The Sixteen Personality Factor Questionnaire (16PF)

25 Pages·2008·0.36 MB·English

9781412946520-Ch07 5/7/08 7:03 PM Page 135
7
The Sixteen Personality Factor
Questionnaire (16PF)
Heather E.P. Cattell and Alan D. Mead
INTRODUCTION research and is embedded in a well-established
theory of individual differences. This ques-
The Sixteen Personality Factor Questionnaire tionnaire’s extensive body of research
(16PF) is a comprehensive measure of normal- stretches back over half a century, providing
range personality found to be effective in a evidence of its utility in clinical, counseling,
variety of settings where an in-depth assess- industrial-organizational, educational, and
ment of the whole person is needed. The 16PF research settings (Cattell, R.B. et al., 1970;
traits, presented in Table 7.1, are the result of H.E.P. Cattell and Schuerger, 2003; Conn and
years of factor-analytic research focused on Rieke, 1994; Krug and Johns, 1990; Russell
discovering the basic structural elements of and Karol, 2002). A conservative estimate of
personality (Cattell, R.B., 1957, 1973). 16PF research since 1974 includes more than
In addition to discovering the sixteen 2,000 publications (Hofer and Eber, 2002).
normal-range personality traits for which the Most studies have found the 16PF to be
instrument is named, these researchers iden- among the top five most commonly used
tified the five broad dimensions – a variant of normal-range instruments in both research
the ‘Big Five’ factors (Cattell, R.B., 1957, and practice (Butcher and Rouse, 1996;
1970). From the beginning, Cattell proposed Piotrowski and Zalewski, 1993; Watkins et al.,
a multi-level, hierarchical structure of per- 1995). The measure is also widely used inter-
sonality: the second-order global measures nationally, and since its inception has been
describe personality at a broader, conceptual adapted into over 35 languages worldwide.
level, while the more precise primary factors
reveal the fine details and nuances that make
each person unique, and are more powerful
in predicting actual behavior. In addition, this HISTORY AND DEVELOPMENT OF THE
factor-analytic structure includes a set of third- 16PF QUESTIONNAIRE
order factors, also discussed in this chapter.
Due to its scientific origins, the 16PF The history of the 16PF Questionnaire
Questionnaire has a long history of empirical spans almost the entire history of standardized
9781412946520-Ch07 5/7/08 7:03 PM Page 136
136 THE SAGE HANDBOOK OF PERSONALITY THEORY AND ASSESSMENT
Table 7.1 16PF Scale Names and Descriptors
Descriptors of Low Range Primary Scales Descriptors of High Range
Reserved, Impersonal, Distant Warmth (A) Warm-hearted, Caring, Attentive To Others
Concrete, Lower Mental Capacity Reasoning (B) Abstract, Bright, Fast-Learner
Reactive, Affected By Feelings Emotional Stability (C) Emotionally Stable, Adaptive, Mature
Deferential, Cooperative, Avoids Conflict Dominance (E) Dominant, Forceful, Assertive
Serious, Restrained, Careful Liveliness (F) Enthusiastic, Animated, Spontaneous
Expedient, Nonconforming Rule-Consciousness (G) Rule-Conscious, Dutiful
Shy, Timid, Threat-Sensitive Social Boldness (H) Socially Bold, Venturesome, Thick-Skinned
Tough, Objective, Unsentimental Sensitivity (I) Sensitive, Aesthetic, Tender-Minded
Trusting, Unsuspecting, Accepting Vigilance (L) Vigilant, Suspicious, Skeptical, Wary
Practical, Grounded, Down-To-Earth Abstractedness (M) Abstracted, Imaginative, Idea-Oriented
Forthright, Genuine, Artless Privateness (N) Private, Discreet, Non-Disclosing
Self-Assured, Unworried, Complacent Apprehension (O) Apprehensive, Self-Doubting, Worried
Traditional, Attached To Familiar Openness to Change (Q1) Open To Change, Experimenting
Group-Orientated, Affiliative Self-Reliance (Q2) Self-Reliant, Solitary, Individualistic
Tolerates Disorder, Unexacting, Flexible Perfectionism (Q3) Perfectionistic, Organized, Self-Disciplined
Relaxed, Placid, Patient Tension (Q4) Tense, High Energy, Driven
Global Scales
Introverted, Socially Inhibited Extraversion Extraverted, Socially Participating
Low Anxiety, Unperturbable Anxiety Neuroticism High Anxiety, Perturbable
Receptive, Open-Minded, Intuitive Tough-Mindedness Tough-Minded, Resolute, Unempathic
Accommodating, Agreeable, Selfless Independence Independent, Persuasive, Willful
Unrestrained, Follows Urges Self-Control Self-Controlled, Inhibits Urges
Adapted with permission from S.R. Conn and M.L. Rieke (1994). 16PF Fifth Edition Technical Manual. Champaign, IL: Institute
for Personality and Ability Testing, Inc.
