Over the past century, dozens of taxonomies of personality traits have been proposed. Many have been merely lists of traits, often based on the intuitions of personality psychologists. As personality psychologist Robert Hogan observed, "the history of personality theory consists of people who assert that their private demons are public afflictions" (Hogan, 1983). Indeed, two editors of a book on personality traits (Londo & Exner, 1978) expressed despair at the lack of agreement about a taxonomy of traits, so they simply listed the traits alphabetically. Clearly, however, we can develop a firme basis for or ganizing personality traits. Thus, the taxonomies of traits presented in the rest of this chapter are not random samplings from the dozens available. Rather , they represent taxonomies that have solid empirical and theoretical justification
Eysenck's Hierarchical Model of Personality
Of all the taxonomies of personality, the model of Hans Eysenck, born in 1916, is most strongly rooted in biology . Eysenck was raised in Germany at the time when Hitler was rising to power. Eysenck showed an intense dislike for the Nazi regime, so at age 18 he migrated to England. Although intending to study physics, Eysenck lacked the needed prerequisites, so almost by chance he began to study psychology at the
University of London. He received his Ph.D. in 1940 and after World War II became director of the psychology department at the Maudsley Hospital' s new Institute of Psychiatry in London. Eysenck' s subsequent productivity was enormous, with more than 40 books and 700 articles to his name. Hans Eysenck was the most cited living psychologist until he died in 1998.
Eysenck developed a model of personality based on traits that he believed were highly heritable (see Chapter 6) and had a likely psychophysiological foundation. The three main traits that met these criteria, according to Eysenck, were extraversion-introversion (E), neuroticism-emotional stability (N), and psychoti-cism (P). Together, they can be easily remembered by the acronym PEN.
Let us begin by describing these three broad traits. Eysenck conceptualizes each of them as sitting at the top of its own hierarchy , as shown in Figure 3.1. Extraversion, for example, subsumes a lar ge number of narrow traits—sociable, active, lively, venturesome, dominant, and so forth. These narrow traits are all subsumed by the broader trait of extraversion because they all covary suf ficiently with each other to load o the same large factor. Extraverts typically like parties, have many friends, and seem to require having people around them to talk to (Eysenck & Eysenck, 1975). Many extraverts love playing practical jokes on people. They also display a carefree and easy manner . They tend also to have a high activity level.
Introverts, in contrast, like to spend more time alone. They prefer quiet time and pursuits such as reading. Introverts are sometimes seen as aloof and distant, but they often have a small number of intimate friends with whom they share confidences Introverts tend to be more serious than extraverts and to prefer a more moderate pace. They tend to be well organized, and they prefer a routine, predictable lifestyle (Larsen & Kasimatis, 1990).
"You should get out more."
Introverts prefer to spend more time alone than extraverts.
Source: By Richard Jolley. Used by permission of Cartoonstock, Ltd.
"You should get out more."
Introverts prefer to spend more time alone than extraverts.
Source: By Richard Jolley. Used by permission of Cartoonstock, Ltd.
(a) The hierarchical structure of psychoticism (P).
Aggressive I Cold
^Col^J Egocen^ Jmpersona^j Impulsive
(b) The hierarchical structure of extraversion-introversion (E).
(c) The hierarchical structure of neuroticism (N).
Eysenck's hierarchical structure of major personality traits. Each "super-trait" (P, E, and N) occupies the highest level in the hierarchy, representing broad personality traits. Each of these broad traits subsumes more narrower traits in the hierarchy. (a) The hierarchical structure of psychoticism (P); (b) the hierarchical structure of extraversion-introversion (E); (c) the hierarchical structure of neuroticism-emotional stability (N).
The trait of neuroticism (N) consists of a cluster of more specific traits, includ ing anxious, irritable, guilty , lacking self-esteem, tense, shy , and moody . Conceptually, narrow traits such as anxious and irritable might be viewed as very dif ferent from each other. Empirically, however, men and women who feel anxious also tend to get irritated. Thus, factor analysis has proven to be a valuable tool in showing that these two narrow traits are actually linked together , tending to co-occur in people.
The typical high scorer on neuroticism (N) tends to be a worrier . Frequently anxious and depressed, the high-N scorer has trouble sleeping and experiences a wide array of psychosomatic symptoms. Indeed, a national study of 5,847 individuals found that those high on neuroticism tend to be especially prone to the disorders of depression and anxiety (Weinstock & Whisman, 2006). One of the hallmarks of the high-N scorer is overreactivity on the negative emotions. That is, the high-N scorer experiences a greater degree of emotional arousal than the low-N scorer in response to the normal stresses and strains of everyday life. He or she also has more trouble returning to an even keel after such an emotionally arousing event. The low-N scorer , on the other hand, is emotionally stable, even-tempered, calm, and slower to react to stressful events; moreover, such an individual returns to his or her normal self quickly after an upsetting event.
The third lar ge trait in Eysenck' s taxonomy is psychoticism (P). As shown in Figure 3.1, P consists of the constellation of narrower traits that includes aggressive, egocentric, creative, impulsive, lacking empathy , and antisocial. Factor analysis proves valuable in grouping together narrower traits. Factor analyses show , for example, that impulsivity and lack of empathy tend to co-occur in individuals. That is, people who tend to act without thinking (impulsivity) also tend to lack the ability to see situations from other people' s perspectives (lack of empathy).
The high-P scorer is typically a solitary individual, often described by others as a "loner." Because he or she lacks empathy , he or she thus may be cruel or inhumane (men tend to score twice as high as women on P). Often, such people have a history of cruelty to animals. The high-P scorer may laugh, for example, when a dog gets hit by a car or when someone accidentally gets hurt. The high-P scorer shows insensi-tivity to the pain and suf fering of others, including that of his or her own kin. He or she is aggressive, both verbally and physically , even with loved ones. The high-P scorer has a penchant for the strange and unusual and may disregard danger entirely in pursuit of novelty . He or she likes to make fools of other people and is often described as having antisocial tendencies. In the extreme case, the individual may display symptoms of antisocial personality disorder (see Chapter 19).
