The “gold” model of open access has been championed most enthusiastically not in Europe, the epicentre of the Plan S scheme to make all journal publications freely available, but in less research-intensive regions of the world.
And a new analysis suggests that schemes such as Plan S have a muted impact unless authorities take active steps to monitor and enforce compliance.
Study leader Karl Huang, a statistician at Curtin University in Perth, said the findings would surprise some people. “Our analysis is at an institutional level, [so] it’s difficult to say how whole regions perform.
“But Latin American and African universities have popped up as some of the top performers – better than institutions in Europe, where they’re pushing really hard on Plan S.”
The study, published in the journal eLife, evaluated the effectiveness of open-access strategies around the world. Rather than restricting the analysis to a single bibliometric source, the team mined services including Microsoft Academic, Web of Science, Scopus, the non-profit linking agency Crossref, the Global Research Identifier Database (Grid) and the open database Unpaywall.
Dr Huang said this helped to broaden coverage of research in some regions and languages while sidestepping differences in the ability to access and integrate data. “We’re combining multiple data sources to hopefully give us a more comprehensive view.”
He attributed Latin America’s high levels of gold open access – a pay-to-publish rather than pay-to-read model – to the ubiquity of the Scientific Electronic Library Online (SciELO) project, which supported open access through the provision of publishing infrastructure.
“This suggests that increasing levels of open-access publishing through article-processing charges is potentially expensive compared to the costs of providing infrastructure,” the paper says.
Dr Huang said Africa’s high levels of gold access reflected its reliance on funding from organisations such as the Bill and Melinda Gates Foundation and the Wellcome Trust, “where there are mandates on open access”.
He said universities in some Asian countries also achieved high levels of open access, highlighting Indonesia’s “big push” for universities to host their own journals.
In European and North American universities, by comparison, few institutions offered gold open access to more than 40 per cent of their research publications – a level that changed little between 2014 and 2018.
The analysis found an emphasis on repository-based open access in the UK, in line with Research Excellence Framework requirements, but access levels had tailed off as funding declined.
Compliance action plays a pivotal role in encouraging adherence to open-access policies, the analysis found. “In the Netherlands and the UK, in particular, but also in the US – where funder policies have moved from encouragement, to mandates, to monitoring with sanctions for non-compliance – there are substantial shifts in overall levels of open access,” the paper says.
“In countries where policy remains effectively at the level of a recommendation, such as Australia, levels of open access lag significantly.” This could change over the next two years due to “recent increases in reporting requirements by Australian funders”.
Dr Huang acknowledged limitations in the analysis. “Even though we are combining multiple data sources, we’re probably still missing a big part of disciplines like humanities, arts and social sciences and African output which is not necessarily captured by these databases,” he said. “There’s still a lot of work to be done.”