personality measurement. Instead of being hydrogen and oxygen). For psychology to
developed to measure preconceived dimen- advance as a science, he felt it also needed
sions of interest to a particular author, the basic measurement techniques for personality.
instrument was developed from the unique Thus, through factor analysis – the powerful
perspective of a scientific quest to try to new tool for identifying underlying dimen-
discover the basic structural elements of sions behind complex phenomena – Cattell
personality. believed the basic dimensions of personality
Raymond Cattell’s personality research could be discovered and then measured.
was based on his strong background in the Over several decades, Cattell and his col-
physical sciences; born in 1905, he witnessed leagues carried out a program of comprehen-
the first-hand awe-inspiring results of sci- sive, international research seeking a
ence, from electricity and telephones to auto- thorough, research-based map of normal per-
mobiles, airplanes, and medicine. He wanted sonality. They systematically measured the
to apply these scientific methods to the widest possible range of personality dimen-
uncharted domain of human personality with sions, believing that ‘all aspects of human
the goal of discovering the basic elements of personality which are or have been of impor-
personality (much as the basic elements of the tance, interest, or utility have already become
physical world were discovered and organ- recorded in the substance of language’
ized into the periodic table). He believed that (Cattell, R.B., 1943: 483). They studied these
human characteristics such as creativity, traits in diverse populations, using three differ-
authoritarianism, altruism, or leadership skills ent methodologies (Cattell, R.B., 1973):
could be predicted from these fundamental observation of natural, in-situ life behavior or
personality traits (much as water was a L-data (e.g. academic grades, number of traffic
weighted combination of the elements of accidents, or social contacts); questionnaire
9781412946520-Ch07 5/7/08 7:03 PM Page 137
THE SIXTEEN PERSONALITY FACTOR QUESTIONNAIRE (16PF) 137
or Q-data from the self-report domain; and research in each new country. Introduction of
objective behavior measured in standardized, Web-based administration in 1999 allowed
experimental settings or T-data (e.g. number international test-users easy access to admin-
of original solutions to problem presented, istration, scoring, and reports in many differ-
responses to frustrations). Eventually, this ent languages, using local norms
research resulted in the 16 unitary traits of
the 16PF Questionnaire shown in Table 7.1.
From the beginning, Cattell’s goal was to
investigate universal aspects of personality. CATTELL’S THEORY OF PERSONALITY
Thus, his University of Illinois laboratory
included researchers from many different
Primary and secondary-level traits
countries who later continued their research
abroad. Ongoing collaborative research was From its inception, the 16PF Questionnaire
carried out with colleagues around the world, was a multi-level measure of personality
for example, in Japan (Akira Ishikawa and based on Cattell’s factor-analytic theory
Bien Tsujioka), Germany (Kurt Pawlik and (Cattell, R.B., 1933, 1946). Cattell and his
Klaus Schneewind), India (S. Kapoor), South colleagues first discovered the primary traits,
Africa (Malcolm Coulter), England (Frank which provide the most basic definition of
Warburton, Dennis Child), and Switzerland individual personality differences. These
(Karl Delhees). more specific primary traits are more power-
Since its first publication in 1949, there ful in understanding and predicting the com-
have been four major revisions – the most plexity of actual behavior (Ashton, 1998;
recent release being the 16PF fifth edition Judge et al., 2002; Mershon and Gorsuch,
(Cattell, R.B. et al., 1993). The main goals of 1988; Paunonen and Ashton, 2001; Roberts
the latest revision were to develop updated, et al., 2005).
