Cookies

We use cookies to ensure that we give you the best experience on our website. You can change your cookie settings at any time. Otherwise, we'll assume you're OK to continue.

Durham University

University Library

Citations: How do I?

This page provides guidance and links to tutorials for how to get access to, use and correct some of the key metrics you be familiar with, are asked to provide (by journals, colleagues, funders) or come across elsewhere. If you need any further help, please contact the Academic Liaison Librarian for Research.


Article-level indicators

Why might you want to do this?

  • To keep up to date with newly published research citing a known publication, and how it has been considered within wider scholarly debate.
  • To be kept up to date with who is reading and citing your own publications.
  • Each citation might be from a potential collaboration opportunity or competitor.

Key sources

There are four key sources of publication citation data:

  • Scopus (provided by Elsevier)
  • Web of Science (Provided by Clarivate Analytics)
  • Google Scholar
  • Open Citations via Crossref

For more information, see ‘Datasources’ on our metrics overview page.


How do I do this?

See our page on Citation Searching, or check out our quick links for Scopus/Web of Science:


Things to consider

  • No single database indexes all scholarly output.
  • Most citation indexes are less comprehensive in their coverage of the arts & humanities than for STEM subjects.
  • Most citation indexes are less comprehensive in their coverage of non-journal publication formats.
  • Google Scholar's coverage beyond STEM subjects may be better, and will also offer better coverage of grey literature, but inclusion criteria are less stringent and less transparent than most academic databases: content includes some 'predatory' journals, duplicate citations and false citations. See our blog post on citation counts in Google Scholar.
  • A publication may cite another publication for a range of reasons. Citations are not equal.
  • Citations should be seen as an approximation of the impact, or reach, of a publication, rather than a measure for the 'quality' of a paper.

Why might you want to do this?

  • Citations in traditional scholarly publications can take time to appear in the published record.
  • Traditional 'scholarly impact' metrics provide information on who has cited a work in a peer-reviewed article, but do not consider who else is reading, sharing, discussing or commenting outside of traditional publication channels.
  • Knowing who is discussing and sharing your research can help identify potential academic and non-academic collaborations.
  • Indicators of use outside of academic circles can help identify pathways to (and potential sources of evidence) for the wider impact of your research.
  • It’s often nice to know someone is reading/talking about your research!

Key Sources

  • Altmetric.com: provides various tools for researchers, including:
    • Plugins for websites to highlight attention your publications have received.
    • A browser 'bookmarklet' which allows you to quickly see the altmetric data for publications you discover on the web.
  • PlumX Metrics: now integrated witin Scopus and ScienceDirect, provide authors and researchers information on how a publication has been shared, mentioned, used and cited.
  • Kudos: Free to use, providing a simple profile page to promote your output, and track where and how engagement is taking place.

How do I do this?

Altmetric.com and PlumX Metrics aallow an author or researcher to view or display altmetrics for an individual article.


Things to consider

  • Altmetrics can help a researcher understand how publications are being shared or communicated, and the level of attention outside of academic circles it might be receiving.
  • High levels of attention does not mean the article is important or of high quality.
  • Articles which are open access or free to read are easier to share and comment on than articles locked by subscription journal paywalls.
  • Papers talking about diet, religion, popular archaeological discoveries (think Pyramids, Richard III or human evolutionary discoveries), cancer and other common health ailments are regular features in the altmetric top 100. Papers talking about chemical structures, complex mathematical modelling and theoretical physics - less so.
  • Altemetrics are often useful indicators of pathways to impact, or sources of impact. They are not (usually) evidence of impact in and of themselves.

Why you may want to do this?

  • Quantitative data may give you additional supporting evidence of which (of several articles) has had a greater impact within a particular field or discipline.
  • You may need to include a sample of research publications on an application, cv or web page and want some quantitative data to support or challenge qualitative data (e.g. peer review, subjective opinion) as to which to include.
  • However, such comparisons come with significant health warnings – you can rarely simply compare the number of citations received between 2 or more publications.

Things to consider

  • Many factors impact the rate and volume of citations:
    • the venue (e.g. journal or conference proceeding) of publication
    • the type of research and format of publication
    • the discipline and field of study
    • recognised bias in how other authors cite research based on the gender and career stage of an author.
  • It is important, therefore, not to try and compare the 'value' of different publications based simply upon the number of citations they have received at a given point in time.
  • There may be times where you wish to try get some measure of how the citation impact of different articles compare, and it is important when doing so to use normalised or field-weighted indicators to help make such a comparison.
  • However, even field-weighted metrics do not take into account all factors which might affect the rate at which a publication accrues citations. Field-weighted metrics are often reliant on:
    • A subject classification created by the data source (Scopus, Web of Science) which may not accurately reflect the specific field of research of the individual article.
    • This is because it is usually this is based on the subject classification of the journal in which an output has been published, rather than a more granular classification based upon the subject matter of the article itself.

