Supporting Bibliometrics
Download as PPTX, PDF2 likes593 views
The document discusses the role of librarians in research evaluation, focusing on the relevance of bibliometrics for researchers at the University of Warwick. It outlines how services like expert guidance, workshops, and repository management are offered to support researchers in understanding and utilizing bibliometric measures for publication and grant applications. The importance of various metrics, such as h-index and citation counts, in assessing academic performance and departmental benchmarking is also highlighted.
Supporting Bibliometrics
- 1. JIBS - “Research evaluation - is it our business? Librarians' roles in the brave new world of research evaluation” Supporting BibliometricsJenny DelasalleAcademic Support Manager (Research)
- 2. I will cover…Some background information about University of Warwick and about me.Bibliometrics in use: how they can be relevant to researchersUnderstanding bibliometrics: stuff I tell researchers!
- 3. About me & my role at WarwickAcademic Support Librarian (Research)Head ofAcademic Services Research Exchange Advisers, Co-ordinator, Assistant., Ambassadors
- 5. What the Library offers to ResearchersSubject expert ASLs who liaise with departments about their needs & deliver information skills training, enquiries support, etcSpecialist collections, including MRC (Archive)Information Management strand on the Research Student Skills ProgrammeOne-off, tailored workshops and presentations.WRAP repository & NEW Publications ServiceExpertise on advanced information skills… including bibliometrics, copyright, publishing, Web 2.0The Wolfson Research Exchange
- 6. Liaison with ….IT ServicesStudent Careers and Skills - for training students (including PhD students)Graduate School – administration of post-graduate students (including PhD students)Institute for Advanced Studies – early career fellowshipsLearning and Development Centre - part of HR, for training staff.Research Support Services – help with grant applications but also managing fincances throughout projectManagement & Planning office – handle REF return but also provide HoDs with information.Communications Office – promoting Warwick researchLegal Compliance – guidance on copyright, data protection, etc
- 7. Routes to reach researchers at WarwickRSF – Research Staff ForumResearch CommitteeFaculty Research Forum (chairs)(Heads of) Research Centres/ Institutes (Heads of) Doctoral training centresDirectors of Graduate studies PhD supervisers’ trainingDirect e-mail (list from HR)Via ASLs, Graduate SSLCs and Library repsCommunications Office newsletter every week & monthly magazine & the Uni IntranetLDC newsletterWRAP newsletterLibrary website, Research Exchange website, blogs, Facebook
- 8. An interest in bibliometrics!Journals to keep/withdrawRepository management: advocacy messages that OA publishing boosts citations.Does repository deposit make a difference to citations?Will the REF include citation measurements?Web of Knowledge training free, through MimasJiscmailLis-bibliometrics list.
- 9. Our Approach to bibliometricsUnderstanding of bibliometrics is a key information skillBest not to refer to the term “bibliometrics” itself?Cover elements of these in other workshops:“Getting Published”“Literature Searching”Use the controversy around publication “strategy” to engage audience: workshop at a doctoral centre.Publicise our expertise: amongst researchers and University management.Online material to complement our offer. REx website: Researcher to Researcher approach
- 14. Bibliometrics in useConsidering where to publish.What do we do about the REF?Benchmarking a department.Evidence for a job application/grant application/salary review, etcMeasurements in University rankings.
- 15. Appraising journals when considering publicationLibrarians: Know the data sources (WoK Ifs, SciMago SJR, RePec, Anne-WilHarzing’s website)Know about Open AccessKnow where journals are indexed & which indexes are important to a disciplineOther than impact factors, we have knowledge of:Features of journals such as letter, blogs, reviews, etcCopyright agreementsPublicising your work using Web 2.0Other publication channels than journals: how to measure them!Cross-disciplinary perspective.
- 16. Impact Factors : JCR and SciMago
- 17. What can we say about REF 2014 and bibliometrics?Not much – yet!Watch for announcements this summer.Work with department which gathers our return.Ensure that publications data is properly recorded.Medicine, Science and Engineering most likely candidates.
- 18. Benchmarking a departmentNo. of papers published by a department in a given period, for example:Source: http://www.science-metrix.com/pdf/SM_CETC_Energy_Report_V05_F3.pdfNB how do you define which papers to be included for each set?No. of citations: total or per paper?
