1 See sidebar What Makes a Good Indicator? for a brief and high-level summary of the data sources used in the Science and Engineering Indicators (Indicators) report and the data quality issues that influence the interpretation and accuracy of the information presented in Indicators.

Workers with S&E Skills

1 An additional complexity, as data from the United States show, is that a direct correlation often does not exist between an individual’s degree and occupation. S&E degree holders report applying their S&E expertise in a wide variety of jobs, including S&E and non-S&E jobs. This indicates that the application of S&E knowledge and skills is widespread across the technologically sophisticated U.S. economy and is not just limited to jobs classified as S&E. For more information on this and the U.S. S&E workforce, see National Science Board (2015).

R&D Expenditures and R&D Intensity

1 Business spending and government spending as reported here are defined by international guidance. As recommended in the Frascati Manual 2015 (OECD 2015), R&D funding from government-run businesses is to be reported as funding from the business sector. Actual sector classification may differ somewhat by the circumstances in specific countries.

2 These terms are defined in the chapter Glossary.

Research Publications

1 For more information on the developing and developed economy classification, see the International Monetary Fund classification of countries, available at, accessed 4 December 2017. According to the IMF, “This classification is not based on strict criteria, economic or otherwise, but instead has evolved over time with the objective of facilitating analysis by providing a reasonably meaningful organization of the data.”

2 The implications of these differences in top citations should be drawn with care because the data used for the analysis require that article abstracts be provided in the English language. Many publications from China have English-language abstracts but Chinese-language text, limiting their accessibility and likelihood of citation for researchers not fluent in Chinese.

Invention, Knowledge Transfer, and Innovation

1 Information and communications technologies consists of communication processes, computers, digital communications, information technology management, semiconductors, and telecommunications.

2 Testing, measuring, and control consists of analysis of biological materials, control, measurement, and optics.

3 Chemistry and health consists of pharmaceuticals, biotechnology, basic material chemistry, organic chemistry, macromolecular chemistry, chemical engineering, and medical technologies.

4 Materials and nanotechnology consists of materials and metallurgy, microstructural and nanotechnology, and surface technology and coating.

5 For a broader discussion of this trade and the role of intellectual property protection, see the White House (2015:Box 7-1).

Knowledge- and Technology-Intensive Economic Activity

1 Boeing sources about 70% of the parts from U.S.-based companies and 30% from companies outside the United States to produce its advanced 787 airliner and other similar models (CNN Money 2013).

2 China had 18 of its supercomputers listed in the world’s top 500 supercomputers in November 2016 (

3 Public KI services—health and education—are much less market driven than other KTI industries. Additionally, international comparison of these sectors is complicated by variations in the size and distribution of each country’s population, market structure, and degree of government involvement and regulation. As a result, differences in market-generated, value-added data may not accurately reflect differences in the relative value of these services. The Overview presents other indicators for education such as data on degrees awarded in Chapter 2.