Content uploaded by Douglas R. Lindsay
Author content
All content in this area was uploaded by Douglas R. Lindsay on Jan 12, 2017
Content may be subject to copyright.
The Journal of Positive Psychology, April 2007; 2(2): 80–98
Implicit theories of courage
CHRISTOPHER R. RATE
1
, JENNIFER A. CLARKE
2
, DOUGLAS R. LINDSAY
3
,
& ROBERT J. STERNBERG
4
1
Yale University, USA,
2
United States Air Force Academy, USA,
3
Pennsylvania State University, USA, and
4
Tufts University, USA
Abstract
What is courage? This question garners significant interest and attention but little empirical research. An operational
definition of courage is essential to good research, yet no consensus definition has fully emerged. This article systematically
investigates people’s conceptions of courage and courageous behavior through a series of studies employing well-grounded
implicit methodologies. The organizational structure and components of courage are investigated using a response-
generating task (Study 1) and alternate methods (Studies 2 and 3), followed by an experimental approach to determine
if people actually use their implicit theories in their evaluations of others (Study 4). Collectively, these studies reveal an
organizational structure of people’s implicit theories of courage. Further, they indicate that people apply their implicit
theories accurately in evaluating others.
Keywords: Courage, courageous behavior, implicit theories, prototypes
Then, Laches, suppose that we first set about determining
the nature of courage ...Tell me, if you can, what is
courage
Socrates (Plato, trans. 1987)
Introduction
In Denver, Colorado, a 9-year-old boy disregarded
his mother’s desperate plea to run as two masked
men attempted to break into their family home.
Filled with terror, yet displaying selfless disregard,
he threw himself over his 4-year-old brother to shield
him from the invaders. The boy was shot in the back,
but his younger brother was not hit (‘‘Denver Boy
Honored,’’ 2005). Halfway across the country in
Cedar Grove, Wisconsin, a 44-year-old man tried to
save his 6-year-old daughter who had fallen through
the thin layer of ice covering a skating pond. During
his attempt, he fell in. Both father and daughter
tragically died (‘‘Wisconsin Father, Daughter Die,’’
2005). As we ponder these true-life events, we might
ask ourselves: were these acts courageous? If so, why?
And if not, why not?
What is courage? As one’s perceptions of the
events above show, courage is at least, in part, in the
eye of the beholder. What one person might view as
courageous (jumping into an ice-covered lake)
another might view as foolhardy. In some instances,
such as spying, what one group views as courageous
another might view as treacherous. And in cases of
suicide bombers, what appears as courageous to one
group might be viewed as evil to other groups. More
to our point, then, is the question of what people
think courage is. The current investigation is an
attempt to bring us toward an understanding of
courage, its nature, and its use.
Definitions of courage
Scholars, politicians, and laypeople have all entered
the debate over how to define courage. Some believe
that American culture over the last 30 years or so
has defined courage. That is, courage has been
‘‘attributed to all manner of actions that may indeed
be admirable but hardly compare to the conscious
self-sacrifice on behalf of something greater than self-
interest’’ (McCain & Salter, 2004, p. 13). At the
same time, others would argue that courage can be
attributed not only to the occasional and isolated act
of rescue or self-sacrifice, but also to everyday acts
(Evans & White, 1981; Putman, 2001; Woodard,
2004). Defining courage in terms of character
Correspondence: Christopher R. Rate, Department of Psychology, Yale University, New Haven, CT 06520, USA.
E-mail: Christopher.rate@yale.edu
ISSN 1743-9760 print/ISSN 1743-9779 online ß2007 Taylor & Francis
DOI: 10.1080/17439760701228755
strengths (bravery, persistence, integrity, and vitality)
of human goodness and excellence, the positive
psychology movement endeavors to view courage
through the lenses of positive subjective experiences,
positive individual traits, and institutions that enable
positive experiences and positive traits (Peterson &
Seligman, 2004; Seligman & Csikszentmihalyi,
2000).
In the course of investigating the concept of
courage, we uncovered numerous definitions or
descriptions of courage of which 29 are listed in
Table I. Hemingway’s description of courage as
‘‘grace under pressure’’ (cited in Lopez, O’Byrne, &
Peterson, 2003, p. 191) marks the simplest, yet
eloquent, example. A more complex, multidimen-
sional definition of courage as ‘‘the disposition to
voluntarily act, perhaps fearfully, in a dangerous
circumstance, where the relevant risks are reasonably
appraised, in an effort to obtain or preserve some
perceived good for oneself or others recognizing that
the desired perceived good may not be realized’’
(Shelp, 1984, p. 354) anchors the other extreme.
Though numerous definitions of courage provide
a rich foundation from which to build, we remain at
a loss for an operational definition of this construct
on which to base sound explicit theories. Even with
all of the attempts to define courage, we have not
advanced the domain to an agreed-upon conceptual
definition. Lopez and colleagues correctly assert that
‘‘though we have been able to parse out the different
types of courage by establishing between-brand
differences, we have been less successful at determin-
ing the elements or components of courage. Thus,
what is common to all brands remains unclear’’
(Lopez et al., 2003, p. 189). Operational definitions
of constructs of interest are essential to good research
and a lack of coherence in courage definitions is an
obstacle to the advancement of research. Therefore,
the search for a common structure of courage
(apparently underlying the disagreement as to
people’s implicit theories of courage) forms a major
focus of the current investigation.
Theories of courage
Why are there so many definitions of courage, and
why is courage so haphazardly defined? This should
come as no surprise in the context of implicit-theory
studies which demonstrate a diffuse understanding of
implicit theories. Theories of courage can be divided
into explicit- and implicit-theoretical categories,
and both are important to the advancement of the
psychological field. One might consider, however,
that explicit theories of courage may simply be the
implicit theories of courage researchers (Ardelt,
2005).
Explicit theories of courage
Explicit theories are constructions of psychologists
that are based on data collected from people
performing tasks presumed to measure the construct
under investigation. In an explicit approach to the
measurement of courage, Cox, Hallam, O’Conner
and Rachman (1983) measured decorated (identified
as courageous) bomb operators’ physiological
responses to fear and stress compared to non-
decorated (noncourageous) operators’ responses.
They found distinctive physiological responses
under stress for decorated and nondecorated bomb
operators, indicating that past courageous behavior
in a particular situation will reduce one’s physiolo-
gical responses to fear and stress in similar situations.
Replication experiments sampling bomb-disposal
operators (O’Connor, Hallman, & Rachman, 1985)
and follow-up experiments studying British veteran
paratroopers (Macmillan & Rachman, 1987)
supported these initial findings.
Implicit theories of courage
According to Sternberg (1985) and Sternberg,
Conway, Ketron and Bernstein (1981), implicit
theories are people’s own cognitive constructions.
Such theories reside in people’s heads, and need
to be discovered rather than invented because they
already exist. The goal of research on implicit
theories is to find out the form and content of
people’s informal theories. The data of interest are
people’s communications (in whatever form) regard-
ing their notions as to the nature of the construct.
Implicit theories are important in psychology and in
the world because most judgments people make of
each other are based on their implicit theories
(Sternberg, 1987). For example, there is no formal
test of courage; people evaluate each other’s courage
on the basis of their implicit theories of what courage
means. Even constructs which can be measured
using explicit theories, such as in the assessment
of intelligence, the overwhelming majority of judg-
ments are based on implicit theories, such as when
people evaluate others’ intelligence on the basis of
job interviews, colloquium presentations, published
articles, interactions on dates, and so forth.
The study of naturalistic concepts has furthered
understanding of several psychological constructs,
such as love and commitment (Barnes & Sternberg,
1997; Fehr, 1988), wisdom (Holliday & Chandler,
1986; Sternberg, 1985), creativity (Sternberg, 1985),
intelligence (Sternberg, 1985; Sternberg et al.,
1981), and moral exemplars (Walker & Hennig,
2004), and is now beginning to further our under-
standing of courage (Evans & White, 1981; O’Byrne,
Lopez, & Peterson, 2000; Woodard, 2004).
Implicit theories of courage 81
Table I. Selected definitions and descriptions of courage.
Source Definitions and descriptions
American Heritage
Dictionary
The state or quality of mind or spirit than enables one to face danger with self-possessions, confidence, and
resolution; bravery (1985).
