Open science has come to be seen as virtuous by default, especially when it comes to the agencies funding social science research. But Professor Rosalind Edwards asks whether the rationales behind seeing open science as something to be desired are flawed, especially when it comes to qualitative research.

The drive to promote open science across the social sciences and humanities has become an accepted part of the research landscape in the UK, with publicly funded agencies in particular, but also other research funders placing requirements on researchers to make their original data accessible to the wider research community. For example, UKRI refers to ensuring open research for projects it funds as supporting a healthy research and innovation culture.

Their website states that open research is integral to UKRI’s mission to deliver economic and social benefits, for accessible, transparent, reproducible and cooperative research, which are pretty standard rationales across bodies addressing the issue internationally.   

As someone who’s generated qualitative longitudinal data for reuse and undertaken secondary analysis of data sets, I’m committed to archiving data as important, not only for immediate re-enlightenment but as historical records far into the future. But I want to raise a question concerning the justification currently put forward for compelling this practice:  

“If open science is the answer, what’s the question?”

Questioning the value of the public value metric

This blog has previously hosted about the specifics of prevailing conceptions of open science for qualitative research in particular, looking at the practice and ethics of open science for qualitative data, the affordances of qualitative data reuse, and urging humanities scholars to adopt the replication agenda. But I want to raise a different issue, one that relates to all forms of research data: quantitative, qualitative, visual, etc. My challenge concerns the notion of “public value” as a rationale. 

Viewed through the neoliberal economic calculation lens, paying for the process – for the research – means owning the output: the data.

A core element of the argument that’s put forward for why there should be open data access is that the research is publicly funded. Public funding should mean open research to maximise public value, according to this rationale. This justification sounds so reasonable it’s hard to take issue with. But the reason for that is that neoliberal logic infuses our understanding, ordering our social reality, regulating our practice and beliefs. Viewed through the neoliberal economic calculation lens, paying for the process – for the research – means owning the output: the data.   

It can also mean, for the good of the economy, feeding that data through into supporting AI development, training AI models etc., with economic value extracted in social research from making private lives “public” for private companies. This leaves aside other ways we might ask questions about and consider what’s of public value. The environmental costs of the open science agenda, creating and sustaining the digital infrastructure and resources necessary for long term preservation and processing of data, might, for example, give us pause for thought. We might also take a more humanistic and socially just perspective when thinking about open science and how we pursue it, that would make available other justifications and rationales for the accessing and sharing of data. 

Another issue with the rationale that public funding means open science in order to maximise public value is that it hides the labour of production. Social research data of whatever sort overwhelmingly comprises research participants’ lives and entirely researchers’ intellects. But the lives of people and their participation in research about those lives that is a foundation of the data creation is concealed, while highlighting the source of funding. In this, the intellectual labour of the original researchers who generate, analyse and make available that research and its data is also obscured. Both participants and researchers are reduced to data generators for extracting value. We might want to value the labour of production in alternative justifications for open science. In turn, this may lead us to think about inequalities, power and social justice in who lays a claim to data.   

Alternative perspectives on open science

One example of an alternative perspective is the international Indigenous data sovereignty movement. This refers to Indigenous peoples controlling their own data rather than institutional bodies, and has a more collective notion of ownership and benefit sharing from secondary analysis. The CARE Principles for Indigenous Data Governance, for instance, center fundamental notions of collective ownership, stewardship, and benefit grounded in respectful values and ethics that go beyond regulation. Linked to this, decolonizing approaches to research raise important questions equally relevant to open science: who owns the research? Who initiates it? In whose interests is it conducted? Who controls it? And so on.

Qualitative research and open science

Open science is often seen as a way to support the reproducibility of results, and this brings me to another element of the open science question: is open science valuable when we are dealing with qualitative research? The justification that open data access works in the service of rigour is grounded in the idea that other researchers can reproduce the analysis of the original data as a form of quality control. They can verify the accuracy and reliability of the research findings. Or they can reproduce or replicate the study, usually with a new but comparative sample. 

The term “sample” is defined in the Oxford and Merriam dictionaries, respectively, as ‘a small part or quantity intended to show what the whole is like’ and as related to the word representative, and “a representative part from a larger whole or group” as well as a “finite part of a statistical population whose properties are studied to gain information about the whole”.  In qualitative research, even if we merge several qualitative data studies together, this doesn’t mean that multiplying the number of small “unrepresentative” (in quantitative terms) samples will add up to a representative sample. It won’t. What it will do is to strengthen theoretical generalisations in the form of claims to understanding how social processes work. 

The idea of data sharing for replication and verification is based on the notion that data is free-standing and that anyone correctly analysing the same data set will come to the same results.

Indeed, it’s odd to talk of samples, or preferably cases, in qualitative social research without a context since whom and where you research depends on the nature and design of your study. So the term “sample” is inappropriate for qualitative research: firstly because the main focus of data generation is on process rather than a numerical end point, and secondly, because the notion of sample invokes the idea of social phenomena as somehow independent of the researcher’s account of them, to be picked – that is sampled – from a pre-existing social context. The idea of data sharing for replication and verification is based on the notion that data is free-standing and that anyone correctly analysing the same data set will come to the same results.

I’m not sure this is the case for quantitative data in the social sciences, but I’m certain that it isn’t the case for qualitative research. For example, given a single qualitative data set, a researcher with a psychosocial analytic approach may come up with quite different research findings to one who approaches the data from a critical realist or interpretive perspective. Neither will be “wrong”; they are providing different angles on the data.

That isn’t necessarily an argument against data sharing of course. It’s just that, as the case of using open science as a way of justifying the value of public funding, the rationale for requiring data sharing needs to start from a different basis. Qualitative social research data is generated in and through context, not free-standing from it. It is reflexivity and transparency about this that is rigour, transparency about epistemological position, theoretical lens, methods and research processes.

Where we are currently with justifications for open science however, shuts down on humanistic approaches and social justice, and reveals the driving force of an economic rationality that obscures participants, researchers and epistemologies.

The content generated on this blog is for information purposes only. This Article gives the views and opinions of the authors and does not reflect the views and opinions of the Impact of Social Science blog (the blog), nor of the London School of Economics and Political Science. Please review our comments policy if you have any concerns on posting a comment below.

Image Credit: StunningArt on Shutterstock.