Cardiovascular Research 2002 56(2):175-177; doi:10.1016/S0008-6363(02)00651-X
© 2002 by European Society of Cardiology
Copyright © 2002, European Society of Cardiology
Productivity in science: more more and more?
Tobias Opthofa,* and
Ruben Coronelb,c
aDepartment of Medical Physiology, University Medical Center, P.O. Box 85060, 3508 AB Utrecht, The Netherlands
bDepartment of Cardiology, University Medical Center, Utrecht, The Netherlands
cDepartment of Experimental and Molecular Cardiology, Academic Medical Center, Amsterdam, The Netherlands
t.opthof{at}med.uu.nl
* Corresponding author. Tel.: +31-30-253-8900; fax: +31-30-253-9036.
Received 23 August 2002; accepted 29 August 2002
 |
1. Introduction
|
|---|
During the first 60–70 years of the last century it was
more or less possible to read everything that was published
worldwide in whatever scientific speciality. This is no longer
true. De Jong and Schaper
[1] published a paper on the number
of publications in clinical cardiovascular science including
the citation of those papers over a 12-year period from 1981
to 1992. Of the 137 000 papers published during that period
46% remained uncited with a remarkable geographical variability
from 31% for Norway to 69% for Japan. It should be realized
that even a single citation by the authors themselves (self-citation)
during such a long period would classify a paper in the cited
category. Low citation is not synonymous with low scientific
quality over brief periods (as used for popular bibliometric
parameters as the ‘impact factor’)
[2,3]. However,
zero citation of about 62 000 papers over a 6-year period after
publication
[1], raises the question whether the scientific
community should try to limit this flood of publications. This
question becomes even more relevant when it is realized that
electronic publishing will probably alleviate the burden of
page limitations imposed on journal editors, and may lead to
rejection rates that are more in agreement with reviewer’s
recommendations
[4]. It has been pointed out previously that
rejection rates are much higher than what is actually advised
by reviewers
[4].
|
2. More journals—more papers—more references
|
|---|
Fig. 1 shows the number of journals (left ordinate) and the
number of papers published by these journals (right ordinate)
in the Life Sciences as covered by the Journal of Citation Reports,
a product of the Institute for Scientific Information (ISI,
Philadelphia, USA). The number of journals covered by the ISI
grew by 27% from 4464 in 1990 to 5684 in 2000. It is possible
that these numbers may point to a better coverage of what is
published by the ISI and that it may overestimate the real growth
in the number of scientific journals. The number of papers increased
by 38% from about 507 000 in 1990 to 702 000 in 2000.
Fig. 2 shows that the number of citations almost doubled (+96%). This
increase is partly based on the increase in published papers
(
Fig. 1), but to a larger extent on an increase in the references
per paper (16.5 in 1990 and 23.3 in 2000). It is important to
note that the references per paper are relevant to the citing
papers, not to the cited papers. It is, theoretically, possible
that in 2000 the 16.4 million citations in 702 000 citing papers
went to only 24 papers. This would be the case if the 702 000
citing papers with on average each 24 references would all have
cited the same 24 papers.
View larger version (21K):
[in this window]
[in a new window]
[Download PowerPoint slide]
|
Fig. 1 Number of journals covered by the Institute for Scientific Information (Philadelphia, USA) in the yearly published Journal Citation Reports at the left ordinate and number of papers published by those journals at the right ordinate from 1990 till 2000.
|
|
View larger version (19K):
[in this window]
[in a new window]
[Download PowerPoint slide]
|
Fig. 2 Total number of citations in papers covered by the Institute for Scientific Information (Philadelphia, USA) in the yearly published Journal Citation Reports at the left ordinate and references per paper at the right ordinate from 1990 till 2000.
