On Mon, Nov 1, 2010 at 10:24 AM, Michael Peel <email(a)mikepeel.net> wrote:
On 31 Oct 2010, at 23:08, John Vandenberg wrote:
We should be careful with new studies even when
published in respected
journals, until the citation count rises to the point that we feel
comfortable that the study has been accepted by the academic
community.
The citation count isn't the only measure within academic journals, though - the
reputation of the author should also be borne in mind, i.e. (speaking generally) the
reliability at which their previous works have been rated, and hence the likelihood that
the new work that they have been published should also be considered worthwhile of
attention. And, of course, the level of peer review that the article has undergone -
different journals require higher standards of review, and hence will have different
initial levels of acceptance/trust from the academic community. Relying on citations alone
is definitely a flawed measure, and is not something that we should rely on in solitude if
we're interested in covering the latest scientific findings.
I agree that citation count isn't a good measure of 'truth'. As you
point out, it is much more complex than that.
The funding is almost inconsequential when considering
these other metrics, given that they're based almost entirely on alternative sources
of reliability (or should be within an ideal information/scientific-based world).
In addition, requiring that we describe the funding in our articles is
going to be an exercise in madness for us, and useless information for
the reader. For example, it is quite common for funding to be split
between a competitive grant and an industry partner. Often the
competitive grant _requires_ an industry partner, as the public purse
wants to know that there is industry demand for the research, and also
wants industry to pay part of the costs. In these cases we would need
to say 'According to a 2009 study funded by [x] and [govt body]...',
which then begs the question of how much of it was paid for by whom.
--
John Vandenberg