Aquinas* Defined fortitude as firmness in mind in enduring or repulsing whatever makes steadfastness outstandingly difficult,
that is, particularly serious dangers, primarily sustaining action to overcome fears of bodily harm and death and
secondarily in persevering in attacking.
Aristotle* Defined andreia (military courage) as the disposition to act appropriately in situations that involve fear and
confidence: rationally determined mean between cowardice and foolhardiness.
Cavanagh and
Moberg
Courage, also called fortitude or bravery, is the ability to endure what is necessary to achieve a good end, even in the
face of great obstacles (1999, p. 2).
Clancy Courage is likely defined as a willingness to face tough choices as well as overcoming the fear associated with them
(2003, p. 132).
Evans and White An empirical definition of courage probably involves three important attributional dimensions: (a) the fear level of
the person making the attribution; (b) the perceived fear level of the attributee; and (c) salient features of the
situation e.g., objective risk involved and so on (1981, p. 420).
Finfgeld* Being courageous involves being fully aware of and accepting the threat of a long-term health concern, solving
problems using discernment, and developing enhanced sensitivities to behavior consists of taking responsibility
and being productive.
Gergen and Gergen* To be courageous, then, is to remain steadfast within the bosom of those relationships from which one’s sense of
personal esteem and identity are derived.
Gould Courage is revealed in three dimensions: (1) fear; (2) appropriate action; and (3) a higher purpose.
Haitch* Courage is two-sided: there is an aspect of standing firm or fighting, and an aspect of accepting intractable
realities ...courage is the psychic strength that enables the self to face danger and death.
Hemingway* Grace under pressure.
Hobbes* The contempt of wounds and violent death. It inclines men to private revenges, and sometimes to endeavor the
unsettling of public peace.
Kant* Defined fortudido as the capacity and the resolved purpose to resist a strong but unjust opponent; and with regard to
the opponent of the moral disposition within us.
Kennedy* (Describing senators with political courage) men whose abiding loyalty to their nation triumphed over personal and
political considerations.
Kilmann, O’Hara
and Strauss
A courageous act in an organization includes five essential properties: (1) member has free choice to act;
(2) member experiences significant risk; (3) member assess the risk as reasonable; (4) member’s contemplated
act pursues excellence or other worthy aims and (5) member proceeds despite fear with mindful action (2005).
Klein and Napier Courage involves five factors: candor (speak and hear the truth), purpose (pursue lofty and audacious goals), rigor
(invent disciplines and make them stick), risk (empower, trust, and invest in relationships), and will (inspire
optimism, spirit, and promise) (2003).
Kohut* Oppose the pressures exerted on them and remain faithful to their ideals and themselves.
McCain and Salter Defined courage as an act that risks life or limb or other very serious personal injuries for the sake of others or to
uphold a virtue: a standard often upheld by battlefield heroics but one that is certainly not limited to martial valor
(2004, p. 14).
Mencius (Mengzi) Distinguished between types or courage, seeing some as ‘‘petty,’’ those concerned exclusively with personal honor;
and ‘‘great,’’ those grounded in and oriented toward the good. ‘‘Those who know that they are in the right are
justified in their cause and this provides them with the motivation to confront and engage even the greatest of
dangers’’ (cited in Ivanhoe, 2002, p. 68).
O’Byrne, Lopez
and Peterson*
Dispositional psychological courage is the cognitive process of defining risk, identifying and considering alternative
actions, and choosing to act in spite of potential negative consequences in an effort to obtain ‘‘good’’ for self or
others recognizing that this perceived good may not be realized.
Plato* The ability to remember what is worth prizing and what is worth fearing.
Putman* Facing the fears associated with the loss of psychological stability.
Rachman Willing and able to approach a fearful situation despite the presence of subjective fear and psychophysiological
disturbances (1990, p. 12).
Seligman* The capacity to rise to the occasion.
Shelp* The disposition to voluntarily act, perhaps fearfully, in a dangerous circumstance, where the relevant risks are
reasonable appraised, in an effort to obtain or preserve some perceived good for oneself or others recognizing that
the desired perceived good may not be realized.
Shepela et al. Courageous resistance: selfless behavior in which there is high risk/cost to the actor, and possibly the actor’s family
and associates, where the behavior must be sustained over time, is most often deliberative, and often where the
actor is responding to a moral call (1999, p. 789).
Snyder* Extraordinary behavior in ordinary times.
Walton Courage consists of three characteristics: (1) careful presence of mind and deliberate action, (2) difficult,
dangerous, and painful circumstances, and (3) a morally worthy intention.. .at the agent’s personal risk and
suffering (1986, p. 3).
Woodard Courage is defined as the ability to act for a meaningful (noble, good, or practical) cause, despite experiencing the
fear associated with perceived threat exceeding the available resources (2004, p. 174).
Note: The definitions attributed to sources annotated by a star (*) are from Lopez, O’Byrne and Peterson (2003). Copyright 2003 by the
American Psychological Association. Adapted with permission of the author.
82 C. R. Rate et al.
Implicit-theory studies of courage. In a response-
generation study examining how the average person
views courage, O’Byrne et al. (2000) asked 97 people
the question, ‘‘What is courage?’’ They found that
laypeople’s views of courage varied considerably.
Participants described courage as ‘‘taking action,’’
‘‘standing up for what one believes in,’’ ‘‘sacrificing,’’
and ‘‘facing threats/fears/challenges and overcoming
obstacles.’’ Philips (2004) asked 65 people the same
question, ‘‘What is courage?’’ discovering similar
themes regarding people’s notions about courage.
Courage requires sacrifice, risk, and overcoming
fear for a good purpose. To have courage is to
endure, persevere, and overcome. These studies, not
unlike the present article’s Study 1, provide an
empirical rationale for the numerous definitions of
courage. The responses provide only foundation
from which we can ascertain the components by
which people evaluate courage, and the organization
underlying individuals’ conceptions of courageous
behavior. If left as stand-alone studies the
results would indicate the potential for many
subjective definitions of courage, however, the
underlying structure of courage would remain
undetermined.
In efforts to develop a measurement scale of
courage, Woodard (2004) also utilized implicit-
theories methodologies. Ten experts with varied
areas of specialty in the field of psychology generated
statements considered representative of assessing
the construct of courage, defined as ‘‘the ability to
act for a meaningful (noble, good, or practical)
cause, despite experiencing the fear associated with
perceived threat exceeding the available resources’’
(Woodard, 2004, p. 174). Two hundred participants
then rated their agreement with the statements on
a 5-point scale. Each statement was also accompa-
nied by a fear-rating question to establish the level of
fear that the respondent might associate with the
situation depicted in the statement, and by a third
question asking the respondents whether they had
experienced the situation posed in the statement.
Factor analysis revealed four dimensions labeled:
(a) endurance for positive outcomes—including
behaviors such as ‘‘acts despite bullying as a
minority’’; (b) dealing with groups—including beha-
viors such as ‘‘help grieving family’’ and ‘‘rejection
by others for goals’’; (c) acting alone—including
behaviors such as ‘‘accept job despite criticism’’
and ‘‘avoid confronting my own pain’’; and
(d) physical pain/breaking social norms—including
behaviors such as ‘‘intervene in domestic dispute’’
and ‘‘endure pain for political secrets.’’ The factors
were found to have a strong correlation with
each other, suggesting a relationship among
the ideas that contributed to deciding whether to
act in a courageous manner. The concepts
represented by the factors have further illuminated
the basic manner in which people conceive of the
construct of courage.
In a series of interview-based studies in the nursing
domain, chronically ill adolescents (Haase, 1987),
middle-aged adults (Finfgeld, 1998), and older
adults (Finfgeld, 1995) were asked to describe
a situation in which they thought they were
courageous. They were instructed to describe their
thoughts, feelings, and perceptions as they remem-
bered experiencing them. These findings point to
the development of attitudes and coping methods,
highlighting the process of becoming and being
courageous in the face of chronic illness.
Additional studies of courage have also implemen-
ted implicit-theories methodologies. Evans and
White (1981) focused on the role of fear in people’s
implicit theories of bravery (used synonymously with
courage). They concluded that courage probably
involved, in part, not only the perceived fear of the
target actor, but also the fear level of the person
making the evaluation. Szagun (1992) and Szagun
and Schauble (1997) used structured interviews to
investigate juvenile, adolescent, and adult concep-
tualizations of courage. In their developmental
approach, they revealed that courage was viewed
as less physical and increasingly non-physical with
age. Themes of deliberate or intentional cognitive
processes, overcoming or persisting despite fear,
risk, and moral acts were indicated.
