Overall Scope and Temporal Coverage:
- Timespan: 2017-2025: This is a relatively recent dataset, covering 8 years. This recency is important because it means the data reflects current trends in the field.
- Annual Growth Rate %: 13.94: A growth rate of nearly 14% per year signifies a rapidly expanding field of research. This could indicate increasing interest, new discoveries, or increased funding in this area. This rapid growth might warrant further investigation into the specific drivers of this expansion.
- Document Average Age: 2.99 years: This low average age indicates the literature is current. This is expected given the timespan, but it also underscores the dynamism of the research area. Findings might be quickly superseded, and researchers need to stay abreast of the latest publications.
Source and Document Characteristics:
- Sources (Journals, Books, etc.): 1195: The collection draws from a wide range of sources. This breadth suggests the research area is interdisciplinary or that the topic is discussed in many different venues. Consider analyzing the source list to identify the most influential journals, publishers, or conference proceedings. Are there specific journals that dominate the field?
- Documents: 4909: Almost 5,000 documents represent a substantial body of work. This provides a good foundation for in-depth bibliometric analysis.
- Average citations per doc: 29: This suggests a moderate impact, on average. Each document in the collection has been cited 29 times. This is important for a relative value comparison. Is this high, low, or average for the specific field that the documents belong to? Comparing this number with the average citations in similar fields can help evaluate the relative impact of the collection.
- References: 299672: The high number of references suggests a well-connected network of research. It also provides a rich dataset for citation network analysis to identify key foundational works and influential papers within the field.
Author and Collaboration Patterns:
- Authors: 10666: A high number of authors contributing to this field. It signifies a large and active research community.
- Authors of single-authored docs: 368: This implies that a portion of authors contribute alone to the field. This also provides a basis of comparison with the remaining authors.
- Single-authored docs: 436: The number of single-authored documents is relatively small compared to the total number of documents, indicating a collaborative research environment.
- Co-Authors per Doc: 3.23: This relatively high number (over 3 authors per document) confirms the collaborative nature of the research. Teamwork seems to be prevalent.
- International co-authorships %: 35.83: Over a third of the papers involve international collaboration, highlighting the global nature of the research. This global collaboration can lead to diverse perspectives and increased impact. You might want to investigate which countries are the most frequent collaborators.
Keywords and Focus Areas:
- Keywords Plus (ID): 7395:
- Author’s Keywords (DE): 10428: The distinct numbers of Keywords Plus and Author’s Keywords shows different emphasis and focus within the documents. Keywords Plus are assigned by the database (Scopus in this case), while Author’s Keywords are provided by the authors themselves.
Key Implications and Potential Further Analysis:
1. Dynamic and Growing Field: The high annual growth rate and recent timespan suggest a rapidly evolving area of research. Researchers in this area need to continuously update their knowledge.
2. Collaborative Research: The high co-authorship rate and international collaboration percentage point to a strong emphasis on teamwork and global knowledge sharing.
3. Moderate Impact: The average citations per document provide a baseline. Comparing this to other fields or subfields will provide context.
4. Scope and Breadth: The high number of sources and documents suggests a well-established field with diverse perspectives.
Next Steps for Deeper Analysis (using Biblioshiny or other tools):
- Identify Key Authors and Institutions: Who are the most prolific and highly cited authors and institutions? This can reveal the leading players in the field.
- Keyword Analysis: Explore the most frequent keywords (both author-provided and database-assigned). This will help identify the core themes and emerging topics.
- Citation Network Analysis: Map the citation relationships between documents to identify influential papers, research clusters, and the evolution of ideas.
- Thematic Evolution: Analyze how research themes have changed over time. This can reveal emerging trends and shifts in focus.
- Source Analysis: Identify the most influential journals or conferences in this research area.
- Co-authorship Network Analysis: Visualize the collaboration patterns between authors and institutions.
By combining these initial observations with more in-depth analyses, you can gain a comprehensive understanding of the research landscape and identify opportunities for future research. Remember to consider the limitations of Scopus as a data source and potential biases in the data. Good luck!