Empirically, the P-scale predicts a number of fascinating criteria. Those who score high on P tend to show a strong preference for violent films and rate violen scenes from films more enjoyable and even more comical than those who score lo on P (Bruggemann & Barry , 2002). High-P individuals prefer unpleasant paintings and photographs more than do low-P individuals (Rawling, 2003). Men, but not women, who score high on Machiavellianism (which is highly correlated with P) endorse promiscuous and hostile sexual attitudes—they are more likely than low scorers to divulge sexual secrets to third parties, pretend to be in love when they are not in love, ply potential sex partners with alcoholic drinks, and even report trying to force others into sex acts (McHoskey, 2001). Those who are low in P tend to be more deeply religious, whereas high-P scorers tend to be somewhat cynical about religion (Saroglou, 2002). Finally , high-P scorers are predisposed to getting into severe and life-threatening events, such as violence and criminal activity (Pickering, Farmer , Harris, Redman, Mahmood, Sadler, & McGuf fin, 2003)
As you might imagine, the labels Eysenck has given to these super -traits, especially P, have generated some controversy . Indeed, some suggest that more accurate and appropriate labels for psychoticism might be "antisocial personality" and "psychopathic personality." Regardless of the label, P has emerged as an important trait in normal-range personality research.
Let's look more closely now at two aspects of Eysenck' s system that warrant further comment—its hierarchical nature and its biological underpinnings.
Hierarchical Structure of Eysenck's System
Figure 3.1 shows the levels in Eysenck' s hierarchical model—with each super -trait at the top and narrower traits at the second level. Subsumed by each narrow trait, however, is a third level—that of habitual acts. For example, one habitual act subsumed by sociable might be talking on the telephone; another might be taking frequent cof fee breaks to socialize with other students. Narrow traits subsume a variety of habitual acts.
At the very lowest level in the hierarchy are specific act (e.g., I talked on the telephone with my friend and I took a coffee br eak to chat at 10:30 a.m.). If enough specific acts are repeated frequentl , they become habitual acts at the third level. Clusters of habitual acts become narrow traits at the second level. And clusters of narrow traits become super -traits at the tops of the hierarchy . This hierarchy has the advantage of locating each specific personality-relevant act within a precise neste system. Thus, the fourth-level act I danced wildly at the party can be described as extraverted at the highest level, sociable at the second level, and part of a regular habit of party-going behavior at the third level.
There are two aspects of the biological underpinnings of Eysenck' s personality system that are critical to its understanding— heritability and identifiable physiologica substrate. For Eysenck a key criterion for a "basic" dimension of personality is that it has reasonably high heritability . The behavioral genetic evidence confirms that al three super-traits in Eysenck' s taxonomy—P, E, and N—do have moderate heritabil-ities, although this is also true of many personality traits (see Chapter 6 for more discussion of heritability of personality).
The second biological criterion is that basic personality traits should have an identifiable physiological substrate—that is, that one can identify properties in th brain and central nervous system that correspond to the traits and are presumed to be part of the causal chain that produces those traits. In Eysenck' s formulation, extraversion is supposed to be linked with central nervous system arousal or reactivity . Eysenck predicted that introverts would be more easily aroused (and more autonom-ically reactive) than extraverts (see Chapter 7). In contrast, he proposed that neuroti-cism was linked with the degree of lability (changeability) of the autonomic nervous system. Finally, high-P scorers were predicted to be high in testosterone levels and low in levels of MAO, a neurotransmitter inhibitor .
In sum, Eysenck's personality taxonomy has many distinct features. It is hierarchical, starting with broad traits, which subsume narrower traits, which in turn subsume specific actions. The broad traits within the system have been shown to be moderately heritable. And Eysenck has attempted to link these traits with physiological functioning— adding an important level of analysis not included in most personality theories.
Despite these admirable qualities, Eysenck' s personality taxonomy has several limitations. One is that many other personality traits also show moderate heritability , not just extraversion, neuroticism, and psychoticism. A second limitation is that
Eysenck may have missed some important traits in his taxonomy—a point ar gued by other personality psychologists, such as Raymond B. Cattell, and more recently by authors such as Lewis Goldberg, Paul Costa, and Robert McCrae. Since he was a contemporary of Eysenck's, we'll turn first to a discussion of Cattell s taxonomy.
Cattell's Taxonomy: The 16 Personality Factor System
Cattell was born in England in 1905. A precocious student, he entered the University of London at age 16, where he majored in chemistry . He pursued graduate study in psychology to gain an understanding of the social problems of the times. During his graduate education, Cattell worked closely with Charles Spearman, the inventor of factor analysis. Cattell viewed factor analysis as a powerful new tool for developing an objective, scientifically derived taxonomy of personalit . He devoted much of his career to developing and applying factor analytic techniques to understanding personality.
Cattell came to the United States in 1937 to become the research associate of Edward Thorndike (a famous psychologist) at Columbia University in New York. Cattell retired from University of Illinois in 1973, moved to Hawaii, and continued to write books and articles. Cattell, similar to Eysenck in many ways, also died in 1998.
Early in his career, Cattell established as one of his goals the identification an measurement of the basic units of personality . He took as an example the biochemists who were, at that time, discovering the basic vitamins. Cattell followed vitamin researchers by naming with letters the personality factors he discovered. Just as the biochemists named the first vitamin vitamin A, the second vitamin B, and so on, Cattell named the personality factors A, B, and so forth in the order in which he was convinced of their existence.
Cattell believed that true factors of personality should be found across dif ferent types of data, such as self-reports (S-data) and laboratory tests (T -data) (see Chapter 2). In contrast to Eysenck, who developed one of the smallest taxonomies of personality, as judged by the number of factors (3), Cattell' s taxonomy of 16 is among the lar gest in the number of factors identifie as basic traits. Much research has been conducted on the personality profiles of persons in various occu pational groups, such as police of ficers, researc scientists, social workers, and janitors. Descriptions of the 16 PF (personality factors) are presented in Table 3.3 and include information about occupational groups that score high or low on those scales.
Cattell, like Eysenck, published an extensive volume of work on personality , including over 50 books and 500 articles and chapters (e.g., Cattell, 1967, 1977, 1987). During his most productive period (the mid-1960s), there were times when he published over 1,000 pages a year . Cattell can be credited with developing a strong empirical strategy for identifying the basic dimensions of personality and with stimulating
Table 3.3 The 16 Personality Factor Scales
1. Factor A: interpersonal warmth. Warmhearted, personable, easy to get along with, likes being with other people, likes helping others, adapts well to the needs of others rather than has others adapt to his or her needs; this is similar to Eysenck's extraversion.
