What is bibliometrics, and what can it contribute to a better understanding of the research landscape?

Bibliometrics is a set of methods that use data from databases indexing records of scientific publications and patents, as well as other R&I outputs of growing interest such as designs and trademarks, to derive new insights into these outputs’ corresponding funding and performance. Within bibliometrics, the branch of scientometrics examines the records of research publications to measure scientific activity. It is also increasingly looking at related types of outputs such as research data sets and protocols. The branch of technometrics focuses on patent records as a proxy measure for innovation. It is being actively expanded to cover other forms of innovation that are not well captured by patents, such as those covered by designs and trademarks.

Bibliometrics can be used to inform the following study goals and more:

  • Measuring scientific and technological production and benchmarking R&I performance at multiple levels (researcher, program, organization, network, R&I infrastructure, state/region and country) and across several dimensions (output volume (raw and per capita), specialization, impact and more)
  • Assessing the international impact of R&I within and beyond the scientific community, such as through citation impact within the scientific, patent and policy-relevant literature
  • Identifying leading organizations and competitors in a particular area of research
  • Identifying who is collaborating with whom, and how, in a variety of S&T fields
  • Mapping collaboration networks and identifying mutually beneficial collaboration opportunities
  • Assessing the impact of R&I funding on the scientific and technological output of graduate students and researchers across different career stages
  • Assessing the impact of funding for R&I infrastructure
  • Monitoring research trends in specific S&T areas and sectors
  • Identifying emerging research topics
  • Supporting the design and monitoring the outcomes of national and international S&T strategies and policies
  • Supporting the design and evaluation of funding programmes
  • Supporting organizational strategic review
  • Providing other evidence in support of organizational decision-making and outcome assessment
  • Supporting peer-review processes

Science-Metrix can conduct purely bibliometric studies, but we also have a long history of integrating bibliometrics with other quantitative and qualitative methods for evaluations and mixed-method studies at the international level. We develop advanced bibliometric indicators to extract, synthesize, analyze and report on useful information to help our clients achieve their study goals.

We can provide clients with a detailed databook to explore and extract insight from, or we can provide in-depth analytical reports explaining the significance and limitations of the data and methods, as well as the findings and their implications for policymaking or strategic decision-making.

Contact us to discuss how we can tailor a bibliometric study to suit your needs.


Scientometrics, to give a basic description, takes the metadata and information contained within scientific papers and extracts and combines those data to provide rich insight into various aspects of scientific activity, as measured by various indicators. Metadata may include author names, the publishing date, the references cited by the publication, the keywords as provided by the authors, and the authors’ institutional affiliations. Information may include text extracts that are then used to identify the key topics covered within the paper.

These data are then cleaned to remove errors, suspect entries and outliers, and indicators are used to calculate various aspects of scientific activity, such as output volume, growth in production, scientific impact, peer collaboration, areas of specialization, open access to scientific publications and the participation of women in science.

Of course, there are many caveats and limitations to factor into such work, but at Science-Metrix we are very transparent about such limitations when communicating with our clients to make make sure they understand their implications. We are also continually developing and refining our scientometric indicators in response to client requests for customized studies, developments in our field and our own exacting professional standards. More recently, we have enriched the metadata of scientific publications with complementary data sources (e.g., PlumX, Overton, PATSTAT, NICE clinical guidelines) to track knowledge valorization or translation towards the longer-term socioeconomic impacts of research—what is broadly referred to as “altmetrics”. Scientometrics is also increasingly examining related types of outputs such as research data sets and protocols.


Technometrics is often seen as a proxy for measuring innovation activities because patents are a formal protection for new technological progress, which is more and more often built upon formal research. Previous analyses have established statistically significant relationships between patent counts and innovation. As a result, patents can be used to characterize companies’ and countries’ innovation performance in terms of new technologies, new processes or new products.

Patent counts are an established indicator for measuring innovation activities, and recent work on technometrics has focused not so much on new indicator development as on reinforcing the data treatment processes and analyses necessary to arrive at these counts. Patent counts are increasingly complemented by trademark and design application counts—forms of registration that capture other dimensions of the innovation process—although these methodologies are currently only used at the country and regional levels.