Annual Scientific Production
| 2017 | 244 | |
| 2018 | 304 | 24.59% |
| 2019 | 338 | 11.18% |
| 2020 | 440 | 30.18% |
| 2021 | 457 | 3.86% |
| 2022 | 663 | 45.08% |
| 2023 | 770 | 16.14% |
| 2024 | 1000 | 29.87% |
| 2025 | 693 | -30.70% |
Three-Field Plot
Overall Structure:
The plot is a network graph showing the relationships between three metadata fields:
- Left Field (CR): Cited References. These are the articles that are being cited *by* the documents in your Scopus collection.
- Central Field (AU): Authors. This lists the authors of the documents *in* your Scopus collection.
- Right Field (KW\_Merged): Keywords. These are the keywords associated with the documents *in* your Scopus collection. “Merged” likely indicates that the Scopus keywords and author-supplied keywords have been combined.
The connections (lines) show which authors have cited which references, and which authors are associated with which keywords.
Interpretation and Insights:
1. Dynamic Capabilities Focus:
* The most prominent connection is around “dynamic capabilities”. The cited references by Teece (various articles) strongly link to authors like Vrontis, Chatterjee, and Chaudhuri, which in turn link to the keyword “dynamic capabilities.” This clearly identifies dynamic capabilities as a central theme within your Scopus collection.
2. Key Influences and Foundational Work:
* The “CR” field reveals the core publications that are foundational to the research represented in your collection. Teece’s work on dynamic capabilities stands out. This suggests that this is a key theoretical or empirical starting point for many of the authors in your collection.
* The presence of Wernerfelt’s “resource-based view” and Barney’s work on “firm resources” suggests a connection to broader strategic management theories.
3. Author Clusters and Research Streams:
* The “AU” field suggests potential clusters of authors working on similar topics. Authors connected to the same cited references and keywords likely have overlapping research interests. For example, Wang J may be researching Absorptive Capacity. Further investigation into their publications would be needed to confirm this.
4. Keyword Themes:
* The “KW\_Merged” field shows the prominent themes being explored in the collection. Beyond “dynamic capabilities,” you see keywords like “innovation,” “sustainability,” “sustainable development,” “enterprise resource management”, and “digital transformation”. This indicates the breadth of topics related to the core theme.
5. Emerging Trends:
* The presence of “COVID-19” as a keyword suggests that some research within the collection is addressing the impact of the pandemic.
How Elements are Interconnected:
- Citation Patterns: The links between “CR” and “AU” reveal which authors are building upon or engaging with specific prior work. Heavily cited works are central to the intellectual foundation of the field as represented in the collection.
- Author-Keyword Associations: The links between “AU” and “KW\_Merged” show the topics that authors are actively researching and publishing on. This allows you to understand the research agenda of specific authors or clusters of authors.
- Reference-Keyword Associations (Indirect): While not directly linked, the plot indirectly shows the keywords associated with influential cited references. This can reveal the conceptual origins of different research areas.
Further Analysis and Critical Discussion:
1. Central Authors: Identify the authors with the most connections in the “AU” field. These are central figures in the network and may be key opinion leaders or researchers in the field.
2. Citation Classics: Determine the most frequently cited references in the “CR” field. These are foundational publications that have significantly influenced the field.
3. Thematic Evolution: Analyze the keywords in “KW\_Merged” to identify emerging trends and shifts in research focus over time.
4. Network Structure: Examine the overall network structure to identify clusters of authors, references, and keywords. This can reveal sub-disciplines or research communities within the collection.
5. Database Limitations: Recognize that the Scopus database is not exhaustive. The results may be biased towards publications indexed in Scopus.
By using this interpretation as a starting point, you can delve deeper into the specifics of your research collection and gain valuable insights into the structure and dynamics of the field.
Most Relevant Sources
Sources’ Local Impact
Sources’ Production over Time
Most Relevant Authors
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
Reference Spectroscopy
Overall Interpretation:
The RPYS plot illustrates the historical development of the intellectual foundations of the research area represented in your Scopus-derived dataset. The black line visualizes the overall citation frequency of publications from different years, providing a general sense of the age distribution of cited literature. The red line highlights years that have an unexpectedly high impact, given the citation rates in the preceding five years. Peaks in the red line identify “seminal” years – those contributing disproportionately to the intellectual heritage of the field.
Key Observations and Interpretations Based on the Data:
1. Dominated Reference Age: The steep increase in the black line after 1990 suggests that the knowledge base is heavily influenced by research published from the 1990s onwards. This might indicate a relatively young field or a period of rapid development and paradigm shifts. This also can be due to data limitations from the Scopus database, if there is minimal pre 1990 publication coverage.
