Hoofdstuk 9 uit The Cambridge handbook of personality psychology van
Corr en Matthews
INTRODUCTION
There is little doubt that the Five-Factor Model (FFM) of personality traits
(the “Big Five”) is currently the dominant paradigm in personality research
and one of the most influential models in all of psychology. Digman’s 1990
review on the topic has become one of the most highly cited articles in the
history of the Annual Review of Psychology, with over 5,800 citations.
Barrick and Mount’s 1991 meta-analysis of job performance and the FFM –
itself cited 11,000 times – brought personality back into the mainstream of
Industrial/Organizational Psychology. The FFM has led to novel and
compelling reformulations of the personality disorders that have
influenced DSM-V (Widiger & Trull, 2007). Cross-cultural collaborations
have shown the universality of the FFM and demonstrated pervasive
fallacies in national character stereotypes (Terracciano et al., 2005).
Social Psychologist Harry Reis (personal communication, April 24, 2006)
characterized the FFM as “the most scientifically rigorous taxonomy that
behavioral science has,” and for his research on the FFM, Paul Costa was
selected by the Division of General Psychology of the American
Psychological Association to present the 2004 Arthur W. Staats Lecture for
Contributions towards Unifying Psychology.
What is it that researchers from so many disciplines have come to
appreciate? As Digman and Inouye (1986) put it, “If a large number of
rating scales is used and if the scope of the scales is very broad, the
domain of personality descriptors is almost completely accounted for by
five robust factors” (p. 116). In other words, these five factors provide a
structure in which most personality traits can be classified. This structure
arises because traits covary.
For example, people who are sociable and assertive tend also to be
cheerful and energetic; they are high on the Extraversion (E) factor, which
is said to be defined by sociability, assertiveness, cheerfulness, and
energy. However, people who are sociable and assertive may or may not
be intellectually curious and imaginative. Those latter traits define a
separate factor: Openness to Experience (O). Neuroticism versus
Emotional Stability (N), Agreeableness versus Antagonism (A) and
Conscientiousness (C) are the remaining factors (Widiger, 2017).
There is a widespread consensus that these five factors are necessary and
more-or-less sufficient to account for the covariation of most personality
traits, and it is this comprehensiveness that chiefly accounts for the utility
of the FFM. Researchers who wish to conduct a review of the literature on
personality correlates typically find that many different scales and
1
,instruments have been used to assess personality. If each is assigned to
one of the five factors, their results can be meaningfully combined.
Consequently, the FFM provides a framework for systematic exploratory
research. Suppose, for example, one wished to study the effects on
personality of growing up in East versus West Germany (Angleitner &
Ostendorf, 2000). One might hypothesize that East Germans would be,
say, higher in Conscientiousness and lower in Openness, and administer
only measures of those two factors. But if the real differences turned out
to be in levels of Neuroticism and Agreeableness, such a design would not
reveal it. By administering measures of the full FFM, one can be sure that
important traits have not been overlooked.
Background to the Five-Factor Model of Personality
The origins of the FFM can be traced back to Sir Francis Galton and the
beginnings of trait psychology, and the details of its history have been
recounted by a number of authors (Digman, 1990; John, Angleitner &
Ostendorf, 1988). Perhaps the most interesting historical question,
however, is why the FFM was not widely adopted until the end of the
twentieth century. There were a number of contributing causes to this
delay.
It was not immediately clear how one could specify the full list of traits in
order to determine what structure was needed to organize them. The
solution came with the adoption of the lexical hypothesis, which argues
that traits are so important in human affairs that common words will have
been invented to name them all; an unabridged dictionary would provide
an exhaustive listing of traits.
which could be sorted out into a basic structure. Several researchers
worked on this idea (Tupes & Christal, 1961, 1992) and reanalyzed data
(Griggs & Church, 1961), and mainstream Objections (1963) to the FFM,
using the exact concept of personality psychologists at that time, were
finally resolved. The growing belief in the five-factor model was based on
many studies that identified a similar set of factors, with some researchers
proposing additional factors, especially a sixth: “positive versus negative
valence” (e.g., McCrae & Costa, 1985). Researchers focused on
understanding how the FFM could predict job performance (Barrick &
Mount, 1991), clinical problems (O’Connor & McCrae, 2004), and biological
processes (see DeYoung, 2010). Recent reviews emphasized the value of
the model in capturing the basic structure of traits observed across
cultures. Even in the 1980s a series of studies confirmed that all these
instruments showed associations with the FFM. For example, inventories
labeled under different trait theories — like the 16PF, Eysenck’s
2
,dimensions, and others — still covered the five FFM factors: E
(Extraversion), N (Neuroticism), O (Openness), A (Agreeableness), and C
(Conscientiousness).
