The journal impact factor has
numerous flaws, which makes it highly irresponsible for the UGC to rely on it
to evaluate a teacher’s research performance and decide whether she gets a job
or not.
University Grants Commission.
Credit: PTI
18/Apr/2018
Since 2012, nearly 12,000
individuals and 500 organisations have signed the San Francisco Declaration on
Research Assessment (DORA). This includes India’s Department of Biotechnology
(DBT). In fact, in their joint Open Access Policy, the Department of Science
and Technology (DST) and the DBT quote the central recommendation of the
declaration verbatim, which is that journal-based metrics like the journal
impact factor (JIF) should not be used as “a surrogate measure of the quality
of individual research articles, to assess an individual scientist’s
contributions, or in hiring, promotion, or funding decision.”
And yet this enlightened attitude
has only partially filtered down to the ground.
The JIF is a simple metric, originally
designed to help librarians decide what journals to buy for their libraries. It
is the total number of citations received by a journal in the preceding two
years divided by the total number of citable items published in those years.
In their book chapter preprint, Vincent Larivière and
Cassidy R. Sugimoto lay out six major critiques of the JIF. The first is the
inclusion of citations for “front matter” such as editorials, news reports,
obituaries, letters to the editor, etc. in the numerator while not actually
counting these items in the denominator as they aren’t ‘citable items’. The
authors show how large journals like Nature and Science have used
front matter to boost their impact factors. The second critique is the
inclusion of self-citations (when a paper from a journal cites other papers
from the same journal), which has led to documented cases of manipulation by
unscrupulous editors.
The third critique is the
arbitrariness of the two-year window for citations which favours certain
disciplines over others. Larivière and Sugimoto write that, while “physics
papers generate more citations than psychology papers within the first five
years, the reverse is true for the following 25 years. … [U]sing a 30 – year
citation window, we find that the first two years captures only 16% of
citations for physics papers, 15% for biomedical research, 8% for social
science papers, and 7% in psychology.”
Another related critique is that the
JIF does not take into account the differences between fields and disciplines.
Because of this, the indicator cannot be used to compare across
disciplines. The difference in publication and reference practices means that
“medical researchers are much more likely to publish in journals with high JIFs
than mathematicians or social scientists.”
The fifth critique is the skewness
of science research, i.e. that a small percentage of papers accounts for the
majority of citations. Their analysis shows that for the large majority of
journals indexed in the Journals Citations Report 2016, only 20-40% of papers
receive as many citations as the JIF suggests.
The last critique is the systematic
inflation of average JIFs, caused by a number of factors, including the rise in
number of papers and references per paper. Some 56% of journals increased their
JIFs between 2014 and 2015. Larivière and Sugimoto write, “As there is no
established mechanism for acknowledging inflation in reporting, editors and
publishers continue to valorise marginal increases in JIFs which have little
relation to the performance of the journal.”
The JIF in India
In a recent paper, titled
‘Evaluation of research in India – are we doing it right?’, Muthu Madhan,
Subbiah Gunasekaran and Subbiah Arunachalam discuss how the “the answer to the
question in the title cannot be anything but ‘no.’”
(Muthu Madhan and Subbiah
Arunachalam are affiliated with the DST Centre for Policy Research, Indian
Institute of Science, Bengaluru, and Subbiah Gunasekaran with the CSIR-Central
Electrochemical Research Institute, Karaikudi.)
They systematically go through the
evaluation and promotion frameworks of a number of different regulatory
agencies in India and critique their use of JIF and other metrics. The DBT and
DST still include cumulative impact factor as a criterion for awards like the
Ramalingaswami Reentry Fellowship, the Tata Innovation Fellowship, the
Innovative Young Biotechnologist Award and the National Bioscience Awards for
Career Development.
The Indian Council of Medical
Research “routinely uses average IF as a measure of performance of its
laboratories.” Laboratories of the Council of Scientific and Industrial
Research on the impact factor and number of papers published to assess
scientists.
The National Assessment and
Accreditation Council uses various bibliometrics including impact factors in
its accreditation process. It also asks for the “h-index of each paper”, which
the authors describe as “patently absurd” because it betrays a fundamental
misunderstanding of what the h-index means. Only individuals can have an
h-index.
Business schools have instituted
monetary incentives for publishing in high impact-factor journals. Dinesh
Kumar, the chairperson of research and publications at IIM-Bangalore, told the Wall Street Journal
in 2011 that the institute had been giving a cash award since 2006 to any
faculty member whose paper was published in an ‘A-grade journal’.
The National Academy of Agricultural
Sciences assigns impact factors of its own to journals and uses these scores to
select fellows. This impact factor is calculated in an opaque and seemingly
arbitrary manner. The authors argue, “The Annual Review of Plant Biology
had an IF of 18.712 in 2007, which rose to 28.415 in 2010. Yet, the NAAS rating
of this journal recorded a decrease of four points between the two years.”
But the most problematic deployment
of the JIF is its use in the appointment and promotion of teachers by the
University Grants Commission (UGC) and the All India Council for Technical
Education. The UGC calculates an Academic Performance Indicator (API) score
which includes points for research. According to UGC policy, teachers earn more
points for papers published in journals with a higher JIF. The authors of the
‘Evaluation of research’ article summarise the rules thus:
The API score for papers in refereed
journals would be augmented as follows: (i) indexed journals – by 5 points;
(ii) papers with IF between 1 and 2 – by 10 points; (iii) papers with IF
between 2 and 5 – by 15 points; (iv) papers with IF between 5 and 10 – by 25
points.
As discussed above, the JIF has a
number of flaws. It fluctuates erratically from year to year because of the
two-year window. It favours certain journals and disciplines. It doesn’t take
into account any kind of field-normalisation. It doesn’t predict citations. And
it suffers from a creeping inflation. This makes it highly irresponsible for
the UGC to rely on JIFs to evaluate a teacher’s research performance and decide
whether she gets a job or not.
The UGC has further compounded the
arbitrariness of its policy by formulating awarding points based on the ranges
of impact factors. As Madhan et al write,
Take the hypothetical case of a
journal whose IF is around 2.000, say 1.999 or 2.001. No single paper or author
is responsible for these numbers. If a couple of papers receive a few more
citations than the average, the IF will be 2.001 or more and the candidate will
get a higher rating; if a couple of papers receive less than the average number
of citations the IF will fall below 2.000 for the same paper reporting the same
work.
In 2010, Anthony van Raan, director
of the Centre for Science and Technology Studies at Leiden University, the
Netherlands, told Nature that,
“If there is one thing every bibliometrician agrees, it is that you should
never use the journal impact factor to evaluate research performance for an
article or for an individual – that is a mortal sin.”
Metric-based assessments discourage
risk-taking and long-term thinking among young scientists. It tells them that
they can’t afford working on something that won’t lead to citations and papers
immediately. Institutions need to follow the lead of the DBT and consider
signing the DORA. It would be the first step in signalling to young
researchers, as the declaration states, “that the scientific content of a paper
is much more important than public metrics or the identity of the journal.”
Source: The Wire / The Sciences dated 18 April, 2018
(accessed on April 23, 2018 at 2.45pm)
