Chapter 17: Cattell: Scientific Approach to Trait Theory

Part 2: Basic Concepts of Cattell’s Theory

Cattell studied a variety of personality types and personality traits.  Of particular interest to Cattell was how to assess personality, and his work is heavily influenced by the systematic collection of scientific data.  This is quite different than many of the psychodynamic and humanistic theorists, who based their theories on clinical observation, but it is similar to the learning theorists, who also value careful, objective observation and the collection of scientific data.  Neither approach is inherently better, since they each serve a different purpose. Cattell’s approach, however, has had a dramatic effect on psychological testing.

Types and Traits

A psychological type refers to a broader description of personality than a psychological trait, and is often associated with abnormal psychology.  According to Cattell, a type can only be understood in terms of personality traits. For example, a villain is a type based on a pattern of associated traits such as immorality, cruelty, and disregard for the law and the rights of others.  Cattell considered types to fall into one of five principal categories: temperamental characteristics, interests and character, abilities, disposition, and disintegration and disease processes.  As further examples, and in accordance with Cattell’s type categories, we can include the ancient personality types of Hippocrates (sanguine, choleric, melancholic, phlegmatic), the oral-erotic and anal-erotic types of Sigmund Freud, musical vs. mathematical geniuses, unrestrained vs. restrained personalities, and various neurotic and psychotic syndromes (Cattell, 1946, 1950a,b, 1965).

Cattell believed that clinical psychologists always took personality traits for granted, but focused their attention on the patterns of traits that defined clinical syndromes (or types).  However, if one wishes to conduct a thorough description and measurement of personality, traits must be the target of that investigation. Thus, Cattell focused his attention on the details of understanding and describing traits.  He agreed with Allport’s description of individual vs. common traits, though he preferred the use of the term unique traits to describe the former.  Cattell described a trait as a collection of reactions or responses bound by some sort of unity, thus allowing the responses to be covered by one term and treated similarly in most situations.  The challenge lies in identifying the nature of the unity, which has been done in different ways throughout the history of studying personality. Thus, a trait guides behavior in a specific direction, by connecting all aspects of that trait into a unit (whether the process is directed outward, a response, or the result of external stimuli, a reaction).  Since an understanding of an individual’s traits would allow us to predict the nature of such responses or reactions, Cattell offered a rather simple definition of personality: “Personality is that which permits a prediction of what a person will do in a given situation” (pg. 2; Cattell, 1950b).

According to Cattell, traits and types are not fundamentally different, but rather opposite extremes of the same statistical measures.  The fundamental, underlying traits are known as source traits.  Source traits often combine and/or interact in ways that appear, on the surface, to indicate a single trait.  For example, in the area of abilities, a unitary intelligence shows itself in good academic performance, such a child who does well in school.  Of course, children who do well in school typically do well in most areas, such as math, English, social studies, etc. What may now appear to be a type, a “good student,” can also be described as a surface trait (Cattell, 1950b).  As useful as surface traits, or types, may be descriptively, in order to truly understand personality, one must address the source traits.  First, however, they must be identified.

Source Traits and Factor Analysis

Cattell used the factor-analytic technique to identify sixteen source traits.  He often uses the terms source trait and factor interchangeably.  Factor analysis is a statistical technique that determines a number of factors, or clusters, based on the intercorrelation between a number of individual elements.  Cattell considered factor analysis to be a radical departure from the personality research that preceded his, because it is not based on an arbitrary choice as to which variables are the most important.  Instead, the factor-analytic technique determines the relevant variables, based on the available data:

…the trouble with measuring traits is that there are too many of them!…The tendency in the past has been for a psychologist to fancy some particular trait, such as ‘authoritarianism’, ‘extraversion’, ‘flexibility-vs-rigidity’, ‘intolerance of ambiguity’, etc., and to concentrate on its relations to all kinds of things…individual psychologists lead to a system which tries to handle at least as many traits as there are psychologists! (pg. 55; Cattell, 1965)

