G20 research and innovation scorecard 2025

Understanding the diversity of the G20

The G20 members are not only large and influential but also remarkably diverse. This figure highlights this diversity, setting the stage for the rest of the scorecard.

Key features:

• Using trusted third-party sources, we provide insights about the overall population of G20 economies, their number of researchers (including female researchers), gross domestic product, gross domestic expenditures on R&D, business enterprise expenditures on R&D, and patents filed.
• Using Web of Science data, we enhance G20 member profiles with the overall insights about research productivity (number papers, both overall and relative), citation impact (CNCI, Collab-CNCI), internationalization or open access publication.

More granular data and analyses are available on other figures in this scorecard.

*For data from Web of Science, Mainland China includes data from Hong Kong SAR and Macau SAR. For data from the OECD, UNESCO, WIPO and the World Bank, Mainland China excludes these special administrative regions.

Output and citation impact of G20 research

This figure shows the trends in output and citation impact of research produced by each G20 member, offering a detailed look at how often research papers are referenced in subsequent studies.

Key indicators:

• Category Normalized Citation Impact (CNCI): This metric measures the significance of a research paper based on how often it is cited, normalized against the global average for its document type, field and year of publication. A CNCI value greater than 1.0 indicates that a paper is cited more frequently than the world average.
• Collaboration CNCI (Collab-CNCI): Acknowledging the growing trend of research collaboration, this metric adjusts for various collaboration types: domestic single institution, domestic multi-institution, international bilateral, trilateral and quadrilateral plus. It provides a more nuanced view of citation impact by factoring in the collaborative nature of research. Again, a value of 1.0 is the world average.
• % > World average: The percentage of papers cited more than the global average for their document type, field and year of publication. Equivalently, the percentage of papers with a CNCI greater than 1.0.
• % in top 10%: The percentage of papers in the top 10% for citations by document type, field and year of publication. By definition, a value of 10% is the benchmark for world average.

Visualization tools:

• Impact Profile: This profile shows the distribution of CNCI and Collab-CNCI values for a set of papers, offering a comprehensive perspective beyond a single average. Papers are categorized as uncited or assigned to various citation categories, allowing for direct visual comparison with the world average.

Explore Figure 2 to understand the trend in output and citation impact of G20 research.

* Mainland China includes data from Hong Kong SAR and Macau SAR.

Visualizing research impact

This figure provides a detailed look at the disciplinary breakdown of each member’s research. Research is categorized into the six major OECD fields, for ease of comparison with data from other sources, with each major field composed of various minor fields.

Key features:

• Scatter chart: The scatter chart on the left shows the percentage of papers associated with each OECD field and how these percentages compare with the average percentages across all G20 members. This makes it easy to see which fields each member is focusing on, and which fields they’re focusing on more than other members.
• Bar chart: The bar chart on the right shows CNCI and Collab-CNCI, together with G20 averages by field.
• Trend lines: For the major fields, trend lines show output by field (relative to 2015) and CNCI. This helps to clarify which fields are growing the fastest for each member and whether this research is maintaining its impact.
• Subject breakdown: Clicking on a major field in the key in the lower right shows a breakdown for the member’s output and impact for the minor fields associated with that field. You can use the arrows or keyboard arrow keys to switch between pages.

By comparing the profiles between the various charts, you can identify a member’s relative strengths and weaknesses in various research disciplines.

Explore the profiles in Figure 3 to gain insights into the disciplinary focus and impact of G20 members on the global research stage.

* Mainland China includes data from Hong Kong SAR and Macau SAR.

Research Footprints and the UN Sustainable Development Goals (SDGs)

Contribution to the SDGs

This figure presents Research Footprints that highlight each member’s contribution to research aligned to the UN Sustainable Development Goals (SDGs). Adopted in 2015, the 17 SDGs are part of the UN’s 2030 Agenda for Sustainable Development.

Key features:

• SDG output: The left chart shows how each member’s SDG-related research output compares to the G20 average.
• Citation impact: The right chart shows how each member’s SDG-related research performs in terms of citation impact. It compares two metrics – CNCI and Collab-CNCI – against G20 and global averages, helping to assess the reach and influence of each member’s research.
• Societal impact: Research has the greatest value when its impact reaches beyond academia. The left chart includes two indicators that reflect how research connects with society:
  • Non-academic collaboration: The number of publications co-authored by academics and non-academics, compared with the G20 average – highlighting direct engagement beyond academia.
  • Non-academic citations: The number of publications cited at least twice by non-academics, compared with the G20 average – indicating broader societal influence.
• By comparing the profiles between the two charts, you can identify where each member’s research priorities lie and assess how effectively their research contributes to addressing global challenges.

Explore Figure 4 to see how G20 members are advancing research towards the UN Sustainable Development Goals and contributing to solving the world’s most pressing issues.

* Mainland China includes data from Hong Kong SAR and Macau SAR.

The impact of open access across the G20

This figure looks at trends in open access (OA) research publication, which is becoming more widespread due to demands from research funders, including governments. In OA, the author or funder pays for publication, making the research freely accessible to all, unlike traditional subscription-based models.

• OA output: The two charts on the left show the trend in output, broken down into three broad classes of OA output. The data are presented both in terms of the percentage of all research output and the relative distribution of OA papers between the three broad classes (full definitions available here).
  • Journal OA includes: Gold, Gold Hybrid, Free to Read (Bronze) open access
  • Self-archived OA includes: Green open access
  • Both OA includes: Papers that are self-archived by the author and published open access in a journal.
• Impact Profile: The right figure provides an Impact Profile comparing OA and non-OA research against the member’s overall performance. This shows whether open access research has more influence or not.

Explore Figure 5 to see details on the rise of open access research.

* Mainland China includes data from Hong Kong SAR and Macau SAR.

Trends in international research collaboration

Growth in collaboration

Most of the world’s most highly cited research now involves authors from multiple countries and regions. This figure examines these collaborations, illustrating the partnerships between G20 members and their most frequent collaborators.

Key features:

• Collaboration trend: Figure 6a shows the overall trend in each member’s collaboration profile broken down into five collaboration classes: whether researchers are from a single institution (domestic single); from multiple domestic institutions (domestic multi); or collaborate through bilateral, trilateral or quadrilateral or larger international relationships.
• Impact distribution: On the right is an impact profile showing Collab-CNCI broken down into these collaboration classes. Papers are categorized as either uncited or assigned to various citation categories, allowing for a direct visual comparison with the world average.
• Collab-CNCI: This metric normalizes citation impact against research field, document type, publication year and collaboration. Values greater than 1.0 indicate that a paper is cited more often than the world average. It’s important to remember that each collaboration class is compared against other papers from the same class globally. This means that while domestic single institution papers typically receive fewer citations than internationally collaborative papers, the chart shows how a country/region’s domestic output compares against domestic output worldwide. For the African Union and European Union, the collaboration classes are based on the perspective of individual countries/regions in the union. For example, a bilateral collaboration between Egypt and Nigeria would contribute to the bilateral international class for the African Union.
• Collaboration partners: Figure 6b can be seen via the page navigator at the bottom or via keyboard arrow keys. It shows the ten most frequent collaborative partners and the trend in collaboration rate and Collab-CNCI for the top five partners. The charts for the countries/regions of the African Union and European Union show their external partners, providing a clear view of their international collaborative landscape.
• Collaboration by discipline: Figure 6c shows the proportion of output and Collab-CNCI for each of these five leading partnerships broken down by OECD field.

Explore Figure 6 to understand the dynamic nature of international research collaboration and see key partnerships across G20 nations.

* Mainland China includes data from Hong Kong SAR and Macau SAR.