Key sources

  • Scopus (provided by Elsevier)
  • Web of Science (Provided by Clarivate Analytics)
  • SciVal (Citation Analysis tool provided by Elsevier, using data from Scopus)

How do I do this?

Scopus

Scopus displays both the number of citations an output has received within Scopus, and the Field-Weighted Citation Impact (FWCI) of the article.

For example, an article which had an FWCI of 1.35 indicates that the article has received 35% more citations than the average number of citations attracted by similar articles (by age, output type and field of study). It does not, however, give any information as to why that article has attracted more citations, which could be for a variety of reasons.


Web of Science

Web of Science allows you to filter results to those it has highlighted as "hot papers" or "highly cited papers", based upon the number of citations they have received in comparison to other articles published in journals under the same subject classification and of the same age.

Web of Science Hot Papers


SciVal

The University has access to SciVal, a Citation Analysis tool provided by Elsevier which uses citation and publication data from the Scopus database (see 'data sources' in our overview of metrics for a comparison of the different major sources available).


Why might you want to do this?

Viewing the most highly cited papers in a subject area can help give you an idea of key authors in the field, key articles in the field or journals in which you may wish to send your own publications.


Key sources


How do I do this?

Google Scholar

Whilst it is not possible to filter or order your search results based on the citation impact of results returned within Google Scholar's native interface, the citation count and how recently an article has been cited appear to form part of the searching algorithm, meaning these articles are most likely to appear towards the top of your list of search results if ranked by 'relevance' (although this is not the only factor taken into account).

Alternatively, you can use free software such as Publish or Perish to search for and download references from Google Scholar, and then rank these by either the Citation Count, or the number of citations received by year since publication (software and guidance can be found via our Publish or Perish pages).

Publish or Perish


Scopus

From your search results within Scopus, you can select to order those results to show the most highly cited at the top of the list.

Publish or Perish

From within an article abstract record in Scopus, you can also vie the SciVal Topic Prominence information (below the article abstract). This can highlight additionally citation information about core articles and publishing authors within that identified topic area.

Publish or Perish


Web of Science

Within Web of Science, you have two options available to you.

  • You can choose to order your search results, similar to in Scopus, by choosing to rank your results by "Times Cited -- highest to lowest ".
  • You can filter results to those it has highlighted as "hot papers" or "highly cited papers", based upon the number of citations they have received in comparison to other articles published in journals under the same subject classification and of the same age.

Web of Science Hot Papers


Things to Consider

  • No single database indexes all scholarly output, and so what might be shown as the "most cited" is only the "most cited" within that dataset.
  • A publication may cite another publication for a range of reasons. Citations are not equal.
  • Citations should be seen as an approximation of the impact, or reach, of a publication, rather than a measure for the 'quality' of a paper.


Author-level metrics

Why you may want to do this?

  • The h-index is a widely recognised metric, calculated from a measure of both the productivity and citation impact of, usually (but not exclusively), an individual author.
  • The h-index is intended to provide a simplified metric for the 'impact' of an author within their field.
  • Many academics, particularly in the Sciences and Social Sciences, will know their h-index and may include it in funding, recruitment or progression applications.
  • You may find you are asked for your h-index in various circumstances, from submitting an application for promotion or employment, to funding applications and even when submitting a manuscript for publication.
    • You should always feel confident to question why you are being asked to provide your h-index, as it may not be appropriate or relevant as an indicator of your research or productivity potential.
    • Authors should be aware that the h-index is widely criticised over limitations in what it does not measure (or misunderstandings of what it is intended to measure), and its appropriateness as an indicator that can give any meaningful or useful comparisons between any two authors.

Data Sources

  • See our overview of metrics for a definition of the H-index.
  • Your h-index will be dependent upon the source of publication and citation data. To find you h-index according to data held in Scopus, use Scopus.
  • Other key sources of data include Web of Science and Google Scholar.

How do I do this?

Using the Author Details page

Scopus author details page


For a custom set of publications

  • You can create a citation overview from any set of outputs which you have found within Scopus. This will include a calculation of the h-index for that publication set.
  • This might be useful for calculating the h-index for a research group, or a set of outputs resulting from a single research project which don't all have a single author in common.
  • Scopus Support: How do I create a Citation Overview?