2. Factor B: intelligence. A rough indicator of intellectual functioning or efficiency of processing information.
3. Factor C: emotional stability. A high level of emotional resources with which to meet the challenges of daily life, able to work toward goals, not easily distracted, good emotional control, able to "roll with the punches," tolerates stress well; this is similar to Eysenck's neuroticism factor (reverse scored).
4. Factor E: dominance. Self-assertive, aggressive, competitive, forceful and direct in relations with others, likes to put own ideas into practice and have things own way; occupational groups scoring high on this dimension include athletes and judges, and low-scoring groups include janitors, farmers, and cooks.
5. Factor F: impulsivity. Happy-go-lucky, lively, enthusiastic, enjoys parties, likes to travel, prefers jobs with variety and change; occupational groups scoring high on this dimension include airline attendants and salespersons; adults scoring high on impulsivity tend to leave home at an earlier age and to move more often during their adult lives.
6. Factor G: conformity. Persistent, respectful of authority, rigid, conforming, follows group standards, likes rules and order, dislikes novelty and surprises; military cadets score above average, along with airport traffic controllers; university professors, however, tend to be below average on conformity.
7. Factor H: boldness. Likes being the center of attention, adventurous, socially bold, outgoing, confident, able to move easily into new social groups, not socially anxious, has no problems with stage fright.
8. Factor I: sensitivity. Artistic, insecure, dependent, overprotected, prefers reason to force in getting things done; high scorers are found among groups of employment counselors, artists, and musicians, whereas low scorers are found among engineers.
9. Factor L: suspiciousness. Suspecting, jealous, dogmatic, critical, irritable, holds grudges, worries much about what others think of him or her, tends to be critical of others; accountants are one group scoring high on this dimension.
10. Factor M: imagination. Sometimes called the "absent-minded professor" factor; unconventional, impractical, unconcerned about everyday matters, forgets trivial things, not usually interested in mechanical activities; high-scoring groups include artists and research scientists; high scorers are more creative than low scorers but also tend to have more automobile accidents.
11. Factor N: shrewdness. Polite, diplomatic, reserved, good at managing the impression made on others, socially poised and sophisticated, good control of his or her own behavior; high scorers may appear "stiff" and constrained in their social relations.
12. Factor O: insecurity. Tends to worry, feels guilty, moody, has frequent episodes of depression, often feels dejected, sensitive to criticism from others, becomes upset easily, anxious, often lonely, self-deprecating, self-reproaching; extremely low scorers come across as smug, self-satisfied, and overly self-confident; low-scoring persons may not feel bound by the standards of society and may not operate according to accepted social conventions, (i.e., may be somewhat antisocial).
13. Factor Q1: radicalism. Liberal attitudes, innovative, analytic, feels that society should throw out traditions, prefers to break with established ways of doing things; high scorers
Table 3.3 (continued)
tend to be effective problem solvers in group decision-making studies; however, high scorers, because they tend to be overly critical and verbally aggressive, are not well liked as group leaders.
14. Factor Q2: self-sufficiency. Prefers to be alone, dislikes being on committees or involved in group work, shuns support from others; social workers tend to be below average on this dimension; accountants and statisticians tend to be high, with Antarctic explorers among the highest groups ever tested on self-sufficiency.
15. Factor Q3: self-discipline. Prefers to be organized, think before talking or acting, is neat, does not like to leave anything to chance; high-scoring persons have strong control over their actions and emotions; airline pilots score high on this dimension.
16. Factor Q4: tension. Anxious, frustrated, takes a long time calming down after being upset, irritated by small things, gets angry easily, has trouble sleeping.
Source: Adapted from Krug, 1981.
and shaping the entire trait approach to personality . Nonetheless, Cattell's work, especially the model of 16 factors of personality , has been criticized. Specificall , some personality researchers have failed to replicate the 16 separate factors, and many argue that a smaller number of factors capture the most important ways in which individuals differ.
People have been fascinated with circles for centuries. There is something elegant about circles. They have no beginning and no end, and they symbolize wholeness and unity. Circles have also fascinated personality psychologists as possible representations of the personality sphere.
In the twentieth century, the two most prominent advocates of circular representations of personality have been Timothy Leary (also known for his LSD experiments at Harvard) and Jerry Wiggins, who formalized the circular model with modern statistical techniques. (Circumplex is simply a fancy name for circle.)
Wiggins (1979) started with the lexical assumption—the idea that all important individual dif ferences are encoded within the natural language. But he went further in his ef forts at taxonomy by ar guing that trait terms specify dif ferent kinds of ways in which individuals differ. One kind of individual dif ference pertains to what people do to and with each other— interpersonal traits.
Other kinds of individual dif ferences are specified by the following types of traits: temperament traits, such as nervous, gloomy, sluggish, and excitable; character traits, such as moral, principled, and dishonest;
Gregarious- 45° extraverted
Source: Adapted from "Circular Reasoning About Interpersonal Behavior" by J. S. Wiggins, 1989, Journal of Personality & Social Psychology, 56, p. 297. Copyright 1989 by the American Psychological Association. Reprinted with permission.
material traits, such as miserly and stingy; attitude traits, such as pious and spiritual; mental traits, such as clever , logical, and perceptive; and physical traits, such as healthy and tough.
Because Wiggins was concerned primarily with interpersonal traits, he carefully separated these from the other categories of traits. Then, based on the earlier theorizing of Foa and Foa (1974), he defined interpersonal as interactions between people involving exchanges. The two resources that define social exchange are love and status: "interpersonal events may be defined as dyadic interactions that have r ela-tively clear -cut social (status) and emotional (love) consequences for both participants" (Wiggins, 1979, p. 398, italics original). Hence, the dimensions of status and love define the two major axes of the Wiggins circumplex, as shown in Figure 3.2.
