[GOAL] Counting researchers - some results

[Arthur Sale]

I recently posed to this list the question ‘How many researchers are there in
the world?’ and gave some rough estimates that bounded the result N by 1M < N 
<
10M.  I have received several very useful pieces of data (and some non-useful
responses). The question is clearly relevant to the production rate of articles.

 

My best estimate of N is now 3.60M researchers in 2012. This is based on the
UNESCO Science Report 2010 which details researchers by country, and this is an
extrapolation from 2 354 851 in 2002 and 2 979 913 in 2007. Note: This figure is
to be treated with caution, because of the following factors:

1.      I use the FTE counts which are higher than the headcounts. On
inspection, many countries (such as Canada, USA and Australia) did not supply
UNESCO with headcounts. I could have fudged the two categories together but the
precision of the data did not seem to warrant that.

2.      The raw data is itself subject to various errors. The footnote to 
the
Table states “Text Box: – Text Box: – Text Box: – Text Box: – Text 
Box: – ‑n/+n
= data refer to n years before or after reference year; a = university graduates
instead of researchers; b = break in series with previous year for which data
are shown; e = estimation; g = underestimated or partial data; h = overestimated
or based on overestimated data.”

3.      Not all of these researchers are what I call ‘producing 
researchers’:
researchers who (co)author articles which could be made open access. It is
difficult to determine this factor though use of article-based author-lists or
author IDs may be useful. This is probably the biggest uncertainty in the data,
and means that 3.60M is probably an over-estimate.

 

One of the reasons I wanted to know this value is to see how large the Mendeley
count of users is – they report 1.43M at time of writing.  Some of these are 
not
‘producing researchers’, but are people searching literature for work, 
hobby or
medical purposes, but private communication suggests this is a relatively small
fraction of the total. In any case, just to do the raw numbers: 1.4M / 3.60M =
40%. If point 3 above dominates, this is an under-estimate of Mendeley’s
penetration as a researcher tool.

 

What this implies is that 40% (or whatever) of researchers in the world are
using Mendeley, and have the potential to make their work open access by simple
actions. Les Carr has blogged that the level of people doing this is about the
same as the level achieved in his University of Southampton departmental
mandated repository. That’s good news in itself.  However, it now poses a new
set of questions: are the researchers in Mendeley different from those
represented in institutional repositories, the same ones, or what is the
overlap? Surely this will vary by discipline?

 

If the user sets their own works to be OA, and the users are disjoint from
repository users, then that implies that the Titanium Road (social networking
OA) is making significant progress in the OA campaign in its own right, and
growing at about 37% from June 2011 to January 2012. The complementary approach
to institutional repositories may be valuable.

 

[IMAGE]

 

The same question may be asked of articles, but it is more difficult to draw
conclusions. An article may be put into a repository and made OA by one
co-author, and into Mendeley and made OA by another. I argue this is a net
benefit - the more copies of an article on the Internet the better (within
reason) though not as useful as a new article made OA. Some however may simply
be focussed solely on different article counts and think of this as a waste of
effort. No matter – it seems that social networking tools are proving useful 
in
achieving OA.

 

Arthur Sale

University of Tasmania, Australia

 





    [ Part 1.2, Image/PNG (Name: "image001.png") 189 bytes. ]
    [ Unable to print this part. ]


    [ Part 1.3, Image/PNG (Name: "image002.png") 186 bytes. ]
    [ Unable to print this part. ]


    [ Part 1.4, Image/PNG (Name: "image003.png") 186 bytes. ]
    [ Unable to print this part. ]


    [ Part 1.5, Image/PNG (Name: "image006.png") 21 KB. ]
    [ Unable to print this part. ]


    [ Part 2: "Attached Text" ]

_______________________________________________
GOAL mailing list
GOAL@eprints.org
http://mailman.ecs.soton.ac.uk/mailman/listinfo/goal