Bibliometrics is a key discipline for the quantitative and qualitative analysis of scientific output. This exercise enables the identification of trends, relationships between concepts, and key actors in a specific knowledge area. Below, we explore the main metrics, tools, and methodologies used in bibliometrics, ending with a practical example.
Common Metrics in Bibliometrics
| Metric | Description |
|---|---|
| Co-citation | Analyzes how often two documents are cited together, identifying thematic relationships. |
| Author Collaboration Network | Studies collaboration networks between authors, visualizing co-authorships and working groups. |
| Country Collaboration Network | Measures international collaborations, representing interactions between countries. |
| Keyword Collaboration Network | Examines relationships between keywords, showing thematic areas and their evolution. |
| Citation Analysis | Evaluates the impact of documents based on the number of citations received. |
| Bibliographic Coupling | Identifies documents that cite common references, revealing thematic affinity. |
Tools and Resources for Bibliometrics
There are various tools and platforms that facilitate the development of bibliometric exercises. Among the main ones are some that use artificial intelligence to enhance analyses:
Libraries:
bibliometrix: R package for bibliographic data analysis and visualization.
igraph: Package used to create and analyze complex networks.
vosviewer: Software tool for bibliometric network analysis and visualization, especially citations and co-citations.
Bibliographic Databases:
Web of Science (WoS): Bibliographic database providing access to high-quality articles and scientific citations.
Scopus: Multidisciplinary database offering metrics and citation analysis.
PubMed: Database focused on medical and biological sciences.
Google Scholar: Academic literature search engine including articles, theses, and books.
Online Platforms:
CiteSpace: Platform facilitating visualization of scientific patterns in the literature.
Publish or Perish: Tool for extracting and analyzing citations from various sources.
Dimensions AI: Bibliometric analysis platform with AI algorithms.
Connected Papers: Interactive application to explore related scientific documents through citation networks.
ResearchRabbit: Literature exploration tool based on machine learning.
Scite.ai: Tool analyzing citations with context and classifying them as supportive, contradictory, or neutral.
These tools and resources enable detailed literature analysis, generating interpretative graphs and tables with a user-friendly interface. It is worth noting that users have created codes to delve deeper into the traditional knowledge area of bibliometrics, such as Tree of Science.
Methodology: Tree of Science
The Tree of Science(ToS) is a methodology developed by Zuluaga et al. (2022) that structures scientific literature into three hierarchical levels. This methodology adds value by facilitating the identification of conceptual pillars in a research field, enabling efficient prioritization and organization of knowledge. By categorizing literature into hierarchical levels, ToS not only helps understand the evolution of knowledge but also fosters informed decision-making in research and development processes. It is particularly valuable for identifying research gaps and suggesting new exploration avenues.
Roots: Fundamental articles that originated the field of study. These are usually highly cited publications.
Trunk: Studies building on the roots, consolidating foundational knowledge.
Leaves: Recent publications exploring specific applications or new thematic areas.
The ToS organizes documents by relevance and impact, facilitating an understanding of how a knowledge field evolves.
How to Implement It:
Extract data from Web of Science and/or Scopus in a compatible format.
Process the data using tools like bibliometrix.
Identify tree levels based on citations and co-citations.
Visualize results to interpret key trends and relationships.
Examples
Bugge et al. (2016). What Is the Bioeconomy? A Review of the Literature
Bugge, Hansen, and Klitkou (2016) conducted a bibliometric exercise to identify the main paths of the bioeconomy. This article classifies the bioeconomy into three main visions:
Biotechnology Vision: Focused on research and commercialization of biotechnology to generate economic growth and employment.
Biological Resources Vision: Highlights the use of biological raw materials and the establishment of new value chains.
Bioecology Vision: Centered on sustainability, ecological process optimization, and biodiversity conservation.
While countries and authors redefine the concept of bioeconomy, most adopted the paths described in the article as a guide for public policies. This underscores the importance of integrative visions enabling an orderly transition to the bioeconomy, prioritizing economic, social, and environmental objectives in a balanced manner. Therefore, finding insights from multiple information sources is useful to standardize processes, build metrics, and drive strategies to promote new alternatives like the bioeconomy.
Using Tree of Science
In this exercise, templates provided by Tree of Science in its GitHub were used to analyze a collection of scientific articles related to three areas of interest associated with the bioeconomy (nature-based scientific tourism, regenerative livestock, and natural ingredients), using Scopus and WoS bibliographic databases. This method structures literature into roots, trunk, and leaves, facilitating the understanding of conceptual bases, current developments, and emerging areas in the field. Below are three PDF files containing the selected articles for this analysis in a template that facilitates result interpretation.
Some Tips
✅ Define a precise search equation: Before starting, ensure the search equation adequately reflects keywords, logical combinations (AND, OR, NOT), and thematic areas you wish to analyze.
✅ Select appropriate databases: Evaluate which database is best for your study area, as some are stronger in certain disciplines (e.g., PubMed for medical sciences).
✅ Use data cleaning tools: Make sure to remove duplicates and normalize author, institution, and keyword names for consistent results.
✅ Visualize results correctly: Use tools like VOSviewer or bibliometrix to generate comprehensible and aesthetically clear network maps.
✅ Interpret the data: Beyond generating graphs, spend time interpreting what the trends, networks, and collaborations mean.
✅ Document the process: Keep a detailed record of the steps followed, from the search to the analysis, to facilitate reproducibility and result verification.
APA
Amaya Guzmán, B. (2024, December 29). How to perform a bibliometric analysis? brianamaya.co. https://brianamaya.co/post_website/2024-12-29%20bibliometric_post/en/index.html
BibTeX
@misc{amaya2024biblio,
author = {Amaya Guzmán, Brian},
title = {How to perform a bibliometric analysis?},
year = {2024},
month = dec,
day = {29},
url = {https://brianamaya.co/post_website/2024-12-29%20bibliometric_post/en/index.html},
note = {Blog post, brianamaya.co}
}