There are three clear advantages to the Wiggins circumplex. The first is that i provides an explicit definitio of interpersonal behavior. Thus, it should be possible to locate any transaction in which the resources of status or love are exchanged within a specific area of the circumplex pie. These include not just giving love (e.g., giving a friend a hug) or granting status (e.g., showing respect or honor to a parent). They also include denying love (e.g., yelling at one' s boyfriend) and denying status (e.g., dismissing someone as too inconsequential to talk to). Thus, the Wiggins model has the advantage of providing an explicit and precise definition of interpersona transactions.
The second advantage of Wiggins's model is that the circumplex specifies th relationships between each trait and every other trait within the model . There are basically three types of relationships specified by the model. The first is adjacency, or how close the traits are to each other in the circumplex. The variables that are adjacent, or next, to each other within the model are positively correlated. Thus, gregarious-extraverted is correlated with warm-agreeable. Arrogant-calculating is correlated with hostile-quarrelsome.
The second type of relationship is bipolarity. Traits that are bipolar are located at opposite sides of the circle and are negatively correlated with each other . Thus, dominant is the opposite of submissive, so the two are negatively correlated. Cold is the opposite of warm, so they are negatively correlated. Specifying this bipolarity is useful because nearly every interpersonal trait within the personality sphere has another trait that is its opposite.
The third type of relationship is orthogonality, which specifies that traits tha are perpendicular to each other on the model (at 90 0 of separation, or at right angles to each other) are entirely unrelated to each other . In other words, there is a zero correlation between such traits. Dominance, for example, is orthogonal to agree-ableness, so the two are uncorrelated. This means that dominance can be expressed in a quarrelsome manner (e.g., I yelled in or der to get my way ) or in an agreeable manner (e.g., I organized the gr oup in or der to get help for my friend ). Similarly, aggression (quarrelsome) can be expressed in an active/dominant manner (e.g., I used my position of authority to punish my enemies ) or in an unassured/submissive way (e.g., I gave him the silent tr eatment when I was upset). Thus, orthogonality allows one to specify with greater precision the dif ferent ways in which traits are expressed in actual behavior.
The third key advantage of the circumplex model is that it alerts investigators to gaps in investigations of interpersonal behavior . For example, whereas there have been many studies of dominance and aggression, personality psychologists have paid little attention to traits such as unassuming and calculating. The circumplex model, by providing a map of the interpersonal terrain, directs researchers to these neglected areas of psychological functioning.
In sum, the Wiggins circumplex model provides an elegant map of major individual differences in the social domain. Despite these positive qualities, the circum-plex also has some limitations. The most important limitation is that the interpersonal map is limited to two dimensions. Some have ar gued that other traits, not captured by these two dimensions, also have important interpersonal consequences. The trait of conscientiousness, for example, may be interpersonal in that persons high on this trait are very dependable in their social obligations to friends, mates, and children. Even a trait such as neuroticism or emotional stability may show up most strongly in interpersonal transactions with others (e.g., He overreacted to a subtle interpersonal slight when the host took too long to acknowledge his pr esence, and he insisted that he and his partner leave the party ). A more comprehensive taxonomy of personality that includes these dimensions is known as the five-factor model, t which we now turn.
In the past two decades, the taxonomy of personality traits that has received the most attention and support from personality researchers has been the five-facto model— variously labeled the five-factor model, the Big Five, and even in a humorous vei The High Five (Costa & McCrae, 1995; Goldberg, 1981; McCrae & John, 1992; Saucier & Goldberg, 1996). The broad traits composing the Big Five have been provisionally named: I. surgency or extraversion, II. agreeableness, III. conscientiousness, IV. emotional stability, and V. openness-intellect. This five-dimensional taxonomy o personality traits has accrued some persuasive advocates (e.g., John, 1990; McCrae & John, 1992; Saucier & Goldber g, 1998; Wiggins, 1996), as well as some strong critics (e.g., Block, 1995b; McAdams, 1992).
The five-factor model was originally based on a combination of the lexica approach and the statistical approach. The lexical approach started in the 1930s, with the pioneering work of Allport and Odbert (1936), who laboriously went through the dictionary and identified some 17,953 trait terms from the English language (whic then contained roughly 550,000 separate entries). Allport and Odbert then divided the original set of trait terms into four lists: (1) stable traits (e.g., secure, intelligent), (2) temporary states, moods, and activities (e.g., agitated, excited), (3) social evaluations (e.g., charming, irritating), and (4) metaphorical, physical, and doubtful terms (e.g., prolific, lea ).
The list of terms from the first categor , consisting of 4,500 presumably stable traits, was subsequently used by Cattell (1943) as a starting point for his lexical analysis of personality traits. Because of the limited power of computers at the time, however, Cattell could not subject this list to a factor analysis. Instead, he reduced the list to a smaller set of 171 clusters (groups of traits) by eliminating some and lumping together others. He ended up with a smaller set of 35 clusters of personality traits.
Fiske (1949) then took a subset of 22 of Cattell' s 35 clusters and discovered, through factor analysis, a five-factor solution. Howeve , this single study of relatively small sample size was hardly a robust foundation for a comprehensive taxonomy of personality traits. In historical treatments of the five-factor model, therefore, Fiske i noted as the first person to discover a version of the five-factor model, but he is n credited with having identified its precise structure
Tupes and Christal (1961) made the next major contribution to the five-facto taxonomy. They examined the factor structure of the 22 simplified descriptions i eight samples and emer ged with the five-factor model: surgency, agr eeableness, conscientiousness, emotional stability, and culture. This factor structure was subsequently replicated by Norman (1963), then by a host of other researchers (e.g., Botwin & Buss, 1989; Goldber g, 1981; Digman & Inouye, 1986; McCrae & Costa, 1985). The key markers that define the Big Five, as determined by Norman (1963), are show in Table 3.4.
The past 20 years have witnessed an explosion of research on the Big Five. Indeed, the big five taxonomy has achieved a greater degree of consensus than an other trait taxonomy in the history of personality trait psychology . But it has also generated some controversy. We consider three key issues: (1) What is the empirical evidence for the five-factor taxonomy of personality? (2) What is the identity of the
Talkative-silent Sociable-reclusive Adventurous-cautious Open-secretive
IV. Emotional stability
Talkative-silent Sociable-reclusive Adventurous-cautious Open-secretive
IV. Emotional stability
Intellectual-unreflective/narrow Artistic-nonartistic Imaginative-simple/direct Polished/refined-crude/boorish
Good-natured-irritable Cooperative-negativistic Mild/gentle-headstrong Not jealous-jealous
Responsible-undependable Scrupulous-unscrupulous Persevering-quitting Fussy/tidy-careless
Source: Norman (1963).
fifth factor? (3) Is the Big Five taxonomy really comprehensive, or are there majo trait dimensions that lie beyond the Big Five?
