Describe the emergence of the trait approach to psychology and identify some contribu- tions of important historical figures, such as Allport, Cattell, and Eysenck.
Fuse/Thinkstock
Learning Objectives
After reading this chapter, you should be able to:
• Explain trait theory and how it emerged as a dominant force in personality theory.
• Describe the emergence of the trait approach to psychology and identify some contribu- tions of important historical figures, such as Allport, Cattell, and Eysenck.
• Describe how the taxonomy of traits was developed based on language and how they are organized into a hierarchy of factors (e.g., using three-, five-, and sixteen-factor models).
• Explain how factors relate to behaviors in a hierarchy.
• Describe some of the important outcomes that have been predicted by traits such as neuroti- cism, extraversion, optimism, and locus of control.
• Describe the stability of traits over the lifespan and across cultures and languages.
A Trait Approach to Personality 8
Chapter Outline Introduction
8.1 Trait Theory in Historical Perspective • Traits as Building Blocks of Personality • Hippocrates and Galen: The Ancient Greeks
and Humoral Theory • Carl Jung’s Introduction of Introversion and
Extraversion • Gordon Allport and the Analysis of Language • Raymond Cattell and the Statistical Approach
to Personality • Eysenck’s Model of Personality
8.2 Convergence on the Big Five • Openness to New Experience • Agreeableness • Conscientiousness • Big Five in Cultural Context • Heritability of the Big Five • The Big Five Over the Lifespan
• Characterize Mischel’s critique of the trait approach and the field’s response to that critique (i.e., the person- situation debate).
• Describe the novel approaches to conceptualizing and assessing traits, such as the act-frequency approach.
• Characterize the complementary contributions of the goal approach, which examines traits in the context of our lives.
• Describe some of the commonly used measures of traits.
Lec81110_08_c08_225-252.indd 225 5/21/15 12:40 PM
CHAPTER 8Introduction
8.3 The Person-Situation Debate • Responses to Mischel’s Critique of Trait
Psychology • Understanding Situational Strength,
Domain Breadth, and Trait Relevance • The Role of the Fundamental Attribu-
tion Error • Summary of Person-Situation Debate
8.4 Supplementing the Big Five With Complementary Approaches • Projects, Life Tasks, Concerns, Strivings,
and Goals: An Idiographic Approach • Act-Frequency Approach
8.5 Assessment Methods From the Trait Perspective • The Family of NEO™ Scales • The Big Five Inventory • The HEXACO Inventory • Eysenck’s Personality Questionnaire • Cattell’s 16 Personality Factor
Questionnaire • Myers Briggs Type Indicator®
Summary
Introduction John is presenting a lecture, and as is his custom, he keeps the students enter- tained and engaged with his wit, smooth dialogue, and animated body language. Given the reaction of the students, this is not a lecture, but an hour of informative entertainment. After class, the students are drawn to John because of his gre- garious and friendly demeanor during class, but he is nowhere to be found. John has a habit of retreating from public after giving a lecture because he feels both exhausted and overwhelmed. His favorite place to hide is a stall in the men’s room; it affords the best protection from interactions with others. After 30 minutes or so, he emerges feeling somewhat recovered.
You see, John’s job requires that he engage in an activity that is not especially pleasant for him. John is, in fact, somewhat isolative, but not shy. John doesn’t look at a group of people and long for their attention or for more social inter- actions. Rather than attend a party or be in large groups, he prefers to read a book under a tree or some other solitary, tranquil place. John is what some people call a “pseudo-extravert.” That is, he is actually an introvert, but he engages in extra- verted behavior in order to meet the demands of his life or important life goals.
Why is it that some people are hard-wired to enjoy stimulation, whereas others appear just as hard-wired to dislike and avoid it? Why are some individuals prone to worry? For example, you may know someone who always worries about their exam performance, yet they typically set the curve on every exam. Why do many people gravitate toward taking (or at least wanting) control over a situation, whereas others prefer to give up that control and let others decide matters?
In this chapter, we will examine what is known as the trait approach to personal- ity. We will examine the theorists who initiated the movement, some of the more interesting research findings on traits, the challenges to trait theory, and the field’s responses.
Lec81110_08_c08_225-252.indd 226 5/21/15 12:40 PM
CHAPTER 8 8.1 Trait Theory in Historical Perspective
iStockphoto/Thinkstock
Over 2,000 years after the descriptive terms were introduced, we still use the term “choleric” to refer to an easily angered child.
8.1 Trait Theory in Historical Perspective
Trait theory is a popular approach for studying personality; it is closely tied to the everyday concept of personality that many people hold because traits are commonly employed in everyday language and are widely understood. We will begin by defining the concept of traits and identify some of the earliest contributors to the trait approach to personality, including Hippocrates, Galen, Carl Jung, Gordon Allport, Raymond Cattell, and Hans Eysenck.
Traits as Building Blocks of Personality At its most fundamental level, a trait is a unit of analysis to describe, predict, and explain human thought, affect, and behavior. From a distance, it appears as though there are a great many terms (traits) that are used to characterize human activity, but extensive research suggests that these traits can be organized into coherent and meaningful patterns and even enveloped by a smaller number of broad trait categories, thereby simplifying the trait approach.
Hippocrates and Galen: The Ancient Greeks and Humoral Theory The earliest documented work on humoral theory dates back to ancient Greece—and the belief that the body was compromised of four basic fluids and the balance of these fluids could deter- mine behavioral and emotional tendencies (and disease susceptibility). Based originally on early writings in medicine by Hippocrates and later expanded by Galen (On the Temperaments), humoral theory focuses on blood, phlegm, black bile, and yellow bile; these are the four basic fluids (humors) that were thought to be within the human body.
According to the theory, the four humors have to be in balance to achieve and maintain health, and this likewise pre- dicted imbalances in emotion and behavior. An ideal temperament was associated with a balance of the four humors. Excessive blood (sanguine) was associated with a cheerful disposition, excessive black bile (melancholic) was associated with a sad disposition, exces- sive yellow bile (choleric) was character- ized as irritable, and too much phlegm (phlegmatic) meant an unemotional temperament.
Although contemporary personality researchers do not relate the humors to character, the descriptive terms are still employed. For example, an irritable infant is still referred to as choleric, and
the term melancholy still applies to sadness. Moreover, as we shall see when reviewing the work of Eysenck, the basic terminology for characterizing all human behavior has been somewhat con- sistent for more than 2,000 years, suggesting that there may be a core set of personality traits
Lec81110_08_c08_225-252.indd 227 5/21/15 12:40 PM
CHAPTER 8 8.1 Trait Theory in Historical Perspective
expressed by humans that have been stable throughout much of modern history. Moreover, humoral theory also established the framework for connecting traits to biological functioning, and this tradition also continues today (see especially Eysenck’s work on extraversion and research on heritability coefficients for personality).
Carl Jung’s Introduction of Introversion and Extraversion Carl Jung was first and foremost a central contributor to psychodynamic theory. However, he was also one of the first to popularize the terms introversion and extraversion, and these remain two of the most popular and widely recognized trait terms. Jung described the outward manifestation of behavior in very similar ways to modern-day psychologists. For example, extraversion meant someone who is interested in other people and things and is focused on them, whereas introver- sion meant being withdrawn and focusing on the subjective experience of the world. Although Jung did not develop a formal measure of introversion-extraversion, researchers subsequently developed a measure that was based on Jung’s typology, called the Myers Briggs Type Indicator® (MBTI®) (see Chapter 1).
Gordon Allport and the Analysis of Language One of the more enduring early contributions in trait psychology was the method of studying language as the very basis of traits. Gordon W. Allport is often considered the first trait psycholo- gist, and he was interested in classification. Allport adopted traits as his basic unit of analysis. He believed that traits are closely tied to the nervous system, and that they account for behavioral consistency across time and situations (Allport & Allport, 1921). Allport believed that a trait would predict which behavior would manifest with high frequency, with intensity, and over a wide range of situations (see also Allport, 1937).
Beginning an important tradition, Allport used Webster’s New International Dictionary and culled almost 18,000 words, each of which described some aspect of human behavior (Allport & Odbert, 1936). In adopting this methodology, Allport made the assumption that any descriptive character- istics that are important will become part of our language and that language will evolve such that single words will emerge to capture those important constructs. Thus, Allport’s theory is based on the associated meanings of words, as he thought these meanings transcend the word itself and instead speak to human nature. This approach of studying language to understand person- ality was referred to a lexical analysis, and the underlying theory was referred to as the lexical hypothesis (see also John, Angleitner, & Ostendorf, 1988). The general thesis of this work is that by understanding how different adjectives that are used to describe human behavior are related to each other, one can then understand at least two basic questions: What is the minimum num- ber of different personality traits or factors needed to capture all of the adjectives, and what are the best labels for these traits or factors? A third question that often arises focuses on how the different traits/factors relate to each other (i.e., are they independent or correlated?).
Cardinal, Central, and Secondary Traits
Allport believed that traits are the characteristic manner in which we respond to stimuli in our environment. Some traits are prominent, and they dominate personality. Others are minor and
Lec81110_08_c08_225-252.indd 228 5/21/15 12:40 PM
CHAPTER 8 8.1 Trait Theory in Historical Perspective
are less obvious to others. The way our traits are patterned reflects our unique personality struc- ture and determines our behavior.
