When a country wants to know how well its scientific research and technological capacity are paying off, it turns to Science and Technology (S&T) indicators. These are the statistics that tell policymakers how much is being spent on research, how many researchers are at work, how many patents and publications are produced, and how all of this connects to economic growth. The problem is that most of the standard indicators were designed for wealthy, industrialised economies. When developing nations borrow these frameworks wholesale, the numbers often miss the realities on the ground. This post explores why S&T indicators must be tailored for developing countries, what makes the task so difficult, and how these measurements shape socio-economic development and national competitiveness.

Table of Contents

What S&T indicators actually measure

S&T indicators are quantitative measures used to assess the inputs, outputs, and impact of scientific research and technological activity within a country. Inputs include things like gross expenditure on research and development (GERD), the number of full-time researchers, and the share of GDP devoted to R&D. Outputs include scientific publications, patents, and high-technology exports. Together they give decision-makers an evidence base for planning budgets, designing policy, and comparing performance internationally.

The global standard for measuring R&D is the Frascati Manual, first issued by the OECD in 1963 and revised several times since. UNESCO adopted its basic concepts for worldwide surveys, which is why most countries today report R&D data in broadly comparable terms. India, for instance, follows UNESCO and OECD guidelines so its statistics can be compared across borders. But the Frascati framework was built for, in its own words, developed countries with market economies, and that origin creates real friction when it is applied elsewhere.

Why standard indicators fall short in developing nations

Developing countries face socio-economic and structural conditions that differ sharply from those of advanced economies. Their scientific communities are smaller, funding is tighter, infrastructure is uneven, and the focus of research is often different. Applying indicators designed for a different context can produce misleading conclusions and, worse, push policy in the wrong direction.

The informal economy goes uncounted

A large share of economic and even innovative activity in developing countries happens in the informal sector. Much grassroots problem-solving, adaptation, and incremental innovation never appears in a formal R&D survey. As researchers reviewing R&D measurement have noted, informal behaviours and contributions are central to how R&D systems actually work in these settings, yet they remain largely hidden from official statistics. A purely Frascati-style count therefore underestimates the real innovative effort of the country.

Concentration in a few big projects

National science in a developing country can be dominated by one or two very large projects, often in defence, space, or atomic energy. This concentration affects both the financing and the personnel figures, distorting indicators that assume research is spread broadly across many institutions. India illustrates this pattern: among the central government’s major scientific agencies, the Defence Research and Development Organisation alone accounts for the largest single share of R&D expenditure, with space and agricultural research following close behind, according to the national R&D statistics compiled by the Department of Science and Technology.

Weak statistical capacity

Producing reliable indicators requires trained statisticians, regular surveys, and the institutional muscle to collect data from thousands of enterprises and laboratories. Many developing countries simply lack this capacity. The same constraint shows up in related fields: when the international community agreed on a core set of ICT indicators, national statistical systems found that even producing the underlying figures carried serious organisational, technical, and resource implications. R&D surveys face the same hurdles, which means gaps, delays, and patchy coverage are common.

Goals that GERD does not capture

Indicators built around economic output assume that the purpose of science is to feed a competitive market economy. But development economists recognised decades ago that growth of income alone is an inadequate measure of development, and that reducing poverty and meeting basic human needs should also show up in how we measure progress. For a developing country, a low-cost technology that delivers clean water or improves crop yields may matter more than a headline patent count, yet standard indicators rarely reward such outcomes.

Approaches to tailoring indicators

Recognising these gaps, international bodies and individual nations have developed ways to adapt rather than abandon the standard framework.

The Frascati Manual annex for developing countries

The most significant institutional response came from the UNESCO Institute for Statistics, which proposed methodological guidelines on applying Frascati concepts in developing-country settings. After worldwide consultations, the OECD’s group of national experts approved an annex to the Frascati Manual specifically addressing R&D measurement in developing countries. The strategy here is deliberate: keep the Frascati standard so that international comparability is preserved, while adding guidance that helps capture the realities these economies face. It is adaptation, not replacement.

Complementary surveys, descriptors, and narratives

Numbers alone cannot describe a research system that runs partly on informal effort and is shaped by unique institutional histories. The argument from R&D measurement specialists is that indicators should be supplemented with complementary qualitative surveys that produce descriptors and narratives. These richer accounts explain the dynamics behind the figures, such as why a particular sector is strong, where the barriers lie, and how informal activity feeds into formal results.

Customising innovation indicators

Innovation measurement faces a parallel challenge. Standard innovation surveys, modelled on those used in OECD economies, assume firms with formal R&D departments. Researchers at the UNU-MERIT institute have made the case for the need to customise innovation indicators in developing countries, where much innovation is informal, imitative, or based on absorbing and adapting existing technology rather than inventing new products. Capturing this “doing, using, interacting” mode of innovation requires indicators built for the local context, not imported unchanged.