2. Early Seminal Years (1981, 1991, 1997):
* 1981: The prominence of Fornell and Larcker’s work on structural equation modeling (SEM) indicates that SEM is a fundamental methodological or theoretical framework used in this research area. Their work likely laid the groundwork for many subsequent studies. A critical discussion point could be whether current research sufficiently acknowledges or builds upon the nuances and limitations discussed in these early SEM publications.
* 1991: The appearance of Barney’s work on resource-based view (RBV) and March’s work on exploration vs. exploitation suggests a strong interest in strategic management and organizational learning within the research field. This early focus on RBV might indicate a concern with internal capabilities and competitive advantage.
* 1997: The repeated appearance of Teece, Pisano, and Shuen’s work on Dynamic Capabilities indicates that this is a key theoretical area.
3. Dynamic Capabilities Core (2000, 2003, 2007, 2009, 2018, 2020): The clustering of peak years around the topic of dynamic capabilities after the year 2000 provides strong evidence for the importance and sustained influence of this theory within the analysed research field. The presence of influential papers from Eisenhardt & Martin (2000), Winter (2003), Teece (2007), Ambrosini & Bowman (2009) shows the theoretical roots of the research field. The more recent papers by Teece (2018), Schilke et al. (2018), Laaksonen and Peltoniemi (2018), Ferreira et al. (2020) and Qiu et al. (2020) signal active discourse and development within the field. A critical discussion point may be whether the Dynamic Capabilities framework has been interpreted and applied consistently, or if there are significant variations and debates in its use. Also, one should be wary that the algorithm could have highlighted some recent papers due to the recent uprise in Dynamic Capabilities popularity but these citations may not be as ‘impactful’ as the original papers.
4. Methodological Considerations (2003, 2015):
* 2003: Podsakoff et al.’s work on common method biases (CMB) indicates a concern within the field for methodological rigor, particularly in behavioral research.
* 2015: Henseler et al.’s work on discriminant validity in structural equation modeling reflects an ongoing concern with the appropriate use and validation of statistical techniques. These might indicate a critical self-awareness within the field regarding research methods.
Questions for Critical Discussion:
- Theoretical Diversity: To what extent does the research field rely on a limited set of theoretical frameworks (e.g., Dynamic Capabilities, Resource-Based View)? Are there alternative perspectives that are under-represented?
- Methodological Innovation: Does the field actively engage with emerging research methods, or is it heavily reliant on established techniques like SEM?
- Contextual Specificity: How well do the dominant theories and methods translate to different contexts (e.g., different industries, countries, or types of organizations)?
- Maturity of the Field: Is the field experiencing diminishing returns from continued reliance on established theories, or are there ongoing efforts to refine and extend these frameworks?
Further Investigation:
- Content Analysis: Conduct a more in-depth content analysis of the most cited publications to identify key themes, concepts, and debates.
- Citation Network Analysis: Examine the citation relationships between the seminal publications to understand how ideas have evolved and spread within the field.
- Author Co-citation Analysis: Identify influential authors and research groups, and explore their contributions to the field.
By combining the RPYS plot with a careful examination of the cited literature, you can gain valuable insights into the intellectual history, current state, and future directions of your research area. Remember to consider the limitations of your data source (Scopus) and the potential biases inherent in citation analysis.
Most Frequent Words
TreeMap
Trend Topics
Overall Interpretation:
This plot visualizes the evolution of research topics in your SCOPUS dataset over time. The terms displayed represent the keywords that have the highest median frequency within the documents published in each year. The position of the bubbles on the x-axis (Year) indicates when a term was most prominent. The size of each bubble reflects the overall frequency of that term in the given year, providing a sense of its importance. The lines extending from the bubbles (the interquartile range) give an idea of how consistent the frequency of the term was throughout that year. A wider line suggests more variability in frequency.
Key Observations and Potential Insights:
1. Temporal Trends:
* Emerging Topics: Look at the topics that appear in the later years (2023-2025). These are likely the emerging areas of research. For example, “sustainable development goals,” “business, management and accounting” and “green development” appear as new prominent keywords in the later years. This may reflect an increasing focus on sustainability-related research in your dataset. The prominent appearance of keywords like “artificial intelligence” and “digital transformation” in recent years signals their growing importance in the literature captured by your query.