To reflect these different types of traits, Allport established a hierarchical structure to reflect the variability in the prominence of different traits in different individuals (Allport, 1937). Allport referred to cardinal traits as those that are so pervasive and enduring that they manifest in virtu- ally every aspect of an individual’s life and serve as primary motivators of action. Later in this chap- ter we will discuss the big five factors, and some of these (e.g., extraversion, neuroticism, etc.) can be examples of cardinal traits. Interestingly, an abundance of such cardinal traits can be seen as quite problematic, and, in fact, such persistence of behavior independent of the situation is often the hallmark of personality disorders. Central traits were defined as less pervasive than cardinal traits, but still manifesting in a limited range of situations. Central traits were thought to be more commonly observed in everyone. Finally, secondary traits were considered the least durable
over time and across situations, and the combination of second- ary traits is what contributes most to the individual’s unique- ness. Researchers would later debate the stability of personal- ity over situations (the person- situation debate discussed in this chapter), but Allport already had an answer for this issue— suggesting that not all traits are equally generalizable across situations.
Raymond Cattell and the Statistical Approach to Personality Despite the significant advance that came from Allport’s analysis of language, a major shortcom- ing remained. Given the large number of traits that had been identified (over 4,000), the problem was determining how to best organize those terms. Although early contributions were made by Thurstone’s factor theory (1938), it was Raymond Cattell who emerged as one of the primary researchers of an organizational framework for personality. He applied powerful statistical proce- dures to the taxonomy of traits in an attempt to find an underlying structure.
Most of Cattell’s contributions occurred in the 1940s and 1950s, as he reduced Allport’s 18,000 descriptive terms to a smaller number of clusters, which in turn, were reduced to a smaller num- ber of factors (e.g., Cattell, 1943, 1945, 1957). To understand Cattell’s theory, it is critical to have a clear understanding of the nature of a factor.
Cattell studied under Charles Spearman, who was developing the technique of factor analysis in order to better understand the basic structure of human abilities. The basic premise involved examining ratings of items to determine whether ratings for one item (e.g., if I rated myself as out- going) or set of items are associated with ratings on another item (e.g., if I rated myself as friendly) or set of items. When several items appear to be rated in similar ways by many people, then they are likely to reflect an underlying factor (e.g., the factor of agreeableness for the above two items). Thus, in essence, factors are super-traits that stand toward the top of the organizational structure and can define a large number of other traits. As we will see later, there has been con- siderable debate as to how many factors underlie all traits, with estimates ranging from as many
Beyond the Text: Classic Writings
In this classic 1927 paper, “Concepts of Trait and Personal- ity,” Allport introduces his conceptualization of traits as a “statistical” unit of analysis and highlights their central role in the study of personality. Read the paper at http://psych classics.yorku.ca/Allport/concepts.htm.
Reference: Allport, G. W. (1927). Concepts of trait and per- sonality. Psychological Bulletin, 24, 284–293.
Lec81110_08_c08_225-252.indd 229 5/21/15 12:40 PMhttp://psychclassics.yorku.ca/Allport/concepts.htmhttp://psychclassics.yorku.ca/Allport/concepts.htm
CHAPTER 8 8.1 Trait Theory in Historical Perspective
as 16 to as few as 2. Cattell also believed that the various traits were hierarchically organized, and he referred to source traits as the underlying psychological factors (e.g., see top level of Figure 8.2) and surface traits as those that are subsumed by the different factors and are most directly translatable to behaviors (e.g., see second level from the top in Figure 8.2).
This statistical approach to identifying the underlying factor structure of personality was seen as an advantage over organizational techniques that were more theory-driven (as was the case with the early contributors to the interpersonal circumplex, discussed in Chapter 7). Of course, the statistical reduction of the data was not atheoretical, as numerous assumptions would have to be made that would affect the number of factors that would be extracted—and even the degree to which the factors would be related to one another.
Understanding Factor Analysis
Suppose you were to tell me about a television show that you watch. Let’s use a well-known exam- ple such as American Idol. If you were to summarize in bullet points the primary theme or themes of American Idol, what would you say? Perhaps it might be characterized as a singing competition, with the winner getting a record deal. Perhaps it might be described as a show that highlights human triumph and failure. The emphasis could also be on the fact that it is a reality show or that the audience is involved by voting for and, ultimately, selecting the winners. The bottom line is that one could generate a number of basic themes to describe the show, and some themes may overlap, while some may be quite distinct. These themes would be considered a summary of the show, and they provide some organization to all of the data points describing the show.
In the same way, factor analysis starts with a large amount of directly measurable data and then reduces it down to a smaller number of unobserved units called (latent) factors, which are con- structed by grouping/organizing the items. The goal is to get to the fewest number of latent factors that capture the largest amount of the observable data. Factor analysis groups items by identifying items that are statistically (quantitatively) related, whereas in the above example, the groupings were thematically (qualitatively) related. Factors emerge when the item-level data are related to each other.
Two questions are often addressed using factor analysis:
1. What is the smallest number of factors needed to capture the majority of data? 2. If there are multiple factors (more than one), how are those factors related to each other?
Both of these questions were at the forefront of Cattell’s work (and were noted in the discus- sion of lexical analysis) and have continued as a source of discus- sion (e.g., Goldberg, 1993) and debate (see Borkenau & Osten- dorf, 1990; Church & Burke, 1994; Vassend & Skrondal, 1997; cf. Marsh et al., 2010).
Beyond the Text: Classic Writings
In this important text, L. L. Thurstone writes about the use of factor analysis in finding underlying factors in personality and ability testing. Read “The Vectors of the Mind” (1934) at http://psychclassics.yorku.ca/Thurstone/.
Reference: Thurstone, L. L. (1934). The vectors of the mind. Psychological Review, 41, 1–32.
Lec81110_08_c08_225-252.indd 230 5/21/15 12:40 PMhttp://psychclassics.yorku.ca/Thurstone/
CHAPTER 8 8.1 Trait Theory in Historical Perspective
Multitrait-Multimethod Assessments
Cattell believed that in order to thoroughly test the trait model, it was necessary to sample not only a wide range of traits, but also to collect data using different methodologies. Accordingly, he specified three sources of data that should be sampled:
1. Questionnaire data (Q-data), which takes the form of the traditional self-report invento- ries commonly used in psychology.
2. Life data (L-data), involving data culled from naturalistic settings. This can take the form of observations of behavior in the real world, or even objective information, such as number of divorces, arrests, or college degrees earned, to name a few.
3. Experimental data (T-data) involves data that is collected from standardized experiments. This represents the most objective and standardized data, and because the experimenter manipulates one of the variables of interest using random assignment, it is the only method to allow for causal conclusions.
Finding 16 Personality Factors
By factor analyzing data using each of these methodologies, Cattell argued that the weaknesses of one methodology are offset by the strengths of the other methodologies, thereby providing a more comprehensive picture of the individual’s personality.
Cattell began his analysis by paring Allport’s characteristics down to 171; he removed what he considered to be either redundant or rarely used terms. He then conducted numerous factor ana- lytic studies on trait assessments from each of the three methods, spanning a period of several decades, and concluded that there are 16 fundamental traits (or factors) that can be organized into a hierarchical framework and cover all human trait descriptors (Cattell, 1943):
1. Abstractedness 2. Apprehension 3. Dominance 4. Emotional stability 5. Liveliness 6. Openness to change 7. Perfectionism 8. Privateness
9. Reasoning 10. Rule consciousness 11. Self-reliance 12. Sensitivity 13. Social boldness 14. Tension 15. Vigilance 16. Warmth
Research also suggests that the 16 personality factors (PFs) can be captured in other cultures and languages as well (Prieto, Gouveia, & Fernandez, 1996; Schneewind & Graf, 1998), providing fur- ther evidence of their robustness. Cattell also suggested that these 16 personality factors apply throughout the lifespan, and accordingly, he developed the 16 Personality Factor Questionnaire (16PF®) measures for children, adolescents, and adults.
Because these are considered fundamental factors of personality, Cattell argued that everyone can be characterized by some combinations of these factors. Although Cattell published a number of papers on this topic, other researchers who followed suggested that traits could be reduced even further. Researchers have focused on a smaller number of traits, and these models have garnered more support and use in the field. One of those models was forwarded by Hans Eysenck.
Lec81110_08_c08_225-252.indd 231 5/21/15 12:40 PM
CHAPTER 8 8.1 Trait Theory in Historical Perspective
Eysenck’s Model of Personality Hans J. Eysenck was one of the most controversial and prolific researchers of the 20th century. Eysenck made significant contributions to a number of areas, but none more so than the area of personality psychology. Arguably one of his most lasting legacies was the founding of the journal Personality and Individual Differences, which was and continues to be the official journal of the Society for the Study of Individual Differences.
