An approach to personality description originated by Raymond B. Cattell, based on an analysis of characteristics into surface traits and source traits.A trait is a reasonably consistent personality characteristic which may be used to distinguish an individual from others. Trait names are actually a form of shorthand, each implying a certain pattern of behavior—for instance, if we say a girl is shy, we can predict that she will not show a normal interest in the opposite sex for someone of her age, will be reluctant to speak out in a group or talk about her personal life, etc. Trait names are therefore comprehensive ways of characterizing people. An unabridged dictionary contains about four thousand trait names (forthrightness, humility, boisterousness, etc.), as well as about eighteen thousand adjectives used to describe how people think, perceive, and act (Allport and Odbert, 1936).Investigators concerned with describing personality in terms of traits devise tests, or scales, to determine the degree to which any individual possesses a particular trait or set of traits. Their major problem is to decide which of the huge number of traits to select. In 1946 Cattell attempted to solve this problem by first eliminating synonymous or nearly synonymous items, and bringing together those which appeared to be closely related. This procedure reduced the list to 171 traits, which he called “the personality sphere” since he believed they were all we need to make a complete description of personality. Next, he brought together in a single cluster traits which were highly correlated in the same individual. This reduced the number to thirty-five clusters —for example, warmth versus aloofness includes responsiveness versus unresponsiveness, affectionate versus cold, and even-tempered versus sensitive.Cattell called these thirty-five traits “surface traits,” since they were found by correlational analysis, and this process merely shows that they were discovered together, somewhat as a group of symptoms forms a syndrome in medical diagnosis. TO cut the number down still further, he had judges rate groups of subjects on each of the traits, and then made a factor analysis of the results. In this way he uncovered twelve basic factors, or “source traits,” from which it was possible to predict a person’s score on each of the thirty-five surface traits. Each of them is a summary label for a group of traits—for instance, a person who has the source trait “cyclothymia” will be emotionally expressive and changeable, while “schiz- othymia” implies that an individual is anxious, close-mouthed, and reserved, and “bohemian unconcernedness” means that the person is eccentric, unconventional, and subject to hysterical upsets.The source traits are conceived as underlying features which combine in various ways to determine the many surface traits. Cattell believed they stem from basic, unitary influences which he described as “physiological factors, temperament factors, degrees of dynamic integration, and exposure to social institutions.” These influences, however, were never spelled out. Furthermore, although the traits were obtained in a variety of ways—by objective tests, self- ratings, life record scores—they were bound to be somewhat subjective. For this reason some of the trait descriptions appear forced, and the list as a whole can only be considered tentative (Anastasi, 1961).Cattell and his associates developed a number of personality inventories based on these investigations. The most comprehensive, the Sixteen Personality Factor Questionnaire, for Ages 16 and Over, yields sixteen scores on such traits as aloof versus warm, submissive versus dominant, glum versus enthusiastic, emotional versus calm. This instrument has not proved reliable because of the shortness of the subtests. Similar inventories have been constructed for ages eight to twelve and twelve to eighteen, as well as separate instruments for such personality characteristics as anxiety, introversion-extra- version, and neuroticism. The latter are the IPAT Anxiety Scale, the Contact Personality Factor Test, and the IPAT Neurotic Personality Factor Test. They are all considered experimental since they have not been sufficiently standardized and validated. Trait analysis has been criticized on a number of scores. It is essentially an atomistic approach to personality. It yields specific ratings on the traits that are emphasized, but these ratings tend to be confined to specific persons in specific circumstances, and cannot be generalized to other persons. Finally, the method can be used to summarize specific responses, but it cannot give a picture of the personality as a whole, or uncover basic principles of personality.