Building national statistical machinery

Tailoring also means investing in the institutions that produce the data. India offers a useful example. Its National Science and Technology Management Information System (NSTMIS), housed within the Department of Science and Technology, has been generating S&T databases since 1973, conducting periodic national surveys, and adopting UNESCO and OECD definitions for comparability while expanding coverage to include multinational companies and enterprises previously left out. This kind of sustained institutional effort is exactly what produces indicators that are both locally meaningful and internationally credible.

How indicators shape socio-economic development

Tailored indicators are not just an academic exercise. They directly influence how a nation allocates scarce resources and how it positions itself in the global economy.

Evidence for policy and planning

When indicators are accurate, they become the foundation for evidence-based policymaking. India’s gross expenditure on R&D has grown substantially over the past two decades, and tracking this alongside output measures like publications and patents lets planners judge where investment is working. India’s national R&D statistics reveal both strengths, such as a fast-rising share of global scientific publications, and persistent concerns, such as GERD remaining well below one percent of GDP. Indicators that surface these patterns help governments decide whether to push more public funding, incentivise private R&D, or strengthen higher education.

Measuring competitiveness on the global stage

Composite indices translate S&T indicators into a competitiveness story that policymakers and investors watch closely. The Global Innovation Index, published by the World Intellectual Property Organization, ranks economies using around 80 indicators grouped into innovation inputs and outputs. India climbed to 39th among 133 economies in 2024, ranking first among lower-middle-income economies and first in the Central and Southern Asia region. Notably, India performs better on innovation outputs than on inputs, a signal that it is converting relatively modest investment into strong results. The same exercise also flags weaknesses, which is precisely the diagnostic value of well-chosen indicators.

Identifying gaps within the country

Indicators can also expose internal disparities that national averages hide. A benchmarking study comparing Indian and U.S. regions found that activity is highly concentrated, with most foreign investment and high-technology exports clustered in just a handful of states. This kind of subnational measurement helps governments target support to lagging regions rather than assuming progress is evenly shared. For a large developing country, that granularity is essential for inclusive development.

Linking science to human welfare

Finally, tailored indicators keep the focus on what science is supposed to deliver for ordinary people. By incorporating social and human dimensions alongside economic ones, a developing nation can measure whether its research is improving health, education, agriculture, and access to technology, not just boosting export figures. This broader yardstick aligns scientific effort with national development priorities and ensures that investment in research translates into real improvements in living standards.

Striking the balance

The central tension in this whole exercise is between comparability and relevance. Stick too rigidly to international standards and the indicators miss what is distinctive about a developing economy. Abandon them entirely and the country loses its ability to benchmark itself against the world. The emerging consensus, reflected in the Frascati annex and in the push to customise innovation measurement, is to do both: maintain the core standard for comparison while layering on context-specific surveys, descriptors, and indicators that capture informal activity, basic-needs outcomes, and local innovation patterns. Done well, this gives a developing nation a measurement system that is honest about where it stands and useful for deciding where to go next.

What do you think? If you were advising a developing country, would you prioritise indicators that allow clean international comparison, or ones that capture the messy, informal innovation that standard surveys miss? And how would you measure science’s contribution to human welfare in a way that policymakers would actually act on?

How useful was this post?

Click on a star to rate it!

Average rating 0 / 5. Vote count: 0

No votes so far! Be the first to rate this post.

We are sorry that this post was not useful for you!

Let us improve this post!

Tell us how we can improve this post?

References
  1. https://www.oecd.org/content/dam/oecd/en/publications/reports/2015/10/frascati-manual-2015_g1g57dcb/9789264239012-en.pdf
  2. https://journals.sagepub.com/doi/abs/10.1177/097172180901500104
  3. https://dst.gov.in/sites/default/files/Updated%20RD%20Statistics%20at%20a%20Glance%202022-23.pdf
  4. https://www.researchgate.net/publication/317539063_PRODUCING_ICT_INDICATORS_IN_DEVELOPING_COUNTRIES_CHALLENGES_AND_INITIATIVES
  5. https://www.researchgate.net/publication/4839881_Indicators_of_Development_The_Search_for_a_Basic_Needs_Yardstick
  6. https://www.ricyt.org/en/2011/07/the-oecd-nesti-group-approved-a-frascati-manual-annex-about-rad-measurement-in-developing-countries/
  7. https://ideas.repec.org/h/elg/eechap/14427_19.html
  8. https://dst.gov.in/sites/default/files/R&D%20Statistics%20at%20a%20Glance%202019-20.pdf
  9. https://www.wipo.int/edocs/gii-ranking/2024/in.pdf
  10. https://itif.org/publications/2024/11/15/us-india-subnational-innovation-competitiveness-index/

Comments

Leave a Reply

Your email address will not be published. Required fields are marked *

Informetrics & Scientometrics

1 Information and Measurement

  1. Information Revisited
  2. Framework for Information Exchange
  3. Measurement Techniques
  4. Informativeness
  5. Standardization of Measurement

