Open access, impact, and demand

Peter Suber's comments on open access, impact and demand (BMJ, 14 May
2005) highlighted some key issues with respect to author self-archiving
and raised a number of interesting questions. We have recently collected
data in a large-scale survey that can further inform this debate and go
some way towards producing a coherent appraisal of author self-archiving
behaviour (1).

We know, for example, that 49% of authors have undertaken some form
of self-archiving behaviour, placing copies of their articles on their
personal or departmental websites (27% of authors have done this), in
their institutional repository (20% of authors have done this) or in a
subject-based repository (12% of authors have done this). We know, too,
that the number of people doing these things has grown in the last year
since we carried out a previous, similar, survey: for example, the 20% of
authors who have now deposited a published article in their institutional
repository compares to only 10% twelve months ago.

The proportion of authors now who are not aware of the possibility of
providing open access through self-archiving is 31%. There is obviously a
job to do there in raising their awareness of the issue. Institutions have
their own formal advocacy activities which help, but our data show that of
all the sources of information about self-archiving, word-of-mouth from
peers was the most common (23% of authors found out about the practice
that way). This suggests that as time goes on the good-news message about
increased impact and citations for open access articles filtering further
through the research community will have its own outcome in increased self
-archiving activity.

That still leaves the authors who are aware of the possibility of
self-archiving, but not actually doing it, to be encouraged to take that
step. They are presently discouraged mainly because they think it will
take up time or that it will be technically difficult. Data from authors
who do self-archive show that it takes a few minutes to deposit an article
in a repository and that once this has been done for the first time only
9% of authors subsequently find any degree of difficulty with the process
at all.

Both Jonathan Wren's original study (BMJ 14 May 2005) and Peter
Suber's commentary discuss the fact that authors in high impact journals
tend to provide open access to their work more readily by self-archiving
than those who publish in lower-imapct journals. Our study adds another
dimension: there is a positive correlation between the number of papers
authors publish per year and the level of their self-archiving activity.
Some authors - very productive authors - are extremely determined to get
their work out there and are seizing this simple and effective opportunity
to get it noticed.

(1) Swan, A and Brown, S. (2005) Open access self-archiving: an
author study. To be published by the JISC (Joint Information Systems
Committee), May 2005.

Competing interests:
None declared