* Declining/Constant Topics: Topics that appear early in the plot (e.g., from 2017-2019) and do not continue to the later years are either declining in prominence or have been replaced by newer terms. Keywords appearing throughout the period, but without significant changes in bubble size might represent persistent research areas.
* Sudden Spikes: A keyword that suddenly appears with a large bubble indicates a rapid increase in interest in that topic in that particular year.
2. Topic Clusters:
* Related Terms: Look for keywords that appear close together in time. This might indicate clusters of related research areas. For example, the cluster around “dynamic capabilities”, “enterprise resource management”, and “ambidexterity” in 2021 suggests a possible connection in research themes.
3. Frequency & Consistency:
* Bubble Size: The size of the bubble tells you the relative frequency of the keyword in that year. Larger bubbles indicate a more common term.
* Interquartile Range: The length of the light blue line shows the spread of the data. A longer line means the term’s frequency varied a lot throughout the year. A shorter line suggests more consistent usage of the term.
Interpretation of Specific Terms (Based on the Image):
- “Societies and institutions”, “genetic algorithms” and “cloud computing”: are topics that were most prominent earlier in the studied period, around 2017 and 2019. Their absence in later years might indicate a shift in research focus.
- “Dynamic capability”, “dynamic capabilities”, “enterprise resource management”, “ambidexterity”, “strategic management”: These terms reached their peak around 2021, with some displaying a considerable annual frequency. The interquartile range appears to be relatively contained, suggesting consistent usage throughout the year.
- “Sustainability”, “sustainable development”, “innovation”, “artificial intelligence” and “digital transformation”: these keywords display high prominence in the more recent years, specially around 2023.
- “sustainable development goals”, “business, management and accounting” and “green development”: these keywords display high prominence in the most recent years, around 2025.
Critical Considerations & Further Investigation:
- Keyword Merging (`KW_Merged`): Be mindful of how the keywords were merged. Are there any inconsistencies or potential issues with how terms were combined?
- SCOPUS Bias: Remember that this analysis is based on SCOPUS data. The results might not be representative of all research in the field, as SCOPUS has its own coverage biases.
- Query Specificity: The trends you’re seeing are directly related to the search query you used to download the data from SCOPUS. Changing the query will change the results.
- Contextual Understanding: To fully interpret these trends, you need to have a good understanding of the research domain and the specific context of the keywords.
- Qualitative Analysis: Consider supplementing this quantitative analysis with a qualitative review of the publications associated with these trending topics. This will provide deeper insights into the underlying research themes and discussions.
Next Steps for the Researcher:
1. Refine Search Query: If the results are not what you expected, consider refining your search query in SCOPUS.
2. Explore Specific Trends: Choose a few of the most interesting trends and investigate them further. Read the abstracts and full text of relevant articles to understand the context.
3. Compare with Other Datasets: If possible, compare these trends with data from other bibliographic databases (e.g., Web of Science) to see if the patterns are consistent.
4. Consider Alternative Visualizations: Explore other visualization options in Biblioshiny (e.g., co-occurrence networks, thematic maps) to gain different perspectives on the data.
By carefully considering these points, you can use this trend topics plot to gain valuable insights into the evolution of research in your field. Good luck!
Clustering by Coupling
Co-occurrence Network
Overall Structure:
The network exhibits a clear community structure, visually represented by different colored clusters. This indicates the presence of distinct, yet interconnected, research themes within the Scopus collection based on co-occurring keywords. The network’s density varies, with some clusters being more tightly knit (higher density of connections) than others, suggesting stronger associations between keywords within those specific themes. The node sizes represent the frequency of occurrence of the keyword, while the edge thickness represents the strength of the association between the keywords (based on the “association” normalization).
Communities/Topics:
Based on the keywords and their clustering, here’s a possible interpretation of each community:
- Red Cluster (Center-Right): This cluster is prominently centered around “dynamic capabilities”. Other key terms include “enterprise resource management”, “resilience”, “circular economy”, “digitalization”, “open innovation”, “big data”, “supply chain management”, “innovation performance”, “Industry 4.0”, “Knowledge Management”, “Competitive Advantage” and “SMEs”. This community likely represents research focusing on how organizations adapt and innovate in response to changing environments, often leveraging technology and data-driven approaches, especially within the context of supply chain dynamics. The presence of “resilience” suggests an emphasis on coping with disruptions and building robustness.
