[Paper Review] Usage of OpenAlex for creating meaningful global overlay maps of science on the individual and institutional levels

Robin Haunschild, Lutz Bornmann|arXiv (Cornell University)|Apr 3, 2024

Big Data and Business IntelligenceBusiness, Management and Accounting3 citations

TL;DR

This paper proposes a method for generating global overlay maps of scientific output using OpenAlex data, enabling visualization of individual and institutional research performance across disciplines. By normalizing publication counts and leveraging six base maps, the approach reveals field-specific research focus more clearly than raw data, offering a scalable tool for science mapping at global scales.

ABSTRACT

Global overlay maps of science use base maps that are overlaid by specific data (from single researchers, institutions, or countries) for visualizing scientific performance such as field-specific paper output. A procedure to create global overlay maps using OpenAlex is proposed. Six different global base maps are provided. Using one of these base maps, example overlay maps for one individual (the first author of this paper) and his research institution are shown and analyzed. A method for normalizing the overlay data is proposed. Overlay maps using raw overlay data display general concepts more pronounced than their counterparts using normalized overlay data. Advantages and limitations of the proposed overlay approach are discussed.

Motivation & Objective

• To develop a reproducible method for creating global overlay maps of scientific output using OpenAlex data.
• To compare the visual impact of raw versus normalized publication data in overlay maps.
• To demonstrate the utility of the method through case studies of one researcher and his institution.
• To evaluate the strengths and limitations of using OpenAlex for large-scale science mapping.
• To provide six standardized base maps for consistent global science visualization.

Proposed method

• The authors use OpenAlex's comprehensive scholarly dataset to extract publication records by author and institution.
• They apply a normalization procedure to publication counts per field to reduce bias from high-output disciplines.
• Six distinct global base maps are generated, each emphasizing different aspects of scientific geography.
• Overlay maps are created by superimposing normalized or raw publication data onto the base maps.
• The method supports both individual-level and institutional-level science mapping using standardized, open-access data.
• Visualization is validated through case studies of the first author and his research institution.

Experimental results

Research questions

• RQ1 How can OpenAlex data be used to generate meaningful global overlay maps of scientific output?
• RQ2 What differences emerge between raw and normalized publication data in visualizing research performance?
• RQ3 How do the six proposed base maps enhance the interpretability of science maps?
• RQ4 What are the practical advantages and limitations of this approach for individual and institutional science mapping?
• RQ5 Can this method be scaled to provide consistent, reproducible global science visualizations?

Key findings

• Normalization of publication data significantly improves the clarity of field-specific research focus compared to raw data.
• Overlay maps using normalized data reveal more nuanced patterns of scientific output across regions and disciplines.
• The six base maps provide consistent, standardized visual frameworks for global science mapping.
• Raw data tend to overemphasize high-output fields, while normalized data better reflect disciplinary diversity.
• The method enables reproducible, open, and scalable science visualization for individual researchers and institutions.
• The approach is validated through case studies showing clear, interpretable visualizations of research output at both personal and institutional levels.