Emerging from the biological perspective, Eysenck believed that basic biological/genetic mecha- nisms underlie all human traits (see Chapter 4). From the late 1940s to the early 1950s, Eysenck studied monozygotic and dizygotic twins, concluding that neuroticism has a strong genetic com- ponent (Eysenck & Prell, 1951). Personality, explained Eysenck, is the sum of cognition, character, affect, and somatic components. He believed that the study of personality should be concerned with discovering the general laws of the group (nomothetic approach) as opposed to studying the individual (idiographic approach), as was the emphasis in psychoanalysis. In his research, Eysenck, like Cattell, favored the statistical techniques of factor analysis, which allows the researcher to reduce many variables to their essential factors. However, unlike Cattell, Eysenck arrived at a more economical model: a three-factor solution. Eysenck also differed from Cattell in that he examined traits as dichotomies (e.g., emotionally stable vs. unstable, introverted vs. extraverted, etc.).
A Three-Factor Solution
Using factor analysis and basing his model on the four humors described by the ancient Greeks, Eysenck rearranged the four humors on a continuum describable in terms of two personality dimensions: introverted-extroverted and unstable-stable. He placed the first dimension on a hori- zontal axis and the second on an intersecting vertical axis (see Figure 8.1). Eysenck believed that these two factors subsumed the primary descriptive features of normal human functioning and were the essence of personality (Eysenck, 1947). As per Figure 8.1, the term melancholic refers to someone high in neuroticism (N) but low in extraversion (E). The choleric person is high in N and high in E. The sanguine type was characterized as low in N and high in E, and the phlegmatic person was low on both. In subsequent years, Eysenck (Eysenck & Eysenck, 1985) added a third dimension, psychoticism (meaning recklessness, inappropriate emotional expression, hostility, disregard for common sense, and poor contact with reality), which he believed, along with the first two factors, was most relevant to the characterization of abnormal manifestations of per- sonality. Importantly, Eysenck believed that these three factors were statistically independent of each other, meaning that a score on one factor was unrelated to a score on another factor. For that reason, each factor is depicted at a 90-degree angle of the other factors. Thus, psychoticism/ reality contact would be depicted in a three-dimensional image (i.e., the two-dimensional image in Figure 8.1 shows low psychoticism, but variability on N and E). Much of Eysenck’s subsequent research was directed toward understanding the relationship between behavior and his personal- ity factor model.
Lec81110_08_c08_225-252.indd 232 5/21/15 12:40 PM
CHAPTER 8 8.1 Trait Theory in Historical Perspective
Figure 8.1: The “big two” factors and their relation to the four humors
Theorists have suggested that the basic personality trait terms used to characterize human behavior have not changed dramatically in over 2,000 years. The figure depicts how the four humors map on to two of the big five factors.
Source: From Eysenck, H.M. and Eysenck, M.W., Personality and Individual Difference: A Natural Scientific Approach, Plenum Publishing, 1985. Reprinted with kind permission from Springer Science + Business Media.
Connecting the Three Factors to Behavior
Eysenck not only developed a model to capture the breadth of personality functioning, but he also offered a hierarchy to explain the depth of these constructs. Eysenck clearly aligned himself with some of the basic tenets of behaviorists, and not surprisingly, when he developed a hierarchical model, he explicitly linked personality to behavior. At the top of the hierarchy are the factors; in the diagram in Figure 8.2, the factor of extraversion is depicted. The next level down shows the traits that are included in extraversion; in this diagram, sociability and liveliness are depicted (for a more complete list, see the traits highlighted in yellow and red in Figure 8.1). The third level down depicts habitual behavior; in this instance, regularly talking and smiling. At the lowest level, we find individual instances of behavior (i.e., one specific time when the person spoke to someone). In this way, Eysenck depicts how individual behaviors—ones that are habitually evidenced, are related to each other, and occur in clusters with other traits—define the factors. Eysenck believed that the highest levels of this model represent the most static components that are difficult to change. The lower levels of the hierarchy are those that are most easily changed and, if the change is consistent and broad enough, it can result in changes in personality functioning.
Emotionally Unstable (Neurotic)
Leadership
Emotionally Unstable (Neurotic)
Emotionally Stable
Emotionally Stable
ExtravertedExtravertedIntrovertedIntroverted
Melancholic Choleric
SanguinePhlegmatic Sociable
Outgoing Talkative
Responsive
Easygoing Lively
Carefree
Quiet
Moody
Anxious Rigid
Sober
Pessimistic Reserved
Unsociable
Calm
Passive Careful
Thoughtful Peaceful
Controlled Reliable
Even-tempered
Active
Touchy Restless
Aggressive Excitable
Changeable Impulsive
Optimistic
Lec81110_08_c08_225-252.indd 233 5/21/15 12:40 PM
CHAPTER 8 8.1 Trait Theory in Historical Perspective
Figure 8.2: Depicting the hierarchical structure of extraversion
The top of the hierarchy reflects the factors that are considerably less amenable to change. At the lowest level are the basic behaviors that are very amenable to change. Consistent behavioral enactments that occur in thematic clusters reflect the corresponding traits and factors.
In a similar vein, Eysenck defined the hierarchical structures for both neuroticism and psychoti- cism, indicating the specific behaviors, habits, and traits that lead to the factors that comprise them (see Eysenck, 1990).
Connecting the Three Factors to Biology
Eysenck believed that the factors were, in fact, driven by underlying biological mechanisms, thereby providing some rationale for why the factors are stable over time. Although this area was discussed in Chapter 4, the emphasis here will be on the behavioral tendencies and practices that are associated with extroversion, neuroticism, and psychoticism.
Extroversion-Introversion
Eysenck’s (1994) most extensive research focused on the biological mechanisms underlying extra- version. Eysenck believed that it was the brain’s need for stimulation that resulted in different patterns of behavior for introverts and extraverts, and he highlighted the Reticular Activating System (RAS), which regulates arousal in the cortex, as the specific part of the brain that would most readily demonstrate these differences. Eysenck hypothesized that extraverts are chronically understimulated, and they would therefore engage in behavior to stimulate their brains. In con- trast, he believed that introverts were chronically overstimulated, and they would therefore avoid stimulation.
Sociability
Regularly smiles
Regularly talks
Liveliness
Regularly active
Talked to stranger on
the bus
Individual Behavior:
Habit:
Trait:
Factor:
Spoke to Mary
yesterday
Extraversion
Lec81110_08_c08_225-252.indd 234 5/21/15 12:40 PM
CHAPTER 8 8.1 Trait Theory in Historical Perspective
Eysenck’s original theoretical model proved to be close, but not quite accurate, and he subse- quently suggested that arousability (rather than baseline arousal) might be the more central dis- tinguishing feature (Eysenck, 1994). Indeed, the research literature suggests that in the absence of stimulation, there may be few if any differences between introverts and extraverts at baseline (Revelle, Humphreys, Simon, & Gilliland, 1980; for a review, see Matthews & Gilliland, 1999). For example, in one study, researchers had participants imagine being in a positive and a neutral situa- tion, and then they rated the moods of the participants. Those identified as extraverts using a stan- dardized measure rated their mood as more positive relative to introverts in the positive situation, but the introverts and extraverts did not differ in the neutral condition (Larsen & Ketelaar, 1991). Although the effects appear small, the literature does suggest that there are differences between introverts and extraverts in their potential for, or responsiveness to, arousal (i.e., arousability; see Bullock & Gilliland, 1993). For example, introverts work better when exposed to less background noise (Geen, 1984), and extraverted students are more likely to study in environments with more opportunities for stimulation (Campbell & Hawley, 1982). Similarly, extraverts performed better than introverts on a GRE-type test when stimulated with caffeine (Revelle, Amaral, & Turri, 1976; see also Eysenck, 1994). Interestingly, research has also provided a plausible biochemical expla- nation that connects extraversion and the sensory modulation to explain differences in how the external world is experienced (Rammsayer & Stahl, 2004; Stahl & Rammsayer, 2008). However, a recent study using advanced statistical procedures (structural equations modeling) refuted many of the findings (at least those based on EEG technology); suggesting that external factors have only minimal impact on EEG readings and are not significantly related to extraversion (Hagemann & Naumann, 2009). Thus, the findings in the literature remain somewhat mixed.
Neuroticism
Recall that neuroticism is essentially emotional instability, and neurotic individuals have been characterized as having nervous, negative, anxious, and self-pitying qualities (McCrae & John, 1992). Those scoring in the clinical range on neuroticism are also more likely to apply negative interpretations to ambiguous stimuli. For example, when selecting from pairs of homophones, highly neurotic individuals are more likely to choose the aversive option (e.g., “die” s. “dye”) as compared to their less-neurotic peers (Eysenck, MacLeod, & Matthews, 1987).
Eysenck (1967, 1990) hypothesized that neurotic individuals would show greater responsiveness in the limbic system relative to those who are more emotionally stable. Research findings on neu- roticism indicate that those who score high on neuroticism are more likely to exhibit excitation of the autonomic nervous system, display behavior that is not as readily apparent in others, and have higher drive than normal (for a summary, see Eysenck, 1994). Eysenck (1967) has also sug- gested that variability in the responsiveness of the limbic system is most apparent in emotional situations, resulting in the individual responding neurotically to stress. However, recent empirical tests of Eysenck’s hypothesis have been less favorable (e.g., Beattie & Corr, 2010).