refined item content and collect a large, new Next, these researchers factor-analyzed the
norm sample. The item pool included the primary traits themselves in order to investi-
best items from all five previous forms of gate personality structure at a higher level.
the 16PF plus new items written by the test From this, the broader ‘second-order’ or
authors and 16PF experts. Items were refined global factors emerged – the original Big
in a four-stage, iterative process using Five. These researchers found that the
large samples. The resulting instrument has numerous primary traits consistently coa-
shorter, simpler items with updated lan- lesced into these broad dimensions, each
guage, a more standardized answer format, with its own independent focus and function
and has been reviewed for gender, cultural, within personality, as described in Table 7.2.
and ethnic bias and ADA (Americans With More recently, a similar set of Big Five
Disabilities Act) compliance. Psychometric factors has been rediscovered by other
characteristics are improved, hand scoring is researchers (Costa and McCrae, 1992a;
easier, and the standardization contains over Goldberg, 1990), but using forced, orthogo-
10,000 people. nal factor definitions. The five global factors
Because of its international origins, the also have been found in factor analyses of a
16PF Questionnaire was quickly translated wide range of current personality instruments
and adapted into many other languages. (as Dr. Herb Eber, one of the original 16PF
Since its first publication in 1949, the instru- authors, used to say, ‘These broad factors
ment has been adapted into more than 35 lan- validate across very different populations and
guages worldwide. These are not simply methods because they are as big as elephants
translations, as many questionnaires provide, and can be found in any large data set!’).
but careful cultural adaptations, involving Thus, these five ‘second-order’ or global
new norms and reliability and validity factors were found to define personality at a
9781412946520-Ch07 5/7/08 7:03 PM Page 138
Table 7.2 16PF global factors and the primary trait` make-up
Global Factors
Extraversion/Introversion High Anxiety/Low Anxiety Tough-Mindedness/Receptivity Independence/Accommodation Self-Control/Lack of Restraint
(A) Warm-Reserved (C) Emotionally Stable– (A) Warm–Reserved (E) Dominant–Deferential (F) Lively–Serious
(F) Lively-Serious Reactive (I) Sensitive–Unsentimental (H) Bold–Shy (G) Rule-conscious/Expedient
(H) Bold-Shy (L) Vigilant–Trusting (M) Abstracted–Practical (L) Vigilant–Trusting (M) Abstracted–Practical
(N) Private-Forthright (O) Apprehensive–Self-assured (Q1) Open-to-Change/ (Q1) Open-to Change/ (Q3) Perfectionistic–Tolerates
(Q2) Self-Reliant–Group-oriented (Q4) Tense–Relaxed Traditional Traditional disorder
Primary Factors
9781412946520-Ch07 5/7/08 7:03 PM Page 139
THE SIXTEEN PERSONALITY FACTOR QUESTIONNAIRE (16PF) 139
higher, more theoretical level of personality. likely to come across as warm, modest, and
However, because of their factor-analytic concerned about others, while the second is
origins, the two levels of personality are likely to seem bold, talkative, and attention
essentially inter-related. The global factors seeking (less concerned about others). Thus,
provide the larger conceptual, organizing although both may seek social interaction to
framework for understanding the meaning an equal degree, they do so for very different
and function of the primary traits. However, reasons and are likely to have a very different
the meanings of the globals themselves were impact on their social environment.
determined by the primary traits which con- The primary and global levels of 16PF
verged to make them up (see Table 7.2). traits combine to provide a comprehensive,
For example, the Extraversion/Introversion in-depth understanding of an individual’s
global factor was defined by the convergence personality. For example, although knowing
of the five primary scales that represent basic someone’s overall level of Self-Control/con-
human motivations for moving toward versus scientiousness is important, successfully
away from social interaction. Similarly, motivating that person to accomplish a
the four primary traits that merged to define particular goal depends on also knowing
Tough-Mindedness versus Receptivity whether their self-control is motivated more
describe four different aspects of openness to by strong obedience to societal standards
the world: openness to feelings and emotions (Rule-Consciousness – G+), by a temperamen-
(Sensitivity – I), openness to abstract ideas tal tendency to be self-disciplined and organ-
and imagination (Abstractedness – M), open- ized (Perfectionism – Q3+), or by a practical,
ness to new approaches and ideas (Openness- focused perceptual style (low Abstractedness –
to-Change – Q1), and openness to people M−). Thus, the 16PF Questionnaire can pro-
(Warmth – A). vide an in-depth, integrated understanding of
Cattell’s hierarchical structure is based an individual’s whole personality.
