Contemporary North-South research collaborations are fraught with power relations originating in colonialism. Debates about research ethics have tended to turn around the "procedural ethics" formal model and the "everyday ethics" practical model. We build on that to suggest a second debate that scrutinises ethics and power…

Contends that the high quality of the probability sampling used by the International Association for the Evaluation of Educational Achievement (IEA) is due, in large part, to procedures developed by IEA's first statistical consultant, Gilbert Peaker. Concludes that the Peaker process is a first-class sample design. (CFR)