Things to consider

  • The h-index is not an appropriate metric for comparing researchers of differing career lengths or across different disciplines or fields of study.
  • The h-index does not fully account for very highly cited papers.
  • As an author, your h-index in Scopus may be different to your h-index in Web of Science or Google Scholar: Any reference to an h-index should indicate the source of the data from which it was derived.
  • The h-index does not give any depth or granularity of information to distinguish between many different authors:
    • An author with 500 papers, each of which has received 1 citation, will have an h-index of 1.
    • An author who has only published 1 paper, but that paper has attracted over 3,000 citations, will have an h-index of 1.
    • An author who has published several papers, 1 of which has attracted 30 citations, but the others of which have attracted 0 or 1 citations each, will have an h-index of 1.
    • An author who has not published in 20 years will not see their h-index fall, but may still see their h-index rise.

Why you may want to do this?

  • The h-index is a widely recognised metric, calculated from a measure of both the productivity and citation impact of, usually (but not exclusively), an individual author.
  • The h-index is intended to provide a simplified metric for the 'impact' of an author within their field.
  • Many academics, particularly in the Sciences and Social Sciences, will know their h-index and may include it in funding, recruitment or progression applications.
  • You may find you are asked for your h-index in various circumstances, from submitting an application for promotion or employment, to funding applications and even when submitting a manuscript for publication.
    • You should always feel confident to question why you are being asked to provide your h-index, as it may not be appropriate or relevant as an indicator of your research or productivity potential.
    • Authors should be aware that the h-index is widely criticised over limitations in what it does not measure (or misunderstandings of what it is intended to measure), and its appropriateness as an indicator that can give any meaningful or useful comparisons between any two authors.

Data Sources

  • See our overview of metrics for a definition of the H-index.
  • Your h-index will be dependent upon the source of publication and citation data. To find you h-index according to data held in Web of Science, use Web of Science.
  • Other key sources of data include Scopus and Google Scholar.

How do I do this?

From within Web of Science


From your public-facing ResearcherID profile

  • If you have a ResearcherID, your h-index will be displayed based within your ResearcherID public profile, derived from all publications linked to your ResearcherID.

Things to consider

  • The h-index is not an appropriate metric for comparing researchers of differing career lengths or across different disciplines or fields of study.
  • The h-index does not fully account for very highly cited papers.
  • As an author, your h-index in Web of Science may be different to your h-index in Scopus or Google Scholar: Any reference to an h-index should indicate the source of the data from which it was derived.
  • The h-index does not give any depth or granularity of information to distinguish between many different authors:
    • An author with 500 papers, each of which has received 1 citation, will have an h-index of 1.
    • An author who has only published 1 paper, but that paper has attracted over 3,000 citations, will have an h-index of 1.
    • An author who has published several papers, 1 of which has attracted 30 citations, but the others of which have attracted 0 or 1 citations each, will have an h-index of 1.
    • An author who has not published in 20 years will not see their h-index fall, but may still see their h-index rise.

Why you may want to do this?

  • The h-index is a widely recognised metric, calculated from a measure of both the productivity and citation impact of, usually (but not exclusively), an individual author.
  • The h-index is intended to provide a simplified metric for the 'impact' of an author within their field.
  • Many academics, particularly in the Sciences and Social Sciences, will know their h-index and may include it in funding, recruitment or progression applications.
  • You may find you are asked for your h-index in various circumstances, from submitting an application for promotion or employment, to funding applications and even when submitting a manuscript for publication.
    • You should always feel confident to question why you are being asked to provide your h-index, as it may not be appropriate or relevant as an indicator of your research or productivity potential.
    • Authors should be aware that the h-index is widely criticised over limitations in what it does not measure (or misunderstandings of what it is intended to measure), and its appropriateness as an indicator that can give any meaningful or useful comparisons between any two authors.

Data Sources

  • See our overview of metrics for a definition of the h-index.
  • To find an author's h-index according to data held in Google Scholar;
    • Use Google Scholar's own author profile tool.
    • Use Publish or Perish to calculate the h-index for an author with no Google Scholar profile, or for a custom group of publications.
  • Other key sources of data include Scopus and Web of Science.

How do I do this?

If using Publish or Perish, you can create a custom list of publications, or search for a specific Google Profile based on its Google Profile ID. To find this ID:

  • Open the Google Scholar author profile page for the author your are interested in, e.g. Professor Sarah Atkinson.
  • In the address bar, look at the url for the page. The ProfileID can be found immediately after ?user= and may be followed by further characters, such as &hl=en, as in the example below:

Google Scholar ProfileID


Things to consider

  • The h-index is not an appropriate metric for comparing researchers of differing career lengths or across different disciplines or fields of study.
  • The h-index does not fully account for very highly cited papers.
  • As an author, your h-index in Scopus may be different to your h-index in Web of Science or Google Scholar: Any reference to an h-index should indicate the source of the data from which it was derived.
  • The h-index does not give any depth or granularity of information to distinguish between many different authors:
    • An author with 500 papers, each of which has received 1 citation, will have an h-index of 1.
    • An author who has only published 1 paper, but that paper has attracted over 3,000 citations, will have an h-index of 1.
    • An author who has published several papers, 1 of which has attracted 30 citations, but the others of which have attracted 0 or 1 citations each, will have an h-index of 1.
    • An author who has not published in 20 years will not see their h-index fall, but may still see their h-index rise.