Neuroticism has also been consistently associated with aversive emotional outcomes, such as the incidence of psychiatric diagnoses (Malouff et al., 2005; Saulsman & Page, 2004) and physical health problems and higher rates of mortality (Lahey, 2008). Even measures of skin conductance (Norris, Larsen, & Cacioppo, 2007) and fMRIs (Canli et al., 2004) show greater responsiveness in those high on neuroticism.
Lec81110_08_c08_225-252.indd 235 5/21/15 12:40 PM
CHAPTER 8 8.1 Trait Theory in Historical Perspective
Years of empirical investigation have led researchers to conclude that there are implications for neuroticism in terms of behaviors in school and work-related settings. For example, neuroticism has been associated with poor time management, as neurotic individuals appear to have poorer study habits (Bond & Feather, 1988) and less adaptability in academic settings (Martin, Neiad, Colmar, & Liem, 2013), as well as a higher incidence of negative life events while transitioning through the university setting (Lüdtke, Roberts, Trautwein, & Nagy, 2011). Researchers studying work-related behaviors and attitudes in over 400 employees in different work settings found that neuroticism was inversely related to presenteeism, and resulted in lower productivity (Smillie, Yeo, Furnham, & Jackson, 2006). Neuroticism is also associated with poorer academic perfor- mance (e.g., Chamorro-Premuzic & Furnham, 2008; O’Connor & Paunonen, 2007), though the association may be modest (McAbee & Oswald, 2013).
Because neurotic behavior has important clinical implications, Eysenck also wrote about methods for changing the behaviors associated with neuroticism. Even though neuroticism and the other personality factors were considered to be greatly affected by genetics (e.g., Pedersen, Plomin, McClearn, & Friberg, 1988) and shown to be stable (Costa & McCrae, 1990), Eysenck still believed that the traits could be modified through therapy. Specifically, Eysenck (1947, 1953, 1960a) believed that neurotic patterns are learned and that it is possible to uncondition them (i.e., either through counter-conditioning or extinction; see Chapter 5). He made extensive use of learning theory in his quest to understand and predict human personality dynamics. He thought that per- sonality could be restructured according to the same learning principles on which it was based.
Psychoticism
As noted, Eysenck thought psychoticism was the factor most relevant for distinguishing between normal and non-normal manifestations of personality. It involves acting impulsively and with aggression and is related to contact with reality. Those scoring high on psychoticism are charac- terized as being cold, unemotional, antisocial, paranoid, and lacking in both empathy and insight (Eysenck & Eysenck, 1976). Eysenck hypothesized that psychoticism lies on a continuum, with low psychoticism defining normal functioning, antisocial behavior defining intermediate psychoti- cism, and susceptibility to psychosis (e.g., schizophrenia) defining extreme psychosis (Eysenck & Eysenck, 1976, p. 203). Thus, Eysenck believed that these diagnoses differed only in the degree to which they vary along the factor of psychoticism, rather than assuming that they were dis- tinct entities. Moreover, research suggests that psychoticism has a significant genetic component (Eaves, Eysenck, & Martin, 1989; Gattaz, 1981; Lester, 1989); though the magnitude of the genetic contribution may vary depending on the specific items used from the psychoticism scale of the Eysenck Personality Questionnaire (Heath & Martin, 1990).
Summary of Eysenck’s Work
Overall, it appears that the data are mixed with respect to supporting Eysenck’s views on per- sonality. Eysenck presented a strong challenge to the field of psychotherapy and became a very controversial figure. For one thing, his review of the research led him to the conclusion that cur- rent methods of psychotherapy were ineffective—a conclusion that motivated the psychotherapy research movement to prove him wrong. His writings also seemed to support the belief that intel- ligence is largely inherited—a fact that led to his being accused of being a racist (Buchanan, 2010).
Lec81110_08_c08_225-252.indd 236 5/21/15 12:40 PM
CHAPTER 8 8.2 Convergence on the Big Five
8.2 Convergence on the Big Five
More recently, personality researchers have converged on a small set of personality factors that appear to be recoverable from a wide range of sources, including other measures that purport to assess a larger number of factors (e.g., 16PF®, MBTI®, etc.; Cattell, Cattell, & Cattell, 1993; Noller, Law, & Comrey, 1987).
Collectively known as the Big Five (or Five Factor Model), this model represents a descriptive taxonomy of personality traits. Like previous factorial models, the Big Five provides a simplified framework for understanding personality and for describing situational and temporal consistency (John, Naumann, & Soto, 2008). The Big Five consists of five factors or traits, usually labeled as extraversion, agreeableness, conscientiousness, neuroticism (or its reverse, emotional stability), and openness to experience (or intellectance; Digman, 1990). Extraversion and neuroticism have already been discussed earlier in this chapter, leaving the remaining three factors to be discussed in the subsequent sections. These big five factors have been studied in a wide range of research and applied contexts, and as was the case with earlier reviewed research, the Big Five have been shown to relate to academic accomplishments (e.g., O’Connor & Paunonen, 2007; Poropat, 2009; Richardson, Abraham, & Bond, 2012; Trapmann, Hell, Hirn, & Schuler, 2007) as well as work-related outcomes and performance (e.g., Barrick & Mount, 1991; Tett, Jackson, & Rothstein, 1991).
Openness to New Experience Openness to new experience involves traits such as a sensitivity for aesthetics, imaginativeness, preference for variety, and intellectual curiosity. This is often considered the personality version of creativity and thus is related to play- fulness and a high tolerance for uncer- tainty (Costa & McCrae, 1992b, 2009). Research also suggests that openness is related to sexual behaviors and atti- tudes, as open individuals tend to have more liberal views about sex, have more sexual partners, a wider range of sexual experiences (McCrae, 1994), and greater satisfaction in their sex lives, at least for women (McCrae & Sutin, 2009). Thus, individuals scoring low on this factor are characterized as more traditional, con- ventional, dogmatic, and authoritarian in their thinking and attitudes (Butler, 2000). Those scoring low are also more conservative in their political views, less tolerant of diver- sity, and even more prejudicial in their attitudes (e.g., Sibley & Duckitt, 2000). Unlike some of the other big five factors, openness to new experience is not significantly related to mental health (Malouff, Thorsteinsson, & Schutte, 2005) or quality of life measures (Steel, Schmidt, & Schultz, 2008). Thus, despite clear differences in how people experience the world, in the end, these dif- ferences do not lead to appreciable differences in psychological well-being.
Pinnacle Pictures/Photodisc/Thinkstock
Openness to new experience reflects a personality trait whereby individuals are not only interested in but actually enjoy trying novel and diverse experiences.
Lec81110_08_c08_225-252.indd 237 5/21/15 12:40 PM
CHAPTER 8 8.2 Convergence on the Big Five
Agreeableness The agreeableness factor primarily involves the trait of friendliness and captures the extent to which one is accommodating. Agreeableness involves traits such as trust, altruism, and modesty. In order to minimize conflict with others, agreeable individuals tend to conform with others, and they are more likely to compromise their beliefs and attitudes. In contrast, those scoring low on this factor show less empathy for others, and they are skeptical, suspicious, rude, and unfriendly (e.g., Costa & McCrae, 1991; Graziano & Eisenberg, 1997).
Given the nature of agreeableness, it holds the biggest implications for interpersonal relation- ships. For example, agreeable individuals are more likely to engage in prosocial (helping) behavior across a variety of situations, and even in the absence of external motives for helping—sometimes referred to as altruistic behavior (Graziano, Habashi, Sheese, & Tobin, 2007; Penner, Fritzsche, Craiger, & Freifeld, 1995). One reason agreeable individuals are interpersonally effective is that they appear to be prone to respond emotionally to the needs of others by detecting the emotional needs of others (Tobin, Graziano, Vanman, & Tassinary, 2000).
In addition to manifesting favorable behaviors, emotions, and attitudes, agreeable individuals also minimize the expression of negative reactions and behaviors. Thus, agreeableness is associated with cooperativeness, whereas low agreeableness is associated with competitiveness and conflict (Graziano, Jensen-Campbell, & Hair, 1996; Jensen-Campbell & Graziano, 2001). Some researchers have concluded that agreeable individuals value the relationships over other outcomes (Graziano et al., 1996). Not surprisingly, those low on agreeableness are characterized as hostile and aggres- sive, and these individuals are more likely to experience problems with mental health (Costa & McCrae, 1992b, 2008).
Despite the consistency of the Big Five, the one factor that may be inadequate is agreeableness. For example, researchers have suggested and found corroborative evidence for the fact that traits such as honesty and humility are not adequately captured by the agreeableness factor and merit consideration as an independent (sixth) factor (e.g., Ashton et al., 2004; Ashton, Lee, & Son, 2000). Moreover, it also appears to be the case that the Big Five fail to adequately capture traits such as narcissism and psychopathy (i.e., these are more than simply low agreeableness). Thus, the single factor of agreeableness may be insufficient to capture this variability.