What Is the Empirical Evidence for the Five-Factor Model? The five-factor model has proven to be astonishingly replicable in studies using Englis language trait words as items (Goldber g, 1981, 1990; John, 1990). The five factor have been found by more than a dozen researchers using dif ferent samples. It has been replicated in every decade for the past half-century . It has been replicated in different languages and in dif ferent item formats.
In its modern form, the Big Five taxonomy has been measured in two major ways. One way is based on self-ratings of single-word trait adjectives, such as talkative, warm, organized, moody, and imaginative (Goldberg, 1990), and one way is based on self-ratings of sentence items, such as "My life is fast-paced" (McCrae & Costa, 1999). We will discuss these in turn.
Lewis R. Goldberg has done the most systematic research on the Big Five using single-word trait adjectives. According to Goldberg (1990), key adjective markers of the Big Five are as follows:
1. Surgency or extraversion: talkative, extroverted, assertive, forwar d, outspoken versus shy, quiet, introverted, bashful, inhibited.
2. Agreeableness: sympathetic, kind, warm, understanding, sincer e versus unsympathetic, unkind, harsh, cruel .
3. Conscientiousness: organized, neat, orderly, practical, prompt, meticulous versus disorganized, disorderly, careless, sloppy, impractical.
4. Emotional stability: calm, relaxed, stable versus moody, anxious, insecure.
5. Intellect or imagination: creative, imaginative, intellectual versus uncreative, unimaginative, unintellectual.
In addition to measures of the big five that use single trait words as items, th most widely used measure using a sentence-length item format has been developed by Paul T. Costa and Robert R. McCrae. It' s called the NEO-PI-R: the neuroticism-extraversion-openness (NEO) Personality Inventory (PI) Revised (R) (Costa &
McCrae, 1989). Sample items from the NEO-PI-R are neuroticism (N): I have fr e-quent mood swings; extraversion (E): I don't find it easy to take cha ge of a situation (reverse scored); openness (O): I enjoy trying new and for eign foods; agreeableness (A): Most people I know like me; and conscientiousness (C): I keep my belongings neat and clean .
Your job is to develop a way to measure the Big Five traits in someone you know, such as a friend, a roommate, or a family member. Read the adjectives in Table 3.4 carefully until you have an understanding of each of the Big Five traits. Then, consider the different sources of personality data described in Chapter 2:
1. Self-report—typically, asking questions on a questionnaire.
2. Observer-report—typically, asking someone who knows the subject to report what the subject is like.
3. Test data—typically, objective tasks, situations, or physiological recordings that get at manifestations of the trait in question.
4. Life-outcome data—aspects of the person's life that may reveal a trait, such as introverted people selecting careers in which there is little contact with others.
Your job is to assess your target person on each of the Big Five traits, using a combination of data sources. In your report, you should first list, for each of the five traits, the way in which you measured that trait, such as the items on your questionnaire or interview or the life-outcome data you think indicates that trait. Then, in the second part of your report, indicate how high or low you think your examinee is on each of the five traits.
You might be thinking at this point that five factors may be too few to captur all of the fascinating complexity of personality. And you may be right. But consider this. Each of the five global personality factors has a host of specif "facets," which provide a lot of subtlety and nuance. The global trait of conscientiousness, for example, includes these six facets: competence, order , dutifulness, achievement striving, self-discipline, and deliberation. The global trait of neuroticism has these six facets: anxiety, angry hostility, depression, self-consciousness, impulsivity, and vulnerability. These facets of each global factor go a long way toward adding richness, complexity , and nuance to personality description.
Note that, although the NEO-PI-R traits are presented in a dif ferent order (N, E, O, A, C) than the Goldber g order, and in a few cases the traits are given dif ferent names, the underlying personality traits being measured are nearly identical to those found by Goldberg. This convergence between the factor structures of single-trait item formats and sentence-length item formats provides support for the robustness and replicability of the five-factor model
What Is the Identity of the Fifth Factor?
Although the five-factor model has achieved impressive replicability across sam ples, investigators, and item formats, there is still some disagreement about the content and replicability of the fifth facto . Dif ferent researchers have variously labeled this fifth factor as culture, intellect, intellectance, imagination, openness, openness to experience, and even fluid intelligence and tender-mindedness (see Brand & Egan, 1989; De Raad, 1998). A major cause of these dif ferences is that different researchers start with dif ferent item pools to factor analyze. Those who start with the lexical strategy and use adjectives as items typically endorse intellect as the meaning and label of the fifth factor (Saucier & Goldbe g, 1996). In contrast, those who use questionnaire items tend to prefer openness or openness to experience, because this label better reflects the content of those items (McCrae Costa, 1997; 1999).
One way to resolve these dif ferences is to go back to the lexical rationale to begin with and to look across cultures and across languages. Recall that, according to the lexical approach, traits that emer ge universally in dif ferent languages and cultures are deemed more important than those that lack cross-cultural universality.
What do the cross-cultural data show? In a study conducted in Turkey, a clear fifth factor eme ged that is best described as openness (Somer & Goldberg, 1999). A different Dutch study found a fifth factor marked by progressive at one end and conservative at the other (DeRaad et al., 1998). In German, the fifth factor represent intelligence, talents, and abilities (Ostendorf, 1990). In Italian, the fifth factor i conventionality, marked by the items rebellious and critical (Caprara & Perugini, 1994). Looking across all these studies, the fifth factor has proven extremely di ficul to pin down.
In summary, although the first four factors are highly replicable across culture and languages, there is uncertainty about the content, naming, and replicability of the fifth facto . Perhaps some individual dif ferences are more relevant to some cultures than to others—intellect in some cultures, conventionality in other cultures, and openness in yet other cultures. Clearly , more extensive cross-cultural work is needed, particularly in African cultures and in more traditional cultures that are minimally influenced by Western culture.
