Scorecard shows how research is helping change the world

For the first time, the annual G20 Scorecard: Research performance 2023, which examines research and innovation trends across all G20 member nations, includes data on how each nation’s research base is contributing towards the United Nations Sustainable Development Goals.

Published ahead of the G20 summit in New Delhi, India, on 9-10 September, the report’s new dynamic, online format allows for easy comparisons, giving professionals across academia, government and industry access to wide-ranging data to facilitate research, innovation, policy-making, education and international collaboration across various sectors.

Scorecard author and lead data analyst at the Institute for Scientific Information at Clarivate, Gordon Rogers, talks to University World News (UWN) special correspondent Sharon Dell about what the latest scorecard reveals about the changing patterns of worldwide scientific advancement and its value for G20 countries – and the world.

UWN: The Annual G20 scorecard for 2023 offers a wealth of information and reflects a lot of thought on the part of the report designers. What do you consider to be its key value for universities and-or higher education sectors in individual countries?

GR: The scorecard is based on Clarivate™ trusted and publisher-neutral data in the Web of Science and can therefore be relied upon to provide an accurate source of information about the state of the global research ecosystem. It can help to identify the strengths and weaknesses of each G20 member’s higher education sector, both in terms of the headline data in Figure 1: Context but also, the trends in Figures 2: Impact, 3: Output, and 4: Disciplines.

For example, as policy changes are implemented – such as Plan S in 2018, aiming to increase open access (OA) publishing throughout Europe – the effect of such changes should be visible in the data. Indeed, we see a clear increase in open access output for both Germany and Italy in the following years.

And we can also see above-average rates of open access publishing in mathematics and physical sciences, life sciences, engineering and social sciences for both countries compared with the G20 average. This suggests work needs to be done to improve OA rates in the medical sciences.

Both on a university level and in the higher education sectors in individual countries, this kind of valuable information can be used to inform strategic decisions, foster international collaboration, secure research funding, enhance curriculum relevance, advocate for favourable policies, and position institutions as contributors to global challenges.

UWN: What are the major differences in the way the 2022 data and the 2023 data has been presented? What were these changes responding to?

GR: The most visible difference is the fact that this year’s scorecard is now online and dynamic while last year’s was constrained by the limitations of a static PDF. Given the nature of the data presented in the scorecard, we in the Institute for Scientific Information (ISI)™ had felt over the past few years that a dashboard format would make the data far more engaging and useful. We were pleased that we could take the opportunity to make the change for this year’s report.

This means we have more space to present the data and can therefore present more data more effectively. For example, the contextual information that was presented as a simple row of numbers at the top of each two-page country scorecard in previous reports, has now been replaced by a map and bar chart and expanded to include far more data. And we’ve also added additional data in the disciplinary breakdowns.

UWN: Is this the first year in which the United Nations Sustainable Development Goal (SDG) indicators were added? What was the thinking behind the addition of SDGs as a ‘figure’?

GR: Yes, the data on SDGs was added this year. At Clarivate, sustainability is at the heart of everything we do, and we help our customers solve some of the world’s toughest challenges and achieve their sustainability goals through transformative intelligence and break-through innovations.

In 2015, the UN adopted their 2030 Agenda for Sustainable Development with a commitment to the 17 SDGs. As the G20 represents nearly two-thirds of the global population and spends more than 80% of its GDP, these countries and regions have a disproportionate influence on how likely we are to meet these goals. And much of that influence comes from academic research.

Using InCites Benchmarking & Analytics™ we can identify research articles associated with the SDGs (read our blog for more details).

With the new space available in this year’s report due to the dynamic format, it seemed important to include a figure showing where and how effectively each member nation’s research base is contributing to tackling these issues. This also ties in with the theme of India’s presidency of the G20 this year, Vasudhaiva Kutumbakam or “One Earth, One Family, One Future”, with a focus on sustainability.

We can therefore see, for example, that while Indonesia’s total output is small, ending poverty is of major importance with more than four times the G20 average proportion of papers covering this topic.

On the other hand, India’s focus is on clean water, clean energy and responsible consumption and production which makes perfect sense given its massive population.

UWN: Why is international collaboration such a significant indicator in a report of this kind and what does it reflect about the condition of the individual research ecosystems?

GR: In simple terms, papers that involve more collaboration tend to be more highly cited. Countries/regions with the highest levels of international collaboration – Saudi Arabia, and the United Kingdom, for instance – tend to have the highest levels of Category Normalised Citation Impact or CNCI (an unbiased indicator of impact irrespective of age, subject focus, or document type, allowing comparisons between entities of different sizes and different subject mixes).

Conversely, those collaborating the least – Russia, Türkiye, Japan and India – have the lowest CNCI. The message here is simple: the more you collaborate, the more likely it is that your research will be noticed by the rest of the world.

There are exceptions to this trend, however. Mainland China has the lowest level of international collaboration of all the G20 member nations, yet its average CNCI is above the world average. But its research base is now large enough that it can overcome this collaborative deficit.

UWN: What are the biggest changes or shifts in the data coming through this year? What have been the biggest ‘surprises’ in the data?