- Green Cluster (Bottom-Left): Centered on “innovation,” this cluster connects to terms such as “business”, “manufacturing small and medium-sized enterprise”, “industrial performance”, “sustainability performance assessment”, “competitiveness” and “strategic approach”. This cluster appears to focus on the practical implementation of innovation in business contexts, particularly within manufacturing and SMEs. It emphasizes the link between innovation, sustainability, competitiveness, and overall business performance.
- Blue Cluster (Top-Left): This community includes terms like “decision making”, “structural equation modeling”, “COVID-19”, “China”, “commerce”, “artificial intelligence” and “entrepreneurship”. This cluster seems to represent a set of areas impacted by external factors like the pandemic and global events. The presence of “structural equation modeling” suggests the use of quantitative methods to study relationships between these factors. The inclusion of “artificial intelligence” and “entrepreneurship” hints at the role of technology and new business ventures in navigating these changes.
- Purple Cluster (Left): The “human” term in this cluster has few connections.
- Orange Cluster (Bottom-Right):The “article” term has very few connections.
Most Connected Terms and Their Relevance:
- Dynamic Capabilities: As the largest node and the central point of the most prominent cluster, “dynamic capabilities” clearly represents a core theme in the analyzed literature. Dynamic capabilities refer to an organization’s ability to integrate, build, and reconfigure internal and external competences to address rapidly changing environments. Its high connectivity suggests it’s a central concept linking various aspects of organizational adaptation and innovation.
- Innovation: As the focal point of another significant cluster, “innovation” is another key theme. It’s interconnected with business performance, sustainability, and manufacturing, highlighting its practical relevance in various industries.
- Other Highly Connected Terms (Red Cluster): Terms like “enterprise resource management”, “supply chain management”, “digitalization”, “industry 4.0” and “resilience” are highly connected within the “dynamic capabilities” cluster, indicating their strong association with the core theme. These terms reflect the technological and operational aspects of organizational adaptation and innovation in the modern business environment.
Interpretation Considerations and Potential Research Questions:
- Evolution of Research: Consider comparing this network with networks generated from different time periods to understand how research interests have evolved over time. Are dynamic capabilities becoming more intertwined with specific technologies or industries?
- Bridging Themes: The links between the clusters (“innovation” and “dynamic capabilities”) suggests that there may be studies that explore the role of innovation in fostering or enabling dynamic capabilities. Explore these connecting pathways in the original literature.
- Contextual Factors: The “COVID-19” term in a more weakly connected cluster suggests the influence of external shocks on the field. It would be useful to explore how these external factors have shaped the research landscape related to innovation and dynamic capabilities.
- Specificity vs. Generality: While “dynamic capabilities” and “innovation” are broad concepts, the network reveals specific areas of application (e.g., supply chain, manufacturing, SMEs). This suggests that research is becoming more focused on the practical implementation of these concepts within specific contexts.
Critical Assessment:
- Data Source Limitations: The analysis is based on Scopus data. It is important to remember that this is just one database and the results might differ if other databases were used.
- Keyword Bias: Keyword analysis can be influenced by author keyword choices. Some relevant concepts might not be adequately represented if authors use different terminology.
- Parameter Sensitivity: The network structure and community detection are sensitive to the parameters used (e.g., normalization method, clustering algorithm). Different parameters might yield slightly different results. The ‘walktrap’ community detection algorithm tends to find communities with a relatively balanced size.
In summary, this word co-occurrence network provides a valuable overview of the research landscape related to innovation and dynamic capabilities. It highlights the core themes, their interconnections, and the influence of contextual factors. Further investigation of the underlying literature, considering the limitations discussed above, can provide deeper insights into the evolution and current state of research in this area.
Thematic Map
Factorial Analysis
Overall Structure and Key Parameters:
- Data Source: SCOPUS.
- Field of Analysis: `KW_Merged`, which means the analysis is based on a merged keyword field, likely combining author keywords and keywords plus. This is good because it captures a wider range of relevant terms.
- N-grams: `ngrams: 1`. This indicates that the analysis is considering individual words (unigrams) rather than phrases (bigrams, trigrams, etc.). This makes sense for a broad overview, but it may miss some nuanced relationships.
- Method: `MCA` (Multiple Correspondence Analysis). MCA is suitable for analyzing categorical data, such as keywords, to identify underlying dimensions and relationships between them.