on the idea that all traits are inter-
correlated in the real world (for example,
intelligence and anxiety, although conceptu-
The super factors of personality:
ally quite distinct, are usually strongly inter-
third-order factors
correlated). Because the basic 16PF primary
traits were naturally inter-correlated, they From the beginning, Cattell’s comprehensive
could be factor-analyzed to find the secondary- trait hierarchy was three-tiered: A wide sam-
level global traits. Thus, the data itself deter- pling of everyday behaviors were factor-
mined the definitions of the primary and analyzed to find the primary factors; these
global factors (in contrast to the forced primary traits were factor-analyzed, resulting
orthogonal definitions of factors in the cur- in the five second-order, global traits; and
rently popular Big Five models). then the global factors were factor-analyzed
Thus, the global traits provide a broad into third-order traits at the highest, most
overview of personality, while the primary abstract level of personality organization
traits provide the more detailed information (Cattell, R.B., 1946, 1957, 1973). Factor
about the richness and uniqueness of the indi- analysis of secondary factors to find third-
vidual. For example, two people may have the order factors was practiced first in the ability
same score on global Extraversion but may have domain (e.g. Spearman, 1932), but a few
quite different social styles. Someone who is personality theorists have also looked at
warm and supportive (A+) but shy and modest this highest level of personality structure
(H−) may have the exact same Extraversion (e.g. Eysenck, 1978; Hampson, 1988;
score as someone who is socially bold and Digman, 1997; Peabody and Goldberg, 1989).
gregarious (H+) but emotionally aloof and Because factor-analytic results at each
detached (A−). However, the first person is level depend on the clarity of the traits being
9781412946520-Ch07 5/7/08 7:03 PM Page 140
140 THE SAGE HANDBOOK OF PERSONALITY THEORY AND ASSESSMENT
factor-analyzed, early attempts to find third- Self-Control or conscientiousness); but also
order traits were less reliable. However, the dimensions of internal perceptual sensi-
several independent studies have recently tivity, reactivity, and creativity – openness to
used large-scale samples to investigate the feelings, imagination, esthetics, and new
third-order factor structure of the 16PF ideas (global Receptivity/openness versus
(H.E.P. Cattell, 1996; Dancer and Woods, Tough-Mindedness). Note that higher levels
2007; Gorsuch, 2007; Lounsbury et al., of Self-Control/conscientiousness are related
2004). H.E.P. Cattell (1996) applied a common to lower levels of openness/Receptivity:
factor analysis to the global traits of the Thus, highly conscientiousness, self-con-
16PF Fifth Edition norm sample (n = 2,500), trolled people also tend to be tough-minded
and found two well-defined third-order fac- and less open to emotions and new ideas.
tors. Richard Gorsuch (pers. comm., 12 Conversely, those who are more impulsive
February 2007) applied a common factor and undisciplined also tend to be more
analysis to the 16PF global scores of 11,000 creative and open to feelings and ideas
subjects, and found two very similar third- (and to experience life more vividly). This
order factors. Most recently, Dancer and third-order factor is well illustrated in the
Woods (2007) found very similar results contrasting styles of having a conscientious
working with a sample of 4,405 working focus on concrete, objective, practical
adults, and this factor pattern is presented in tasks, versus occupations that focus on
Table 7.3. abstract, imaginative, and innovative ideas.
