One of the really fascinating points made in the Berstein Report (see here for previous coverage) is that a small proportion of journals account for most the readership within a certain field. We can see this in the graph below:
The data used here is taken from ten universities for readership of articles in the life sciences (they also looked at chemistry, which produced a similar ‘long tail’ dynamic), with the graph showing that the bottom 75% of articles account for 10 to 25% of the readership. This ‘long tail’ of journals is a real elephant in the academic publishing world: for authors it simply means that, unless you’re getting published in a big journal, then it’s unlikely your work is being read. It’d be interesting to see how the data here lines up with the number of citations of a given article.
These findings also have clear implications for academic publishers. As the report notes:
Academic librarians of universities which have discontinued “big deals” with major publishers have found that they could satisfy their core constituencies (faculty, researchers, doctoral candidates) by acquiring a handful of titles, accomplishing significant savings even when forced to pay list prices for the remaining subscriptions.
Assuming the report is correct, then I’m not so convinced that these dynamics will hold in long run, especially as users are shifting to more search-savvy ways of discovering articles that are relevant to them. So, instead of having this long-tail for journals, I would expect to see the long-tail manifest itself in the types of search tools scholars use (e.g. Google Scholar). Of course, if these search tools are heavily biased toward disproportionately prioritising a select number of journals, then we will still be faced with the same problem. Based on my own search habits I haven’t found this to be an issue so far. It’d be interesting to hear other people’s thoughts on the matter.
Tomorrow, I’m going to apply a more critical eye on the Berstein Report’s use of Impact Factor, and how it might not really be telling us much (even if it’s weakly correlated with Author Processing Charges).
- Tags: Impact Factor