- MinDegree: `minDegree: 76`. This is a very important parameter. It means that only keywords appearing in at least 76 documents were included in the analysis. This significantly filters the data and focuses on the *most frequent* terms. Be mindful that this threshold *excludes* less frequent but potentially relevant or emerging topics.
- Clustering: `clust: 1`. A clustering algorithm was applied with a k-max of 8 (maximum number of clusters). The specific clustering method isn’t specified, but clustering helps to identify groups of keywords that tend to co-occur.
- Stemming: `stemming: FALSE`. Stemming reduces words to their root form (e.g., “running” becomes “run”). Since stemming is off, the analysis distinguishes between different word forms.
- Dimensions: The map is displayed on two dimensions (Dim 1 and Dim 2), which explain 35.73% and 25.39% of the variance in the data, respectively. Dimension 1 captures the most variance and would be the most informative.
Interpretation of the Map:
1. Dimension Interpretation (Horizontal – Dim 1: 35.73%):
* Left Side: Terms like “industrial management”, “enterprise”, “competition”, “resource management”, and “supply chains” appear on the left side of the map. This suggests a focus on traditional operational and managerial aspects of businesses, specifically related to the supply chain and resource allocation in a competitive setting.
* Right Side: Terms such as “firm performance,” “dynamic capabilities,” “entrepreneurial orientation,” and “innovation” are positioned on the right. This indicates a focus on strategic outcomes and the factors that drive them, such as innovation and entrepreneurial strategies.
* *Potential Interpretation of Dimension 1:* Dimension 1 seems to contrast operational/tactical management (left) with strategic/performance-oriented management (right).
2. Dimension Interpretation (Vertical – Dim 2: 25.39%):
* Upper Side: Keywords like “manufacturing,” “China,” “sustainability,” “small and medium-sized enterprise,” “sustainable development,” and “circular economy” are located at the top. This cluster suggests a focus on sustainability practices, manufacturing in China, and their application to small and medium enterprises (SMEs).
* Lower Side: The bottom part of the map is where most of the more traditional management terms are located.
* *Potential Interpretation of Dimension 2:* Dimension 2 seems to contrast a focus on sustainability and manufacturing contexts (top) with more general/traditional management (bottom)
3. Clusters and Themes:
* Based on proximity, we can identify several potential thematic clusters:
* Cluster 1 (Top-Left): “Manufacturing,” “China,” “small and medium-sized enterprise”. This suggests research focusing on manufacturing industries, particularly SMEs in China.
* Cluster 2 (Top-Middle): “Sustainability,” “sustainable development,” “circular economy”. This indicates a strong focus on environmental and economic sustainability.
* Cluster 3 (Center): “Artificial intelligence,” “COVID-19,” “digital transformation”. This reflects a more recent trend of research focusing on the impact of AI and digitalization, especially in light of the COVID-19 pandemic.
* Cluster 4 (Right): “Firm performance,” “dynamic capabilities,” “entrepreneurial orientation,” “innovation”. This is a strategy/performance cluster.
* Cluster 5 (Bottom-Left): “Industrial management”, “enterprise”, “competition”, “resource management”, and “supply chains”.
Critical Considerations and Next Steps:
* MinDegree Threshold: The `minDegree = 76` parameter drastically shapes the map. While it highlights the most prominent themes, it omits potentially emerging or niche areas. Consider re-running the analysis with a lower `minDegree` (e.g., 20 or 30) to explore a broader range of topics, but be prepared for a more complex map. You might also consider adjusting this value based on the overall size of your dataset. A higher value is appropriate for very large datasets.
* N-grams: Using n-grams (e.g., bigrams) could capture more complex relationships (e.g., “supply chain management” instead of just “supply” and “chains”).
* Cluster Analysis: The plot only shows the MCA. The number of clusters was determined by k.max. It would be good to know what specific clustering method was used and to visualize the actual clusters on the map to confirm these thematic groupings. The clust parameter set to “1” means that the clustering was performed.
* Database Coverage: Remember that the results are based solely on SCOPUS. Consider comparing with results from other databases (Web of Science, Dimensions, etc.) to assess the robustness of the findings.
* Further Analysis: This map provides a high-level overview. Now you can use this as a springboard for more focused analyses:
* Explore the *relationships* between the clusters. For example, is there research connecting sustainability to firm performance?