Each of these independent studies found Thus, superfactor II might be called self-
the same two-factor solution. The first factor, disciplined practicality versus unrestrained
factor I, involves human activities that are creativity.
directed outward toward the world. This The fifth global factor, Anxiety/neuroti-
includes both Extraversion (movement cism, then loads on both of these third-order
toward social engagement, ‘communion’ factors. This suggests that the distress
or ‘attachment’), as well as Independence described by Anxiety could arise either in the
(mastery/dominance of the social and non- inward/outward engagement domain or in the
social environment). Thus, third-order factor more internalized unrestrained creativity/ self-
I encompasses tendencies to move assertively disciplined practicality domain. Additionally,
outward into the world toward both social high levels of distress may affect either of
connection and toward exploration/mastery these areas. This is consistent with the wide
of the environment, and might be called range of outward and inward human capaci-
active outward engagement. ties that can potentially become unbalanced,
Third-order factor II involves internal or can be affected by stress.
types of processes and events. It includes These results are consistent with Cattell’s
first the age-old dimension of instinctual original belief that these third-order factors
impulsivity versus self-restraint (global may not represent personality traits in the usual
sense, but might reflect some broad, abstract
level of sociological or biological influences
on human temperament (Cattell, R.B., 1957;
Table 7.3 Varimax rotated factor loadings
1973). For example, there may be some
of the second-order factors of the 16PF5
biological/neurological structure that affects
questionnaire (n = 4,405)
outward engagement versus inhibition (super-
Rotated factor I Rotated factor II
factor I), or affects impulse control/
Extraversion 0.821
restraint and perceptual sensitivity/reactivity
Independence 0.669
Anxiety −0.638 −0.522 (superfactor II). Definition and understanding
Self-control 0.816 of these third-order factors await further
Tough-mindedness 0.737
investigation.
9781412946520-Ch07 5/7/08 7:03 PM Page 141
THE SIXTEEN PERSONALITY FACTOR QUESTIONNAIRE (16PF) 141
methods, Costa and McCrae’s results do not
Comparison of the 16PF global
replicate (McKenzie, 1998). Instead, appro-
scales with other five-factor models
priate factoring (see R.B. Cattell, 1978;
For over 50 years, the 16PF has included the Gorsuch, 1983) of the original matrix pro-
broad, second-order dimensions currently duces the five 16PF global factors, rather
called ‘the Big Five’ (Cattell, R.B., 1946; than the three orthogonal NEO factors that
Krug and Johns, 1986). In fact, Cattell located Costa and McCrae chose to use.
three of these five factors in his earliest stud- A range of studies comparing the five 16PF
ies of temperament (1933) – which Digman global factors and the set of NEO Big Five fac-
(1996) called ‘the first glimpse of the Big tors show a striking resemblance between the
Five’. Four of the five current traits were two (Carnivez and Allen, 2005; H.E.P. Cattell,
already described in Cattell’s 1957 book. All 1996; Conn and Rieke, 1994; Gerbing and
five traits have been clearly identified and Tuley, 1991; Schneewind and Graf, 1998).
scorable from the questionnaire since the These studies show strong correlational and
release of the fourth edition around 1970. factor-analytic alignment between the two
Although Cattell continued to believe that models: Between the two extraversion factors,
there were more than five factors, so have between anxiety and neuroticism, between
many other prominent psychologists (Block, self-control and conscientiousness, between
1995; Fiske, 1994; Hogan et al., 1996; tough-mindedness/receptivity and openness-
Jackson et al., 2000; Lee et al., 2005; to-experience, and between independence and
Ostendorf, 1990; Saucier 2001). dis-agreeableness. In fact, the average correla-
The 16PF scales and items also played an tion between the 16PF global factors and their
important role in the development of the other respective NEO five factors are just as high as
Big Five factor models (e.g. Costa and those between the NEO five factors and the
McCrae, 1976, 1985; Norman, 1963; Big Five markers which the NEO was devel-
McKenzie et al., 1997; Tupes and Christal, oped to measure (H.E.P. Cattell, 1996;
1961). For example, the first NEO manual Goldberg, 1992). The alignments among the
(Costa and McCrae, 1985: 26) describes the Big Five models are summarized in Table 7.4.
