The European Union is set to miss its target of having all scientific research freely available by 2020, as progress towards open access hits a ¡°plateau¡± because of deeper problems in how research is assessed.

Sixty to 70 per cent of universities reported that less than a fifth of their researchers¡¯ peer-reviewed publications are freely available, depending on the type of open access, according to a survey of more than 300 members of the European University Association.

Only one in 10 universities said that more than 40 per cent of their research was published as ¡°gold¡± open access, where there is no delay making it public.

In 2016, EU member states¡¯ science and industry ministers, supported by the European Commission, backed a move to full open access in just four years.

ºÚÁϳԹÏÍø

This??asks members about papers published in 2013, 2014 and 2015, so may not capture all progress made to date.?But?it still concludes that to hit the 2020?target ¡°will require greater engagement by all of the relevant stakeholders¡±.

This chimes with an?EU??released at the end of February?which concludes that?¡°100 per cent full open access in 2020 is realistically not achievable in the majority of European countries participating in this exercise in the foreseeable future¡±.

ºÚÁϳԹÏÍø

Lidia Borrell-Damian, the EUA¡¯s director for research and innovation, said that ¡°unfortunately [full open access] is very difficult to achieve¡± and that ¡°we have reached a plateau in which it¡¯s very difficult to move forward¡±.

Open access had taken off in some subjects ¨C like physics, where the?open access?arXiv?pre-print?platform is widely used ¨C?in which?¡°traditional indicators¡± of journal prestige such as impact factors and other measures of citations were ¡°less relevant¡±, she?explained.

But in most disciplines, these measures were still crucial for burnishing researchers¡¯ career prospects, she added, making it difficult for authors to switch to less prestigious, lower impact factor open access journals. ¡°As long as it [research assessment] is based on these proxy indicators, it¡¯s impossible to change the game,¡± Dr Borrell-Damian said.

This is backed up by the survey findings. The biggest barrier to publishing in an open access repository was the ¡°high priority given to publishing in conventional journals¡±, a hindrance cited by more than eight in 10 universities. ¡°Concerns about the quality of open access publications¡± were also mentioned by nearly 70 per cent of respondents.

ºÚÁϳԹÏÍø

In some disciplines, to publish open access, ¡°you have to be a believer or activist¡± and it comes ¡°at the risk of damaging your own career¡±, Dr Borrell-Damian said.

Echoing a long-standing concern in science, she argued that ¡°we need a whole new system¡± of research assessment that does not rely so heavily on citations and impact factors.

The EU¡¯s flagship Horizon 2020 funding scheme requires grant recipients to publish their findings openly, but this was a far from universal policy for national funding bodies, she added.

A spokesman for the EU Council acknowledged that ¡°more efforts will be needed overall to accelerate progress towards full open access for all scientific publications¡±.

ºÚÁϳԹÏÍø

david.matthews@timeshighereducation.com