Caveat to the study of courage
As a complex, multi-dimensional construct, courage
may be better understood as an exceptional response
to specific external conditions or circumstances
than as an attribute, disposition, or character trait
(generally acknowledged as long-term and stable
across situations) which appears to be the intent of
many definitions and descriptions of courage.
Perhaps researchers are on firmer ground when
they label someone as ‘‘courageous’’ based on their
actions in a given situation. In other words, given
a specific situation, researchers can say that one acted
or responded ‘‘courageously,’’ if appropriate.
Conversely, many notable researchers (see
O’Byrne et al., 2000; Peterson, 2006; Peterson &
Seligman, 2006; Shelp, 1984) through explicit
statement, definition, or description of courage
favor a more dispositional courage. For example,
Peterson and Seligman (2004) through their classi-
fication of virtues and character strengths have
provided an extraordinary foundation for looking at
the trait-like qualities of courage. These researchers
have quite nicely described courage as a virtue
achieved through the character strengths of bravery,
Implicit theories of courage 83
persistence, integrity and vitality. Peterson and
Seligman (2004) do note, however, that their
conceptualization of bravery (what many courage
researchers and philosophers label ‘‘courage’’) is
difficult to measure because of the tonic or phasic
nature of different ‘‘types’’ of courage. Their Values
in Action Inventory of Strengths (VIA-IS) ‘‘asks
about typical responses rather than those situated in
the appropriate settings for the display of given
strengths. Further, no one can reasonably nominate
friends or acquaintances as paragons of given
strengths unless they have happened to observe
them in the appropriate settings’’ (Peterson &
Seligman, 2004, p. 633). This appears to indicate
that even if researchers hold the view of dispositional
courage, the situation still matters.
Therefore, while not excluding the possibility of
a courageous disposition or a relationship with
character, there does seem to be limited transfer-
ability between a ‘‘courageous act’’ and a ‘‘coura-
geous actor.’’ We should not necessarily conclude
that individuals have a courageous disposition based
on their actions in one situation. We can not
appropriately infer that the same individual will act
courageously in a different situation. ‘‘In short, to say
that someone is a ‘courageous person,’ when the
speaker means to generalize to long-term disposi-
tions or character habits, may be very shaky
conjecture indeed on the basis of a single action’’
(Walton, 1986, p. 221).
Summary
Why are we studying implicit theories of courage?
Sternberg et al. (1981) remind us that implicit
theories tell us about individuals’ views of what a
construct such as courage is. Wegner and Vallacher
echo this point, stating that ‘‘the layman ...lives by
his theory .... The systems we call ‘implicit theories’
are the individual’s reality’’ (Wegner & Vallacher,
1977, p. 21). Generally, implicit theories are theories
of word usage and the lay understanding of the
construct behind the word. In this case, courage is
of extreme interest to a vast number and variety
of people, from academic scholars, to military
members, and to the ordinary citizen. It is important
to know people’s notions of courage’s nature and if
it is both possible and beneficial to cultivate courage.
Implicit theories are one approach toward developing
a comprehensive scheme for understanding and
describing positive subjective experiences, positive
individual traits, and positive institutions (Peterson,
2006; Peterson & Seligman, 2004). In sum, our
motivation was to systematically investigate people’s
conceptions of courage and courageous behavior
through a series of studies employing well-grounded
implicit methodologies (Bluck & Glu
¨ck, 2005).
Although similar studies have used response-
generation (O’Byrne et al., 2000; Philips, 2004)
and item-rating tasks (Woodard, 2004) to investigate
the construct of courage, our studies greatly expand
upon this pioneering work. In Study 1 and Study 2,
we employed independent, non-overlapping samples
from two very distinct graduate and undergraduate
populations (i.e., Yale University and the United
States Air Force Academy [USAFA]) to ascertain
the nature and use of courage. These samples were
obtained from the Yale University population at large
and through two classes required by all USAFA
students. Furthermore, Study 3 extended beyond
what others have accomplished by introducing
alternative methods to confirm an underlying struc-
ture of courage. To our knowledge, Study 4 is the
first study to investigate the application of courage
and its components in the evaluation of others. The
essential first step is to determine an underlying
structure of courage. Does this structure hold,
however, when people actually use these components
in their evaluation of others’ levels of courage? Study
4 attempted to address this question. In sum,
we investigated the organizational structure and
components of courage using a response-generation
task (Study 1) and alternate methods (Study 2 and 3)
followed by an experimental approach to determine
if people actually use their implicit theories in their
evaluations of others (Study 4).
This sequence of studies provided an understand-
ing of people’s implicit theories of the psychological
construct of courage in everyday terms, while serving
as a launching point to further investigate the implicit
organization of courageous behavior. Core beha-
vioral components were revealed through these
studies. As we uncover and understand this parti-
cular psychological construct, and understand impli-
cit-theory convergence with more recent explicit and
scholarly definitions, such as courage as the emo-
tional character strengths (bravery, persistence,
integrity, and vitality) that involve ‘‘the exercise of
will to accomplish goals in the face of opposition,
either external or internal’’ (Peterson, 2006, p. 143;
Peterson & Seligman, 2004, p. 199; Snyder & Lopez,
2007, p. 19), we can begin to look at the potential
benefits of creating programs to address the devel-
opment of courageous behavior in individuals.
Knowledge of courage and potential developmental
programs could have a significant impact on
the ordinary citizen, business professionals, and
individuals in the military, in fire-fighting, and in
law enforcement, as courageous behavior could be
and often is called upon in our daily lives and in the
line of professional duty.
84 C. R. Rate et al.
Study 1
Study 1 was conducted to compile a master list of
behaviors people regard as descriptive of an ideally
courageous individual. A free-listing task was accom-
plished by a total of 175 undergraduate and graduate
students (109 male and 56 female) from the USAFA
and Yale University. Participants included 107
undergraduate students (79 men and 28 women) at
the USAFA. Their ages ranged from 18 to 26 years
of age (M¼20.0 years, SD ¼1.7). In addition,
68 undergraduate and graduate students at Yale
University (30 men and 38 women) participated in
this study. The Yale contingent ranged from 18 to 38
years of age (M¼22.5 years, SD ¼3.7). We
recruited both undergraduate and graduate students
at Yale to ensure adequate sample sizes.
Participants were asked to spend 5 to 10 minutes
completing a free-listing task in which they were
instructed to imagine observing behaviors of an
ideally courageous person. They were then asked
to list as many behaviors as they could think of that
are associated with this ideal individual. It was
emphasized that we are interested in behaviors, not
attributes. We want to know what courageous people
do (observable actions that make one courageous),
not how they are or what they are like (inferred
attributes/traits). Participants were provided with the
following example: ‘‘Saying that a courageous person
‘is unflappable’ would not really describe behavior;
however, saying this person ‘remains calm in
pressure situations’ would describe behavior.’’ A
raw-form master list of 1118 behaviors was compiled
from the participants’ responses. As a result of the
substantial size of the master list, judgment rules
were implemented to aid in distillation of the
behaviors. Taking a conservative approach to avoid
experimenter bias, we retained behaviors listed even
once, although obvious redundancies were elimi-
nated. (We also anticipated additional item reduc-
tion in Study 2.) The following judgment rules
adapted from Walker and Hennig (2004) were
implemented: (a) compound phrases were split
if each behavioral descriptor could stand alone;
(b) modifiers were dropped; (c) attributes were
converted to behaviors, if possible; and (d) non-
sensical, irrelevant, or inappropriate responses were
eliminated (e.g., wears Right GuardÕ, wrestles a
bear). The final list consisted of 639 items that
participants deemed illustrative of an ideally coura-
geous individual’s behaviors. The responses indi-
cated a substantial number of listed behaviors, which
included willful actions involving risk, sacrifice,
doing the right thing, and acting despite fear.
The final list appears to provide a foundation from
which we will begin to ascertain the structures,
factors, or dimensions by which people evaluate
courage, and the organization underlying individuals’
conceptions of courageous behavior itself in subse-
quent studies.
Study 2
In Study 1 we were able to obtain a large number
of behaviors associated with courageous behavior.