- 19. Evidence about an individual – h-indexh-index of 4 means there are 4 published papers each with at least 4 citationsOR: Professor X has a total of 10 publications Publication 1 20 citesPublication 2 18 citesPublication 3 11 citesPublication 4 7 cites----------------------------------------------------------- h-index: 4Publication 5 4 citesPublication 6 3 citesPublications 7,8,9,10 0 cites
- 20. Evidence about an individual (2)Egghe's g-index – aims to improve on the H-index by giving more weight to highly-cited articles. Contemporary h-index - aims to improve on the H-index by giving more weight to recent articles and therefore rewards academics who maintain a steady level of activity.Be wary of others’ declared h-index scores: which data sets are they based on? Context is everything!
- 21. Measurement in University RankingsWhich ones matter to your University?Academic Ranking of World Universities (ARWU) - also known as ‘The Shanghai Rankings’ Times Higher Education (THE) World University Rankings The QS World University Rankings Webometrics Ranking of World Universities A good ranking… helps universities attract the most talented researchers.university can set higher fees and attract more funding for research.BUT how they are calculated might not mean that a high h-index is the thing to ask of your researchers…!
- 22. Understanding BibliometricsNot one “bibliometric” measure but many: Paper countsjournal impact factors The h-indexcitation scores at article level visitor numbers (or other info) for online articles Motivations for publication.Motivations for citation.Effects of collaboration? Discipline? Differences in data sources: their coverage/collection, the measures they publish and the graphs/analysis they offer.
- 23. RAE: what was submitted?All institutions 2008 RAE journal articles: 167,831All institutions 2008 RAE all types: 222,177https://www.jisc-collections.ac.uk/Project-Merit/Merit-Search/
- 24. RIN (September 2009). Communicating knowledge: How and why UK researchers publish and disseminate their findings, Research Information Network; JISC.
- 25. Collaboration effectsOther researchers – own discipline, or other disciplines? Peer review practices vary: in the Arts can take upwards of 6 months.With industry – publication good for them?Agreement on what to publish & whereAcknowledging contributions: variety of practiceNegotiation skills!
- 26. Some motivations for citationsPaying homage to expertsEspecially those likely to be peer reviewers!Credit to peers whose work you have built uponLend weight to own claimsProvide background readingCriticising/correcting previous workSign-posting under-noticed work (own paper which would affect your h-index!)Self citations!
- 27. Citation patternsMost publications have little or no citations.Variety across the disciplines.Therefore comparisons within a discipline are most useful.Percentages against a world average within each discipline are more useful than basic numbers.
Editor's Notes
- #4: I am a Librarian! Former repository manager and former Project officer and former Assistant Subject Librarian. Also a secondment to our Careers service.
- #16: Beyond our expertise?: standing in the field, relevance to the article, publication lead times, rejection rates, etc
- #19: Elsevier SciVal or TR InCites can make this much quicker/easier to do.
- #20: Scimago publishes an h-index for a journal. Easier to generate a report on WoK, Scopus or Publish or Perish and have this analysis done for you!
- #22: Example: better to have one article cited a hundred times or ten articles each cited ten times?
- #25: What your peers are doing: do you want to do the same?“We investigated the kinds of outputs being produced by activeresearchers through a bibliometric analysis of the outputsproduced in 2003 and 2008 by a sample of authors who wereincluded in the last two RAEs. A key point to be stressed is howmany of these researchers, across all disciplines, did not produceany publications at all in those two years. Despite intensivesearches across a wide range of sources, bibliographic databasesand websites, we could find no traceable outputs for 52% of oursample in 2003, and for 45% in 2008. Somewhat surprisingly,the proportions of non-publishers were as high in the life sciencesand physical sciences as in the humanities and social sciences.”Also: (LISU slides) report a slight increase in the no. of publications per author in 2008 compared to 2003, but a significant increase in no. of multiple-author works. These are multi-institutional and international. No apparent difference in citation behaviours.
- #26: NB Collaborative publication practices: great variety in acknowledgement through order of names: might be greatest contribution 1st or alphabetical or most senior grant holder last. Leads to problems when crossing disciplines?