Uncovering the Path from Mendeley Readership to Citations - - - PowerPoint PPT Presentation

Uncovering the Path from Mendeley Readership to Citations

• Work-in-progress -

Fereshteh Didegah* – Juan Pablo Alperin – Rodrigo Costas

* f.didegah@ubc.ca

Altmetrics18 Workshop – London, UK – 25 September 2018

Bac ackg kground

• Weak to moderate correlations between Mendeley saves and citation

counts of research articles (Costas, Zahedi, & Wouters, 2014; Thelwall, Haustein, Larivière, & Sugimoto, 2013)

• Most relevant to current study: 85% of the users saved the articles to

their libraries with a tendency to cite them in their publications (Mohammadi, Thelwall & Kousha, 2016)

Uncovering the Path from Mendeley Readership to Citations

Ai Aims and objectives

• To technically investigate the extent to which Mendeley users cite the

saved articles from their libraries in their publications through matching Mendeley user profiles with Scopus author profiles.

• To find a mechanism to understand how scholars use Mendeley in

their workflows.

• To examine whether there is a direct pathway between saving articles

in a library to citing them in future publications.

Uncovering the Path from Mendeley Readership to Citations

Data ta & & method

Web of Science:

• Articles from 3 journals: Scientometrics, Journal of Informetrics and Journal
• f the Association for Information Science and Technology
• Published in 2012
• 452 articles with DOIs
• 5,053 unique citing authors, citing the 452 articles between 2012-2016

(Authors’ information including first name, last name, institutional affiliation, discipline)

• Author profiles were matched with Scopus to retrieve their Scopus IDs

Uncovering the Path from Mendeley Readership to Citations

Data ta & & method

Uncovering the Path from Mendeley Readership to Citations

Mendeley:

• 5,019 users (profile IDs) saved the articles to their Mendeley library between

2012-2016.

• The profile IDs were searched through Mendeley API to get users’ information

(including first name, last name, institutional affiliation, discipline, Scopus ID and ORCID number).

• The Mendeley users were matched with the citing authors for each article first

via systematic matching of Scopus IDs and then through matching first names, last names and institutional affiliations, which was carried out manually.

How m many us user ers c s cited ed a articles i es in t n thei heir publ publications? s?

Through matching the Scopus IDs via Scopus and also matching first names and last names of citing authors (5,053 authors) with the Mendeley users (5,019 users), only 270 users (~5%) were identified in the list of citing authors, citing at least one of the articles they saved to their libraries.

Uncovering the Path from Mendeley Readership to Citations

Multi tidisciplinary a approach

Cited articles % Saved articles % Information Science & Library Science 63.57 Management 17.08 Management 7.06 Information Science & Library Science 9.16 General Biomedical Research 6.1 General Biomedical Research 7.15 Computers 3.7 Computers 4.95 Science studies 1.6 Cancer 4.69 Sociology 1.29 Economics 3.43 General Psychology 1.28 General Physics 2.98 General Biology 1.12 Ecology 2.97 Communication 1.12 Biochemistry & Molecular Biology 2.07 Electrical Engineering & Electronics 0.96 Sociology 1.9 Uncovering the Path from Mendeley Readership to Citations The 270 matched scholars:

• Saved 8,182 unique DOIs to

their libraries (from 132 fields)

• Cited 623 unique DOIs (from

52 fields) A stronger disciplinary focus in citing publications while more disperse range of fields when it comes to save publications in their libraries

Concluding ng r remarks

• Technically, only ~5% of Mendeley scholars cited their saved library

documents in their publications while 85% scholars surveyed by Mohammadi, Thelwall & Kousha (2016) intended to cite the saved documents later.

• Scholars saving behaviour differs from their citing behaviour

according to different features.

• Future work: to explore a larger dataset and compare saving vs. citing

behavior across a range of years and various subject domains.

Uncovering the Path from Mendeley Readership to Citations

THANK YOU!

Fereshteh Didegah* – Juan Pablo Alperin – Rodrigo Costas

* f.didegah@ubc.ca

Altmetrics18 Workshop – London, UK – 25 September 2018

Other r characteri risti tics

Uncovering the Path from Mendeley Readership to Citations

Features Saved articles Cited articles Highly cited 26% 17% Self-citation 49% 54% Co-author credit 32% 51%