As more funders implement or consider mandates, we were curious whether the data show any connection between funders’ open access policies and open access uptake.

Introduction

The Registry of Open Access Repository Mandates and Policies (ROARMAP) is a searchable international registry charting the growth of open access mandates and policies adopted by universities, research institutions, and research funders. It tracks the policies requiring or requesting researchers to provide open access to their peer-reviewed research article output. It focuses on policies about use of repositories, but also captures information about OA publishing.

By comparing the number of policies in ROARMAP with proportion of output that is open access, we can see whether a relationship exists between policy activity and OA uptake. We define OA to include articles that are both free to read and reuse, in all journal types. We exclude public access (“bronze”). As we want to examine market activity, we also exclude repository-only versions of articles.

Policies and Uptake

Figure 1, below, shows how numbers of policies and the proportion of output that is open access track over time. The blue bars show the cumulative number of policies (left axis). The orange line shows the proportion of articles that are OA (right axis).

Sources: ROARMAP, Crossref, Unpaywall, Delta Think Analysis.

Clearly both are growing and there is a strong correlation overall. However, the two sets of numbers follow slightly different trajectories:

• The numbers of policies appear to follow a classic S-shaped curve. A burst of steep growth from around 2009 to 2018 is flattening off. For example, in 2013 and 2014 growth was around 20% per year. Last year it was 1%.
• The proportion of OA articles appear to follow more of a shallow hockey stick. Its initial growth curve has given way to a straighter line.

Although both curves are broadly similar, it seems that they diverge in the detail. The number of OA policies grew faster than OA uptake around 2001 to 2017. But we have since seen a turnaround. The proportion of OA continues to increase apace, even though the numbers of policies have almost reached a steady state.

Growth can be hard to judge from absolute numbers. For example, the straight line above actually points to slowing growth. (Constant growth compounds and would show as an exponential curve.) Figure 2 below compares the annual changes in both measures.

Again, we see a superficial similarity but with differences in the detail.

• With only a few exceptions, growth in proportion of OA output has consistently out-paced growth in OA policies.
• We can see that although growth in OA adoption is slowing, growth in policies is falling faster.
• Note: Slight changes in timing across the calendar year-end can cause the jagged charts. For example, if output shifts across the year-end (perhaps due to the way the data were captured that year), then one year will see unusually low growth, the following unusually high growth (as the previous year’s papers are included), and the year after that will see low growth (as the data comes back on to trend from the high previous year).

Conclusion

Are policies driving OA adoption?

At first glance: Yes. Clearly there is a strong overall correlation between numbers of policies and OA uptake.

However, correlation is not causation. Policies do not necessarily mandate actions. Different funders and institutions may apply different incentives to researchers. Policies take time to take effect – as we see with Plan S. And, while the longest-standing and most robust policies are likely to see highest compliance, compliance rates are highly variable.

Whatever the correlation with policy numbers, perhaps the OA market has taken on a life of its own. It continues to grow regardless of policy numbers and – in many cases – of policy strength. Anecdotally, we can say that many publishers view OA as “the direction of travel” and are increasing their OA options often in advance of mandates.

We have focused here on “Gold” OA. It is worth noting what we haven’t included: Green. ROARMAP data looks primarily at manuscript deposition, as do flagship policies such as the NIH’s public access mandate. If authors can comply through the green route, then why bother with variants on gold OA at all? Yet, our underlying data show repository-only articles hovering at a steady state of around 5% of annual output; with articles in gold fully OA journals driving the increase in OA uptake.

Perhaps this is a sign that publishers are having an influence too. Many may facilitate manuscript deposition, but would prefer a gold route to OA. The numbers suggest they are getting their wish.

This article is © 2021 Delta Think, Inc. It is published under a Creative Commons Attribution-NonCommercial 4.0 International License. Please do get in touch if you want to use it in other contexts – we’re usually pretty accommodating.

TOP HEADLINES

Springer Nature signs joint letter with 50+ other publishers – February 3, 2021

“The undersigned share with cOAlition S the goal to expand Open Research and are committed to supporting cOAlition S-funded researchers through the various paths provided for Open Access. However, we are unable to support one route to compliance offered by Plan S, the “Rights Retention Strategy”, in its current form.”

Portugal’s national funding agency for science, research and technology joins cOAlition S – January 26, 2021

“cOAlition S is pleased to announce that the Foundation for Science and Technology of Portugal (FCT) is the latest national funding agency for science, research, and technology to join the coalition and demonstrate its commitment to the realisation of full and immediate Open Access.”

PLOS and the Big Ten Academic Alliance Announce Publishing Deal – January 21, 2021

“The Big Ten Academic Alliance (BTAA) and the Public Library of Science (PLOS) today announced an agreement for BTAA members to participate in PLOS’ Community Action Publishing (CAP) program. The agreement ensures time to test and evaluate this new community-based model and allows BTAA researchers to publish in PLOS Biology and PLOS Medicine without incurring fees.”

Focused on author choice & research quality, AAAS announces new OA policy – January 15, 2021

“The American Association for the Advancement of Science (AAAS), the nonprofit publisher of the six Science family journals, is announcing an update to its terms of open access publication. On a trial basis, AAAS will allow authors funded by cOAlition S organizations to place a CC BY or a CC BY-ND license on their accepted manuscripts.”

New Open Access Platform for Mathematics launched online – January 13, 2021

“With its funding decision in 2019, the Joint Science Conference of the Federal Government and the Länder had paved the way for zbMATH Open. Now it is done: In an elaborate work process, the information service was transformed into an open access platform. This means that much of the data can be used freely for research purposes and for linking to other non-commercial services.”

OA JOURNAL LAUNCHES