Summary
The general problem of comparability of standards has engaged the interest of educational researchers, teachers and laymen for many years. The Rasch model is one attempt at solving it which has been widely canvassed recently, notably in connection with the work of the Assessment of Performance Unit. The method is in principle very simple, in the sense that its assumptions and main features can be outlined in a few sentences. But its complexity in conceptual terms is considerable, and the practical details for the user are none too easy. This paper attempts to set down the main features of the method, and the conditions in which its use is appropriate, in a form which is comprehensible to the non‐specialist.