The methods used to evaluate the quality of research can be far more accurate and far-reaching, according to a new doctoral thesis on bibliometrics from Karolinska Institutet in Sweden. "A common pitfall is that bibliometricians assess the average quality of journals instead of the individual scientific articles," says PhD Jonas Lundberg.
Bibliometrics is used to describe and assess the quality of research, and to give an idea of the influence a research group or university has on a particular field. As research becomes all the more international and competition between researchers stiffens, more exacting systems are needed to assess the quality of research.
"I usually say that I research into research," says Jonas Lundberg at the Department of Learning, Informatics, Management and Ethics (LIME). "There are a number of bibliometric methods, the simplest of which measures how often a researcher is published and cited in scientific journals."
His thesis points out the shortcomings of current methods, and shows how bibliometrics can be used more accurately and effectively. He has also developed the method that is today considered the best.
"There are two prevailing opinions on bibliometrics," he says. "One places too much faith in it and attaches great importance to simple bibliometrics. There are many pitfalls with today's measurement methods. Those who don't believe in bibliometrics tend to dismiss it out of hand, when in actual fact it is very useful. You just have to use it correctly and to develop the methods."
One of the most common methodological failings is that instead of reviewing each individual article, bibliometricians simply assess the average quality of journals. Furthermore, many scientists publish their articles in popular science magazines rather than in specialist periodicals, something that regular bibliometric methods fail to take into account.
"We know, for example, that 70 per cent of cancer research is published in magazines other than specialist oncology journals," says Jonas Lundberg.
Thesis: Bibliometrics as a research assessment tool: impact beyond the impact factor. ISBN: 91-7140-965-3.
For further information, please contact:
PhD Jonas Lundberg
Phone: +46-(0)8-524 866 89
Email: jonas.lundberg@ki.se