When Cattell applied factor analysis to the list of words identified by Allport and Odbert, he identified 16 personality factors, more or less.  The reason for saying more or less is that any statistical technique is subject to known probabilities of error. Thus, Cattell considered his sixteen factors to be only an estimate of the number of source traits (Cattell, 1952).  As potential source traits were identified that Cattell found difficult to put into words, he assigned them a Universal Index (U.I.) number so that they could be kept for consideration until they could be studied and explained. Cattell identified as many as forty-two personality factors (see Cattell, 1957).  By 1965, when Cattell wrote The Scientific Analysis of Personality, he had included three additional factors to his primary list, giving him nineteen personality factors, and kept thirteen of the remaining factors on his list as yet to be confirmed (though each one had a tentative name).

In the late 1940s, Cattell and his colleagues developed the Sixteen Personality Factor Questionnaire (commonly known as the 16-PF), based on the 15 factors they considered best established by their data, plus general intelligence as the sixteenth factor (see Cattell, 1956).  The sixteen factors are described in the table below. In a very interesting chapter written by Heather Cattell (Cattell’s third wife), the 16-PF profiles are presented, and compared, for a married couple in which the husband was undergoing therapy with Heather Cattell (see H. Cattell, 1986).  She described how the profiles offer insight into the problems occurring for Mr. A (as the husband is identified in the chapter), both in his personal life and in his marriage. Although the marriage ended in divorce, a subsequent follow-up found both Mr. A and Mrs. A seemingly doing well in their separate lives.

The Sixteen Personality Factor Questionnaire Dimensions
Factor Low Score Description High Score Description
A. Warmth Reserved – detached, critical, aloof, stiff Outgoing– warmhearted, easy-going, participating
B. Intellect Less Intelligent – concrete-thinking More Intelligent – abstract-thinking, bright
C. Emotional Stability Affected By Feelings – emotionally less stable, easily upset, changeable Emotionally Stable – mature, faces reality, calm
E. Aggressiveness Humble – mild, easily led, docile, accommodating Assertive – aggressive, stubborn, competitive
F. Liveliness Sober – taciturn, serious Happy-Go-Lucky – enthusiastic
G. Dutifulness Expedient – disregards rules Conscientious – persistent, moralistic, staid
H. Social Assertiveness Shy – timid, threat-sensitive Venturesome – uninhibited, socially bold
I.Sensitivity Tough-Minded – self-reliant, realistic Tender-Minded – sensitive, clinging, overprotected
L. Paranoia Trusting – accepting conditions Suspicious – hard to fool
M. Abstractness Practical – “down-to-earth” concerns Imaginative – bohemian, absent-minded
N. Introversion Forthright – unpretentious, genuine but socially clumsy Astute – polished, socially aware
O. Anxiety Self-Assured – placid, secure, complacent, serene Apprehensive – self-reproaching, insecure, worrying, troubled
Q1. Open Mindedness Conservative – respecting traditional ideas Experimenting – liberal, free-thinking
Q2. Independence Group-Dependent – a “joiner” and sound follower Self-Sufficient – resourceful, prefers own decisions
Q3. Perfectionism Undisciplined Self-Conflict – lax, follows own urges, careless of social rules Controlled – exacting will power, socially precise, compulsive
Q4. Tension Relaxed – tranquil, unfrustrated, composed Tense – frustrated, driven, overwrought

The Types of Data Used in the Assessment of Personality

In a rather obvious statement, Cattell noted that in order for a psychologist to study correlations, there must be two measures available to be correlated.  The systematic measure of various aspects of the mind, including personality, has led to the development of a specific branch of psychology known as psychometry.  In order for a psychometrist to get a complete and unbiased measure of personality, they must have a concept of the individual’s total behavior, what Cattell called the personality sphere.  Cattell believed this could best be accomplished by taking a sample 24-hour period in the person’s life and collecting three types of data: measures of the individual’s “life-record,” or L-data; information provided by questionnaires, or Q-data; and data on their personality structure provided by objective tests, or T-data (Cattell, 1965).