development of the questionnaire as beginning However, there are important differences
with cluster analyses of 16PF scales, which between the two models. Although propo-
these researchers had been using for over nents of the other five-factor models have done
20 years in their own research. However, this much in the last decade to try to bring about
origin, or even acknowledgement of the exis- a consensus in psychology about the exis-
tence of the five 16PF global factors, does not tence of five global factors, their particular
appear in any current accounts of the develop- set of traits have been found to be problem-
ment of the Big Five (Costa and McCrae, atic. In the development process, the NEO
1992a; Digman, 1990; Goldberg, 1990). Big Five factors were forced to be statisti-
Furthermore, when the 16PF correlation cally uncorrelated or orthogonal for reasons
matrix, which was used in the original devel- of theoretical and statistical simplicity.
opment of the Big Five, is re-analyzed However, few have found this as a satisfactory
using more modern, rigorous factor-analytic approach for defining the basic dimensions
Table 7.4 Alignments among the three main five-factor models
16PF (Cattell) NEO-PI-R (Costa and McCrae) Big Five (Goldberg)
Extraversion/Introversion Extraversion Surgency
Low Anxiety/High Anxiety Neuroticism Emotional stability
Tough-Mindedness/Receptivity Openness Intellect or culture
Independence/Accommodation Agreeableness Agreeableness
Self-Control/Lack of Restraint Conscientiousness Conscientiousness or dependability
9781412946520-Ch07 5/7/08 7:03 PM Page 142
142 THE SAGE HANDBOOK OF PERSONALITY THEORY AND ASSESSMENT
of human personality. For example, Big Five (Wiggins, 2003), the five-factor model has no
supporter Jack Digman (1997) stated: ‘The factor that centrally includes either domi-
apparent orthogonality of the Big Five is a nance or warmth. Rather factor analyses of
direct result of the general employment of the NEO-PI-R show that the central traits of
varimax rotation, a procedure that imposes dominance and warmth are widely dispersed
rather than finds independent factors.’Additi- and spread thinly among several of the five
onaly, Loevinger writes: factors, particularly extraversion
and agreeableness (H.E.P. Cattell, 1996;
There is no reason to believe that the bedrock of
personality is a set of orthogonal ... factors, unless Child, 1998; Conn and Rieke, 1994; Costa
you think that nature is constrained to present us and McCrae, 1992).
a world in rows and columns. That would be con-
However, in the 16PF Questionnaire, the
venient for many purposes, particularly given the
Independence global factor is organized
statistical programs already installed on our com-
around traits of assertiveness and influence
puters. But is this realistic? (1994: 6)
in the world (high scorers are dominant,
The decision to impose orthogonal loca-
independent-minded and innovative, low
tions had fundamental effects on the resulting
scorers are deferential, cooperative, and
factors and their meanings. In his analysis
agreeable). Thus, the 16PF global Independ-
of this basic issue of factor analysis, Child
ence factor is defined around traits of domi-
states:
nance or ‘agency’, while in the NEO model,
Oblique solutions can spread the common vari- the basic trait of dominance is split and
ance between and within factors; orthogonal rota-
relegated to small roles in several factors
tion can only spread variance between factors.
including extraversion and dis-agreeableness
That is why it is so important to carry out an
(where dominance is centered in a negative,
oblique solution, to allow no escape of important
variance ... Unfortunately, the orthogonal compro- hostile context).
mise disguises both the relationship between In a similar way, factor-analyses of the
domains and the number of factors which could
NEO-PI-R have found that the basic trait of
possibly be present in hyperspace. (1998: 353–354)
warmth (or communion) is also divided, with
In contrast to the orthogonal definitions low loadings on several factors including
that were fundamental to the development of extraversion and agreeableness (H.E.P. Cattell,
the NEO factors, recent studies have found 1996; Child, 1998; Conn and Rieke, 1994;
that the NEO five factors are actually sub- Smith et al., 2001). However, in the 16PF,
stantially inter-correlated (Carnivez and Allen, Warmth plays a central role in Extraversion,
2005; Goldberg, 1992; Smith et al., 2001). the factor that focuses on the basic dimensions
Even the latest NEO-PI-R manual (Costa of interpersonal relating. Additionally, these
and McCrae, 1992: 100) shows neuroticism factor analyses of the NEO-PI-R indicate that
and conscientiousness to inter-correlate − the openness trait (called ‘intellect’ in
0.53, and extraversion and openness to inter- Goldberg’s model) tends to focus more on cog-
correlate 0.40. Goldberg’s Big Five markers nitive or intellectual curiosity, rather than
also show substantial inter-correlations. equally measuring the whole domain, which
These inter-correlations contradict the origi- includes openness to feelings, emotions, and
nal premise on which the NEO Big Five fac- esthetics. Also, the Big Five factor ‘conscien-
tors were defined. tiousness’ appears to be narrower in content
The forced orthogonal factor locations of than 16PF Self-Control and doesn’t include the
the five-factor model have had substantial whole domain of human methods for self-
effects on the meanings of the traits. For control and self-restraint versus impulsivity
example, although the basic traits of domi- (Roberts et al., 2005).