There was also growing support for the FFM. Types and classifications such
as Type A personalities were publicly used in clinical, social, or even
political spheres, but they lost ground as the Five-Factor model became
more dominant. Theorists (like Eysenck in 1991) had believed in a limited
set of traits or “types,” but McCrae and Costa argued for a broad range of
personality. In 2007, Ruisel referred to the FFM as the “core of modern
psychology.” McAdams (1994) made an important distinction: he described
three levels of understanding personality — traits, personal concerns (like
goals or values), and life stories. The FFM mostly covers the first level.
Some researchers, like Mischel, focused on how behavior changes
depending on the situation, while others emphasized personality stability.
Over time, the FFM grew in acceptance, partly because it could explain
both stable and changing aspects of personality.
Meanwhile, studies on development showed how traits like
Conscientiousness increase with age. Work on genetics also confirmed that
traits are partly inherited. Neuroscientists identified patterns in the brain
that link to personality traits. More recently, Roberts and Jackson (2008)
proposed models of personality development that integrate both biology
and environment.
The FFM is now widely used in research, clinical assessment, and even in
applied fields like business. Most agree that it gives a comprehensive and
fairly stable model of personality. Critics point out that it doesn’t explain
everything about the individual — but it remains one of the strongest and
most researched tools for summarizing an individual’s personality.
Later, the FFM was subsequently found using Q-sort methods, in which
people sort trait words from least to most descriptive of themselves (Block,
1965; McCrae, Costa, & Busch, 1986), and in sentence ratings. Since that
time, the five-factor structure has been confirmed in a wide range of
samples and with a range of techniques (e.g., McCrae & Costa, 1997).
Questionnaires based on the FFM became popular and widely used. To
measure the FFM, McCrae & Costa (1985) developed the NEO Personality
Inventory (NEO-PI), which was later revised into the NEO-PI-R, and then
the NEO-PI-3 (McCrae & Costa, 2005). These instruments include 240
items and assess the Big Five along with 6 narrower traits (facets) within
each factor. The shorter 60-item NEO-FFI version of the NEO-PI-R is
commonly used in research. The NEO instruments have been translated
and validated in dozens of languages, making them one of the most widely
3
, used and internationally researched measures of the FFM in a variety of
languages.
RESEARCH DISCOVERIES
Armed with a solid measurement model and a variety of valid and reliable
instruments, personality psychologists began conducting more systematic
research into traits. This work and repeated results have yielded a body of
knowledge with broad implications across domains like clinical,
educational, and workplace psychology. Trait research has been
particularly useful in describing changes across the lifespan. Studies show
that certain traits change with age in relatively consistent and predictable
ways. For example, people tend to become more conscientious, more
agreeable, and less neurotic as they age (McCrae et al., 1999; Roberts et
al., 2006; Soto et al., 2011). Costa & McCrae (2006) described a model in
which traits are seen as both stable and open to change. Traits are
considered part of the “basic tendencies” of personality, whereas changes
in how those tendencies are expressed in behavior (called “characteristic
adaptations”) reflect influences like experiences or cultural demands
(McCrae & Löckenhoff, 2010).
There are also known differences across generations in personality traits.
These differences appear to reflect not only age and cohort effects but
also social and cultural change. For instance, recent generations of young
people in the U.S. have reported higher levels of traits like anxiety and
self-esteem than earlier generations. These results support the idea that
personality is shaped by both nature and nurture.
Other research has shown consistent patterns across the world. The basic
five-factor structure appears in most cultures where it has been tested,
although small differences sometimes emerge. A few researchers have
argued that a sixth or even seventh factor might be needed to fully
capture personality in some cultures. However, overall, the FFM has
proven to be remarkably consistent globally. McCrae and colleagues (e.g.,
McCrae et al., 2005) documented the universality of the FFM, especially in
their studies of 50+ cultures. Researchers still adapt and adjust NEO-PI-R
versions using insights from local norms.
A recent study replicated age-related trends using a large sample of
college-age and adult targets from fifty cultures. The same study
replicated gender differences in self-reports (although the age and gender
trends were a bit weaker in some of the individual samples). Figure 4.1 (in
the original book) shows the replicated gender differences seen in the self-
reports of Americans and other countries (e.g., Costa et al., 2001).
