SLA2005 - Open Access - Tuesday 13:30

Open Access: Evaluating Quality and Participation

Marie E. McVeigh
Citation Development Manager at Thompson Scientific
isiHighlyCited.com etc.

George Kendell?
National Academy of Sciences
will address publisher's business models

Peter Suber
Open Access Project Director at Public Knowledge
Senior Researcher at SPARC
well-known Open Access News weblog

talks start

Marie E. McVeigh
Citation Analysis of Open Access Journals

how many open access journals are part of the Thomson citation database
- What is OA
- Thomson coverage of OA journals
- Measuring the effect of OA on a journal, and on an article

Nine flavours of OA
- unqualified
- partial
- delayed
- author-fee
- dual mode
- co-op
- OA abstracts
- per-capital OA
- article-level archiving

Thomson normally selects 10% of journals for coverage, they are covering roughly 20% of OA journals.

Lists of OA journals
www.doaj.org
www.scielo.br
www.jstage.jst.go.jp

Total 1638 journals listed.  270 journals covered in Web of Science.

"an incredibly vibrant, changing population of journals"

* Effect of Open Access at the journal level

quick definitions on immediacy index, impact factor, cited half-life

chart on citation performance
- open access journals do appear in top 10% by impact factor and immediacy
- more higher rated by immediacy

focus on medicine - 83 OA journals
- there is a suggestion that a higher percentage of these journals are being cited more in the first year (immediacy)

many journals only have partial OA content online - medicine journals on average 37 years in print, but only 7.4 years of OA content available (I may be mangling this point)

Summary
- net increase in # of journals using OA
- OA journals appear at all impact factor levels, most subjects have at least one OA journal in top impact factor ranking
- some point I missed

chart showing different combinations of OA, partial OA, delayed OA etc.

examined 174 journals in clinical and research medicine
approx 60% of articles 1992-2003 are avail electronic, approx 20% is OA

count increases if you consider additional OA models

chart showing distribution of journals across different OA models

* Effect of OA at the article level

Lawrence's work on increased citations for freely available articles.
However, limited to conference papers in computer science field.
This result may not be representative.

study by Kristin Antelman?
found an effect across several disciplines, for traditional journal articles

articles freely available have a higher number of citations

* Which articles are OA

Jonathen Wren (2005)
- Probability of finding an article freely available correlated with how recently the article was published and with the impact factor of the original source journal

Brief Mention of Access Issue

---

next speaker

Open Access and PNAS

PNAS overview
website 1.5 million hits per week
11.4 million PDFs downloaded

graph of usage decay
most popular in first few months

at six months, content becomes freely available

PNAS Open Access Survey
willing to pay 50/50
max amount $500 (80%), $1000 (15%)

they chose $1000 surcharge

experiment began May 2004
immediate access at PNAS and PubMed Central vs after 6 months

In 2005: if your institution subscribes, they charge you $750

uptake roughly 16-17%

uptake per classification: genetics 21%, neuroscience 19%, down from there

Are OA papers read more and sooner than subscription access papers?
- Preliminary data indicate OA papers are read by almost 50% more in the first month

Are OA papers cited more and sooner?
- Preliminary data indicate no

Does OA cause subscription cancellation?
- don't know

In first month, OA articles have about 900 hits, subscription articles just under 600 hits.

Chart showing no citation impact difference.

they have made a clear distinction on their site (visually) between OA and non-OA articles
have also marketed OA

eligible authors were not using discount, so they are now manually checking

* Conclusions and further steps

- survey author community
- weigh risk and opportunity
- a couple more I missed

---

next presenter
Peter Suber
Faculty Interest in Open Access

Vehicles of OA
- archives or repositories
- journals
- other (e.g. personal websites)

Faculty
- wider access (more stuff to read)
Library
- save money
- make research more accessible

Awkward background facts

Librarians understand the problem better than faculty.
Faculty have more control over the solution than librarians.

Faculty control

- they decide whether to submit to OA journal
- they decide whether to deposit in repository
- they decide whether to transfer copyright

More awkward facts

Faculty are slow to act.  They are focused on their research.
- Only about 6% journals are fully OA
- Only about 20% of faculty have self-archived to an institutional repository

Why are faculty slow to act?

- don't understand it
- fear that it will take too much time

Need to convince faculty that this is a career-building activity.

Providing OA to their own work: Why?

- OA increases impact
- further studies opcit.eprints.org/oacitation-biblio.html

- article research Harnad and Brody D-Lib 2004 (for Physics)
- 250% to 550% impact ratio increase

faculty options:
- submit to OA journal or submit to non-OA journal allowing postprint archive
- you may need to retain specific rights

discussion of OA journals

OA archives or repositories

- preprints (no permission needed) - but some journals still disagree, consider it prior publication
- postprints - need permission from rights-holder, but most already give permission in advance
- institutional repositories
- disciplinary repositories
- should be OAI-compliant

He has a list www.earlham.edu/~peters/fos/lists.htm

Retain key rights

For the cases where the journal does not permit OA, the author needs to retain key rights.
- grant the right of first print and electronic publication, and retain the rest
- retain the right of postprint archiving and grant everything else
- many journals will negotiate if asked
- they need to know what authors want
- use retained rights to consent to OA

What about prestige?

- OA journals can have prestige, but many of them don't
- Faculty can help make OA journals prestigious
- enlighten hiring, tenure, promotion committees

Real excellence will bring reputed excellence with it, after a time lag.

Why remove permission barriers?

- or why open access and not just free access?  why permit usage beyond "fair use"?
- interest of authors
-- free users from the delay of asking permission
-- free users from fair-use judgment calls
-- free archivists to migrate content to new media and formats
-- unleash readers, researchers, teachers, translators, archivists

How librarians can help

- Launch an OA institutional repository
- support OA journals
- encourage faculty

Launch an IR
Fill the IR
- help faculty deposit their work
-- examples
--- MIT (wandering 3 FTEs)
--- St. Andrews (just email us, we will do the rest)

- administrators should adopt a policy to encourage or require deposit

Support OA journals: (1) Publish them

- like Philosophers' Imprint from U Michigan
- like J of Insect Science from U Arizona
- Cornell hosts Project Euclid

Support OA journals: (2) recognize them

Educate faculty

- copyright issues
- trying is risk-free
- experts can help
www.arl.org/sparc/author/addendum.html

Faculty may not know...

- which OA journals exist in their field
- that there are OA archives, not just OA journals
- OA archiving is compatible with publsihing in a non-OA journal
- 80% of surveyed non-OA journals permit postprint deposit in an OA repository
- may have to retain rights to provide OA
- that OA significantly increases citation impact

---

Q for Peter Suber: Has anyone actually convinced admin committees to change their evaluation criteria for promotion etc.?
A: No.

Comment from ATT Labs person: Google Scholar people (Steve Lawrence?) may be interested in impact factor info.  Mention that in corporate context, the corporation owns the copyright.
Peter Suber says: Why is ATT transferring copyright to publications, if they want to retain their intellectual property?

Guy from Caltech: published conference papers Open Access, hugely successful.
Discussion with PNAS guy about author charges.

Eliizabeth Knight, science libriarian: Q: Creative Commons?  Does that play a role?
A (Peter Suber): Yes.  Creative Commons removes permission barriers and makes it clear that those barriers have been removed.  They provide human-readable, lawyer-readable, and machine-readable versions of the permissions.