ASIST2017 METRICS PM

Oct 27 2017 Published by under bibliometrics, Conferences

Edwin Henneken, Alberto Accomazzi, Sergio Blanco-Cuaresma, August Muench, Lars Holm Nielsen Asclepias – Capturing Software Citations in Astronomy

Asclepias project. Enabling software citation & discovery workflows. To "promote scientific software into an identifiable, citable, and preservable object. " Adding DOI based software citations to ADS. Tracking events.

Collaborative Codebase (GitHub) > Repository (Zenodo) > software broker (harvests repositry events, software citations

example corner.py (triangle.py) - published in JOSS, 60 regular citations to, but also deposited in Zenodo. Citations to every single version of the software and a total of 100 citations.

Journals need to be able to accept software citations (actual citation to the software and not a related article). Just slapping a doi on it isn't enough.

End to end go from original proposal through all the data, papers, software, etc. and have analytics along the way.

Q: difficult to get people doing the right thing with the repositories? yes - but astro is amenable. long history of linking data

Q2: like bigger world of citing things not papers about things? yes

Eto MasakiIncreasing Source Documents of Rough Co-citation to Expand Co-citation Networks for Scientific Paper Searches

rough co-citation is a generation back from co-citation.

a + b cited together, co-citation... a+c cited together infer relationship with b, this did increase information retrieval retrieved documents that didn't exist in the network.

Pei-Ying Chen (speaker), Erica Hayes, Stefanie Haustein, Vincent Larivière, Cassidy R. Sugimoto -  Politics of platforms: the ideological perspectives of social reference manager users on scholarly communication

Looking at Mendeley and Zotero - hypothesis that Zotero users will be more to open data, etc., and Mendeley will be more traditional bcs using corporate platform.

Mendeley provided a stratified random sample of 26k users, response from about 1200. Zotero was an anonymous link advertised by Zotero at conferences.

In survey they didn't provide a category for librarians so they got a lot of "others"

From both groups: all advocate for open source software, all adopter of new technologies, most advocate for open access.

Majority of both think peer review system is broken and publishers aren't necessary for scholarly communication.

Some similarities and differences, but no real clear support for their hypothesis, as far as I could tell.

Q: try to look at the contents of the library to see if more oa or paywall journals?

 

Eshan Mohammadi, Mike Thelwall, Kristi Holmes - Interpret the meaning of academic tweets: A multi-disciplinary survey

Altmetrics - who uses twitter to communicate scholarly info, does twitter play an important role in communicating scholarly info, why, does it depend on discipline

twitter users who re/tweeted academic publications at least once using altmetric.com 4.5m twitter accounts

looked at personal web page urls 1.7 urls

using webmining, identified emailaddresses

sent online survey to 57k twitter users, got 2000 responses.

most respondents tweeting scholarly information were from the social sciences and humanities

most agree:

  • change way to read and disseminate sci info
  • twitter facilitates knowledge flows
  • reflects research impact
  • share academic findings with the general public

motivations for using and type of content shared depend on discipline, occupation and employment sector

They have a paper under review in a journal so stand by.

 

Philippe Mongeon Is there a Matilda effect in academic patenting?

We know men publish more papers than women and their papers are more cited

Now for patenting. Only about 15% of inventors are women. Patent-paper pairs. Same discovery published in a paper and patent

are women less likely to be inventor than men when we control for: position on the byline, discipline, reputation, contribution

Previous studies: no gender difference (Haeusslet & Sauermann, 2013), female more likely excluded from inventorship (Lissoni et al 2013)

all articles with 2 or more authors in wos 1991-2016, uspto patent applications 1986-2015

papers patents -1 to 5 years of app, all inventors on the actors list

text similarity of title and abstract.

discipline - based on discipline of journals cited by the paper

attribution of gender - based on Wikipedia pages (Berube in preparation)

automatic disambiguation of authors

accumulated number of citations at time of app.

contributions - manual extraction, where there were statements coded conception, analysis, performed...

regression models...

turns out place in author list has much more impact than gender, but gender is significant for all but engineering.

When taking contribution into account (many fewer papers), conception role is important  - which makes sense.

Small effect of gender on the attribution of inventorship, gender gap occurring earlier in the research process

 

No responses yet

Leave a Reply