L-data deals with the individual’s actual everyday situations.  Ideally, L-data can be collected without the need for the judgment of a trained psychometrist.  Examples of specific behaviors include things such as their grades in school, the number of automobile accidents a person has had, the number of times they have been arrested by the police, how many organizations they belong to, etc.  Sometimes these data are not so easy to obtain, and must be gathered from someone who knows the person well. For example, we may ask friends or family members to rate the person in terms of how sociable they are in school, how emotionally stable they are when playing sports, or how responsible they are (Cattell, 1965).

Q-data is obtained by having the person fill out a questionnaire, such as the information sheet you fill out when waiting to see a doctor for a medical exam.  Unfortunately, these data are subject to a number of problems, such as distortions due to poor self-knowledge, delusions about the self, or the deliberate intention to fake the outcome of the questionnaire.  Therefore, it is very important that a psychologist choose the right words when developing a questionnaire:

Although a questionnaire looks like a simple series of questions to which a person underlines a brief answer, such as ‘yes’, ‘no’, ‘generally’, [sic] etc., actually a great deal of art enters into the psychologist’s choice of words, the direction of the question, the use of adjectives to ensure that all alternatives are well used, and so on. (pg. 61; Cattell, 1965)

As noted above, T-data is obtained from objective tests.  According to Cattell, questionnaires may seem objective, since their scoring is objective, but the process involves having the individual evaluate themselves.  In truly objective tests, the individual’s specific behaviors or thoughts are directly and precisely measured. It is essential that only closed-ended questions are used, such as multiple choice or Yes-No options.  If open-ended questions are used, such as “How do you feel when you wake up in the morning?” it is possible that two psychologists will interpret the answer quite differently.  If there is a possibility of different interpretations, obviously the test cannot be objective.

In comparing the three types of data, Cattell made some interesting observations regarding L-data.  Although it occurs naturally, measuring it is artificial and somewhat arbitrary. Although it is objective in the sense that it is real behavior, it is neither created nor controlled; it is simply observed.  It is also subject to cultural differences much more so than Q-data and T-data. Of particular concern to Cattell, however, was the commonplace nature of L-data:

Much of the irresponsible theorizing on personality criticized [early in the book] happens to have grown up in the realm of L-data, for this has been the traditional field of observation of the philosopher, the armchair observer, and the clinician, whereas Q- and T-data have been developed by the psychometrist concerned with the more disciplined methods.  L-data is, indeed, the field of behavior that is the common property of everyone…there arises at this point the need for a proper development of measurement techniques particularly as they apply to L-data… (pp. 54-55; Cattell, 1957)

Cattell believed that personality assessment worked best when the psychologist understood a person’s entire personality sphere.  To accomplish this, one needs to measure L-data, Q-data, and T-data. Do these data provide a complete picture of the person? What data do you think might be the most difficult to obtain, and how might that affect the overall personality picture?

Supplemental Materials

Raymond Cattell

This video [47:19] provides a narrated version of the content provided on Cattell on Wikipedia.  Topics covered include his biography, his scientific approach to psychology, multivariate research, factor analysis, personality theory, and his views on race and eugenics.



Text:  Kelland, M. (2017). Personality Theory. OER Commons. Retrieved October 28, 2019, from  Licensed under CC-BY-4.0.

Wikipedia TTS.  (2019, October 4).  Raymond Cattell | Wikipedia audio article.  [Video File]. Retrieved from Standard YouTube License.


Icon for the Creative Commons Attribution 4.0 International License

PSY321 Course Text: Theories of Personality Copyright © by The American Women's College Psychology Department and Michelle McGrath is licensed under a Creative Commons Attribution 4.0 International License, except where otherwise noted.