* Examine the *evolution* of these themes over time. Are some clusters becoming more or less prominent?
* Conduct a *content analysis* of the most cited papers within each cluster to understand the specific research questions and findings.
By critically evaluating the parameters and using the map as a starting point, you can gain valuable insights into the research landscape and identify opportunities for future investigation. Remember that this is just one view of the data, and further exploration is crucial for a comprehensive understanding.
Co-citation Network
Overall Structure:
The network visually displays the co-citation relationships between cited references. A co-citation network connects items that are cited together in other publications. The stronger the connection (i.e., the thicker the edge), the more frequently those two references are cited together. The placement of nodes (references) indicates which references tend to be co-cited more frequently. The node size (size=5; size.cex=TRUE) reflects the overall frequency with which a reference is cited within the analyzed collection (i.e., betweeness centrality).
Communities (Clusters):
The network is colored based on community detection using the Walktrap algorithm (cluster: walktrap). This algorithm attempts to identify groups of nodes that are more densely connected to each other than to the rest of the network. We see several distinct color-coded communities, which likely represent distinct intellectual or thematic clusters within the literature.
- Red Cluster (Dominant): The largest and most central cluster (red) is heavily influenced by `Teece D.J. 1997-1`, which is at the very core of the network. This cluster likely represents a core set of literature focused on a key topic or framework.
- Green Cluster: Appears to link seminal works to `Teece D.J. 1997-1`. Authors like `Barney, J. (1991)` and `Wernerfelt, B. (1984)` are probably linked to the dynamic capabilities framework.
- Purple Cluster: This community seems to be composed of authors such as `Hair J.F. (2019)`. This could be related to quantitative methodologies, since Hair is a renowned author on data analysis techniques.
- Other Clusters (Orange, Blue, Gray): The other clusters represent distinct, but potentially related, areas of research. Further investigation, looking at the content of the papers within each cluster, would reveal the nature of the differences between these areas.
Most Connected Terms:
The parameters specify that the top 50 most connected nodes are labeled (`label.n: 50`). These are the most frequently cited references within this network.
- `Teece D.J. 1997-1` (Central Node): The prominence of “Teece D.J. 1997-1” suggests that this publication (likely related to Dynamic Capabilities) is a central and influential piece in the field. The size of the node indicates that it is the most cited reference in the dataset.
- Other Prominent Authors/Works: Pay attention to other highly connected nodes like `Barney J. (1991)`, `Winter S.G`, `Cohen W.M. (1990)`, and `March J.G. (1991)`. These are seminal works.
Interpretation and Further Investigation:
1. Centrality of Dynamic Capabilities: The strong presence of `Teece D.J. 1997-1` suggests that the research area is heavily influenced by the dynamic capabilities framework.
2. Community Themes: You should examine the publications within each cluster to determine the specific topics, methodologies, or theoretical perspectives that define them. The communities detected via walktrap algorithm should represent distinct areas of research within the collection you have provided.
3. Relationship Between Communities: The connecting edges between communities indicate relationships or overlaps between these areas. Are there some communities that are more strongly linked than others? `Teece D.J. 1997-1` has a very large amount of edges, meaning that its cited references have been cited with the remaining authors of the graph a number of times.
4. Evolution of the Field: Consider the publication dates. Are there older clusters representing foundational work, and newer clusters representing more recent developments? How do the various clusters relate to one another temporally?
5. Database Influence: Remember this analysis is based on SCOPUS data. SCOPUS has a particular coverage profile. Consider if this database is truly representative of the field you are studying.
Critical Discussion Points:
- Limitations: Co-citation analysis has limitations. For example, it only considers cited references.
- Novelty: Does this network reveal any unexpected connections or overlooked areas of research?
- Future Directions: Based on the structure and clusters within this network, what are some promising directions for future research?
By further investigating the content of the key publications and the themes of each community, you can develop a richer and more nuanced understanding of the research landscape represented by this co-citation network. Let me know if you want help with analyzing particular authors, communities or making improvements on the graph!
Historiograph
Collaboration Network
Overall Structure and Key Observations
- Clustering: The network clearly exhibits community structure, as identified by the Walktrap algorithm. There are at least 5 distinct clusters that indicate distinct groups of collaborating authors. The clustering is algorithm-based, so the meaning must be inferred from the authors in each cluster.
