Bibliometric Analysis

  • desk-based
  • week-month timescale
  • low resource requirement.

Bibliometric analysis observes publication data of research projects to understand the dynamics of scientific inquiry in a particular field. A bibliometric approach is an explicitly quantitative approach and uses, to a lesser or greater degree depending on the study, statistical/mathematical methods. At its core bibliometric research is research on research.

  • Should I use Bibliometrics?

    Analyses of scientific literature are useful barometers for the trends and themes within a field. A bibliometric analysis can indicate whether a trend is recent or historical, whether it came from a particular community, or how topics are understood within the scientific literature. Bibliometrics can also serve as a quantitative measure of scientific output for a particular organisation, individual, team, or group.

    One limitation of bibliometric analysis, however, is its time intensity. While there are ways to do a ‘quick’ bibliometric analysis, and some trained individuals can perform a rapid analysis in hours or even minutes, a full bibliometric study takes time and effort. It is therefore important to understand whether the aims of the research require a bibliometric approach or a systematic literature review.

    For research where the main aim is to understand themes of a particular research area and to map gaps in a field, a systematic literature review is sufficient. However, where bibliometrics becomes important is when the very object of study is the science itself. In these cases, where scientific output is a dependent variable, bibliometric analysis is more appropriate.

  • How do I use Bibliometrics?

    There are a range of different means by which a bibliometric analysis can be undertaken to produce insights. The choice of each of these will depend upon the research question being asked. The basic steps, however, are similar across types: set a scope for search, collect data, clean data, analyse trends & patterns.

    Set scope for search by deciding on search terms and database. In the first instance, the database to be queried is an important choice to make because some databases are language-based (e.g. SciELO is a Spanish & Portuguese database), others are regional (e.g AJOL is African region), or field/discipline (e.g. PubMed is Biomedical/ Medicine/ Public Health). Databases also vary by coverage in terms of journals, countries, and years as well as how quickly a publication is indexed.

    On the search term itself, it is important to be aware of the different ways a database can be sectioned and therefore scoped for data collection. Particular date ranges will allow, for example, data collection before/after a particular event or to include only more recent data. Another way to section a database is by field; many databases will have a function to include only Informatics Journals, or Sociology Journals – this will help to avoid capturing publications which have alternative meanings in other fields (e.g. translation in terms of linguistics, translation in biochemical terms, or translation in terms of applications for basic science).

    On keywords, it is important to note that synonyms may exist for your search terms. In some databases, there may be lists of keywords already in place (e.g. PubMed has MeSH terms that allocate papers to particular subjects); these can be helpful in accessing an already curated sub-section of a database. One final point to note is the use of, in many databases, wildcards (often *) representing a set of letters at the end of a word (e.g. interdisciplin* meaning interdiscipline, interdisciplinary, interdisciplinarily, interdisciplinarity, etc.) and Boolean operators like AND and OR which combine search terms in different ways. 

    Collecting data once it had been found in a database is the next important step. One decision that needs to be made here is what data to collect? Many databases have options for basic citation info (author, title, keywords, year, journal, issue, volume, etc.), others have the option to also download abstracts and keywords for each publications, others can download author affiliations or funder information, and others still (often requiring an API and some coding) will allow you to download full-texts.

    Cleaning data is a key step because, while many databases are expertly maintained, there are many inaccuracies to data search or collection and many possibilities for error in the database. A brief look at some random papers in the dataset will give you a good idea of some issues. Common problems include authors with similar (or the same) names, funders with multiple programmes/name variations, affiliations with multiple centres/name variations, and even papers that are outside of scope. Problems like this can often be rectified manually or, in the case of very large datasets, with programmes like Google’s OpenRefine.

    A second part of data cleaning may include extracting more easily analysable data from thicker data like full-texts or abstracts. This can often be done manually, but will include a good deal of time-investment to ensure that all publications are classified/extracted from.

    Data analysis can take many forms either through a quantitative analysis of key terms, analysis of trends, or network analysis. Considering quantitative analysis, this can be done through multiple programmes like excel, SPSS, R, etc. What a quantitative analysis can show is the frequency of publications in particular fields, or trends over time in publications. Network analysis can be done through programmes like VosViewer, or through quantitative methods using excel or R, or by producing a relational matrix that can be viewed in network software like gephi or pajek.

  • Examples of Bibliometrics in Sustainability Research

    ERTZ, M. and LEBLANC-PROULX, S., 2018. Sustainability in the collaborative economy: A bibliometric analysis reveals emerging interest. Journal of Cleaner Production, 196, pp. 1073-1085. 

    GARRIGOS-SIMON, F.J., BOTELLA-CARRUBI, M.D. and GONZALEZ-CRUZ, T.F., 2018. Social capital, human capital, and sustainability: A Bibliometric and visualization analysis. Sustainability (Switzerland), 10(12), 

    JIA, Q., WEI, L. and LI, X., 2019. Visualizing sustainability research in business and management (1990-2019) and emerging topics: A large-scale bibliometric analysis. Sustainability (Switzerland), 11(20),.

    LEETCH, A. and HAUK, M., 2017. A Decade of Earth in the Mix: A Bibliometric Analysis of Emergent Scholarly Research on Sustainability Education and Ecopsychology in Higher Education.

    RUIZ-REAL, J.L., URIBE-TORIL, J., VALENCIANO, J.D.P. and GÁZQUEZ-ABAD, J.C., 2018. Worldwide research on circular economy and environment: A bibliometric analysis. International Journal of Environmental Research and Public Health, 15(12),.

  • References and Resources

    ANDRÉS, A (2009). Measuring Academic Research: How to undertake a bibliometric study. Cambridge: Woodhead Publishing. 

    ROUSSEAU, R; EGGHE, L; GUNS, R (2018). Becoming Metric-Wise: A Bibliometric Guide for Researchers. Cambridge, MA: Chandos Publishing. 

    DAIM, T.U., RUEDA, G., MARTIN, H. and GERDSRI, P., 2006. Forecasting emerging technologies: Use of bibliometrics and patent analysis. Technological Forecasting and Social Change, 73(8), pp. 981-1012.

    HICKS, D., WOUTERS, P., WALTMAN, L., DE RIJCKE, S. and RAFOLS, I., 2015. Bibliometrics: The Leiden Manifesto for research metrics. Nature, 520(7548), pp. 429-431.

    VAN ECK, N.J. and WALTMAN, L., 2010. Software survey: VOSviewer, a computer program for bibliometric mapping. Scientometrics, 84(2), pp. 523-538.

Suggested citation: Moon, J.R. (2019). Bibliometric Analysis [online] Sussex Sustainability Research Programme Research Methods for Sustainability Catalogue. Available at: http://www.sussex.ac.uk/ssrp/resources/research-methods/bibliometrics.