|
|
Fig. 3 shows that the increase in references per paper coincided
with a gradual decrease in the number of pages of scientific
papers at least in Circulation which is the top journal in the
category Cardiac and Cardiovascular System of the Journal Citation
Reports. In 1970 the number of pages per paper was about 10
and it declined gradually to just over 6 in 2000. If this process
continues at the same speed a paper published by Circulation
will be no longer than one page in 2020! It is of interest to
note that Seglen
[5] found a strong correlation between the
number of pages of an article and its citation 4 years after
publication, i.e. well beyond the time window used for the calculation
of the impact factor
[2]. Of course, the decrease in the number
of pages per paper may point to less verbosity, but it is also
possible that it partly reflects a trend to less information
per paper (minimal publishing unit).
|
3. Productivity of individual scientists
|
|---|
The increase in the numbers of published papers probably results
from a larger number of scientists involved in research and
from increased pressure to publish. Individual authors produce
different numbers of papers during their career. Interestingly,
Alfred J. Lotka scored the number of papers of authors with
names beginning with A or B in a decennial index of Chemical
Abstracts covering the years 1907–1916 and published his
results in 1926
[6].
Fig. 4 shows that about 60% of the authors
produced only one paper during that period. He formulated the
‘inverse square law’ which says that the fraction
of scientists producing
n papers equals 1/
n2 of the fraction
of scientists producing only one paper. In other words, about
7% of scientists produced three or more first author papers
and about 1% of scientists produced 8 or more papers (
Fig. 4).
View larger version (21K):
[in this window]
[in a new window]
[Download PowerPoint slide]
|
Fig. 4 Percentage of authors (with surname starting with A or B) who published one or more papers in Chemical Abstracts between 1907 and 1916 [6] and the percentage of authors that had one or more cited papers in the Science Citation Index of 1996. For the latter analysis the first 500 author surnames (all starting with Aa) were scored according to their number of cited papers.
|
|
As stated in the previous section there has been an enormous
increase in the number of scientific journals and papers during
the last decade. It is not known whether the increased urge
to publish scientific papers (‘publish or perish’)
has actually increased productivity per (first) author in a
relevant way. Of course assessment of productivity of authors
has both quantitative and qualitative aspects.
Fig. 4 reproduces
the original data from Lotka
[6] on productivity per author
together with the computed best fit (open circles).
Fig. 4 further
shows the fraction of scientists with one or more papers cited
in the Science Citation Index of 1996. There was a technical
reason to choose for the 1996 edition of the Science Citation
Index, because it permits such an analysis by alphabet more
easily than more recent products of the ISI. We simply scored
the first 500 authors from the database (all with family names
starting with AA). The cited papers could have been published
in any year, but there were hardly papers older than 10 years
as was the case in the analysis of Lotka in 1926
[6]. The agreement
between the data set of 1907–1916 (published, but not
necessarily cited) and the data set of 70 years later (published
and cited) is obvious. Thus, the large increase in scientific
publications (
Fig. 1) is based on the involvement of more scientists.
It is suggested, however, that the productivity of relevant
(cited) scientific information per author has not changed very
much. Maybe this paper should not have been published, so please
cite it.
|
References
|
|---|
- De Jong J.W., Schaper W. The international rank order of clinical cardiology. Eur Heart J (1996) 17:35–42.[Abstract/Free Full Text]
- Opthof T. Sense and nonsense about the impact factor. Cardiovasc Res (1997) 33:1–7.[Abstract/Free Full Text]
- Opthof T., Coronel R. The most frequently cited papers of Cardiovascular Research (1967–1998). Cardiovasc Res (2000) 45:3–5.[Free Full Text]
- Coronel R., Opthof T. The role of the reviewer in the editorial decision making. Cardiovasc Res (1999) 43:261–264.[Abstract/Free Full Text]
- Seglen P.O. Evaluation of scientists by journal impact. Weingart P., Sehringer R., Winterhager M., eds. (1992) Representations of science and technology. Proceedings of the international conference on science and technology indicators, Bielefeld, Leiden. Leiden University: DSWO Press. 240–252.
- Lotka A.J. The frequency distribution of scientific productivity. J Washington Acad Sci (1926) 16:317–323.
CiteULike 
Connotea 
Del.icio.us What's this?
Top
1. Introduction
2. More journals--more papers-...