structure of the co-citation network. Be mindful of potential biases and limitations in your data.
* The parameters used to generate the network (e.g., `edges.min`, `alpha`) affect the visual representation and interpretation of the results. Justify your choice of parameters and consider sensitivity analyses.
5. Suggestions for Further Exploration:
* Examine the content of the most highly cited papers (central nodes) and the key papers in each community. This involves reading the papers themselves to understand their arguments, methodologies, and findings.
* Combine co-citation analysis with other bibliometric techniques, such as co-word analysis or author co-citation analysis, to gain a more comprehensive understanding of the research field.
* Use the network as a starting point for a literature review. Explore the connections between different communities and identify potential research gaps or opportunities for future research.
By carefully examining the structure of your co-citation network, identifying the key publications and communities, and considering the limitations of the analysis, you can gain valuable insights into the intellectual structure and dynamics of your research field. Remember to always ground your interpretations in a thorough understanding of the underlying literature and to critically evaluate the assumptions and limitations of the bibliometric methods you employ. Good luck!
Collaboration Network
Overall Structure:
The network displays a fragmented structure, consisting of several distinct clusters or communities with limited connections between them. This suggests relatively siloed collaboration patterns within this research field. Instead of a large, interconnected web of researchers, we see smaller, more self-contained groups.
Community Detection (Walktrap Algorithm):
The Walktrap algorithm was used for community detection. This algorithm identifies communities based on the idea that random walks within a network tend to get “trapped” within dense regions, indicating community membership. The different colors in the network represent the different communities identified by the algorithm.
- The presence of multiple communities (indicated by different colors) suggests distinct research sub-areas or collaboration circles within the larger field. These communities likely have their own focused research questions, methodologies, or geographical concentrations.
Most Connected Authors (Label.n = 50):
The labels on the network show the most connected authors (top 50). Let’s analyze the most relevant ones.
* Parida V: It appears to be the most central node within the network. It suggests that Parida V is a key player in connecting different researchers or has collaborated extensively within the field. Their position suggests they might be a central figure in one or more research areas represented by the dataset.
* Wang Z, Liu Y: these appear to be key players in connecting different researchers within the blue and purple clusters respectively, that or they could have collaborated extensively within the field. Their positions suggest they might be central figures in one or more research areas represented by the dataset.
* The fact that these nodes are labeled indicates that they have a high degree of connectivity (number of connections to other authors) within the network.
Implications and Interpretation:
1. Collaboration Patterns: The network highlights the specific collaboration patterns within the field represented by your SCOPUS data. The fragmented structure implies that while some researchers are highly connected within their groups, there’s a lack of broader inter-community collaboration.
2. Identifying Key Players: Identifying the most connected authors (e.g., Parida V, Wang Z, Liu Y) allows you to pinpoint individuals who are either central to the field’s development, act as bridges between research groups, or lead influential research teams. Further investigation into their publications and research areas could provide insights into the dominant themes and directions within the field.
3. Community-Specific Research: The distinct communities suggest the existence of different research streams or sub-disciplines. Understanding the research focus of each community could reveal the diversity of topics investigated within the broader field and potential areas for cross-disciplinary collaboration.
4. Potential for Increased Collaboration: The relatively low connectivity between communities could indicate opportunities for fostering greater collaboration across these groups. Initiatives aimed at connecting researchers from different communities might lead to new insights and innovative research directions.
Critical Discussion Points:
- Database Bias: The structure of the collaboration network is dependent on the data source (SCOPUS). The coverage of SCOPUS might influence which authors and collaborations are represented. Consider comparing networks generated from other databases (e.g., Web of Science) to assess the robustness of the observed patterns.
- Timeframe: The analysis represents collaborations within the timeframe of your SCOPUS data. Collaboration patterns might evolve over time.
- Normalization: The normalization method (“association”) influences how co-authorship is weighted. Association strength reflects the degree to which two authors co-author more frequently than expected by chance, given their individual publication frequencies.
- Granularity: The author level is a specific choice. One could analyze collaboration at the institutional or country level to reveal broader patterns.
- Walktrap Limitations: Walktrap has a tendency to identify overly fine-grained community structure. Other algorithms might reveal broader clusters.
Further Steps:
- Examine the publications of the most connected authors to understand their research focus and how they connect to different research areas.
- Analyze the keywords associated with each community to identify the specific research topics they address.
- Investigate the institutions and countries affiliated with each community to understand the geographical distribution of research activity.
- Consider using different network analysis metrics (e.g., betweenness centrality) to identify authors who act as bridges between communities.
By carefully considering these points, you can gain a deeper understanding of the collaboration dynamics within your research field and identify opportunities for future research and collaboration.