In Study 2, we asked participants to rate the degree
to which these behaviors were associated with
courage. This allowed us to integrate the myriad
behaviors into a set of empirically driven factors that
could be used in subsequent studies. Study 2 also
allowed us to omit behaviors that did not appear to
be generally associated with people’s core percep-
tions of courage.
Method
Participants. A total of 126 undergraduate and
graduate students participated in this second study.
Participants included 59 undergraduate students,
41 men and 18 women, at the USAFA. Their ages
ranged from 18 to 24 years of age (M¼20.75,
SD ¼1.15). In addition, a sample of 67 under-
graduate and graduate students from Yale University
participated. This sample included 19 men and 48
women. Their ages ranged from 18 to 56 years of age
(M¼22.99, SD ¼5.10). Participants received either
US$10 compensation or course credit for participa-
tion in this study.
Materials and procedure. Participants received an
informed consent document and completed a back-
ground questionnaire eliciting responses to gender,
age, and educational status. Participants were asked
to spend 30 to 45 minutes completing the back-
ground questionnaire and rating task. Prior to
the rating task, we instructed participants to imagine
an ideally courageous person (i.e., someone who has
just performed a courageous act) forming their own
impressions of the ideal. We then asked them to rate
how ‘‘distinctively characteristic’’ the 639 behaviors
on the master list are in describing their ideally
courageous person. All behaviors required ratings of
1 (low) to 9 (high).
Results
Missing data were considered missing at random and
were therefore entered by group mean imputation.
No single item was missing more than 3% of the
data. Furthermore, no participant failed to rate more
than 5% of the listed behaviors.
Implicit theories of courage 85
Basic statistics and prototypicality ratings. Given that
a mix of undergraduate (n¼36) and graduate
(n¼31) students at Yale University participated in
this study in order to attain an appropriate sample
size, a t-test was performed to see if this sample could
be collapsed for future analyses. No significant
differences were found for the mean distinctively
characteristic ratings between undergraduate
(M¼6.06, SD ¼1.15) and graduate (M¼6.08,
SD ¼1.21) students. In addition, the correlation
between Yale undergraduates and graduates across
the distinctively characteristic ratings was 0.91.
Thus, both the level and pattern of the responses
were largely the same.
Overall, the mean rating of the items was 6.25
(SD ¼1.16). Ratings were also approximately equal
between USAFA (M¼6.46, SD ¼1.13) and Yale
(M¼6.07, SD ¼1.17) participants and between
male (M¼6.18, SD ¼1.18) and female (M¼6.32,
SD ¼1.16) participants. A 2 (academic
institution) 2 (gender) ANOVA was conducted.
Both main effects and the interaction failed to reach
significance. Ratings were extremely reliable, mid- to
high-90s, both from the standpoint of Spearman–
Brown split halves of participants (subject reliability)
and of Spearman–Brown split halves of items (item
reliability). The correlations between academic
institutions and between genders across the distinct-
ively characteristic ratings indicated strong associa-
tions, r¼0.91 and r¼0.95, p50.01, respectively.
Thus, the rating levels and pattern of ratings were
similar for the sample populations.
As a result of the cumbersome nature of the
original 639 items as input into a final analysis and of
our recognized small sample size in comparison, only
distinctively characteristic items with mean ratings of
7.0 or higher on the 1–9 distinctively characteristic
scale were retained for further analysis. The number
of items was reduced to a more manageable
131 items. Sternberg et al. cautioned that not
all ‘‘characteristic’’ behaviors are important beha-
viors; for example, ‘‘eating is highly characteristic of
intelligent people, but few people would see eating as
central to defining their conceptions of such people’’
(Sternberg et al., 1981, pp. 43, 44). In the present
study, however, laypeople deemed these particular
behavioral items as ‘‘distinctive’’ or unique to
courage (i.e., more central, core, or important)
based on their ratings, rather than generally char-
acteristic behaviors of an individual, courageous
or not.
Structure and content of people’s conceptions of
courage. Although the participants-to-variables
ratio was noted as statistically problematic,
exploratory factor analyses were performed using
principal-components extraction followed by vari-
max rotation (with Kaiser normalization) of the
factorial axes. Separate analyses were first conducted
for Yale and USAFA samples. Although there were
differences between the exact factor structures
obtained by the academic institutions, the factor
solutions were indeed quite similar, mirroring the
level and pattern ratings of the two samples,
indicating remarkable similarities in their perceptions
of courage. Based on this observation a combined
factor analysis was then performed. Three factors
were extracted (on the basis of scree analysis,
strength, and interpretability) accounting for 23%,
19%, and 12% of the variance in the data,
respectively, for a total of 54%. Table II shows the
results of the factor analysis of the participants’
distinctively characteristic ratings of an ideally coura-
geous person’s behavior. The three factors were
labeled: (I) self-focused response to affect and external
circumstances; (II) non-physical/social-oriented acts—
internal noble motivation; and (III) selfless sacrifice/
risk—external noble motivation. Behaviors with load-
ings above 0.63 (indicating 40% overlapping var-
iance between the behavioral items and factor) were
chosen for interpretability. Although any cut-off is
subjective, Comrey and Lee (1992) suggested that
interpretability of items loading in excess of 0.63 is
very good. The highest loading items on each factor
are listed in Table II.
The first factor, self-focused response to affect
and external circumstances, included behaviors such
as ‘‘endures tough situations,’’ ‘‘perseveres in face of
obstacles,’’ ‘‘does not give in to fear when making
decisions,’’ and ‘‘acts even in the presence of fear.’’
The second factor, non-physical/social-oriented
acts—internal noble motivation, included behaviors
such as ‘‘stands up to unjust social practices because
of what one thinks is right,’’ ‘‘maintains honesty
no matter others’ opinions,’’ ‘‘demonstrates integ-
rity,’’ and ‘‘speaks up for one’s beliefs even under
controversy.’’ The third factor, selfless sacrifice/
risk—external noble motivation, included behaviors
such as ‘‘sacrifices one’s life for another person’s
safety,’’ ‘‘risks life to protect others,’’ and ‘‘risks
one’s life for a greater good.’’
Discussion
Analysis of data suggests that, for the most part,
the listed behavioral items were viewed as quite
characteristic of courageous individuals. No signifi-
cant differences between mean ratings by academic
institutions or genders were found. Further, the
ratings were extremely reliable across both partici-
pants and items. All subject reliabilities exceeded
0.92 and all item reliabilities exceeded 0.97.
86 C. R. Rate et al.
This high degree of internal consistency across
participants (subject reliability) tells us there was
considerable agreement as to what constituted
courageous behavior. The high degree of internal
consistency across items (item reliability) indicated
that we avoided a hodgepodge of unrelated items;
in other words, the items constituted a coherent set
of behaviors.
The results of our factor analysis are intriguing.
The analysis provided an underlying structure for
how people judge others in terms of courageous
behavior. These factors seem to point to general
categories such as external circumstances (e.g., risk),
affective (e.g., fear), and motivational (e.g., noble/
worthy purpose) components. Furthermore, this
analysis also provides compelling evidence of core
components of courage against which definitions of
courage may be evaluated. As implicit theories are
about word usage, we can use the emergent structure
to help determine the adequacy of previous defini-
tions. For example, Shelp’s (1984) description of
courage seems to fully encapsulate the three emer-
gent factors in this analysis. Other definitions are
perhaps less fully adequate or complete in compar-
ison. For instance, Rachman’s (1990) definition of
a courageous person as one ‘‘willing and able
to approach a fearful situation despite the presence
of subjective fear and psychophysiological distur-
bance’’ ( p. 12) corresponds to just one emergent
factor, self-focused response to affect and external
circumstances; whereas Putman’s (1997) definition
of courage as ‘‘facing fears associated with the loss of
psychological stability’’ also corresponds only to the
self-focused response to affect and external circum-
stances factor, while introducing a ‘‘psychological
stability’’ component undefined through the implicit
theories of our population.
To conclude, this study shows that people do seem
to have prototypes corresponding to the concept of
courage that are organized into sensible factors.
Based on this analysis, we would suggest that
definitions of courage should reflect, in some form,
the components of risk,fear, and noble/worthy purpose.
This study also provides evidence for an underlying
structure of courage against which definitions of
courage may be evaluated. It also strengthens the
foundation upon which further research may be
based to reveal more about people’s implicit theories
of courage and how they use them.