nance (or agency) and warmth (or communion) Thus, the imposed orthogonality of the
have long been seen as two of the most fun- NEO has had multiple impacts on its
damental dimensions of human personality factor definitions. Furthermore, researchers
9781412946520-Ch07 5/7/08 7:03 PM Page 143
THE SIXTEEN PERSONALITY FACTOR QUESTIONNAIRE (16PF) 143
have found that when oblique methods are just as well with other Big Five domains than
used on the NEO-PI-R items, allowing the data their own (even the test authors stated that
itself to determine factor definitions, the result- the 1992 revision of the NEO was prompted
ing factor definitions are different, and show by the fact that the facets for neuroticism and
more clarity and simple structure than do the extraversion did not cohere psychometrically
current NEO-PI-R factors (Child, 1998). (McCrae and Costa, 1992)). For example,
However, the biggest difference between Roberts et al. (2005) found that three of the
the two approaches is the method of develop- six conscientiousness facets do not adhere to
ment of the primary level traits. In the 16PF that domain, but are as strongly related to
Questionnaire, the first-order primary trait other Big Five domains as they are to consci-
definitions are based on decades of scientific entiousness.
research, and have been confirmed in a wide Overall, the strong correlations of many
range of independent studies (see the section facets with theoretically unrelated domains
on Validity). In contrast, the NEO-PI primary- and facets bring into question the definition
level personality facets were decided by of the Big Five factors. This lack of adher-
consensus among a small group of psycholo- ence of the NEO facets to their assigned
gists (who selected what they felt should domains is inconsistent with the basic model
appear in each NEO domain). Child (1998) of the questionnaire (and probably a result of
comments: the non-empirical origins of the facets).
Thus, a number of important issues have
It does seem miraculous that the personality
been raised about the integrity of the NEO
domains divided exactly into six facets. Of course,
model, as a result of both the arbitrary choice
as the NEO PI-R is a “top-down” theory, the
researchers can choose whatever number they of facet trait meanings and orthogonal global
wish before tying up the parcel. The snag with this factor definitions.
procedure is its arbitrary nature and proneness to
Another important distinction between the
creating factors or traits to fit a theory. (1998: 352)
16PF and other questionnaires is the contex-
This method of selecting the fundamental tualized nature of its items. For example,
facets of personality raises some basic ques- items on the NEO-PI-R involve a high degree
tions about the NEO model. First of all, this of transparent self-rating or self-assessment
arbitrary approach to choosing the facets of traits (e.g. ‘I’m an even-tempered person’;
leaves them open to debate by every other ‘I am dominant, forceful, and assertive’; ‘I am
psychologist who happens to conceptualize known as a warm and friendly person’).
