Reproducibility and research integrity report published

Crisis? What crisis? The House of Commons Science, Innovation and Technology Committee has taken a look at reproducibility in academia

James Coe is Associate Editor for research and innovation at Wonkhe, and a partner at Counterculture

The House of Commons Science, Innovation and Technology Committee report Reproducibility and Research Integrity is out and it makes the case for the value of the hard grind of confirmation, replication, and transparency.

The report starts from the premise that given the increase in funding for science and research, it should follow that there is increased attention paid to the probity of the ways in which research is produced and disseminated.

There is an entire debate on what reproducibility means with different views on the use of the term and its scope. The committee settles on a common sense view that this term should refer to the quality and transparency of research. In essence, whether there is sufficient openness in the way research is produced to allow for its effective interrogation by peers.

The report stops short of stating that there is a reproducibility crisis in academia. On the basis that it is too blunt and grand a statement to characterise the various challenges that afflict the wider research ecosystem. This is not to say the committee does not believe there are challenges but that to consider it a crisis frames separate but linked issues as systematic, impossible to resolve, or abstract, when none of these things may be true.

Instead, the report points toward some contingent factors that inhibit the reproducibility of research. There are some cases of bad actors that are engaged in academic misconduct where the committee recommends clearer roles and responsibilities for infrastructure bodies in taking action. The committee believes the pressure to publish can prevent the timely sharing of material in such a manner that makes research replication easy and accessible. And there is a skills question on the extent to which research teams are statistically literate enough to engage with issues of reproducibility, and the extent to which universities are training research staff with a view to enhancing the interrogation of research methods.

Elsewhere there are more structural issues with the research ecosystem. There is a question of whether journals prioritise novel research and in doing so encourage the search for striking findings above confirmatory studies. There are also considerations of the link between longer contracts and the opportunities to spend time on confirmatory work. And, there is a nod to the need to consider how transparency can be integrated into measures like REF and enhanced through funders like UKRI.

Leave a Reply