Study 3
In Study 3, we asked participants to sort prototypical
behaviors of courage based on their similarity in
an attempt to reaffirm the organizational structure
underlying the individual’s conception of courageous
behavior. Using a card-sorting task, analysed with
Table II. Factors underlying people’s implicit conceptions of
courage: ‘‘distinctively characteristic’’ ratings of an ideally
courageous person.
Factor Factor
loading
(I) Self-focused response to affect and external circumstances
Endures tough situations 0.83
Can handle rough situations 0.82
Handles tough, high-pressure situations 0.80
Perseveres in face of obstacles 0.80
Remains focused in high-stress environments 0.78
Remains composed in dangerous situations 0.78
Does not give in to fear when making decisions 0.77
Remains steadfast even when task requires
unpleasant actions
0.77
Perseveres under pressure 0.76
Overcomes obstacles 0.76
Does not shy away when the situation demands
action
0.75
Rises above pressure 0.74
Overcomes fears to perform dangerous tasks 0.74
Fights through pain and suffering 0.74
Acts even in presence of fear 0.74
Overcomes challenges 0.74
Refuses to bow to pressure 0.73
Does not freeze in pressured situations 0.73
Does not quit when ‘‘the going gets tough’’ 0.72
Faces that which causes fear 0.71
Remains calm when afraid 0.69
Pushes through stress 0.68
Does not back down when scared 0.68
Does not give up 0.67
(II) Non-physical/social-oriented acts—internal motivation
Stands up to unjust social practices because of
what one thinks is right
0.76
Stands up for the ‘‘right thing’’ against opposition 0.73
Maintains honesty no matter others’ opinions 0.72
Acts upon one’s values even if the values are
not the norm
0.71
Demonstrates integrity 0.70
Does what is right even if no one will know 0.70
Stands up for what is right even if the majority
disagrees
0.70
Does not fear to follow up on one’s ideals 0.70
Acts based on moral principles rather than
self-interest
0.70
Does what is right even if it is unpopular 0.70
Speaks up for one’s beliefs, even under controversy 0.68
Does the right thing 0.68
Stands up for others’ rights without thought
of reward
0.68
Lives life according to personal ideals 0.68
Stands up for the rights of others 0.67
Stands up for what one believes is right 0.67
(III) Selfless sacrifice/risk—external motivation
Risks life to protect others 0.86
Sacrifices self to save another 0.85
Risks one’s own life to save others in need 0.82
Sacrifices one’s life for another person’s safety 0.80
Sacrifices self for good of someone else 0.75
Sacrifices one’s life for another person’s survival 0.74
Sacrifices health for another person’s survival 0.74
Risks death to do something one knows is necessary 0.71
Risks their life for a greater good 0.67
Note: Proposed factor labels are italicized. The three factors
account for 53.8% of the total variance in the data.
Implicit theories of courage 87
classical multidimensional scaling (CMDS) and
cluster analysis, we confirmed the implicit and
theoretical underpinnings of courage. Again, the
components of risk, fear, and noble/worthy purpose
emerged.
Method
Participants. In this second study, 65 participants
completed a web-based ‘‘card-sorting’’ task. Tullis
and Wood (2004) found that card sorts using a
sample size of 30 correlated at 0.95 to a full-
participant study (N¼168), with diminishing returns
as the sample size increased (e.g., a 0.98 correlation
for n¼60 with N¼168). Based on these observa-
tions, we recruited two samples of about 30
participants each. Thirty-five undergraduate students
(26 men and 9 women) at the USAFA participated.
In addition, participants included 30 undergraduate
and graduate students (8 men and 22 women) from
Yale University. Both undergraduates and graduates
from Yale were recruited for this study to ensure an
adequate sample size. The combined ages ranged
from 17–29 years (M¼20.98, SD ¼2.23).
Participants received either course credit or US$5
compensation for their participation.
Materials. Participants received an informed
consent document and completed a background
questionnaire eliciting responses to gender, age, and
educational status. Next, the participants accessed
WebSortÕto complete a web-based, computerized
similarity-sorting task in which they sorted
60 descriptive behaviors into as many or as few
groups as they deemed appropriate. These beha-
viors, listed in Table III, were generated in the
Study 1 free-listing task, and were identified from
Study 2 as the top 60 prototypic behaviors
of courage based on their mean distinctively
characteristic ratings.
Procedure. Participants spent 20 to 30 minutes
completing the background questionnaire and
sorting task. All participants received the same set
of 60 prototypic behaviors, and they were instructed
to sort these behaviors into as many or as few
categories as they deemed appropriate on the basis
of which behaviors were ‘‘likely to be found
together’’ in a person. The participants were not
told in advance what the behaviors had in common
(i.e., that they are descriptive behaviors of courage).
All responses were electronically recorded.
Table III. Prototypicality (characteristicness) ratings of the
behavioral items for an ideally courageous person.
Behavioral descriptors M
Risks one’s own safety to save others in need 8.15
Does the right thing even if it is hard to do 7.97
Fights for what one believes in 7.89
Does what is right even if no one will know 7.82
Acts even in presence of fear 7.75
Risks one’s own life to save others in need 7.71
Stands up for what one believes in the face of strong
opposition
7.71
Stands strong in the face of adversity 7.69
Sacrifices one’s life for another person’s safety 7.63
Does the right thing regardless of consequences to self 7.60
Stands up for what is right even if the majority disagrees 7.60
Persists in one’s actions despite the risk of personal harm or
detriment
7.57
Perseveres in spite of great personal risk 7.56
Does not back down from worthy causes 7.54
Fights for what is true and just 7.53
Does things in spite of personal sacrifice 7.52
Does not give up 7.52
Takes responsibility for actions, even when difficult 7.52
Acts bravely in dangerous situations 7.51
Does not back down when scared 7.50
Acts bravely 7.50
Takes risks for a noble end 7.49
Sacrifices one’s life for another person’s survival 7.48
Does not quit when ‘‘the going gets tough’’ 7.46
Endures hardship if necessary to achieve ideals 7.45
Sacrifices self to save another 7.45
Follows through even if one is scared 7.44
Stands up for what one believes is right 7.44
Remains steadfast in one’s beliefs in the face of persecution 7.44
Explores unfamiliar territory despite fear 7.43
Risks their life for a greater good 7.43
Stands up to unjust social practices because of what one
thinks is right
7.43
Does what is right even if it is unpopular 7.43
Accomplishes something despite personal fear 7.40
Does not abandon others in dangerous situations 7.40
Acts in the face of adversity 7.40
Endures hardship if necessary to achieve mission 7.39
Risks death to do something one knows is necessary 7.39
Behaves in accordance with one’s beliefs, even if the beliefs
are unpopular
7.39
Overcomes challenges 7.37
Acts upon one’s values even if the values are not the norm 7.36
Defends those who cannot defend themselves 7.36
Says no to what is wrong when it is not popular 7.36
Stands one’s ground even if it goes against popular opinion 7.36
Fights through pain and suffering 7.35
Stands up for what one believes no matter what the
circumstances
7.33
Takes responsibility 7.33
Risks life to protect others 7.33
Takes responsibility for one’s mistakes 7.33
Stands up for the ‘‘right thing’’ against opposition 7.33
Displays perseverance 7.33
Tells the truth even if there may be negative consequences 7.32
Stays and helps when others run 7.32
Confronts danger 7.31
Endures tough situations 7.30
Speaks up for one’s beliefs, even under controversy 7.30
Does not surrender under coercive conditions 7.29
Makes the best possible choice to better everyone,
not just self
7.28
Perseveres in face of obstacles 7.28
Rises to face daunting challenges 7.27
88 C. R. Rate et al.
Results
Based on remarkable similarities between Yale
University and USAFA samples ascertained in the
previous studies, both hierarchical cluster analysis
(HCA) and CMDS used combined samples to
confirm previous findings (also based on combined
samples) regarding the implicit underpinnings of
courage. HCA and CMDS were applied to the same
dissimilarity matrix of variances. HCA provides an
empirical method for determining the hierarchical
structure of prototypical categories or concepts
(Rosch, 1978). CMDS allows one to determine
the perceived relative representation, based on an
appropriate number of dimensions, of the items in a
matrix (Hair, Anderson, Tatham, & Black, 1998).
We used both analytic techniques to discover under-
lying structures in our data. HCA and CMDS
do not, however, explain why the clusters or
dimensions exist.