GR: That’s an interesting question as something that at first glance seems surprising may have a more mundane explanation behind it.

In some respects, the fact that India has now overtaken Mainland China as the most populous country or region globally and yet its GDP and research output lag so far behind may be surprising. But India’s population has grown more rapidly in recent years, while Mainland China’s has levelled off.

This means India has a far younger population that will take time to realise its potential. Hopefully our annual G20 scorecards over the next 10 or 15 years will show the country’s progress in doing so.

Indonesia and South Africa stand out as star performers when it comes to citation impact. Indonesia’s CNCI is higher than any other G20 member nation in Asia, and South Africa’s impact is even higher than Indonesia’s.

Again, these statistics seem surprising, but the scorecard helps you to dig further into the data and see that these high CNCI values are driven by international collaboration.

Looking at their Collaborative CNCI values, which compare papers according to the nature of their collaboration, their performances are less impressive. This is why looking at citation impact alone can give an incomplete impression and it’s important to look at a more complete picture to really understand the inherent quality in research.

UWN: Generally speaking, is it possible to say what the scorecard says about the general state of global research performance?

GR: The scorecard presents research data for the largest and most impactful nations, and therefore it gives an important overview of what works and what doesn’t. Of course, there’s a recurring theme here that international collaboration leads to higher citation impact.

But that picture is fairly simplistic: collaboration by researchers from the United States has increased over the past decade, and yet its citation impact has fallen. And Mainland China’s collaboration has remained consistently low while its impact has increased. This suggests an evolving picture as Mainland China expands its research base and an ever-larger share of research output moves away from North America and Europe to the rest of the world.

The disciplinary picture, however, is more complex. Mainland China’s output is focused more in the mathematical and physical sciences and engineering, while the United States publishes more papers in medicine. The impact of this medical research has remained more static over the past decade, while impact in other fields has fallen. It’s therefore clear that this redistribution of research isn’t happening in a consistent way.

UWN: The data on female researchers is intriguing. In your view, why is this an important indicator? The summary highlights Japan's percentage of female researchers as being below one-fifth, yet the actual number of female researchers seems quite high – it is third after the UK and Germany. Why are these kinds of nuances in the data important when considering the health of research systems?

GR: Diversity is an important feature of any successful group, with a variety of ideas helping to avoid the dangers of groupthink. While such diversity can come from a spectrum of different cultural backgrounds, it can also, of course, be provided by ensuring a strong mixture of male and female contributions.

This lack of diversity in Japan’s research community is one of several symptoms of a wider problem this country has in terms of the health of its research system.

While Japan’s gross national expenditure on R&D (GERD) to GDP ratio is high at 3.3% – many countries or regions in the G20 have figures below 3% – this isn’t translating into productive research. They rank among the lowest of the nations in the scorecard for papers per researcher or papers per GERD, with below-average levels of international collaboration and consequently below-average impact, whether measured by CNCI or Collaborative CNCI.

While it’s not possible to conclusively prove cause and effect here, South Korea has similar issues. Its percentage of female researchers is only just above one-fifth, and it also has low levels of productivity and a low level of international collaboration. However, the impact of its output has been improving over the last few years.

The data we’ve provided can therefore help to identify issues in each country or region’s research ecosystem. While individual governments should have a more complete picture of their own economic health – the economic and population data we’ve used ultimately come from them – the research data we’ve provided will help to broaden that perspective and help them gain an overview of how their financial strengths and weaknesses are impacting on their research.

UWN: Why is Saudi Arabia's CNCI so high?

GR: This is a great example of how the array of data in the scorecard can be combined to help discover and then explain an interesting story. As mentioned earlier, Saudi Arabia’s very high level of international collaboration has led to higher quality output that picks up more citations. This is partly due to the country making effective use of its wealth to attract some of the best researchers from elsewhere in the world to help improve its domestic research base.

However, the picture is slightly more complex than this. Figure 2: Impact in the scorecard shows that Saudi Arabia’s CNCI has trended higher over the past decade – it’s only in the last few years that its CNCI has been particularly high. However, Figure 3: Output shows that its output has also increased significantly over the decade, with output in 2022 three times higher than it was in 2016.

Consequently, its high CNCI is partly due to these recent papers that have so far performed much better than the world average. As more citations come in, it will remain to be seen whether these papers manage to maintain such a high level of impact.

UWN: Finally, how will the scorecard data inform the G20 summit in India next month?

GR: We released our annual G20 scorecard ahead of the G20 Summit in India to maximise the impact and usefulness of the insights and data it contains. The rich array of data we present in the scorecard has the potential to influence policy priorities, establish benchmarks and accountability, facilitate informed decision-making, identify best practices, guide discussions, promote global cooperation, and enhance public awareness.

This data will help G20 member nations focus on strengths and weaknesses, encourage competition and effective policy implementation, and foster collaboration while showcasing transparency and commitment to global challenges.

We do hope that those participating in the summit will take note of the scorecard data and that it will help them to make informed decisions that will ultimately transform our world for the betterment of society.

To access the scorecard, click here.