Why you may want to do this?

Whilst the 'number of citations' an author has attracted is a very limited indicator (as a simple number would not take any account of discipline, particular research focus, career length or any other contributing factor), tracking the number of citations received may be of interest to an author to see:

  • which of their publications, or publications in particular research areas, are attracting the most interest.
  • identifying who is citing their research, or where citations are originating from.
  • both of the above reasons may help identify key areas to direct effort or key strengths to support funding applications.
  • maintaining a general sense of self-worth that someone, somewhere, might be reading their research output.

Data Sources

  • Google Scholar
  • Scopus
  • Web of Science

When using Google Scholar publication and citation data, where the author does not have a Google Scholar Profile you may find it easier to use software such as Publish or Perish.


How do I do this?

You can view the citations to any group of publications, e.g. all publications of a particular author, as below:

For further help and guidance, contact your academic liaison librarian.


Things to consider

There is no single source of citation data, so bare in mind that:

  • It is unlikely any source will cover all publications by an author, or all citations to those publications.
  • Some sources will often not yet index recent citations and publications.
  • Citation cultures vary across disciplines and subject specialisms, and other factors can also affect citation rates, meaning it is difficult to make any useful comparison based on citation count to an authors portfolio of work alone.

Why you may want to do this?

Scopus will automatically create a Scopus Author ID whenever it identifies (or it believes it has identified) a unique author linked to a paper indexed in the database. It is possible that the algorithm may incorrectly link a publication to your Scopus Author ID, or misidentify a publication authored by you and link this to another Scopus Author ID (either an another author, or create a new Author ID which may require to be merged with the existing Scopus Author ID). It is important that you correct this as:

  • Scopus Author Profiles are public facing: they provide an outward view of your scholarly information not only to Scopus, but to other systems using Scopus data
  • This may include potential employers, collaborators and funders.
  • Having two Scopus Author IDs will mean your publication output, and citations linked to those outputs, are split and neither profile will fully represent your scholarly output to interested parties.

(Whilst Scopus Author IDs are not used directly in university rankings, it is important that the affiliation recorded in your publications and your Scopus Author ID includes Durham University).


Data Sources


How do I do this?


Things to consider

No single source indexes all published content, nor provides access to all available citation data. If you identify a publication of yours which cannot be found in Scopus, you can ask Scopus to index the output (if you believe it has been ommitted in error), or suggest the title (journal of publication, book or monograph) is included in the Scopus index


Why you may want to do this?

The Open Researcher and Contributor ID (ORCID) is an open, non-profit, community-driven initiative providing you, as a researcher, a unique persistent identifier. It is used by funders, publishers and research organisations to try and link together sources of funding with the outputs of research, and for you as a researcher can help you:

  • Avoid mistaken identify based on name-similarities and ensure all your publications (and their citations) are correctly credited to you.
  • "Enter once, re-use often" - ORCID aims to help reduce the data entry you can be asked to do in multiple systems. Rather than trying to search for all your publications using name/subject/affiliation details to try and build a citation profile, just search for all publications linked to your ORCID.
  • Improve the discoverability of your research: help others find, access, read and hopefully cite your work.

Many systems use ORCID - this includes Scopus (where you can search for publications linked to an ORCID, and link your Scopus Author ID to your ORCID) and Web of Science (where you can search for publications linked to an ORCID, and link your ResearcherID to your ORCID).

See our ORCID FAQs for further details.


Data Sources


How do I do this?

Registering for ORCID takes only a few minutes.

Or see our ORCID pages for further information.


Things to consider

  • ORCID is a non-proprietary system, meaning it can (and is) integrated into many systems. This allows you to choose which information to share with each system, without having to manually type in or upload the same information repeatedly.
  • ORCID is increasingly required to submit a manuscript to publishers, apply for funding from funders, or to share information with university CRIS systems.
  • ORCID is strongly recommended for REF2021 for all Category A staff included in the submission.
  • From August 1st 2018, Durham University Research Committee requires all University staff and postgraduate research students who have not already registered for an ORCID to do so.

Your Academic Liaison Librarian

James Bisset

Academic Liaison Librarian
Research Support

james.bisset@durham.ac.uk

0191 334 1589

DU Library Blog

Metrics Top Tips

  1. Always use quantitative metrics together with qualitative inputs, such as expert opinion or peer review.
  2. Always use more than one quantitative metric to get the richest perspective. 
  3. If comparing entities, normalise the data to account for differences in subject area, year of publication and document type.

See our pages on Responsible Metrics for further information.