What Are the Empirical Correlates of the Five Factors? Over the past 15 years, a tremendous volume of research has been conducted on the empirical correlates of each of the five factors. This section summarizes some of the most recent interesting findings
Surgency or extraversion. Extraverts love to party—they engage in frequent social interaction, take the lead in livening up dull gatherings, and enjoy talking a lot. Indeed, recent evidence suggests that social attention is the cardinal feature of extraversion (Ashton, Lee, & Paunonen, 2002). From the perspective of the extravert, "the more the merrier ." Extraverts have a greater impact on their social environment, often assuming leadership positions, whereas introverts tend to be more like wallflowers (Jensen-Campbell & Graziano, 2001). Extraverted men ar more likely to be bold with women they don' t know, whereas introverted men tend to be timid with women (Berry & Miller , 2001). Extraverts tend to be happier , and this positive af fect is experienced most intensely when a person acts in an extraverted manner (Fleeson, Malanos, & Achille, 2002). But there are also downsides— extraverts like to drive fast, listen to music while driving, and as a consequence, tend to get into more car accidents, and even road fatalities, than their more introverted peers (Lajunen, 2001).
Agreeableness. Whereas the motto of the extravert might be "let' s liven things up," the motto of the highly agreeable person might be "let' s all get along." Those who score high on agreeableness favor using negotiation to resolve conflicts; low agreeable persons try to assert their power to resolve social conflicts (Graziano Tobin, 2002; Jensen-Campbell & Graziano, 2001). The agreeable person is also more likely to withdraw from social conflict, avoiding situations that are unharmonious Agreeable individuals like harmonious social interaction and cooperative family life. Agreeable children tend to be less often victimized by bullies during early adolescence (Jensen-Campbell et al., 2002). As you might suspect, politicians, at least in Italy, tend to score high on scales of agreeableness (Caprara, Barbaranelli, Consiglio, Picconi, & Zimbardo, 2003).
At the other end of the scale of agreeableness lies aggressiveness. In a fascinating study of daily acts, Wu and Clark (2003) found that aggressiveness was strongly linked to many everyday behaviors. Examples include: hitting someone else in anger; blowing up when things don' t work pr operly; slamming doors; yelling; getting into ar guments; clenching fists; raising voices; being intentionally rude damaging someone 's pr operty; pushing and hitting others; and slamming down the phone. So the next time you think about getting into an ar gument with someone, you might want to find out where they are on the agreeable-aggressivenes disposition.
Agreeable individuals, in short, get along well with others, are well liked, avoid conflict, strive for harmonious family lives, and may selectively prefer professions i which their likeability is an asset. Disagreeable individuals are aggressive and seem to get themselves into a lot of social conflict
Conscientiousness. If extraverts party up and agreeable people get along, then conscientious individuals are industrious and get ahead. The hard work, punctuality , and reliable behavior exhibited by conscientious individuals result in a host of life outcomes such as a higher grade point average, greater job satisfaction, greater job security, and more positive and committed social relationships (Langford, 2003).
Those who score low on conscientiousness, in contrast, are likely to perform more poorly at school and at work. The fact that highly conscientious individuals succeed in the work domain is likely due to two key correlates. They do not procrastinate, in contrast to their low-conscious peers whose motto might be "never put of f until tomorrow what you can put of f until the day after tomorrow" (Lee, Kelly , & Edwards). And those high in conscientiousness are exceptionally industrious, putting in the long hours of diligent hard work needed to get ahead (Lund et al., 2006). Furthermore, low C is linked with risky sexual behaviors such as failing to use condoms (Trobst, Herbst, Masters, & Costa, 2002), and being more responsive to other potential partners while already in an existing romantic relationship (Schmitt & Buss, 2001). Among a sample of prisoners, low-C scorers tend to have frequent arrests (Clower & Bothwell, 2001). The high-C individual, in sum, tends to perform well in school and work, avoids breaking the rules, and has a more stable and secure romantic relationship.
Emotional stability. Life poses stresses and hurdles that everyone must confront. The dimension of emotional stability taps into the way people cope with these stresses. Emotionally stable individuals are like boats that remain on course through choppy waters. Emotionally unstable people get buf feted about by the waves and wind and are more likely to get knocked of f course. The hallmark of emotional instability or neuroticism is variability of moods over time—such people swing up and down more than emotionally stable individuals (Murray , Allen, & Trinder, 2002). Perhaps as a consequence, emotionally unstable individuals experience more fatigue over the course of the day (De Vries & Van Heck, 2002). Psychologically, emotionally unstable individuals are more likely to have dissociate experiences such as an inability to recall important life events, feeling disconnected from life and other people, and feeling like they've woken up in a strange or unfamiliar place (Kwapil, Wrobel, & Pope, 2002). Have you ever had thoughts about committing suicide? Those high on neuroticism also tend to have more frequent suicidal ideation than those low on neuroticism (Chioqueta & Stiles, 2005). Those high on neuroticism report poorer physical health, more physical symptoms, and fewer attempts to engage in health-promoting behaviors (W illiams, O'Brien, & Colder, 2004). They also engage in health-impairing behaviors, such as drinking alcohol as a means of coping with, and attempting to for get about, their problems (Theakston et al., 2004).
Interpersonally, those high on neuroticism or emotional instability have more ups and downs in their social relationships. In the sexual domain, for example, emotionally unstable individuals experience more sexual anxiety (e.g., worried about performance) as well as a greater fear of engaging in sex (Heaven, Crocker , Edwards, Preston, Ward, & Woodbridge, 2003; Shafer, 2001). And with highly stressful events, such as an unwanted loss of a pregnancy , emotionally unstable individuals are more likely to develop "post-traumatic stress disorder ," in which the psychological trauma of the loss is experienced profoundly and for a long time (Englelhard, van den Hout, & Kindt, 2003).