Hierarchical cluster analysis. Cluster analysis was
performed on the variance dissimilarity matrix
using complete linkage (furthest neighbor) amalga-
mation rules. While there is no established objective
procedure for determining the optimal number of
clusters to retain from HCA, one can look at the
coefficients in the agglomeration schedule. The
coefficients indicate the similarity/dissimilarity
between two clusters (or items) joined at each
stage. One looks for a jump in the coefficients. The
stage before the sudden change indicates the optimal
stopping point for merging clusters (Norus
ˇis, 2005).
As with factor analysis and CMDS, interpretability
of the clusters is another important consideration
when deciding on the optimal number of clusters.
Following this guidance, we found a three-cluster
solution to be appropriate.
Three distinct clusters and their associated beha-
viors form the cluster solution for courage. The first
cluster was labeled self-sacrifice/risk for others (n¼14
behavioral items), and included behaviors such as
‘‘sacrifices self to save another’’ and ‘‘risks one’s life to
save others in need.’’ The second cluster was labeled
non-physical/social-oriented acts for noble ends (n¼24).
It included behaviors such as ‘‘speaks up for one’s
beliefs, even under controversy,’’ ‘‘acts upon one’s
values even if the values are not the norm,’’ and ‘‘does
what is right even if it is unpopular.’’ The third cluster
was labeled self-focused perseverance despite fear (n¼22).
This cluster included behaviors such as ‘‘does not back
down when scared,’’ ‘‘accomplishes something
despite personal fear,’’ ‘‘displays perseverance,’’ and
‘‘endures hardship if necessary to achieve ideals.’’
To determine the relative importance of each
cluster to the participants, we averaged the
prototypicality ratings for the behaviors (generated
in Study 2) within a cluster. The means for the three
clusters were 7.48, 7.49, and 7.43, respectively.
A one-way analysis of variance was conducted on
the three cluster means. Results indicated no sig-
nificant difference between the perceived importance
of each cluster.
Classical multidimensional scaling. We analysed the
scaling of the dissimilarity matrix using CMDS with
the ALSCAL program. A non-metric, Euclidean
distance model was specified, along with ordinal,
matrix conditional, untied data. The appropriate
number of dimensions was determined by Kruskal’s
Stress Formula 1, which shows the degree of
‘‘badness’’ of fit between the model and scaled data
and R
2
, which measures the proportion of matrix
data variance accounted for by the model. In
addition, Kruskal and Wish (1978) recommended
that the number of items minus one should be at
least four times as great as dimensionality for
ensuring statistical stability of the analysis.
Table IV shows the CMDS scaling for courage.
Three dimensions accounted for 94% of the variance
in the data, with stress of 0.09. The three-dimen-
sional solution tended to yield bipolar dimensions in
which the positive and negative polarities lent
themselves to separate interpretations. Such bipolar
dimensions are common in this kind of work
(Sternberg, 1985). The first dimension, means to
achieve noble ends, yielded two interpretations:
physical risk/sacrifice for the positive polarity (e.g.,
risks one’s own life to save others in need; sacrifices
self to save another; risks one’s life for a greater good)
and non-physical/social-oriented action for the nega-
tive polarity (e.g., acts upon one’s values even if the
values are not the norm; stands up for what one
believes in the face of strong opposition).
The second dimension, self versus other—
motivation for action, also yielded two bipolar
interpretations: other-focused motivation for the
positive polarity (e.g., does the right thing regardless
of consequences to self; sacrifices self for a cause;
risks one’s life for a greater good) and self-focused
motivation for the negative polarity (e.g., explores
unfamiliar territory despite fear; follows through even
if one is scared; overcomes challenges).
The third dimension, responsibility and enduring
action, further yielded two bipolar interpretations:
responsibility for actions for the positive polarity
(e.g., takes responsibility even when difficult) and
endurance/perseverance for the negative polarity
(e.g., endures hardship if necessary to achieve
ideals; displays perseverance; does not give up).
Figure 1 shows the conjoining of the HCA and
CMDS techniques.
Implicit theories of courage 89
Table IV. Nonmetric multidimensional scaling solutions for courageous behaviors.
Data label Scaling solutions Stimulus coordinates
Dimension 1 (Means to achieve noble ends)
Positive polarity: Physical risk/sacrifice
V60 Risks one’s own life to save others in need 1.87
V1 Risks one’s own safety to save others in need 1.84
V55 Sacrifices one’s life for another person’s survival 1.82
V56 Sacrifices self to save another 1.80
V8 Sacrifices one’s life for another person’s safety 1.80
V59 Risks one’s life to protect others 1.75
V35 Risks death to do something one knows is necessary 1.55
V51 Sacrifices self for a cause 1.54
V31 Does not abandon others in dangerous situations 1.49
V29 Risks one’s life for a greater good 1.37
Negative polarity: Non-physical/social-oriented action
V36 Behaves in accordance with one’s beliefs, even if the beliefs are unpopular 1.83
V46 Speaks up for one’s beliefs, even under controversy 1.81
V33 Acts upon one’s values even if the values are not the norm 1.81
V6 Stands up for what one believes in the face of strong opposition 1.74
V39 Stands one’s ground even if it goes against popular opinion 1.74
V10 Stands up for what is right even if the majority disagrees 1.68
V57 Says no to what is wrong when it is not popular 1.65
V37 Stands up for what one believes no matter what the circumstances 1.64
V26 Does what is right even if it is unpopular 1.61
V25 Stands up to unjust social practices because of what one thinks is right 1.61
Dimension 2 (Self vs. other—motivation for action)
Positive polarity: Other-focused motivation
V50 Makes the best possible choice to better everyone, not just oneself 1.72
V9 Does the right thing regardless of consequences to self 1.63
V41 Defends those who cannot defend themselves 1.36
V43 Tells the truth even if there may be negative consequences 1.36
V29 Risks one’s life for a greater good 1.28
V4 Does what is right even if no one will know 1.13
V18 Does things in spite of personal sacrifice 1.11
V51 Sacrifices self for a cause 1.09
V8 Sacrifices one’s life for another person’s safety 1.04
V60 Risks one’s own life to save others in need 1.01
Negative polarity: Self-focused motivation
V24 Explores unfamiliar territory despite fear 1.69
V38 Overcomes challenges 1.61
V17 Does not give up 1.56
V21 Does not quit when ‘‘the going gets though’’ 1.48
V28 Follows through even if one is scared 1.48
V20 Does not back down when scared 1.44
V30 Accomplishes something despite personal fear 1.43
V53 Rises to face daunting challenges 1.42
V44 Displays perseverance 1.37
V48 Endures tough situations 1.36
Dimension 3 (Responsibility and enduring action)
Positive polarity: Responsibility for actions
V14 Takes responsibility for actions, even when difficult 1.47
V40 Takes responsibility for one’s mistakes 1.45
V45 Takes responsibility 1.28
Negative polarity: Endurance/perseverance
V22 Endures hardship if necessary to achieve ideals 1.44
V44 Displays perseverance 1.35
V21 Does not quit when ‘‘the going gets though’’ 1.23
V13 Does not back down from worthy causes 1.18
V48 Endures tough situations 1.01
V17 Does not give up 1.00
Note: Proposed labels italicized; stress (Kruskal’s stress formula 1) ¼0.09 R
2
¼0.94.
90 C. R. Rate et al.
Discussion
We determined several notable aspects of this
study. First, the proportion of variance in the
CMDS three-dimensional solution accounted for in
the data was 0.94, with stress of 0.09. These indices
of fit indicated the scaling was very reliable, as the
model extracted most of the variance from the
proximity data with a small number of dimensions.
It is therefore reasonable to conclude that the
dimensions obtained reflect people’s notions of the
underlying structure of courage rather than spurious
or random fluctuations of the data.
Second, as one should expect, the three-cluster
HCA solution and the three-dimensional CMDS
solution provided complementary evidence for, and
clear, interpretable representations of, people’s
implicit structures of courage. Figure 1 clearly
indicates a complementary relationship between the
two techniques. Further, as evidenced by HCA,
people found each cluster to be of similar importance
in representing the concept of courage. These
analytical techniques uncovered notable behaviors
illustrative of the concept of courage, such as risking
or sacrificing self, persisting despite fear, and acting
for a noble/worthy purpose.
