Psychometrics tries to measure things nobody can see, like intelligence and introversion
The word psychometry was coined in 1842 by an American who used it for supposed paranormal powers. Today it names a serious and argumentative science: working out how to put numbers on hidden qualities such as intelligence or personality, and whether that counts as measurement at all.
Psychometrics deals with latent constructs, traits that cannot be observed directly. Intelligence, introversion, mental disorders and educational attainment are all inferred from how people answer questions on tests and scales, using mathematical models to estimate the underlying level. Its name joins Greek words for soul and measure. Specialists, usually psychologists with advanced training, design and validate tests for universities, testing companies and consultancies.
The field grew from two roots. One ran from Darwin through Francis Galton, often called the father of psychometrics, who was inspired by natural selection to study how people differ and added mental tests to his physical measurements; James McKeen Cattell then coined the phrase mental test. The other came from psychophysics, where Weber looked for the smallest stimulus a sense can detect and Fechner proposed that perceived sensation grows with the logarithm of stimulus strength. Their follower Wilhelm Wundt is credited with founding psychology as a science.
Intelligence testing came first. Alfred Binet and Theodore Simon built an early test in France, which Lewis Terman at Stanford adapted into the Stanford-Binet IQ test. Personality testing followed, although no single theory of personality has won general agreement; the Big Five model and the Myers-Briggs indicator are among the familiar tools. Statistical methods such as factor analysis, advanced by Charles Spearman and L. L. Thurstone, try to uncover the few dimensions hidden beneath many test scores, though users still disagree about how many factors to keep.
A deeper quarrel concerns what measurement means. In 1946 Stanley Smith Stevens defined it loosely as assigning numbers by some rule, partly in response to a British committee of physicists that doubted sensations could be measured. Others insisted psychology should meet the stricter standard of physics, an approach embodied in the Rasch model, which states criteria first and then tests whether data satisfy them.
Source: Psychometrics
