Standards in sociology: Agreeing on principles,disagreeing on ambition

Given that the EAS was founded explicitly to help “turn the tide,” and that the authors acknowledge the tide has not been turned in 25 years, is this really a success story? I will argue that a more demanding stance is warranted, and I do so along two lines: open science as a case in point, and doctoral programmes in sociology as a parallel domain where the same softness is visible.

The open science section begins with Merton’s norm of organised scepticism and uses it to ground a commitment to transparency and replicability. The authors follow this logic into strong normative territory: journals, professional organisations, and universities “should demand sociology to be an open science,” and institutions “must establish the incentives and conditions necessary to sustain these practices.” But the demand is left without operational content. What, specifically, should journals require as a condition of publication? What does being “aligned with these evolving standards” mean in practice, and who determines when that alignment has been achieved?

I therefore argue for a harder version of the open science principle than Aksoy et al. (2026) do. European sociology journals should require, as a condition of publication, that authors document their research materials to the extent that law and ethics permit. A key distinction is needed here: sharing and documenting are not the same thing. Many European sociologists work with data that cannot be shared due to data protection legislation. This is a real constraint, but it does not exempt the researcher from documentation. A researcher working with population registry data can provide a reference to the exact data extract, sufficient to allow independent replication of the data request. From that point, replication access becomes the data provider’s problem, not the researcher’s. The same applies to qualitative data: interview transcripts that cannot be anonymised cannot be shared, but the sampling strategy, coding scheme, and analytic protocol can and should be documented. The researcher’s responsibility for documentation and openness extends to the limits of legislation and research ethics — but it genuinely ends there, not before.

This version of the requirement is operationally tractable and methodologically neutral. It applies equally to survey researchers, ethnographers, and experimentalists. It also protects researchers: if researchers have documented everything they legally and ethically could, they are not vulnerable to reproducibility challenges they cannot meet. The authors note that open science “cannot rest solely on individual scholars” — but documentation itself is an individual act that requires no institutional permission to perform.

EAS has no house journal, but EAS fellows sit on editorial boards of the major European sociology journals — the very journals the statement cites approvingly as having “strengthened their requirements for authors to make their research material openly available.” Strengthened is not sufficient. EAS could formally call on those journals to go further: to make thorough documentation a condition of publication rather than an encouraged practice, and to state explicitly what that documentation must include. This would translate the statement’s own language — institutions “must establish the incentives and conditions necessary” — into something operational and attributable.

The same logic applies to the EAS prizes. The Distinguished Publication Prize and the Raymond Boudon Award for Early Career Achievement recognise outstanding work in the field. Documentation is not an optional extra but a basic condition of scientific quality: research that cannot be independently assessed should not be regarded as outstanding, regardless of its other merits. Making thorough documentation an explicit nomination criterion for both prizes would signal what the discipline values at no institutional cost.

Aksoy et al. note approvingly that doctoral training in sociology has improved considerably across Europe, and cite a number of examples. The question is whether good examples scattered across the continent constitute a disciplinary standard or merely demonstrate that excellence is possible. If the statement in 2026 is that the best programmes are better than they were, that is progress, but it is not the same as saying that standards have risen.

Training in rigorous research methods — in documentation, in transparent analysis, in the relationship between theoretical claims and empirical tests — should be a structural expectation of doctoral programmes in sociology, not an optional feature of the best ones. EAS has leverage here through its connections to graduate programmes and accreditation discussions. A concrete step would be to call for a dedicated module on transparency, reproducibility, and documentation across European doctoral programmes. The statement does not take such steps.

The authors are right to argue that causal identification should be balanced against substantively important questions, and right to note that sociological approaches to causality require more than identifying a causal link. I would add that rigour is not the exclusive property of causal designs. A theoretical argument is rigorous when it derives testable predictions from stated premises. Qualitative work is rigorous when sampling and interpretation are systematically documented and claims are appropriately bounded. What rigour is not agnostic about is transparency, logical coherence, and the relationship between evidence and claim. On this, I am sure Aksoy et al. agree — I simply argue for making it a requirement rather than an aspiration.