A third notable aspect of Study 3 results was their
relationship to this article’s Study 2 results. The
clusters and dimensions complemented the factors
obtained in Study 2 with remarkable consistency,
which used a different methodology (factor analysis
of the distinctively characteristic ratings), a different
set of participants, and a superordinate set of 131
behaviors. The three HCA clusters (self-focused
perseverance despite fear; non-physical/social-
oriented acts for noble ends; and self-sacrifice/risk
for others) corresponded quite consistently with the
three Study 2 factors (self-focused response to affect
and external circumstances; non-physical/social-
oriented acts—internal noble motivation; and selfless
sacrifice/risk—external noble motivation). The con-
cern then, that participants merely sorted behaviors
based on the presence or absence of key words
(e.g., fear, risk) appears unfounded in light of this
remarkable consistency between studies. Overall, the
implicit and theoretical underpinning of courage as
obtained in Studies 2 and 3 are very consistent.
Fourth, and arguably most important, the dimen-
sions and clusters appear clearly to represent people’s
notions about the nature of courage. The purpose of
the present study’s analysis, as Sternberg (1985)
would say, is ‘‘not to go beyond what we already
know, but rather to bring what we know into the
open for scientific inspection’’ ( pp. 617, 618). When
we characterize someone’s behavior as courageous or
evaluate someone as possessing ‘‘courage,’’ we must
be prepared to support what we mean by the use of
these terms.
Study 4
In this study, we sought to ascertain the extent
to which people actually used behaviors associated
with courage to evaluate other people’s courage,
V60
V59
V57
V56
V55
V53
V51
V50
V48
V46
V45
V44
V43
V41
V40
V39
V38
V37
V36
V35
V33
V31
V30
V29
V28
V26
V25
V24
V22
V21
V20
V18
V17
V14
V13
V10
V9
V8
V6
V4
V1
Cluster 1
Cluster 2
Cluster 3
Figure 1. Conjoining hierarchical cluster analysis and multidimensional scaling dimensions. Three distinct clusters are represented.
Data labels correspond to behaviors listed in Table IV.
Implicit theories of courage 91
particularly when they are presented with written
descriptions.
Method
Participants. In this study, 169 undergraduate and
graduate students from the USAFA and Yale
University participated. The sample included
92 men and 72 women; ages ranged from 18 to
36 years of age (M¼21.9, SD ¼2.9). Participants
received either course credit or US$10 compensation
for their participation.
Materials. Participants received an informed-consent
document and completed a brief online question-
naire eliciting demographic information. Participants
then completed the vignette-rating task. All partici-
pants received the same 33 vignettes. The vignettes
described scenarios of varied situations and indivi-
dual behaviors and differed from one another only
in that different pairing of names with descriptions
of people were used for different participants (i.e.,
a given vignette was paired half of the time with a
male name and the other half with a female name).
Vignettes were developed, in part, from past real-
world events, while other scenarios were entirely
fictitious. Each vignette was developed with varying
levels of four dimensions (i.e., intentionality/delib-
eration; presence of personal fear; noble/good act;
and known substantial personal risk). These thematic
dimensions were revealed in our previous studies.
1
These dimensions were manipulated along a con-
tinuous scale rather than discrete (present/absent)
variations. Not all combinations of varying levels of
the dimensions could be represented in the vignettes;
however, efforts were made to avoid correlated
pairing levels of dimensions. The following two
examples are typical vignettes used in this study:
(1) Mark (Julie) was surfing off the coast of the
Florida panhandle when he witnesses a 14-year-
old girl being attacked by a tiger shark. The
young victim was in severe agony as she
screamed for help. Although in stunned disbelief
and visibly shaken by what he was witnessing,
Mark immediately began paddling as hard and as
quickly as he could to reach the girl. Mark began
to hit, beat, and kick the tiger shark to ward
off further attacks. He was able to beat back the
7-foot attacker. He then pulled the girl onto
the surfboard and paddled to shore.
(2) Susan (Roger) had been climbing mountains for
a long time. Now she had the opportunity to
climb a 14er (mountain over 14,000 feet in
elevation). Four experienced climbers had
died during the previous year attempting this
same feat. This was the last mountain on Susan’s
list. Nothing was going to stop her from making
this climb. The climb was fraught with unstable
conditions, both from a geological and meteor-
ological standpoint, but Susan dauntlessly suc-
ceeded in reaching the summit.
In real life, evaluations of an individual’s coura-
geous behavior would be largely subjective and
require the observer to make certain inferences
(e.g., level of fear of the protagonist). Therefore, we
maintained the ecological validity of the vignettes
by avoiding overly formulaic designs. Vignettes
were descriptive of ‘‘physical,’’ ‘‘moral,’’ and
‘‘social’’ situations; however, this study did not
explicitly compare these different ‘‘brands’’ or types
of courageous behavior.
Procedure. Participants spent approximately 30 min-
utes completing the computerized, web-based, vign-
ette-rating task. In an effort to assure independence
of dimension and courage ratings, one half of the
participants were instructed to read the vignettes and
then to evaluate the presence of four dimensions.
The other half of the participants were instructed to
read the vignettes and then to evaluate the vignette
protagonist’s level of courage. Participants were
randomly assigned to each condition. All ratings
were made on a 1 (low) to 9 (high) scale. All
responses were electronically recorded.
Results
Initially, we conducted 2 22 factorial ANOVAs
separately for the four dimensions and courage
ratings. The independent variables included
academic institution (USAFA vs. Yale), participant
gender (male vs. female), and vignette name (male
vs. female). All main effects and interactions failed to
reach significance.
Next, we collapsed the data across participants
(using vignettes as the unit of analysis) for reliability
and regression analyses. For each vignette the rating
of courage was treated as the dependent variable, and
the ratings of the four dimensions were the indepen-
dent variables. Table V presents the results of this
study. The first panel displays the basic statistics of
the variable ratings along with the Spearman–Brown
split-half subject and item reliabilities of the ratings.
All ratings were extremely reliable in terms of subject
reliability and respectably reliable in terms of item
reliability.
The intercorrelations among the variables,
displayed in the second panel, show strong, positive
associations between the four dimensions and
courage. In particular, we noticed a strong
92 C. R. Rate et al.
association, r¼0.68, p50.01, between the presence
of personal fear (Fear) and known substantial
personal risk (Risk). The correlations between male
and female vignette protagonist ratings for each
variable were all in the mid- to high-90s. This set of
results indicated that the pattern of ratings was very
similar, regardless of the gender of the vignette
protagonist. Furthermore, the mean ratings of
courage were equal to the hundredth place for male
and female names (M¼5.28). The ratings of the four
dimensions were also similar for male and female
names, respectively: intentionality (M¼6.75, 6.67),
Fear (M¼5.22, 5.42), Noble Purpose (M¼6.12,
6.10), and Risk (M¼5.88, 6.06). Therefore, the
analysis shows that the levels of ratings as well as
rating patterns were the same across the genders of
the vignette protagonists.
Finally, multiple regression was used to predict the
ratings of courage from the rated presence of the four
dimensions (Intentionality, Fear, Noble Purpose,
and Risk). Regression analysis results appear in the
third panel. The regression analysis indicated a
significant R¼0.93, F(4, 28) ¼43.86, p50.001.
The total variance in courage accounted for by
knowing the ratings of the four dimensions was 86%.
The precision index, or standard error of estimate for
this model, was 0.73. This regression model did
exceptionally well in predicting levels of courage
of the protagonists depicted in the vignettes.
Three of the four independent variables con-
tributed significantly to prediction of courage.
Intentionality (sr
2
¼0.02), Noble Purpose
(sr
2
¼0.50), and Risk (sr
2
¼0.06) accounted for a
unique variance of 0.59. The four dimensions
combined contributed another 0.27 in shared varia-
bility. Although the correlation between Fear and
courage was marginally significant, r¼0.31,
p¼0.075, Fear did not contribute significantly to
the regression. It is possible that the relationship
between Fear and courage is redundant to, or media-
ted by, the relationship between Risk and courage.
