At a recent conference in Innsbruck, held in memory of Bruno De Finetti, attention focused on one of his aphorisms: analysts should think about things. This paper seeks to do precisely that in the context of public examinations in England and Wales. It attempts to think about those quantitative things that are done to marks in the process of assessing candidate scripts. Public examinations are central to our education system: hundreds of thousands of candidates enter them every year. In assessing the candidates performances, the examination boards must absorb, analyse and combine literally millions of marks. To do so they must use many quantitative procedures. What is the validity of these procedures? What is their purpose and do they have their desired effect?
@article{urn:eudml:doc:40524,
title = {Statistical and decision theoretic aspects of examination assessment.},
journal = {Trabajos de Estad\'\i stica},
volume = {4},
year = {1989},
pages = {33-66},
language = {en},
url = {http://dml.mathdoc.fr/item/urn:eudml:doc:40524}
}
French, Simon. Statistical and decision theoretic aspects of examination assessment.. Trabajos de Estadística, Tome 4 (1989) pp. 33-66. http://gdmltest.u-ga.fr/item/urn:eudml:doc:40524/