Conscientiousness Conscientiousness captures traits such as diligence, organization, punctuality, scrupulousness, self-discipline, thoroughness, deliberation, and need for achievement (Costa & McCrae, 1991). It also includes such traits as perceived self-efficacy, rule consciousness, and internal locus of control (DeNeve & Cooper, 1998). Of the big five factors, conscientiousness is the one most closely tied to favorable outcomes in a number of domains. For example, conscientious individuals are more likely to be successful in school and in work settings relative to their less conscientious peers (Hig- gins, Peterson, Lee, & Pihl, 2007). This effect is driven largely by the fact that conscientiousness predicts hard work and effort in a variety of settings. That is, the predictive validity of conscien- tiousness is not setting or job-specific.
Lec81110_08_c08_225-252.indd 238 5/21/15 12:40 PM
CHAPTER 8 8.2 Convergence on the Big Five
Conscientiousness is also a positive predictor for physical and mental health. Specifically, highly conscientious individuals engage in more proactive health behavior (Roberts, Jackson, Favard, Edmonds, & Meints, 2009) and have a longer lifespan. In fact, over 20 published studies have doc- umented the association between conscientiousness and longevity (Kern & Friedman, 2008). In a recent study that attempted to establish a clear connection between the two variables, research- ers prospectively studied over 700 individuals for a period of 40 years, ranging from childhood to adulthood (Hampson, Edmonds, Goldberg, Dubanoski, & Hillier, 2013). After measuring conscien- tiousness in childhood (mean age 10), researchers tracked the participants, who then had com- prehensive medical exams in adulthood (mean age 51). Childhood conscientiousness was shown to predict objective measures of adult health, such as fasting blood sugar, body mass index, blood pressure, and the like (Hampson et al., 2013; see also Friedman, Kern, Hampson, & Duckworth, 2012).
Conscientiousness has also been related to measures of well-being and life satisfaction (Steel, Schmidt, & Shultz, 2008), and this factor predicts success in relationships as well, with conscien- tiousness linked to marital idealization (O’Rourke, Neufeld, Claxton, & Smith, 2010). Low consci- entiousness is also more likely to result in such problematic behavior and outcomes as homeless- ness, drug use, imprisonment, and unemployment (Roberts et al., 2009).
Big Five in Cultural Context The bulk of the research on the Big Five (and any trait or other factor) comes from standardized self-report measures (e.g., Goldberg, 1981). However, the five factors appear to emerge even when the data are collected from other sources, such as peer ratings and spousal ratings (Costa & McCrae, 1980; Digman & Takemoto-Chock, 1981; McCrae & Costa, 1985, 1990), and even when the peer raters may be less familiar with the person they are rating (e.g., Passini & Norman, 1966).
Despite some methodological discrepancies, it is reasonable to conclude that the Big Five are generally replicable, with only a limited number of exceptions, such as when descriptive words do not have equivalent translations in other languages (e.g., Benet-Martinez & Oishi, 2008; McCrae & Costa, 2008). Indeed, in one of the larger analyses that covered seven European and Asian languages, researchers were able to extract all of the big five factors, though they suggested that agreeableness was better defined as two separate factors (honesty and humility) (Ashton et al., 2004).
Heritability of the Big Five Research supports the conclusion that the big five factors are highly heritable (e.g., Bergeman et al., 1993; Bouchard & McGue, 2003; Yamagata et al., 2006), and this appears especially true for facet scores (Jang, Livesley, Angleitner, Riemann, & Vernon, 2002; Jang, McCrae, Angleitner, Rie- mann, & Livesley, 1998).
Lec81110_08_c08_225-252.indd 239 5/21/15 12:40 PM
CHAPTER 8 8.2 Convergence on the Big Five
It is also important to note that when considering the heritability of the Big Five, the answer does not appear to be the same for all five factors. Specifically, it appears that extraversion and neuroticism (some- times referred to as the “big two”) have significantly higher heritability coefficients relative to the other three. Of course, analyses of the genetic influence on the Big Five also clearly indicate that considerable vari- ability is predicted by environmental influences as well (Plomin, 1989). See Chapter 4 for more detailed cover- age of the biology and genetics of personality.
The Big Five Over the Lifespan From a theoretical standpoint, it is hypothesized that scores on the Big Five (personality) should be stable because genetics (which are stable) contrib- ute significantly to personality and because there will likely be some increasing stability as a func- tion of individuals being able to exercise choice over their environments (e.g., Caspi, Roberts, & Shiner, 2005). Caspi and colleagues have termed this a cumulative continuity hypothesis, whereby individu- als choose environments that are consistent with their
personalities (something that they can do more easily as they move out of childhood), and these environments essentially reinforce their personalities (resulting in greater stability). Thus, an extraverted individual would be more likely to select environments that allow for social interac- tions and, in turn, foster further extraverted behavior. This hypothesis allows for the interaction of genetic and environmental factors to promote personality stability (see also Kandler et al., 2010).
Several meta-analyses (a statistical method for summarizing the effects observed over many dif- ferent studies) have not only found general stability, but, consistent with the cumulative continu- ity hypotheses, have shown that stability typically increases from childhood to adulthood (Fergu- son, 2010; Roberts & Delvecchio, 2000).
One of the more impressive studies examining personality stability involved the collection of cross-sectional data from over 1.25 million individuals from around the world, aged 10 to 65 (Soto, John, Gosling, & Potter, 2011). The authors were able to arrive at several conclusions, including the fact that during adolescence and childhood, the most pronounced changes occur and gender differences begin to emerge (Soto et al., 2011). The data also indicate that neuroticism increases for females as they move from late childhood to adolescence, whereas this trend does not occur for males. Openness to new experience decreases into adolescence, then increases into the col- lege years. Thus, these findings again support the fact that personality stability is greatest in later adulthood.
Fuse/Thinkstock
Extraversion appears to be one of the big five factors with the highest level of heritability.
Lec81110_08_c08_225-252.indd 240 5/21/15 12:40 PM
CHAPTER 8 8.3 The Person-Situation Debate
As suggested by the above-mentioned study, the question of stability for the big five factors across the lifespan may also depend, to some extent, on which of them is being considered. In a 40-year longitudinal study of almost 800 individuals (Hampson & Goldberg, 2006), the highest stability across the lifespan emerged for extraversion and conscientiousness, with lower stability for open- ness to new experience, agreeableness, and neuroticism, in that order (Hampson & Goldberg, 2006).
One final issue to consider is whether the stability coefficients tend to show any further change when considering later life. Two recently published studies suggest that although personality sta- bility increases from childhood to adolescence to adulthood, there may be a trend toward less stability at the last stages in life (see Ardelt, 2000). For example, a recent German study that exam- ined rank-order stability beyond age 60 found that peak stability emerged in adulthood, but then decreased with age thereafter (Lucas & Donnellan, 2011). Similarly, a large-scale study examining over 13,000 Australians found a U-shaped function over the lifespan, with the highest stability occurring in middle age and relatively lower values in adolescence and later life (Wortman, Lucas, & Donnellan, 2012).
8.3 The Person-Situation Debate
If someone wanted to know how you would behave at a party, would it be better to know the context of the party you were to be attending or to know something about you? The stability of personality across situations was a widely accepted assumption when trait research first became popular. However, researchers began to question this assumption, cham- pioning instead the situation, and this culminated in Walter Mischel’s 1968 book, Personality and Assessment, in which he critiqued the trait approach on two grounds:
1. Traits were limited in their predictive utility. 2. Traits were simply descriptors, not explanatory accounts.
The first criticism was drawn in large part from Mischel’s own data examining the cross-situational consistency of children (Mischel, 1968). Mischel concluded that traits can only predict approxi- mately 9% of the variance in behavior (which corresponds to correlations of .30), and he consid- ered this to be a small amount of explained variability in behavior. Mischel concluded that the poor predictive power of traits highlights the fact that the situation is a critical element in predict- ing how the person will act. Mischel’s view was that psychologists should not consider the situa- tion as error variance (i.e., “noise”) when examining traits (this reflected the common practice of trait researchers at that time). Instead, his approach involved considering the situation or context as a relevant variable, equally, if not more, important than the trait. Mischel argued that more robust consistency would be observed within specific situational contexts (e.g., Mischel, 2004).
Lec81110_08_c08_225-252.indd 241 5/21/15 12:40 PM
CHAPTER 8 8.3 The Person-Situation Debate
Mischel’s second critique was perhaps more damaging, in that he correctly noted that much of the trait research occurring was theoretical in nature (Mischel, 1968). That is, to conclude that a particular set of behaviors occurs simply because a person is extraverted or neurotic provides no explanation or theory; rather, it merely labels the behavior. In fact, the end result is a tautology: Why is John behaving in an outgoing and gregarious manner? Because he is extraverted. Why is he extraverted? Because he engages in outgoing and gregarious acts.
These criticisms had a significant impact on the field, and for a short period of time, trait research retreated to the shadows. Of course, trait researchers eventually responded to Mischel’s critiques, especially the first point, and, as we just saw, research on traits continues today. But Mischel’s criticisms played a key role in advancing the field.