Emotional instability augers poorly for professional success. This may be partly due to the fact that emotionally unstable people are thrown of f track by the everyday stresses and strains that we all go through. It may be partly due to their experience of greater fatigue. But it may also be attributable to the fact that they engage in a lot of "self-handicapping" (Ross, Canada, & Rausch, 2002). Self-handicapping is define as a tendency to "create obstacles to successful achievement in performance or competitive situations in order to protect one's self-esteem" (Ross et al., 2002, p. 2). Those high on neuroticism seem to undermine themselves, creating roadblocks to their own achievement. Nonetheless, one study found that those high on neuroticism actually outperformed their more emotionally stable counterparts in performance in an of fic setting when changes in the work needs created an unusually busy work environment (Smillie, Yeo, Furnham, & Jackson, 2006). In sum, the af fective volatility that comes with being low on emotional stability af fects many spheres of life, from sexuality to achievement.
Openness. Would you agree or disagree with the following statements? "Upon awakening during the night, I am unsur e whether I actually experienced something or only dr eamed about it," "I am awar e that I am dr eaming, even as I dr eam," "I am able to control or direct the content of my dreams," "A dream helped me to solve a current problem or concern" (Watson, 2003). If you tend to agree with these statements, the odds are that you score high on the personality disposition of openness. Those who are high on openness tend to remember their dreams more, have more waking dreams, have more vivid dreams, have more prophetic dreams (dreaming about something that later happens), and have more problem-solving dreams (Watson, 2003).
The disposition of openness has been linked to experimentation with new foods, a liking for novel experiences, and even "openness" to having extramarital affairs (Buss, 1993). One possible cause of openness may lie in individual dif fer-ences in the processing of information. A recent study found that those high in openness had more dif ficulty in ignoring previously experienced stimuli (Peterson, Smith, & Carson, 2002). It' s as though the perceptual and information processing "gates" of highly open people are literally more "open" to receiving information coming at them from a variety of sources. Less-open people have more tunnel vision and find it easier to ignore competing stimuli. Those high in openness exhibit less prejudice against minority groups, and are less likely to hold negative racial stereotypes (Flynn, 2005). In sum, the disposition of openness has been correlated with a host of other fascinating variables from intrusive stimuli to possible alternative sex partners.
Combinations of Big Five variables. Many life outcomes, of course, are better predicted by combinations of personality dispositions than by single personality dispositions. Here are a few examples.
• Good grades are best predicted by Conscientiousness (high) and Emotional Stability (high) (Chamorro-Premuzic & Furnham, 2003). One reason might be that emotionally stable and conscientious people are less likely to procrastinate (Watson, 2001).
• Risky sexual behaviors, such has having many sex partners and not using condoms, are best predicted by high Extraversion, high Neuroticism, low Conscientiousness, and low Agreeableness (Miller et al., 2004; Trobst et al., 2002).
• Alcohol consumption is best predicted by high Extraversion and low Conscientiousness (Paunonen, 2003). A study of more than 5,000 workers in Finland found that low Conscientiousness also predicts increases in alcohol consumption over time, that is, who ends up becoming a heavy drinker (Grano et al., 2004).
• Mount Everest mountain climbers tend to be extraverted, emotionally stable, and high on Psychoticism (Egan & Stelmack, 2003).
• Happiness and experiencing positive af fect in everyday life are best predicted by high Extraversion and low Neuroticism (Cheng & Furnham, 2003; Steel & Ones, 2003; Stewart, Ebmeier, & Deary, 2005; Yik & Russell, 2001).
• Proclivity to engage in volunteer work, such as campus or community services, is best predicted by a combination of high Agreeableness and high Extraversion (Carlo et al., 2005).
• When you join the workfor ce, do you think you will join the voluntary union organization or decline to become a member of the union? Those low on Extraversion and high on Emotional Stability have been shown to have a disproportionately low rate of joining work unions (Parkes & Razavi, 2004).
• Forgiveness, the proclivity to for give those who have committed some wrong, characterizes individuals who are high on Agreeableness and high on Emotional Stability (Brose, Rye, Lutz-Zois, & Ross, 2005).
• Leadership effectiveness in business settings is best predicted by high Extraversion, high Agreeableness, high Conscientiousness, and high Emotional Stability (Silverthorne, 2001).
We should not be surprised that combinations of personality variables often do better than single variables in predicting important life outcomes, and we can expect future research to focus increasingly on these combinations.
Is the Five-Factor Model Comprehensive?
Critics of the five-factor model a gue that it leaves out important aspects of personality. Almagor, Tellegen, and Waller (1995), for example, present evidence for seven factors. Their results suggest the addition of two factors— positive evaluation (e.g., outstanding versus ordinary) and negative evaluation (e.g., awful versus decent). Goldberg, one of the proponents of the five-factor model, has discovered that factor such as religiosity and spirituality sometimes emerge as separate factors, although these are clearly smaller in size (accounting for less variance) than those of the Big Five (Goldberg & Saucier, 1995).
Lanning (1994), using items from the California Adult Q-set, has found a replic-able sixth factor, which he labels attractiveness, including the items physically attractive, sees self as attractive, and charming. In a related vein, Schmitt and Buss (2000) have found reliable individual dif ferences in the sexual sphere, such as sexiness (e.g., sexy, stunning, attractive, alluring, ar ousing, sensual, and seductive) and faithfulness (e.g., faithful, monogamous, devoted, and not adulterous). These individual difference dimensions are correlated with the five factors: sexiness is positively correlated with extraversion, and faithfulness is positively correlated with both agreeableness and conscientiousness. But these correlations leave much of the individual variation unaccounted for, suggesting that these individual dif ferences in sexuality are not completely subsumed by the five-factor model
Paunonen and colleagues have identified 10 personality traits that appear to fal outside of the five-factor model: Conventionale , Seductiveness, Manipulativeness, Thriftiness, Humorousness, Integrity , Femininity, Religiosity, Risk Taking, and Egotism (Paunonen, 2002; Paunonen et al., 2003). Other researchers have confirmed tha these traits are not highly correlated with the Big Five, and that they highlight many interesting facets of personality at a more specific level than the "global" factors rep resented by the five-factor model (Lee, Ogunfowora, & Ashton, 2005).
