Taras Hrendash, Štěpán Jurajda, Stanislav Kozubek, Daniel Münich
IDEA, Think-tank of the Economics Institute of the Czech Academy of Sciences
September 2018
The color of country data points distinguishes three types of OECD countries. By clicking on the legend, one can eliminate specific types from the plot.
Data point size corresponds to the share of a particular discipline within a country on total output.
TIP: It is possible to zoom in on any sub-area of each graph (by highlighting the chosen area) and to create PDFs for print.
Graph #1 shows the relationship between the economic performance of countries and their publication performance in the field Economics for the upper half of journals by their influence.
The graph reveals that the UK, the Netherlands, and Denmark produce unusually high volumes of articles per millions of population in highly influential journals relative to their R&D public expenditures per capita.
In contrast, all of the new EU members fall below the fitted regression line, suggesting that they perform below the level corresponding to their economic might.
Furthermore, some of the new EU members (the Czech Republic and Slovenia in particular) appear to produce unusual volume of research output in Economics in low-influence journals.
© Taras Hrendash, Štěpán Jurajda, Stanislav Kozubek, Daniel Münich, 2018
This website was created by Paulína Očkajová and Vít Macháček.
Besides many studies measuring the scientific output of nations, there is also a growing literature addressing the issue of how best to measure cross-country differences in research productivity, that is, of research output relative to inputs. Early studies in this literature simply contrast the citation intensity of nations or their publications per researcher with country-level GDP per capita and with the share of GDP spent on publicly funded R&D (e.g., King, 2004). Recent work uses individual-level information and strives to measure inputs by field (e.g., Abramo and D’Angelo, 2014a, 2014b; Bentley, 2015).
Our aim is to provide a simple contribution to this literature by exploiting a journal citation indicator, the Article Influence Score, to approximate the quality of published articles in scientific disciplines, which are well covered in the Web of Science. Unlike most existing descriptive studies, we therefore provide a comparison for a wide set of countries of their scientific publication performance by its level of expected impact and by discipline, expressed relative to an aggregate country-wide input measure of public R&D expenditures or GDP per capita. Ideally, future work would use discipline-specific input measures.
We contrast countries’ publication performance in high-influence outlets as well as in journals placed in the lower half of their discipline-specific rankings based on their scientific influence. This allows us to ask separately about publication output that is likely to have real scientific influence and which countries are relatively more productive in (concentrate on) publishing in lower-influence journals.
The publication performance of a country in a scientific discipline is measured by the number of articles published during the period 2012-2016 in journals indexed by the WoS assigned to particular disciplines.
An article is credited to a country if at least one of its authors is affiliated with an institution that has an address in that country.
Publication performance is measured per million of population, so that differences in publication output are not due to a country’s size.
The source of data is the WoS.
An article's influence level is inferred from the journal’s Article Influence Score (AIS). Journals are ranked according to their AIS within their WoS Category. When a journal is assigned to multiple WoS Categories, the weighted average percentile rank across those WoS Categories is used.
In the left menu, one can choose five different publication performance indicators according to AIS-based scientific influence:
Economic performance of countries is measured by Gross Domestic Product (GDP) per capita expressed in purchasing power-adjusted USD levels.
The source of the data is the OECD.
R&D expenditures of countries per capita include government budget allocations for R&D in USD as reported by the OECD, within the scope defined in the 2015 Frascati Manual. They are expressed in purchasing power-adjusted USD levels. The source of the data is the OECD.
Government budget allocations for R&D (GBARD) include the R&D expenditure operations conducted by the government at the central (federal), regional (state) and local (municipal) levels, and financed through the budget under standard budgetary approval procedures [§12.5, OECD Frascati Manual 2015].
GBARD encompass all spending allocations met from sources of government revenue foreseen within budgets, such as taxation. Spending allocations by extra-budgetary government entities are within the scope of GBARD measures only to the extent that their funds are allocated through the governmental budgetary process [§12.9, OECD Frascati Manual 2015].
The menu differentiates between 226 academic disciplines based on WoS Categories used to classify journals. If a journal belongs to several WoS Categories, an article published in that journal is credited to each of those WoS Categories.
The application covers OECD countries except Luxembourg, which is excluded from the comparison due to peculiarities that make the country an outlier in the graphs.
OECD countries are classified into three groups distinguished by the color of the plots:
It is fundamentally difficult to identify the causal links between economic performance, public R&D funding, and discipline-specific scientific publication performance due to the complexity and opaqueness of R&D funding across countries, the absence of consistently defined cross-country data on R&D funding by scientific disciplines, and the paucity of natural experiments affecting discipline-specific funding.
Even if we abstract from the likely causal impact of research quality on economic growth, the associations highlighted by our comparisons are likely to be attributable to a mix of phenomena:
The size of individual plots indicates how important the given field is in a particular country in terms of the field’s relative share of publications of all publications produced by a country. The benchmark is the average share of publications in the particular field across OECD countries.
Three degrees of importance are distinguished by the sizes of the plots: Low, Average, High.
Vertical and horizontal dashed lines represent the averages of the corresponding indices for the group of countries currently shown on the graph.
The authors of the interactive application are Taras Hrendash1, Štěpán Jurajda1, Stanislav Kozubek2 and Daniel Münich1.
1 IDEA at CERGE-EI: IDEA is an academic think tank within CERGE-EI, a joint workplace of the Center for Economic Research and Graduate Education, Charles University, and the Economics Institute of the Academy of Sciences of the Czech Republic
2 Institute of the Biophysics of the Czech Academy of Sciences
A study by Jurajda et al. (2017) was the inspiration for this web tool.
Abramo, Giovanni, and Ciriaco Andrea D’Angelo (2014a), How do you define and measure research productivity? Scientometrics, 101: 1129-1144.
Abramo, Giovanni, and Ciriaco Andrea D’Angelo (2014b), Assessing national strengths and weaknesses in research fields, Journal of Infometrics, 8: 766-775.
Bentley, Peter James (2015), Cross-Country Differences in Publishing Productivity of Academics in Research Universities, Scientometrics, 102: 865-883.
Jurajda, Š., Kozubek, S., Münich, D. & Škoda, S. (2017). Scientific publication performance in post communist countries: still lagging far behind, Scientometrics, Volume 112, Issue 1, pp. 315–328.
King, David A. (2004). The Scientific Impact of Nations, Nature, 430.
We are sorry, but this interactive application is not suitable for mobile phones. Reading is useless at lower resolutions (less than 1024x768). We recommend viewing this website on a desktop computer, laptop, or tablet. Alternatively, try to maximize your browser and update the page (F5).
To read this page later, copy the link:
