Every publication on this page passes through the same review before it is posted. The framework below is the standard we hold ourselves to, and the one we will be judged against if we fall short of it.
Research Ethics Committee
The Committee reviews the design of any study before data collection begins, and the draft of any publication before it is released. It has the authority to stop a publication. Where a study is reviewed internally rather than independently, the publication says so on its card — "Internal review" is a weaker signal than independent peer review, and hiding that distinction would be the dishonest option.
Informed consent
Research involving human participants requires informed consent in a language the participant reads fluently, with a plain explanation of what will be collected, how long it will be kept, and how to withdraw. Participants under 18 require guardian consent alongside their own assent. Withdrawal removes the individual record from future analysis; where a figure has already been published, we note the change rather than silently restating it.
Free, Prior, and Informed Consent (FPIC)
Research touching mining-affected or indigenous communities follows FPIC principles. Consent must be sought free of coercion, before the research begins rather than after, informed by a full account of the purpose and the funders, and given by the community's own representative process rather than by whoever is most convenient to reach. A community may decline. If it declines, the research does not proceed and we do not substitute second-hand accounts to fill the gap.
Data sources and corrections
Every publication lists its sources. Where a figure comes from our own programme data we say so, because self-reported programme outcomes deserve more scepticism than independent statistics. If a published figure turns out to be wrong, we correct it in place with a dated note explaining what changed. Corrections are never silent.
Conflicts of interest
Funders and partners have no editorial input into research findings. Where a publication concerns a sector in which a partner operates, the relationship is disclosed in the publication itself.