Responses to Mischel’s Critique of Trait Psychology After Mischel voiced his objections to trait research, new measures for assessing personality emerged that helped to improve predictive validity. Also, alternative statistical and methodologi- cal procedures were developed to help address these criticisms. Moreover, researchers coun- tered with a number of sophisticated points targeting Mischel’s critique, and focused on such issues as Mischel’s omission of better personality studies, where the predictive validity of behav- ior exceeded .30 (e.g., Block, 1977). Researchers also highlighted Mischel’s tendency to predict very specific behaviors using traits, rather than predicting aggregated behavior (Epstein & O’Brien, 1985). This latter point is important and worth elaborating.
Imagine a study in which you objectively measure someone’s behavior in three distinct situational contexts (e.g., a student’s socialization behavior at home with her family, in the classroom, and with her friends while at a party). The researcher then measures the individual’s tendency to socialize and interact with others (e.g., assessing extraversion). The researcher then attempts to predict the individual’s behavior in one of the three settings (e.g., in the classroom) by using his or her extraver- sion score. The likely conclusion is that knowing how people typically behave in the classroom (i.e., the situational information) will provide the best predictor for behaviors in that setting. On the sur- face, this would seem to support the situational perspective. In contrast, if the goal were to predict how the individual would behave on average (i.e., typically over time and over different situations), then the personality trait would be the optimal predictor. The counterviews to Mischel and others are nicely summarized by Kenrick & Funder (1988), as they systematically address and debunk a series of conclusions that would otherwise prove problematic to trait researchers.
Understanding Situational Strength, Domain Breadth, and Trait Relevance Some situations influence an individual’s behavior more than others, and this has been referred to as situational strength (Mischel, 2004). For example, some situations have strong situational scripts, where the cues in the environment strongly and clearly dictate how one should act in that setting. The situational script is a form of psychological pressure on the individual to engage in or refrain from particular behaviors. As an example, consider the situational script for riding in an elevator. Individuals enter, push the button for the desired floor, then stand facing forward and looking at the lights that indicate the current floor. Even the distance at which you stand apart from others is quite prescribed. In all likelihood, a flaming extravert and the most reclusive
Lec81110_08_c08_225-252.indd 242 5/21/15 12:40 PM
CHAPTER 8 8.3 The Person-Situation Debate
introvert would behave similarly in this setting because of the strength of the situational script. Thus, the influence of personality is significantly reduced in this setting.
In contrast, when the situational script is weak or largely absent, then person- ality factors should play a more sig- nificant role in predicting behaviors. To illustrate the contrast between strong and weak situational scripts, consider the example of dating behavior. A first date would have a relatively strong situ- ational script, and the observed behav- ior is, for the most part, going to reflect conservative, cautious, even superficial, but perhaps highly friendly behavior. Again, observing first-date behavior will tell us less about the person. In contrast, a tenth date has a much weaker (more ambiguous) situational script, as there are fewer expectations for what should
or should not occur. As a result, we are likely to see behaviors that are more directly influenced by the individual’s personality.
It is also the case that more specific traits can predict best in specific situations. For example, work LOC was a better predictor of outcomes in the work setting relative to broader non-specific LOC beliefs (Wang et al., 2010), and the same was true for health-specific LOC and health outcomes (Johansson et al., 2001). Higher predictive validity for traits that are domain-specific also bridges the gap in the person-situation debate by identifying the specific situational contexts where traits predict best.
Finally, the person-situation debate can also be addressed by considering the relevance of a trait to the individual (e.g., Bem & Allen, 1974). The argument is that when individuals demonstrate greater cross-situation consistency, it is because their behaviors are being governed by traits with greater personal relevance (see Kenrick, McCreath, Govern, King, & Bordin, 1990; Zuckerman et al., 1988).
The Role of the Fundamental Attribution Error Imagine that you find yourself looking for a parking space on campus on a rainy day. There are long creeping lines of cars searching for spots because, it seems, everyone decided to drive today. Fortunately, you come across a spot—but you may not have been the first one to notice it and signal for it. Nevertheless, you pull into the spot. As you head off to class, other drivers, who are obviously upset with your having taken the spot out of turn, honk their horns at you (and you
MokeyBusiness Images/iStockphoto/Thinkstock
If this were a first date, it is not likely that these individuals would learn much about each others’ personalities. Instead, their behaviors would be driven by the strong situational script of a first date. However, if this were a tenth date, their personalities would likely be much more evident.
Lec81110_08_c08_225-252.indd 243 5/21/15 12:40 PM
CHAPTER 8 8.4 Supplementing the Big Five With Complementary Approaches
can imagine what they are thinking). It’s likely the case that you do not consider yourself a “bad” person; instead you emphasize the situation that resulted in that behavior (e.g., “I don’t normally do that”). Perhaps you even question the legitimacy of the complaining behavior by the other drivers. Now what if you were on the receiving end of this violation of parking etiquette? As you honk your horn at the offending party, you are likely to think quite negatively about the person’s character—instead of thinking that the person was in an urgent rush, or being absent-minded, or perhaps is now even regretful of the decision he just made.
This situation illustrates an interesting phenomenon in the psychological literature. When we attempt to explain our own behavior, there is a tendency to emphasize the influence of situational factors. However, when explaining the behaviors of others, the emphasis tends to be on the influ- ence of traits. This is known as the fundamental attribution error (see Jones & Harris, 1967), and it is relevant in research on traits because it reveals how self-evaluations and evaluations by oth- ers can vary, not just due to error, but because of real and predictable differences in perspective. This concept from social psychology highlights the value of collecting data using multiple sources. Thus, who is assessing traits could affect the determination of whether the situation or the person is emphasized.
Summary of Person-Situation Debate Most personality researchers have now concluded that both the person and the situation contrib- ute to behavior. Specifically, situational variables are more effective when it comes to predicting behavior in specific situations, especially when dealing with settings that have strong situational scripts and traits that are low in relevance to the individual. In contrast, when predicting an aggre- gate of behavior, especially in settings with weak situational scripts and when dealing with highly relevant traits, traits should predict quite well. This reflects the interaction between traits (the person) and the situation.
8.4 Supplementing the Big Five With Complementary Approaches
In addition to the improvement of personality measures and the research regarding the person- situation debate, additional lines of research have emerged as complementary responses that have broadened the scope of how personality can be construed. Two such approaches are briefly reviewed here.
Projects, Life Tasks, Concerns, Strivings, and Goals: An Idiographic Approach This chapter has largely focused on the nomothetic approach to personality, meaning the identifi- cation of discussion of the traits that are common to everyone (e.g., the Big Five). The perspective to be presented here suggests that to fully understand the individual, one must also consider an idiographic approach, which focuses on what is unique to the individual. This approach adopts a broader perspective, as the need to identify what is unique involves capturing the contributions of traits, the demands of the situation, and even the motives of the individual. One approach for capturing personality from this broad perspective (sometimes referred to as a grand systems per- spective), is to focus on units of analysis that at their center assess motivation within the context of life (e.g., McAdams, 1997).
Lec81110_08_c08_225-252.indd 244 5/21/15 12:40 PM
CHAPTER 8 8.4 Supplementing the Big Five With Complementary Approaches
Beginning in the 1970s, and then more prominently in the 1980s, researchers began to develop a series of measures that tap into people’s idiosyncratic manifestations of (motivated) behavior; in other words, personal goal pursuit emerged as a new approach for understanding personality. Also referred to as personal action units (Little, Lecci, & Wakinson, 1992), they include the constructs of current concerns (Klinger, 1977, 1989), personal projects (Little, 1983, 1989), life tasks (Cantor & Kihlstrom, 1987), personal strivings (Emmons, 1986), and personal goals (Karoly & Ruehlman, 1995; see Buss & Cantor, 1989, for a review of these constructs; see also Little, Salmela-Aro, & Phillips, 2007). These units represent intentional actions in the context of our lives and provide a distinct perspective for understanding personality and interpersonal processes. Personal proj- ects are extended sets of personally meaningful behavior, and they range from the simplest of daily activities (e.g., “read Chapter 8”) to the monumental endeavors of a lifetime (e.g., “find the ‘meaning of my life’ ”) (Little, 1989). The theory is that how people undertake the challenges of life, how traits manifest in daily living, and how broad motivations (e.g., achievement, autonomy, etc.) uniquely play out in the individual is a reflection of their personality.
One of the primary reasons for the popularity of the goal perspective is that it addresses the full range of disciplines that have been explored in personality science, including neuroscience, genet- ics, evolution, social psychology, and even more traditional influences, such as the trait approach (e.g., Little, 2005). Idiographic goals provide a conceptual tool that allows for each of these factors to converge on the individual and be represented in a single construct (Little, 2005, 2006).
From a theoretical standpoint, personality traits and personal action units have respectively been equated with the “havings” and “doings” of personality (Can- tor, 1990). According to Cantor, if traits represent what we have to work with (e.g., biological predispositions reflected in basic trait terms), then personal goals rep- resent what we do with what we have. Traits tap what is largely stable, whereas personal goal methodology can reflect what is changeable (Cantor, 1990).
