What are publication points?

The formula for calculating publication points

 

In the funding of Norwegian research institutions we find a special Norwegian phenomenon – points for scientific publications. How does it really work?

One of the main tasks of Norwegian research institutes (in addition to teaching for institutions that do teaching) is participating in and contributing to the research community, and the best way to do so is to publish their research so that others can read it.

Each year, Norwegian scientists publish around 23,000 articles, book chapters and monographs which are considered scientific (you can count for yourself on the Norwegian Center for Research Database website). There is much to keep track of, so since 2004 we have had a system for documenting and premiere scientific publishing (UHR 2004), both to have a certain overview of what Norwegian researchers produce and to encourage actually getting their research written down and shared with the research community.

How can such a system work? A scientific publication is not a uniform matter, but rather depends on several things: is it an article, a whole book or a chapter in an anthology we are talking about? Is it written alone or with many colleagues, maybe with someone from a foreign institution? Is the article in a top journal, or in an obscure journal that few people read?

The most important element in the system is the allocation of points for publications. For each point, a small fraction of the annual state pot for so-called performance-based research funding is awarded, and this is meant to be a financial incentive for research to be published.

Old calculation model

So how are the publishing points calculated? Basically, there were three things that ruled this: how much the individual institution’s share of the authorship of the publication was, what kind of publication it was and whether the publication is in one of the publishing channels (magazine, box series) that is considered among the 20% best channels in their field and thus at level 2 in the Norwegian system.

Thus, it works like counting author addresses to the appropriate institution, dividing the total number of author addresses, and times with a factor read by this table:

	Level 1	Level 2
Article	1	3
Anthology chapter	0.7	1
Monography	5	8

The purpose of having such a division is to reflect that it is considerably more demanding to write and publish a whole book than a chapter in an anthology.

Here is an example of how the calculation took place in the period 2004-2015:

Carlsen et al. (2013) has three authors within five research institutions. NTNU’s share is thus 1/5. It’s an article on level 1, so according to the table over we multiply the author share with one. NTNU gets 0.2 publication points.

New calculation model

The old model gave some interesting results. This especially concerned co-authorship on articles with many co-authors, which resulted in very few points, since the institutional share was very small. In addition, it has been a clear political goal to stimulate more international co-publishing. In 2015, two changes were introduced into the calculation model: firstly, the author’s share is adjusted by taking the square root of the fraction, which makes the proportion a little bigger, and, secondly, introducing a multiplier for internationalization of 1.3. This means that if you have foreign co-authors, your publishing points will be multiplied by 1.3 before they are awarded to the publication.

Here is an example of the new calculation model:

Bawden et al. (2016), an article on level 2, has 19 institutional shares, of which NTNU has 4. After the old model, NTNU would have received 0.63 points (the author’s share 0.21 times the level factor of 3). With the new calculation method, we take the square root of 4/19, which is 0.46, and multiply by three for the level and then by 1.3 for international cooperation. It gives a total of 1.79 – almost three times as much as the old model. For articles with many co-authors, most common in STEM subjects, the new rules have meant a significant boost to points production. Humanities, dominated by sole authorship, get largely unchanged points, and thus a declining share of the total score. This has not been uncontroversial …

That was it?

Now you have what you need to know to calculate the publishing points for Norwegian scientific publications, but there are several things about publishing points that are interesting to explore, such as:

• How does the research process and publishing decisions affect researchers when they are given points based on how they publish?
• Has it led to any changes? What is the reason for rewarding international publications?
• Are there other types of publishing business one would like to reward in the future?
• Is it possible to compare points production over time when the rules change along the way?
• What do the new rules say about the view of the different traditions of traditions with funding agencies?

These are important questions in bibliometry, and we are guaranteed to return to the political sides of the registration and reward system for publications in Norway. But that may be for another post …

PS: Both articles used as an example in this entry are Open Access and free to read for everyone. So take a look!

References

Bawden, L., S. P. Coil, F. Mazzola, J. M. Riley, L.J. Collins-McIntyre, V. Sunko, K. Hunvik, et al. 2016. Spin-Valley Locking in the Normal State of Transition-Metal Dichalocogenide Superconductor. Nature Communications, 7. doi: 10.1038 / ncomms11711.

Carlsen, F., J. Grytten and A. Eskild. 2013. Changes in fetal and neonatal mortality during 40 years by offspring sex: a national registry-based study in Norway. BMC Pregnancy and Childbirth, 13: 101. doi: 10.1186 / 1471-2393-13-101.

Universities Norway. 2004. Vekt på forskning. Nytt system for dokumentasjon av vitenskapelig publisering. Report.