personality differently (e.g. Gough, 1987; Although this type of transparent item may
Hogan et al., 1996; Wiggins, 2003). More do well in research settings, in most assess-
importantly, these facets are now used to ment situations where there are strong moti-
define and calculate scores on the basic vational components, these items tend to be
Big Five factors, which have resulted in vulnerable to distortion. For example, vari-
changed definitions of the Big Five domains ous studies have found that the basic factor
themselves. structure of the NEO-PI-R is different in job
Additionally, many correlational and applicant samples, thus bringing into question
factor-analytic studies have found the under- the validity of the questionnaire in settings
lying factor structure of the NEO facets where motivation and social desirability are
inconsistent and confusing, and that the issues (Schmit and Ryan, 1993; Smith et al.,
domains do not actually hold together (Child, 2001). In contrast, 16PF items tend to be
1998; Church and Burke, 1994; Conn and more indirect and involve more contextualized
Rieke, 1994; Loevinger, 1994; Parker et al., questions about actual behavior or experience
1993; Roberts et al., 2005; Smith et al., 2001). (e.g. ‘When I find myself in a boring situa-
These researchers have found that a large tion, I usually “tune out” and daydream about
proportion of the NEO facets actually correlate other things’; ‘I hardly ever feel hurried or
9781412946520-Ch07 5/7/08 7:03 PM Page 144
144 THE SAGE HANDBOOK OF PERSONALITY THEORY AND ASSESSMENT
rushed as I go about my daily tasks’; ‘I some- ized-ten scores) ranging from 1 to 10, with a
times feel that I need my friends more than mean of 5.5 and a standard deviation of 2.0.
they need me’). The latest standardization includes over
Furthermore, there is substantial research 10,000 people and was published in 2001.
indicating that self-ratings are different Because the questionnaire is un-timed and
from observer ratings in their factor struc- has simple, straightforward instructions,
ture, and that they are only moderately cor- administration requires minimal supervision
related with actual behavior (e.g. Paunonen, in either individual or group settings.
1993; Peabody and Goldberg, 1989). This Administration time is about 35–50 minutes
suggests that much of the variance or mean- for paper-and-pencil format, and about
ing in self-ratings is not explained by the 25–40 minutes for computer administration.
actual trait value, but rather is substantially Easy scoring procedures are provided
affected by self-perception or self-image. for paper-and-pencil, computer, or Internet
For example, self-ratings do not capture the formats. The publisher provides various scor-
important dimensions of personality that are ing services (mail-in, fax, software, and
outside of a person’s awareness or inconsis- Internet) and a range of interpretive reports
tent with their self-image. Therefore, indi- for different applications. Detailed instruc-
rect questions that ask about actual everyday tions for administration and scoring can be
behavior (as 16PF items do) tend to measure found in numerous places (H.E.P. Cattell
personality more accurately, than asking and Schuerger, 2003; Russell and Karol,
a person to rate themselves on the trait – 2002).
particularly where social desirability is The questionnaire is available in many dif-
involved or when no validity scales are ferent languages (international translations
available on the instrument. exceed 35 languages worldwide). Unlike
many commercially available personality
measures, recent 16PF translations are cul-
turally adapted, with local norms and relia-
BASIC FEATURES OF THE 16PF bility and validity information available in
QUESTIONNAIRE individual manuals. Internet administration
also allows use of international norms for
First published in 1949, the 16PF Question- scoring, plus reports in over a dozen different
naire has had four major revisions, in 1956, language groups.
1962, 1968, and the fifth edition in 1993 The 16PF traits are also measured in par-
(Cattell, R.B. et al.). The latest edition con- allel versions for younger age ranges. For
tains 185 multiple-choice items, with a three- example, the 16PF Adolescent Personality
point answer format. Item content is Questionnaire measures the 16PF traits in
non-threatening, asking about daily behavior, 12–18 year olds (Schuerger, 2001). A shorter
interests, and opinions. The short ability scale (20-minute) version of the questionnaire, con-
items (Factor B) are grouped together at the sisting of a subset of somewhat-shortened
end of the questionnaire with separate scales, was developed for use in employee
instructions. The questionnaire is written at selection settings – the 16PF Select (Cattell,
a fifth grade reading level, and meant for R.B. et al., 1999). The 16PF Express
use with people 16 years and older. (Gorsuch, 2006) provides a very short,
The instrument provides scores on the 15-minute measure of all the traits (with
16 primary scales, 5 global scales, and 3 four or five items per factor). The 16PF traits
response bias scales. All personality scales are also appear in the PsychEval Personality
bipolar (have clear, meaningful definitions at Questionnaire (PEPQ; Cattell, R.B. et al.,
both ends), and are given in ‘stens’ (standard- 2003), a comprehensive instrument which