- Disconnected Components: A key observation is the presence of relatively disconnected components. The orange cluster containing “Khan Z” and “Zahoor N” appears isolated from the main network. This suggests that these authors work within a specific subfield or have limited collaboration with the other research groups represented. There are likely distinct bodies of literature for each clusters.
- Hub Authors: Within the larger clusters, some authors appear to be more central and connected (higher degree centrality). Names like “Chaudhuri R”, “Chatterjee S”, “Wang Y”, and “Zhang Y” are larger (indicating the prominence of their contributions to the field) and well connected in this visualization. These are likely leading researchers or brokers of collaboration in their respective communities.
- Data Source: The collection downloaded from SCOPUS means the collaborations in this network are those indexed by SCOPUS. The network may be different if another data source was used.
Community-Specific Insights
1. Central Brown Cluster (Chaudhuri R, Chatterjee S, Vrontis D): This seems to be one of the more prominent clusters, potentially representing a specific research group or a concentration of researchers in a particular area. They may be working on a niche topic or in a location where they have frequent collaborations. Their connections to the green cluster suggest cross-community collaboration.
2. Eastern Red Cluster (Wang Y, Zhang Y, Chen J): This is another large and dense cluster. It also appears to be a group of highly collaborative authors. It appears to be working on a different field than the other clusters, as the interaction is limited.
3. Northern Purple Cluster (Wang I, Zhang Q): This is a small cluster that has limited interaction with the other clusters, suggesting that they are working in an independent field.
4. Green Cluster (Gupta S, Kumar A): They have some collaborations with other clusters (particularly the brown one), and they serve as a bridge for the clusters.
5. Small Blue Cluster(Singh RK, Dubey R): Small cluster working on the same topic, that has little interaction with the other clusters.
6. Isolated Orange Cluster (Khan Z, Zahoor N): As mentioned earlier, their isolation suggests a distinct research focus or a geographic separation that limits collaborations with the other groups.
Implications and Interpretation
- Research Foci: The community structure suggests the existence of distinct subfields or research areas within the overall topic covered by your SCOPUS collection. By examining the publications of authors within each cluster, you can gain a deeper understanding of the specific topics being addressed by each community.
- Collaboration Patterns: The network reveals patterns of collaboration, highlighting which researchers are working together and how different research groups are connected. The thickness of the edges (though set to a fixed size here) typically represents the strength of collaboration (e.g., number of co-authored publications). If variable edge sizes were displayed, you could identify the strongest collaborative relationships.
- Knowledge Diffusion: The links between clusters suggest potential pathways for knowledge diffusion and cross-disciplinary collaboration. Authors who bridge different communities play a crucial role in transferring knowledge and fostering innovation.
- Potential Gaps: The disconnected components might represent areas where more collaboration is needed. Investigating why these groups are isolated could reveal opportunities for fostering new research partnerships.
Critical Considerations
- Data Coverage: Remember that this network is based on data from SCOPUS. Collaborations outside of publications indexed by SCOPUS will not be represented.
- Normalization: The “association” normalization method influences the appearance of the network. Consider what this normalization emphasizes.
- Algorithm Sensitivity: The Walktrap algorithm is one of many community detection algorithms. Different algorithms might reveal slightly different community structures. The parameter `community.repulsion = 0.05` is also algorithm dependent and changes how the clusters are represented.
- Edge Weight Interpretation: Because the `edgesize` is fixed, the edges are not representative of the strength of the collaboration, as this parameter is designed to represent the strength of the collaboration.
Recommendations for Further Analysis
1. Content Analysis: Analyze the publications of authors within each cluster to identify the key themes, research questions, and methodologies being used.
2. Keyword Analysis: Perform keyword co-occurrence analysis on the publications to identify the dominant topics within each community and the relationships between them.
3. Temporal Analysis: Examine how the collaboration network has evolved over time. Are new communities emerging? Are existing communities becoming more or less connected?
4. Author-Level Metrics: Calculate author-level metrics such as degree centrality, betweenness centrality, and PageRank to identify the most influential and central authors in the network.
5. Consider Alternative Visualizations: A different layout algorithm might highlight different aspects of the network structure.
By combining the insights from the network analysis with a deeper understanding of the research content, you can gain valuable insights into the structure, dynamics, and intellectual landscape of the field you are studying.
Countries’ Collaboration World Map