4
Corr en Matthews
INTRODUCTION
There is little doubt that the Five-Factor Model (FFM) of personality traits
(the “Big Five”) is currently the dominant paradigm in personality research
and one of the most influential models in all of psychology. Digman’s 1990
review on the topic has become one of the most highly cited articles in the
history of the Annual Review of Psychology, with over 5,800 citations.
Barrick and Mount’s 1991 meta-analysis of job performance and the FFM –
itself cited 11,000 times – brought personality back into the mainstream of
Industrial/Organizational Psychology. The FFM has led to novel and
compelling reformulations of the personality disorders that have
influenced DSM-V (Widiger & Trull, 2007). Cross-cultural collaborations
have shown the universality of the FFM and demonstrated pervasive
fallacies in national character stereotypes (Terracciano et al., 2005).
Social Psychologist Harry Reis (personal communication, April 24, 2006)
characterized the FFM as “the most scientifically rigorous taxonomy that
behavioral science has,” and for his research on the FFM, Paul Costa was
selected by the Division of General Psychology of the American
Psychological Association to present the 2004 Arthur W. Staats Lecture for
Contributions towards Unifying Psychology.
What is it that researchers from so many disciplines have come to
appreciate? As Digman and Inouye (1986) put it, “If a large number of
rating scales is used and if the scope of the scales is very broad, the
domain of personality descriptors is almost completely accounted for by
five robust factors” (p. 116). In other words, these five factors provide a
structure in which most personality traits can be classified. This structure
arises because traits covary.
For example, people who are sociable and assertive tend also to be
cheerful and energetic; they are high on the Extraversion (E) factor, which
is said to be defined by sociability, assertiveness, cheerfulness, and
energy. However, people who are sociable and assertive may or may not
be intellectually curious and imaginative. Those latter traits define a
separate factor: Openness to Experience (O). Neuroticism versus
Emotional Stability (N), Agreeableness versus Antagonism (A) and
Conscientiousness (C) are the remaining factors (Widiger, 2017).
There is a widespread consensus that these five factors are necessary and
more-or-less sufficient to account for the covariation of most personality
traits, and it is this comprehensiveness that chiefly accounts for the utility
of the FFM. Researchers who wish to conduct a review of the literature on
personality correlates typically find that many different scales and
1
,instruments have been used to assess personality. If each is assigned to
one of the five factors, their results can be meaningfully combined.
Consequently, the FFM provides a framework for systematic exploratory
research. Suppose, for example, one wished to study the effects on
personality of growing up in East versus West Germany (Angleitner &
Ostendorf, 2000). One might hypothesize that East Germans would be,
say, higher in Conscientiousness and lower in Openness, and administer
only measures of those two factors. But if the real differences turned out
to be in levels of Neuroticism and Agreeableness, such a design would not
reveal it. By administering measures of the full FFM, one can be sure that
important traits have not been overlooked.
Background to the Five-Factor Model of Personality
The origins of the FFM can be traced back to Sir Francis Galton and the
beginnings of trait psychology, and the details of its history have been
recounted by a number of authors (Digman, 1990; John, Angleitner &
Ostendorf, 1988). Perhaps the most interesting historical question,
however, is why the FFM was not widely adopted until the end of the
twentieth century. There were a number of contributing causes to this
delay.
It was not immediately clear how one could specify the full list of traits in
order to determine what structure was needed to organize them. The
solution came with the adoption of the lexical hypothesis, which argues
that traits are so important in human affairs that common words will have
been invented to name them all; an unabridged dictionary would provide
an exhaustive listing of traits.
which could be sorted out into a basic structure. Several researchers
worked on this idea (Tupes & Christal, 1961, 1992) and reanalyzed data
(Griggs & Church, 1961), and mainstream Objections (1963) to the FFM,
using the exact concept of personality psychologists at that time, were
finally resolved. The growing belief in the five-factor model was based on
many studies that identified a similar set of factors, with some researchers
proposing additional factors, especially a sixth: “positive versus negative
valence” (e.g., McCrae & Costa, 1985). Researchers focused on
understanding how the FFM could predict job performance (Barrick &
Mount, 1991), clinical problems (O’Connor & McCrae, 2004), and biological
processes (see DeYoung, 2010). Recent reviews emphasized the value of
the model in capturing the basic structure of traits observed across
cultures. Even in the 1980s a series of studies confirmed that all these
instruments showed associations with the FFM. For example, inventories
labeled under different trait theories — like the 16PF, Eysenck’s
2
,dimensions, and others — still covered the five FFM factors: E
(Extraversion), N (Neuroticism), O (Openness), A (Agreeableness), and C
(Conscientiousness).
