The Measurement of Standards

Abstract

STANDARDS ARE AT THE HEART of current educational debates and, if we examine the contexts in which the term is used, we find that the concept is essentially quantitative. Standards are spoken of as high or low and institutions and individuals are ranked according to the level of achievement. Yet, in spite of the ubiquity of the term, there is no simple measuring instrument we can take to a child or a school and read off the appropriate value. Neither is there any natural unit of measurement as there is with some physical quantities. Indeed, the more closely we analyse the concept, the more elusive it seems to be. We find ourselves in the position of St Augustine who is reported to have said of time ‘What is time? If no one asks I know, but if I have to say what it is to one who asks, I know not’. Much the same could be said of standards. The situation we have described is not uncommon in the social sciences; indeed it could be argued that it is the norm. Many of the key variables which occur in the discourse of the social sciences have the characteristic that they cannot be directly observed. Conservatism, alienation, attitude to abortion are further examples of things which are spoken of as if they were real quantities, yet they are not open to direct observation. The abilities and skills with which educational measurement is concerned are of just this kind. In the language of statistics they are latent variables because they are thought of as underlying, and influencing, the world of observation but are not, themselves, directly accessible to observation. Before we. can begin to provide a conceptual framework for