Proponents of the five-factor model are typically open-minded about the poten tial inclusion of factors beyond the five factors, if and when the empirical evidence war rants it (Costa & McCrae, 1995; Goldber g & Saucier , 1995). Nonetheless, these researchers have not found the evidence for additional factors beyond the Big Five to be compelling. Positive and negative evaluation, some have argued, are not really separate factors but, rather, false factors that emer ge simply because raters tend to evaluate all things as either good or bad (McCrae & John, 1992). With respect to the attractiveness factor found by Lanning (1994), Costa and McCrae (1995) ar gue that attractiveness is not ordinarily considered to be a personality trait, although the charming item that loads on this factor surely would be considered part of personality .
One approach to personality factors beyond the Big Five has been to explore personality-descriptive nouns, rather than adjectives. Saucier (2003) has discovered eight fascinating factors within the domain of personality nouns such as: Dumbbell (e.g., dummy, moron, twit), Babe/Cutie (e.g., beauty, darling, doll), Philosopher (e.g., genius, artist, individualist), Lawbreaker (e.g., pothead, drunk, rebel), Joker (e.g., clown, goof, comedian), and Jock (e.g., sportsman, tough, machine). A study of personality nouns in the Italian language revealed a somewhat dif ferent organization than that of the Big Five, discovering factors such as Honesty , Humility, and Cleverness (Di Blas, 2005). As Saucier concludes, "Personality taxonomies based on adjectives are unlikely to be comprehensive, because type-nouns have dif ferent content emphases" (Saucier, 2003, p. 695).
A second approach to personality factors beyond the Big Five has been to use the lexical approach, focusing on large pools of trait adjectives in dif ferent languages. In an exciting development, several studies have conver ged on six rather than fiv factors. One study of seven languages (Dutch, French, German, Hungarian, Italian, Korean, and Polish) found variants of the Big Five, plus a sixth factor Honesty-Humility (Ashton et al., 2004). At one end of the Honesty-Humility factor lies trait adjectives such as honest, sincere, trustworthy, and unselfish; the other end is anchored by adjec tives such as arrogant, conceited, greedy , pompous, self-important, and egotistical. Independent investigators have also found versions of this sixth factor in Greece (Saucier, Georgiades, Tsaousis, & Goldber g, 2005) and Italy (Di Bias, 2005). These findings point to an exciting expansion of the basic factors of personality within th dispositional domain.
In addition to the possibility of discovering dimensions beyond the Big Five, some researchers have had excellent success in predicting important behavioral criteria from within the Big Five using the facets of the Big Five (Paunonen & Ashton, 2001a, b). For example, in predicting course grades in a college class, Paunonen and Ashton (2001a) found significantly greater predictability from the facet subscales o Need for Achievement (a facet of Conscientiousness) and Need for Understanding (a facet of Openness) than from the higher -level factor measures of Conscientiousness and Openness themselves. Similarly , although job performance is well predicted by global measures of Conscientiousness, even better prediction of job performance is attained by including the facet measures such as achievement, dependability , order, and cautiousness (Dudley et al., 2006). Paunonen and Ashton conclude that "the aggregation of narrow trait measures into broad factor measures can be counterproductive from the point of view of both behavioral prediction and behavioral explanation" (Paunonen & Ashton, 2001a, p. 78).
Thus, we are left with an important question: does the five-factor model pro vide a comprehensive description of personality? On the yes side, the five-facto model has proven to be more robust and replicable than any other taxonomy of personality that claims to be comprehensive. Four of the five factors have proven t be highly replicable across investigators, data sources, item formats, samples, languages, and cultures. Furthermore, the five-factor model has been discovered to b the major structure underlying many existing personality inventories. On the no side, claims that the five-factor model is comprehensive may be premature, as th proponents of the five-factor model readily admit. Indeed, the quest for factors beyond the Big Five and the discovery of a replicable sixth factor makes the fiel of personality psychology such an exciting and vibrant discipline (Ashton, Lee, & Goldberg, 2004).
The model has also drawn articulate critics, such as McAdams (1992) and Block (1995b). Block, for example, ar gues that these five factors, although perhaps usefu for laypersons in everyday life, fail to capture the underlying causal personality processes that researchers are really interested in. Describing someone as high on neuroticism, for example, may be useful in social communication or global character descriptions, but it does not capture the underlying psychological processes involved in such things as feeling guilty , obsessing over worst-case scenarios, and worrying excessively when someone fails to respond to an e-mail message.
Proponents of the five-factor model respond to these criticisms by suggestin that the Big Five taxonomy has been proposed merely as a framework for the phe-notypic attributes of personality that have become encoded within the natural language and makes no claims about the underlying personality processes (Goldber g & Saucier, 1995). Debates such as these are the essence of the scientific enterprise an indicate a healthy and thriving field. These controversies can be expected to continue as personality psychologists struggle to develop better, more adequate, and more comprehensive taxonomies of personality .
SUMMARY AND EVALUATION_
This chapter focused on three fundamental issues for a personality psychology based on traits: how to conceptualize traits, how to identify the most important traits, and how to formulate a comprehensive taxonomy of traits.
There are two basic conceptualizations of traits. The first views traits as th internal properties of persons that cause their behavior. In the internal property conception, traits cause the outward behavioral manifestations. The second conceptualization views traits as descriptive summaries of overt behavior . The summary view does not assume that traits cause behavior but, rather, treats the issue of cause separately, to be examined after the behavioral summaries are identified and described.
There have been three major approaches to identifying the most important traits. The first is the lexical approach, which views all the important traits as cap tured by the natural language. The lexical approach uses synonym frequency and cross-cultural universality as the criteria for identifying important traits. The second approach, the statistical approach to identifying important traits, adopts statistical procedures, such as factor analysis, and attempts to identify clusters of traits that covary. The third approach, the theoretical approach, uses an existing theory of personality to determine which traits are important. In practice, personality psychologists sometimes use blends of these three approaches—for example, by starting with the lexical approach to identify the universe of traits and then applying statistical procedures, such as factor analysis, to identify groups of traits that covary and form larger factors.
The third fundamental issue—formulating an overarching taxonomy of personality traits—has yielded several solutions. Eysenck developed a hierarchical model, in which the broad traits extraversion, neuroticism, and psychoticism subsume more narrow traits, such as activity level, moodiness, and egocentricity . Eysenck's taxonomy is based on a factor analysis but is also explicitly anchored in biological underpinnings, including a heritable basis for the traits and the identification o
Was this article helpful?