More Year-End Stats for 2018

For the 2017 Journal Impact Factor, JGR Space Physics increased ever so slightly from 2.7 to 2.8 (rounding to two sig figs). Remember, this is only the second year of the 7 JGR journal titles having their own JIF scores, before that they all shared one, which hovered in the ~3.3 range. Also remember how the JIF is calculated – it is only citations in 2017 to papers published in 2015 and 2016. So, let’s take a closer look at citations to papers in that window.

Here is a line from a chart that Wiley staff put together for us:

JGRA-Dec2018-CitationsPerPaper

These are citations in 2017 to papers published in JGR Space Physics during 2015 and 2016. By the way, I am not revealing a trade secret about the journal here; you could do this yourself if your institution has a subscription to the Clarivate Analytics Web of Science database.

Just under 20% of the papers had zero citations, over half of the papers had 1-3 citations, just over 20% had 4-7 citations, and ~6% had 8 or more citations. Only one paper surpassed the 40 citation mark within the 2017 JIF window, by Kurth et al.

I am not concerned that 300 papers had no citations. Nearly all of these will get cited. Back at the beginning of this blog, I analyzed the Lost Papers and they are only a few percent, even just 5 years after publication. I am certain that this number will drop dramatically in the next few years for this collection of papers.

While I would like more papers to be in the 4-7 range than the 1-3 range, this will take a monumental shift in how the space physics community reads journal articles and cites them. Look at the chart I had in this post – our research community cites papers an average of ~3 times per year for many years and the citations for an average paper grow fairly linearly. I made that chart in January 2017, so the first column (citations to papers published in 2015) is citations of papers from 1.0 to 2.0 years old (so, on average, papers that were ~1.5 years old). After that, it grows at roughly ~3 citations per year for the next 10 years. Furthermore, JGR Space Physics has a citation half life of over 10 years, meaning that 10 years after publication, the average paper can expect its number of citations to more than double as the years progress, with over 60 eventual citations. That’s a lot!

Special side note if you are feeling bad that your decades-old papers do not have 60 citations: remember that these are averages, not medians, and that citation count is a positive definite value with a long-tailed skew. So, most papers will actually have fewer than 60 citations. I have not plotted the histogram and run the numbers; maybe I’ll do that some time.

My point is that space physicists don’t cite papers the way that the creators of the JIF expect researchers to cite papers. The JIF creators expect researchers to cite a paper soon after publication and then the paper declines in importance and relevance with time, perhaps even being forgotten at some point. This might be true in some fields, but not in space physics. To me, it does not seem that we cite articles that way.

Again, remember how the JIF is calculated: it is citations in the year 2017 to papers published in 2015 and 2016. Take the average: a mid-2017 paper citing a paper published at the end of 2015. It is only 1.5 years past its publication date, the same as the first column in my chart in the previous post. So, the average should be ~3. And guess what? To one significant digit, that is the JIF of JGR Space Physics.

If the journal is growing in size, then the paper count is lopsided towards young papers, reducing the citations from this expected average. So, going all the way back to the number of papers with 1-3 citations versus 4-7 citations, I am not surprised that the numbers are what they are. Let me say it one more time – this is how space physicists cite the literature.

Keep on doing what you do, space scientists. Keep reading the literature and citing those papers most relevant to your work: that is, those that you think build the case for a new study; those that contain the descriptions and example uses of the methodology you choose; and those that place proper context around the significance, originality, and limitations of your findings. If those are new papers, so be it. If those are decades-old papers, that’s fine too.

I look forward to seeing your manuscript submissions in the coming year.

 

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s