Numerous studies have been conducted to demon- strate the breath of outcomes that can be predicted using the “doing” side of personality, and they are sim- ilar to those predicted by traits. For example, personal goal constructs can predict academic adjustment (Lit- tle, Lecci, & Watkins, 1992), life satisfaction (e.g., Palys & Little, 1983), meaning in life (e.g., McGregor & Little, 1998), as well as subjective well-being (e.g., Omodei & Wearing, 1990), physical symptoms (King & Emmons, 1990), pain (Karoly & Ruehlman, 1996), health fears (Karoly & Lecci, 1993; Lecci, Karoly, Ruehlman, & Lan- yon, 1996), and even depression (Lecci, Karoly, Briggs, & Kuhn, 1994).
Design Pics/Thinkstock
Are these individuals driven by a need to perform well or to learn? If it’s the former, then grades will be paramount. If it’s the latter, then grades become secondary to the process of learning.
Lec81110_08_c08_225-252.indd 245 5/21/15 12:40 PM
CHAPTER 8 8.4 Supplementing the Big Five With Complementary Approaches
Researchers have also examined different types of goals and how they relate to academic perfor- mance. For example, Dweck and colleagues differentiate performance goals (where the focus is impression management or how you appear to others) from learning goals (where your focus is on accumulation of information to improve the quality of your work) (e.g., Dweck & Leggett, 1988; Olson & Dweck, 2008). Experimental research on students in which either performance or learn- ing goals were induced indicates that performance goals are more likely to lead to anxiety when undertaking a difficult task (see Elliot & Dweck, 1988). This is important experimental work, as it establishes a pathway whereby trait behavior could result in specific goals, and specific goals can reinforce and strengthen specific traits. For example, highly stressful goals with a low probability of working out are more likely to be evidenced by the student high in neuroticism and low in con- scientiousness (see Little et al., 1992).
Researchers have also differentiated approach goals and avoidance goals; the former denotes goals that are desirable outcomes to move toward, whereas the latter denotes goals of avoiding undesirable outcomes (Elliot, 2006). As an example, consider a batter who steps to the plate in the bottom of the ninth inning with two out: “Don’t strike out” would be considered an avoid- ance goal, whereas “Get a hit” would be an example of an approach goal. This may even result in subtle differences in behavior (e.g., swinging defensively versus swinging assertively) and ultimate success. Importantly, however, how goals are formulated predicts both psychological and physical well-being, with avoidance goals being more deleterious (Elliot, Sheldon, & Church, 1997; Elliot & Sheldon, 1998). Not surprisingly, research shows that approach goals are associated with extraver- sion, whereas avoidance goals are associated with neuroticism (Elliot & Thrash, 2002).
Goals have also been differentiated with respect to their authenticity, such that a goal that is highly consistent with one’s core values and sense of self is considered authentic, and this, in turn, results in higher psychological well-being (McGregor & Little, 1998; Sheldon, Ryan, Rawsthorne, & Iardi, 1997) and is positively correlated with openness to new experience (Little et al., 1992).
Acting Out of Character: Defensive Pessimism and Pseudo-Extraversion
Much in the same way that personal action units can illustrate and complement how traits play out in our environments, these units of analysis can also help explain when individuals behave in ways that appear incongruent with their fundamental traits. The assumption is that some indi- viduals engage in motivated behavior (personally defined goals) that helps optimize how their traits can achieve desired outcomes.
For example, perhaps you have a friend who, despite being one of the top grade-earning students, always verbalizes and shows concern over upcoming tests. Despite these concerns, this person repeatedly sets the curve in the class. This approach has been conceptualized as defensive- pessimism, which is when an individual adopts the strategy of firmly believing the worst, despite the fact they have experienced considerable success in a particular domain in life. Importantly, defensive-pessimists differ from both pessimists and optimists in that they verbalize the same negative expectations as pessimists, but they achieve the same favorable outcomes as optimists. The literature has established that optimism and pessimism (dispositions largely subsumed by the factor of neuroticism) do predict a wide range of outcomes, such as measures of health and well- being (e.g., Giltay, Zitman, & Kromhout, 2006; Kim, Park, & Peterson, 2011; Kubzansky, Sparrow, Vokonas, & Kawachi, 2001; Scheier et al., 1989; Tindle et al., 2009), and the pattern of pessimism is similar to that observed for neuroticism. Therefore, it is particularly interesting to note that
Lec81110_08_c08_225-252.indd 246 5/21/15 12:40 PM
CHAPTER 8 8.4 Supplementing the Big Five With Complementary Approaches
some individuals adopt what is usually a maladaptive strategy (pessimism), but they nevertheless succeed.
Researchers studying these individuals in the context of the first-year university experience have found that one of the key variables differentiating the defensive-pessimist is his or her goal pur- suits (Cantor, Norem, Niedenthal, Langston, & Brower, 1987). For example, defensive-pessimists and optimists both find achievement-related goals (life tasks) to be rewarding and important, and they both show great persistence, even when confronted with obstacles to goal achievement (Norem, 1989; see also Martin, Marsh, Williamson, & Debus, 2003). Indeed, defensive-pessimists appear to benefit from the experience of anxiety, possibly self-inducing anxiety as a form of moti- vation, which clearly distinguishes them from optimists (Norem & Cantor, 1986). They also appear to worry about failure to a greater degree than optimists, but they use this negative emotion in constructive ways.
Let us also return to the example of John that was introduced at the start of the chapter. Recall that John presents as an extra- vert, despite the fact that he appears to be hard-wired as an introvert; a combination that might be labeled pseudo- extraversion. Thus, his trait-like disposition is that of an introvert (what was conceptually referred to as the “havings”), but he has important personal goals that allow him to override that social discomfort he experiences, at
least for short periods of time, to engage in the “doings” (Little, 2000; Little & Joseph, 2007). In other words, what John does with what he has is very different than what the typical introvert does with what he or she has. Thus, goals can help account for why similar traits result in different presentations (Cantor, 1990).
Act-Frequency Approach One additional approach for conceptualizing personality is to focus on the frequency with which specific behaviors are enacted, and this perspective is referred to as the act-frequency approach (Buss & Craik, 1983). The act-frequency approach would argue that if one frequently engages in behaviors that most people see as prototypical of a particular trait, then that individual can be characterized by that trait. In contrast to the personal goal approach, which emphasizes act- saliency (i.e., the personally meaningful behavior of the individual), this approach focuses on the frequency of actions. Thus, by frequently enacting any specific set of behaviors, one can come to be labeled by the trait capturing those behaviors. As an example, if one frequently engages in prototypically dominant acts (e.g., cutting people off in conversations, making decisions for oth- ers, etc.), then that person can be assigned the trait label “dominant.”
Beyond the Text: Classic Writings
In this engaging piece, Brian Little (2010) describes how traits and personal projects (goals) interact to reflect the complexities of our lives (including when we seem to act out of character). Read it at www.brianrlittle.com/arti cles/acting-out-of-character-in-the-immortal-profession -toward-a-free-trait-agreement/#more-484.
Reference: Little, B. R. (2010). Acting out of character in the immortal profession: Toward a free trait agreement. Aca- demic Matters.
Lec81110_08_c08_225-252.indd 247 5/21/15 12:40 PMwww.brianrlittle.com/articles/acting-out-of-character-in-the-immortal-profession-toward-a-free-trait-agreement/#more-484www.brianrlittle.com/articles/acting-out-of-character-in-the-immortal-profession-toward-a-free-trait-agreement/#more-484www.brianrlittle.com/articles/acting-out-of-character-in-the-immortal-profession-toward-a-free-trait-agreement/#more-484
CHAPTER 8 8.5 Assessment Methods From the Trait Perspective
8.5 Assessment Methods From the Trait Perspective
Perhaps more so than any other approach to studying personality, the trait approach has a wide range of commonly used tools, though most have been developed and used in the con-text of research. Perhaps this is not surprising, given that the trait approach emerged from an analysis of traits, and this type of analysis not only requires assessment, but it requires numerous forms of assessment in order to provide concurrent validation of the findings.
Because of the large number of measures, only a handful will be reviewed here, but these repre- sent some of the more commonly used ones that are theoretically consistent with the literature.
The Family of NEO™ Scales First developed in 1985 by Paul Costa and Robert McCrae, the various NEO™ measures were developed to specifically assess the big five factors of neuroticism, extraversion, openness, agree- ableness, and conscientiousness (the name of the scale reflects the first three letters from the first three factors; Costa & McCrae, 1985). The original inventory was a 180-item measure. In 1992, Costa & McCrae expanded the measure to 240 items, so as to include six facet scores for agreeableness and conscientiousness, with this version named the NEO™-PI-R. The NEO™ scales have both a self-report version and a version that can be completed by someone who knows the respondent.
In 1989, Costa & McCrae also developed a 60-item short form of the scale that assesses only the five factors (no facets), by focusing on the highest-loading items within each factor. The authors also collected norms for the general adult population, a geriatric sample, and a college student population.