There was also growing support for the FFM. Types and classifications such
as Type A personalities were publicly used in clinical, social, or even
political spheres, but they lost ground as the Five-Factor model became
more dominant. Theorists (like Eysenck in 1991) had believed in a limited
set of traits or “types,” but McCrae and Costa argued for a broad range of
personality. In 2007, Ruisel referred to the FFM as the “core of modern
psychology.” McAdams (1994) made an important distinction: he described
three levels of understanding personality — traits, personal concerns (like
goals or values), and life stories. The FFM mostly covers the first level.
Some researchers, like Mischel, focused on how behavior changes
depending on the situation, while others emphasized personality stability.
Over time, the FFM grew in acceptance, partly because it could explain
both stable and changing aspects of personality.
Meanwhile, studies on development showed how traits like
Conscientiousness increase with age. Work on genetics also confirmed that
traits are partly inherited. Neuroscientists identified patterns in the brain
that link to personality traits. More recently, Roberts and Jackson (2008)
proposed models of personality development that integrate both biology
and environment.
The FFM is now widely used in research, clinical assessment, and even in
applied fields like business. Most agree that it gives a comprehensive and
fairly stable model of personality. Critics point out that it doesn’t explain
everything about the individual — but it remains one of the strongest and
most researched tools for summarizing an individual’s personality.
Later, the FFM was subsequently found using Q-sort methods, in which
people sort trait words from least to most descriptive of themselves (Block,
1965; McCrae, Costa, & Busch, 1986), and in sentence ratings. Since that
time, the five-factor structure has been confirmed in a wide range of
samples and with a range of techniques (e.g., McCrae & Costa, 1997).
Questionnaires based on the FFM became popular and widely used. To
measure the FFM, McCrae & Costa (1985) developed the NEO Personality
Inventory (NEO-PI), which was later revised into the NEO-PI-R, and then
the NEO-PI-3 (McCrae & Costa, 2005). These instruments include 240
items and assess the Big Five along with 6 narrower traits (facets) within
each factor. The shorter 60-item NEO-FFI version of the NEO-PI-R is
commonly used in research. The NEO instruments have been translated
and validated in dozens of languages, making them one of the most widely
3
, used and internationally researched measures of the FFM in a variety of
languages.
RESEARCH DISCOVERIES
Armed with a solid measurement model and a variety of valid and reliable
instruments, personality psychologists began conducting more systematic
research into traits. This work and repeated results have yielded a body of
knowledge with broad implications across domains like clinical,
educational, and workplace psychology. Trait research has been
particularly useful in describing changes across the lifespan. Studies show
that certain traits change with age in relatively consistent and predictable
ways. For example, people tend to become more conscientious, more
agreeable, and less neurotic as they age (McCrae et al., 1999; Roberts et
al., 2006; Soto et al., 2011). Costa & McCrae (2006) described a model in
which traits are seen as both stable and open to change. Traits are
considered part of the “basic tendencies” of personality, whereas changes
in how those tendencies are expressed in behavior (called “characteristic
adaptations”) reflect influences like experiences or cultural demands
(McCrae & Löckenhoff, 2010).
There are also known differences across generations in personality traits.
These differences appear to reflect not only age and cohort effects but
also social and cultural change. For instance, recent generations of young
people in the U.S. have reported higher levels of traits like anxiety and
self-esteem than earlier generations. These results support the idea that
personality is shaped by both nature and nurture.
Other research has shown consistent patterns across the world. The basic
five-factor structure appears in most cultures where it has been tested,
although small differences sometimes emerge. A few researchers have
argued that a sixth or even seventh factor might be needed to fully
capture personality in some cultures. However, overall, the FFM has
proven to be remarkably consistent globally. McCrae and colleagues (e.g.,
McCrae et al., 2005) documented the universality of the FFM, especially in
their studies of 50+ cultures. Researchers still adapt and adjust NEO-PI-R
versions using insights from local norms.
A recent study replicated age-related trends using a large sample of
college-age and adult targets from fifty cultures. The same study
replicated gender differences in self-reports (although the age and gender
trends were a bit weaker in some of the individual samples). Figure 4.1 (in
the original book) shows the replicated gender differences seen in the self-
reports of Americans and other countries (e.g., Costa et al., 2001).
4
