Rethinking Psychometric Measurement

Abstract

I argued in Chapter 2 that Binet illegitimately conflated the distinct concepts of classification (by level of mental development) and measurement of intelligence. Binet constructed an ordered series of items which he thought tested mental performance and, ignoring his own well-grounded theoretical objections, bluntly declared that he had developed a metric scale for the measurement of intelligence. It need hardly be said that those who adapted Binet’s methods saw no problems with his theory. Within a decade of his death entrepreneurially motivated American psychologists had created a huge mental measurement industry without giving more than a passing thought to the question of whether the measurement of intelligence was possible or not. An enclosed tradition was created in which major textbooks on mental measurement and testing, Anastasi’s (1968) is a recent instance, make no mention of measurement theory, and actually do not discuss the concept at all, but simply take for granted the operational identity between testing and measurement. The use of ‘test’ and ‘measure’ as synonyms is so ingrained that most psychology students never realise how many questions are begged by this habit. Where measurement is discussed the presentation is usually limited to an uncritical exposition of Stevens’s four-fold typology of scale types, together with, perhaps, a few comments on the problematic nature of scientific measurement.

Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.