The Big Five Inventory The Big Five Inventory (BFI; John, Donahue, & Kentle, 1991) is also designed to capture the big five factors, though it is a briefer instrument than the NEO™ scales. Specifically, there are 44 items, with each item rated on a 5-point Likert rating scale, ranging from strongly dis- agree to strongly agree. The BFI also allows for the calculation of 10 facets, reflecting more spe- cific traits within each global fac- tor, which match on to the fac- ets assessed by the NEO™-PI-R (see Soto & John, 2009). The BFI demonstrates good psychometric properties (i.e., reasonably strong internal consistency, retest reliability), convergence with other big five measures, and self-peer agreement, and has also been validated in other languages (e.g., Benet-Martı́nez & John, 1998; DeYoung, 2006; John et al., 2008; John & Srivastava, 1999; Rammstedt & John, 2007).
Beyond the Text: Classic Research
In this brief report, the authors put forth a 10-item version of the Big Five. It includes a listing of the 10 items and a response key for the English version, so you can complete the inventory yourself (on p. 8, Appendix A of the article). Read it at: http://www.westmont.edu/_academics/depart ments/psychology/documents/Rammstedt_and_John.pdf.
Reference: Rammstedt, B, & John, O. P. (2007). Measuring personality in one minute or less: A 10-item short version of the Big Five Inventory. Journal of Research in Personality, 41, 203–212. (see p. 8 of this article; Appendix A)
Lec81110_08_c08_225-252.indd 248 5/21/15 12:40 PMhttp://www.westmont.edu/_academics/departments/psychology/documents/Rammstedt_and_John.pdfhttp://www.westmont.edu/_academics/departments/psychology/documents/Rammstedt_and_John.pdf
CHAPTER 8 8.5 Assessment Methods From the Trait Perspective
The HEXACO Inventory Emerging from linguistic analyses of seven languages, researchers have forwarded a version of the Big Five that adds a sixth factor to address the constructs of honesty and humility (Ashton et al., 2004). The theoretical assumption is that for the cultures sampled in this analysis, a separate factor assessing honesty-humility captures meaningful variance because it is better represented in the descriptive terms found in the European and Asian languages that were studied. The facets of the new HEXACO Inventory (Lee & Ashton, 2004) honesty-humility factor includes forgiveness, gentleness, flexibility, patience, and altruism. More recently, a 60-item short form of the HEXACO scales was also developed, with 10 items assessing each of the six main factors (Ashton & Lee, 2009).
Eysenck’s Personality Questionnaire The EPQ was developed to assess the factors of extraversion, neuroticism, and psychoticism using self-reported, forced-choice items (Eysenck & Eysenck, 1975). In addition to assessing these theo- retically independent factors, Eysenck also included a lie scale to help determine the validity of the scores. The authors also revised the psychoticism scale 10 years later; this measure is the EPQ-R, which has both a standard version of 100 items and a short version with 48 items (Eysenck, Eysenck, & Barrett, 1985). Just as with other personality measures, validity scales are especially important under the following conditions: when the respondent has low self-awareness, is unfa- miliar with the construct being assessed, or has a reason to be disingenuous (Burisch, 1984).
Cattell’s 16 Personality Factor Questionnaire The 185-item 16PF® assesses the 16 personality factors identified by Cattell (1946), as well as sec- ondary factors that are consistent with the Big Five (see also Cattell, Eber, & Tatsuoka, 1970). These scales are all bi-polar, in that both ends of each scale can be interpreted. In addition, the test also includes three validity scales assessing impression management, acquiescence, and infrequency (rarely endorsed items) (Cattell, Cattell, & Cattell, 1993). There have been several versions of the 16PF® developed over the years, and the most recent version is the fifth edition, first published in 1993 (Cattell et al., 1993; Russell & Karol, 2002). The most recent edition involves a large norma- tive sample of 10,000 individuals reflecting the U.S. Census for demographic variables. The new edition also has improved psychometric properties. Although the 16PF® was designed for adults aged 16 and older, another version of the test has been developed for adolescents (the 16PF® Adolescent Personality Questionnaire; Schueger, 2001).
Myers Briggs Type Indicator® The MBTI® was originally developed by Myers and Briggs, and intended to be used to assesses the two attitudes (extraversion and introversion) and four functions (sensing, feeling, thinking and intuiting) set forth in the theory of Carl Jung (Myers, McCaulley, Quenk, & Hammer, 1998). The measure has been largely used in industrial-organizational settings in an attempt to match employee attributes with work environments. However, the instrument generally lacks good psy- chometric properties, and has not been well validated, despite it being an older inventory (see Hunsley, Lee, & Wood, 2004). Nevertheless, it is a frequently used measure, due in part to its widespread availability and its relatively simple structure.
Lec81110_08_c08_225-252.indd 249 5/21/15 12:40 PM
CHAPTER 8
Summary
Traits are descriptive terms used to describe human behavior, and they represent one of the more dominant models in personality psychology today. Trait theory was based upon an analysis of language, whereby researchers identified the adjectives that have been used to characterize human behavior. The assumption is that if a trait is fundamentally important, a soci- ety will create terms to represent it. Researchers such as Cattell and Eysenck were among the first to analyze language in order to derive the basic factors that underlie personality; however, they diverged with respect to the minimum number of factors they believed were needed to reflect all traits. Cattell developed the most complex model, with 16 factors, whereas Eysenck suggested 3. Currently, researchers have generally agreed that five factors are sufficient to capture trait vari- ability, and these big five factors have been recovered in different languages and cultures, and they appear to be relatively stable over time; whatever changes do occur are most prominent in childhood to adolescence. It is also the case that traits (the person) predict behavior best when the focus is on traits that are highly salient (important) and situations where the script for how we should act (situational strength) is weakest.
The big five factors of extraversion, neuroticism, openness, agreeableness, and conscientiousness appear to predict important outcomes in life, including measures of well-being, mental health, physical health, and even mortality.
Complementing the trait approach to personality is what has been known as the “doing” side of personality, which focuses on our uniquely defined goal pursuits. Here the emphasis is on traits within the contexts of our daily lives (e.g., control over important goals), and research suggests that how we think and feel about these goals adds significantly to the predictive power of traits. These units of analysis also help to explain why and when people might act out of character, as is seen with pseudo-extraversion.
Finally, we review a sampling of measures of traditional traits, showing that there are sound psy- chometric measures assessing everything from 16- to 3-factor models of personality, though the bulk of the measures assess the Big Five.
Key Terms
agreeableness A personality factor associated with traits such as friendliness, trust, altruism, and modesty.
approach goal A goal that is a desired outcome to move toward.
authenticity The quality of being consistent with one’s core values and sense of self.
avoidance goal A goal of avoiding an undesir- able outcome.
Big Five A model representing a descriptive taxonomy of personality traits assigned to five main factors.
cardinal trait A trait that manifests in every aspect of an individual’s life and that serves as a primary motivator of action.
central trait A trait that manifests itself in a limited range of situations.
choleric The humor of yellow bile.
conscientiousness A personality factor associ- ated with the traits of diligence, organization, punctuality, scrupulousness, self-discipline, thoroughness, deliberation, and need for achievement.
Key Terms
Lec81110_08_c08_225-252.indd 250 5/21/15 12:40 PM
CHAPTER 8Key Terms
cumulative continuity hypothesis The idea that an individual chooses an environment consistent with his or her personality, and this environment reinforces that personality.
defensive-pessimism When an individual adopts a strategy of anticipating the worst, despite the fact that they continually experi- ence success.
extraversion A personality factor associated with an individual’s interest and focus on other people and things.
factor A trait that can describe a large number of other traits.
fundamental attribution error The tendency to attribute our own behavior to situational fac- tors, but the behavior of others to their person- ality traits.
inverse (or negative) correlation A statistical relationship between two variables: When one variable changes, the other variable moves in the opposite direction (e.g., as x increases, y decreases; as x decreases, y increases).
learning goal A goal set by an individual that is focused on the accumulation of information (or practice) to improve the quality of his or her work.
lexical analysis The approach of understanding personality through the study of language.
lexical hypothesis The assumption that any descriptive characteristics that are important will become part of our language and that language will evolve such that single words will emerge to capture those important constructs.
melancholic The humor of black bile.
neuroticism Emotional instability, manifest- ing as negative emotions in the presence of a stressor.
openness to new experience A personality factor associated with traits such as sensitivity for aesthetics, imaginativeness, preference for variety, and intellectual curiosity.
performance goal A goal set by an individual that is focused on impression management and his or her appearance to others.
phlegmatic The humor of phlegm.
positive correlation A statistical relationship between two variables: When one variable changes, the other variable moves in the same direction (e.g., as x increases, y increases; as x decreases, y decreases).
psychoticism A personality factor most rel- evant for distinguishing between normal and non-normal manifestations of personality. It is associated with recklessness, inappropriate emotional expression, hostility, disregard for common sense, and poor contact with reality.
sanguine The humor of blood.
secondary trait A trait that is the least durable over time and across situations and contributes to an individual’s uniqueness.
situational script When cues in the environ- ment dictate how an individual should act in that setting.
situational strength When a situation influ- ences an individual’s behavior more than other situations do.
source trait A trait that is an underlying psy- chological factor.
surface trait A trait that is subsumed by differ- ent factors and is most directly translatable to behaviors.