Abstract
This paper describes the process of designing and administering a sufficiently valid and reliable assessment instrument to measure the progress in attainment of adult learners studying numeracy, and discusses some of the inherent difficulties that were involved. The fieldwork took place during 2003–2005 and involved a sample of 34 teachers and 412 learners aged post‐16. The paper details how the lack of a range of suitable instruments, the heterogeneous nature of adult numeracy teaching, the remit and requirements of the research project and other practical factors, made it difficult to produce an instrument capable of measuring learners’ progress accurately, and hence to produce robust and reliable conclusions, generalisable across the whole sector. Having said this, and despite all the problems faced, the researchers still believe that the test has a sufficient degree of validity for the purposes of the study. The paper ends with some recommendations that may benefit other researchers when it comes to designing future numeracy assessments.