2 Measure of Information

  1. Information and Entropy
  2. Shannon Information
  3. Probabilistic Information
  4. Properties of Shannon Information
  5. Derivation of Shannon Information Formula
  6. Normalization Condition
  7. Relating Semantic Value to Shannon Type Measures
  8. Other Shannon Type Measures of Information
  9. Semantic Information
  10. Fuzzy Information Measure
  11. Other Information Measures

3 Informetrics – Definition, Scope and Evolution

  1. Definitions
  2. Scope
  3. Evolution
  4. Summary

4 Sociology of Science and Scientometrics

  1. Sociology of Science
  2. Growth of Scientific Knowledge
  3. Social Organization in Research Areas
  4. Approaches of Scientometrics to Sociology of Science
  5. Models of Growth of Knowledge

5 Organizations Engaged in Scientometrics and Informetrics Studies

  1. Organizations Engaged in or Supporting Scientometrics/Informetrics Studies
  2. Websites
  3. Research Groups/Discussion Groups
  4. Periodical Publications
  5. Conferences/Seminars/Workshops/Congresses
  6. Individuals Engaged in the Study and Research in Scientometrics/Informetrics

6 Law of Scattering and its Applications

  1. Introduction
  2. Historical Account
  3. Bradford’s Law
  4. Verbal Form of Bradford’s Law
  5. Applications of Bradford’s Law
  6. Graphical Representation of Bradford’s Law
  7. Conditions for Bradford’s Law
  8. Falling Tail of Bradford Curve: The Groos Droop
  9. Ambiguity in Bradford’s Law
  10. Fitting Bibliographic Data to Bradford’s Law

7 Rank and Size Frequency Models

  1. Representations and Organization of Numerical Data
  2. Size – Frequency Approach
  3. Rank – Frequency Approach
  4. Size – Frequency Models
  5. Rank – Frequency Cumulative (Fractional) Models
  6. Rank – Frequency Cumulative (Non-Fractional) Models
  7. Rank – Frequency Non – Cumulative Models

8 Informetrics Phenomena

  1. Terminology and Historical Development
  2. Selected Laws of Bibliometrics and Informetrics
  3. Informetrics Phenomena in Science
  4. Practical Applications of Informetrics

9 Analysis of Library Related Data

  1. Necessity for Analytical Studies in Libraries
  2. Citation Counting: A Versatile Tool for Journal Selection
  3. An Alternative Method of Citation Analysis
  4. Selection of New Source Journals to Eliminate Bias Due to Country, and Language
  5. Weightage Formula to Correct Citation for Post-War Periodicals
  6. Three New Bibliometric Parameters to Re-Rank Scientific Periodicals
  7. Garfield’s Methods for Cito-Analytical Studies
  8. Librametric Analysis
  9. Bibliometric Analysis
  10. Informetrics
  11. Scientometrics: Its Genesis, Scope, Definition, and Applications

10 User Studies

  1. User Studies
  2. Questionnaire Method
  3. Interview Method
  4. Diary Method
  5. Observation Method
  6. Planning a Survey
  7. Classification and Tabulation of Data
  8. Analysis of Data
  9. Presentation of Results
  10. Important User Studies
  11. Application of User Studies

11 Laws of Scientific Productivity

  1. Scientific Productivity – Influencing Factors
  2. Scientific Productivity – Problems in Measurement
  3. Scientific Productivity – Distribution Characteristics
  4. Lotka’s Law
  5. Statistical Distributions or Models
  6. Application of Lotka’s Law
  7. Goodness-of-Fit Test

12 Growth and Obsolescence of Literature

  1. Growth of Literature
  2. Obsolescence of Literature
  3. Growth Vs Obsolescence of Literature

13 Science Indicators

  1. Indicators
  2. Towards Science Indicators
  3. Historical Aspects
  4. Functions of Science Indicators
  5. S&T Indicators for the Developing Countries
  6. Types of Indicators
  7. Validity and Reliability of Indicators
  8. Building S&T Indicators
  9. Literature Based Indicators
  10. Patent Indicators

14 Mapping of Science

  1. Cognitive Mapping
  2. Journal-to-journal Citation Maps
  3. Co-citation Maps
  4. Co-word Maps
  5. Co-classification Maps
  6. Descriptive Mapping

15 Elements of Statistics

  1. Data and Its Measurement
  2. Graphical Representation
  3. Measures of Central Tendency
  4. Measure of Variability
  5. Correlation and Regression

16 Probability Distributions and their Applications

  1. Probability – Definition
  2. Random Variables
  3. Joint Probability Distribution
  4. Conditional Probability Distribution
  5. Some Special Distributions
  6. Applications of Probability

17 Regression Analysis

  1. Simple Linear Regression
  2. Multiple Regression
  3. Stepwise Regression
  4. Regression with Qualitative Explanatory Variables

18 Cluster Analysis and Factor Analysis

  1. Introduction
  2. Cluster Analysis
  3. Factor Analysis
  4. Examples of Cluster and Factor Analysis