Yet another problem with Journal Impact Factors
I was at a meeting this morning in which the vexed issue of the journal Impact Factor (IF) came up. That reminded me of something that struck me when I was checking the NASA/ADS entry for a paper recently published by the Open Journal of Astrophysics, and I thought it would be worth sharing it here. First of all, here’s a handy slide showing how the Impact Factor (IF) for a journal is calculated for a given year:
It’s a fairly straightforward piece of information to calculate, which is one of its few virtues.
Now consider this paper we recently published in the Open Journal of Astrophysics:
As of today, according to the wonderful NASA/ADS system, this paper has 36 citations. That’s no bad considering that it was published less than a month ago. It’s obviously already quite an impactful paper. The problem is that if you look at the recipe given above you will see that none of those 36 citations – nor any more that this paper receives this year – will ever be included in the calculation of the Impact Factor for the Open Journal of Astrophysics. Only citations to this paper garnered in 2024 and 2025 will count to the impact factors (for 2025 and 2026 respectively). There’s every reason to think this paper will get plenty of citations over the next two years, but I think this demonstrates another bit of silliness to add to the already long list of silly things about the IF as a measure of citation impact.
My view of citation numbers is that they do contain some (limited) information about an article’s impact, but if you want to use them you should just use the citations for the article itself, not a peculiar and arbitrarily-constructed proxy like the IF. It is so easy to get article-level citation data that there is simply no need to use a journal-level metric for anything at all.
In the Dark 
November 9, 2023 at 4:21 pm
If the HEI or institute is signed up to DORA – as it should be – then IFs will be irrelevant. Citations can however be taken into account as a ‘quality’ indicator, although that has its problems, as sub-disciplines can have very different citation rates. Citation rates could be used as a secondary indicator of quality for REF, which I personally was not happy with, as I worked in areas that did not generally have high citation rates. At least you don’t have REF to worry about – especially the ‘impact’ stuff….
November 9, 2023 at 4:32 pm
You’re right about DORA but I get the feeling that quite a few institutions have signed up to it but don’t really implement its principles. In other words, they talk the talk but don’t walk the walk.
November 9, 2023 at 4:35 pm
Doesn’t the impact factor of a journal depend on how hard you throw it at the library wall on learning the cost to your institution of subscription?
November 15, 2023 at 10:31 am
Peter: Any updates on when/if OJA will soon get an official Impactfactor?
November 21, 2023 at 4:04 pm
Well, Clarivate reject the last application on the grounds we hadn’t published enough papers. It seems they base their evaluation on completed volumes, so I was planning to reapply when Volume 6 is complete because we’ll have more papers in Volume 6 than in Volumes 3, 4 and 5 put together.