Scientific Excellence at Scale: Open Access journals have a clear citation advantage over subscription journals
— by Mirjam Curno and Stephanie Oeben
Frontiers is the 4th most cited publisher amongst the 20 largest publishers, ranked by average citations over a three-year period (2015-2017). The same analysis also reveals a citation advantage of Open Access journals over subscription journals in the last three years.
In 2017, 16% of peer-reviewed papers worldwide were published in fully Open Access journals according to SCImago (2018). After over two decades of the Open Access revolution, this is a far cry from the aspired 100%. One of the many reasons why the transition to Open Access turned out to be so difficult is the debate whether Open Access can produce scientific excellence.
In this blog post we performed a journal analysis contrasting open access journals and subscription journals based on data from SCImago (2018, SCImago data is based on Elsevier’s Scopus database). The full dataset is available on Figshare here. This data was presented at the EuroScience Open Forum 2018, the powerpoint is available here and the video lecture here.
In 2017, four of the 20 largest publishers are full Open Access publishers (Figure 1), 16 are traditional subscription publishers with a range of subscription, hybrid and fully open access journals.
Looking at average citations over a three-year period (2015-2017) amongst the largest 20 publishers, the performance of fully Open Access publishers is remarkable (Figure 2).
Frontiers ranks 4th most cited for average citations per paper published between 2015 and 2017 amongst the 20 largest publishers. Frontiers’ average citation rate is 3.65 per paper and well above the average of 2.7 for subscription journals and 2.9 for Open Access journals. Amongst the ranked publishers, only Learned Societies specializing in a single, highly-cited field, such as chemistry or physics, rank higher. Among multidisciplinary publishers, Frontiers has the highest average citation rate.
Other fully Open Access publishers also rank above average on citation rates: PLOS ranks 6th with an average citation rate of 3.25 and MDPI 8th with an average citation rate of 3.10 (Figure 2).
In general and across the last three years, Open Access journals receive on average 7% more citations than subscription journals (Figure 3). Interestingly, Open Access journals published by traditional subscription publishers are generally achieving more impact within the same publisher.
If we break out open access and subscription journals by those 20 publishers, Open Access journals see a citation advantage, with an average of 2.9 citations per paper for Open Access journals versus 2.7 citations per paper for subscription journals (including hybrids, Figure 3).
While 10 out of the 16 traditional publishers show a clear citation advantage for their Open Access journals, there are some variations, likely reflecting different strategies with the titles, as well as a publisher’s standing in terms of the content they attract.
These findings are in accordance with a recent analysis by DeltaThink which demonstrated that an increasing number of fully OA journals are attaining higher impact factors at faster rates than their subscription and hybrid counterparts.
The findings are also in line with the citation advantage of Frontiers Open Access journals in the 2017 Journal Citation Reports and 2017 CiteScore.
Many have appreciated the benefits of Open Science for society; accelerated discovery, innovation and economic growth and public access to scientific knowledge. This data demonstrates that Open Access journals outperform traditional subscription journals on citation metrics and deliver higher impact for authors.
Total Documents (2017)
Average cites per paper (2015-17)
Average citations (2015-17): Subscription vs Open Access
- Download 2017 SCImago data https://www.scimagojr.com/journalrank.php
- Add publishers & Open Access classification.
A. Full Open Access are journals in which every article is Open Access. The list is based on DOAJ and Scopus mark-up with additional manual curation to also show those full Open Access journals not marked as such in DOAJ or Scopus.
B. Publisher assignment is based on SCImago and Scopus but manually curated against 2017 journal lists on publisher websites. We also group imprints into one main publisher, e.g. Cell Press is shown as Elsevier. Where no publisher is found, this is shown as unallocated. There are 2951 unallocated documents, which is 0.1% of the SCImago dataset.
- For the Top 20 publishers by volume (Total Documents in 2017) calculate the average citations per paper over a 3 year window (2015-2017).
- Calculate average citations per paper (2015-2017) for open access and subscription journals of those Top 20 publishers.
- Create three charts:
A. Documents (2017) for the top 20 publishers
B. Average cites per paper (2015-2017) for the top 20 publishers
C. Average cites per paper (2015-2017) for open access journals and subscription journals for the top 20 publishers, also showing averages for the top 20 aggregates.
- Study only shows journals as indexed by Scopus at the time the SCImago snapshot was made.
- Subscription journals which allow to make individual article open access (so called Hybrid journals) are counted as subscription journals, as the dataset did not allow to disambiguate content at the individual article level.
SCImago, (2018). SJR — SCImago Journal Rank. Retrieved 03.07.2018, from http://www.scimagojr.com
Shouldn’t the headline read: Open Access journals have a clear citation advantage over commercially published subscription journals. As the graph shows, the major society published journals lead the ‘pack’.