Discussion
One might have expected dissimilar responses and
evaluations of courage from such disparate samples;
a more male-oriented, conservative/traditional
sample (USAFA) and a more female-oriented, liberal
sample (Yale University). Statistical differences were
not demonstrated by our analyses eschewing the
notion that these samples may perpetuate a certain
type of courage. Regardless of gender or academic
institution, our participants similarly used their
implicit theories of courage in evaluating the courage
of others. Their evaluations were based on relatively
brief scenarios and relatively little information, yet
these judgments could be predicted at high and
accurate levels on the basis of their implicit theories
of courage. Furthermore, participants appeared to
use similar implicit theories in evaluating both male
and female protagonists. This finding alone is
interesting as it is in contrast with traditional
Table V. Results for ‘‘vignette-rating’’ study.
Variable MSD Subject reliability
a
Item reliability
a
Basic statistics
Intentionality 6.73 1.55 0.98 0.78
Fear 5.26 1.71 0.98 0.77
Noble purpose 6.13 2.35 0.99 0.70
Risk 5.96 1.85 0.98 0.75
Courage 5.28 1.83 0.99 0.81
Intentionality Fear Noble purpose Risk Courage
Intercorrelations of ratings
Intentionality 1.00 0.40* 0.09 0.55** 0.44**
Fear 1.00 0.03 0.68** 0.31
Noble purpose 1.00 0.16 0.80**
Risk 10.00 0.57**
Courage 1.00
Variables B sr
2
Standard multiple regression coefficients
Intentionality 0.22 0.19* 0.02 Intercept ¼1.65
Fear 0.04 0.04
Noble purpose 0.56 0.72** 0.50
Risk 0.38 0.38** 0.06
Courage (DV)
R¼0.93*** R
2
¼0.86
b
Adj R
2
¼0.84
a
Reliability was calculated using Spearman–Brown split-half formula on randomly assigned split halves.
b
Unique variance ¼0.59; shared
variance ¼0.27. *p50.05; **p50.01; ***p50.001.
Implicit theories of courage 93
images and explicit studies of men as the sole
exemplars of courageous behavior (Aristotle, trans.
1987; McMillan & Rachman, 1987; Miller, 2000;
Plato, trans. 1987; Rachman, 1990).
The regression model indicated that people
evaluate the levels of intentionality of action, nobility
of purpose, and known substantial personal risk
when assessing others’ behaviors as courageous.
However, the presence of personal fear did not
contribute to the regression model, as expected. This
finding is conceptually contrary to previous implicit-
theory studies (Evans & White, 1981; O’Byrne et al.,
2000; Philips, 2004; Woodard, 2004). These results
are also contrary to our own previous findings.
Studies 1 and 2, for example, provided empirical
support for fear as a core component in people’s
implicit theories of courage. Our factor analysis,
HCA, and CMDS all revealed that ‘‘acting despite
fear’’ was integral to people’s conceptions of the
meaning of courage.
At least three possibilities could account for this
inconsistency. First, although people report fear or
acting despite fear as integral to their conceptualiza-
tions of courage, this component may not play a
necessary or core role in evaluating others’ levels of
courage. Walton (1986) dismisses the emotion of
fear as a ‘‘necessary’’ condition of courage, and Shelp
(1984) also indicates by his equivocation, ‘‘perhaps
fear,’’ that fear may not be necessary in determining
or defining what courage is.
Second, possibly we did not correctly anticipate
how fear would be perceived in the vignettes. Fear
and Risk were highly correlated, r¼0.68, p50.01
and perhaps redundant. Participants may have
been unable to separate the affective component
(personal fear) from the conditions or circumstances
(known substantial risk). Future studies should
make every effort to disentangle fear and risk to
ensure participants’ accurate understanding of the
dimensions.
Third, evaluating the level of fear in others may
be too subjective and, by default, one’s own fear of
the situation becomes the contributing factor. In
other words, it may not be the level of fear of the
protagonist that is predictive of courage, rather
the level of fear of the observer projected on to the
vignette. Evans and White (1981) came to a similar
conclusion. Participants in their study attributed
bravery (courage) to the target actor if the actor was
doing something that would have frightened the
participant. Another notable aspect of this study was
the large contribution of noble/good purpose to the
regression model. The regression model indicated
that, in order for our vignette protagonists to be
evaluated as highly courageous, the motivation for
acting must be present and identifiable as noble
(worthy, good) intentions and ends. Some have
argued this ethical dimension is what separates the
concept of courage from other concepts and beha-
viors, such as bravery and intrepidity (Walton, 1986),
recklessness (Woodard, 2004), and thrill seeking
(Gould, 2005).
In sum, people do use their implicit theories, at
least in part, to make judgments and evaluations of
others’ courage, and these judgments can be
predicted from such implicit theories, as was done
in this study. This study also revealed that the role of
an affective component (e.g., fear) in courage needs
to be further evaluated in terms of its disentangle-
ment with risk and in terms of actor versus observer
levels.
General discussion
What do people think courage is? Those investigating
the psychology of courage have yet to reach
consensus. Despite the vast number and wide-
ranging definitions of courage, we endeavored to
unveil a common core to this psychological con-
struct. This line of inquiry sought to discover the
nature and use of people’s implicit theories of
courage.
However, does the implicit approach really
advance the field of courage studies, when explicit/
scholarly definition of courage abound that
demand further study? The answer to this question
is an unequivocal yes. The implicit–explicit dichot-
omy debate has previously emerged in several
domains such as intelligence (Sternberg et al.,
1981) and wisdom (Sternberg, 1985). Good opera-
tional definitions are important. Implicit- and expli-
cit-theory research exemplifies two components of a
more complete operational definition and
approaches the objective of a universal definition of
courage from different angles. In reference to
implicit theories, Bluck and Glu
¨ck (2005) state that
‘‘the notion that an entire culture would carry around
an organized construct of wisdom [courage] that is
wrong, or incorrect, is a slightly bizarre one’’
( pp. 90, 91). It should also be recognized that
explicit definitions of courage often differ among
researchers, and that these definitions are, in part,
the researcher’s implicit theories of the construct.
Though we have not reached consensus, there
remains substantial convergence between implicit
and explicit research on similar sets of dimensions as
central to descriptions and definitions of courage.
People are able to articulate well-developed,
implicit theories of courage. Although the pioneering
work of O’Byrne et al. (2000) indicates myriad views
and descriptions of courage, they recognize emerging
implicit-theory notions such as taking risks, attitudes,
facing challenges, and defending beliefs. Greatly
expanding on their foundational work, our three
94 C. R. Rate et al.
initial studies indeed confirm the presence of four
necessary components of people’s notions of cour-
age; that is, intentionality/deliberation, personal fear,
noble/good act, and known personal risk.
In Study 4 we discovered that people actually use
behaviors associated with courage when evaluating
other people’s courage, particularly when they were
presented with written descriptions. In fact, we
predicted courage very accurately by knowing
people’s ratings on three of the four dimensions
(intentionality/deliberation, noble/good act, and
known substantial personal risk). Protagonist levels
of personal fear did not significantly contribute to the
model.
It is interesting to note that there still remains
a long-standing debate over the role of fear in
courage. Two key historical figures in this debate
were Aristotle and Plato. Aristotle (trans. 1987)
regarded courage as the disposition to act appro-
priately in situations that involve ‘‘feeling’’ fear. On
the other side of the debate, Plato (trans. 1956), from
Protagoras, described courage as ‘‘the knowledge of
what is and is not to be feared.’’ Plato described
a rational understanding of fear rather than an
emotional/affective feeling of fear as a component
of courage. The fact that people have held these
disparate notions of courage for centuries could help
explain the inconsistencies between Studies 1–3 and
Study 4. Recognizing this complexity could help
inform future thinking and study of the role of fear in
courage.
Do we now know what courage is? Perhaps we are
closer than we had previously thought. As we review
the literature which covers a variety of empirical
approaches to people’s conceptions of courage,
we find that at least four aspects of courage emerge
(namely, risk, fear, intentionality/deliberation, and
noble good). The results of the present research seem
to converge with these findings. This also suggests
that certain core behaviors of courage emerge
regardless of the empirical methodology. The results
of the present study also seem to converge on the
more recent explicit definition of courage described
by Peterson (2006), Peterson and Seligman (2004)
and Snyder and Lopez (2007) as ‘‘the exercise of will
to accomplish goals in the face of opposition, either
external or internal.’’ Parsing their definition indi-
cates similar components that are central to the
definition of courage, such as intentionality/delibera-
tion (exercise of will), noble good (accomplish
goals), and known substantial risk (in the face of
opposition). To reiterate, there is at least tacit
agreement that courage is a complex, multi-dimen-
sional construct.
Using our studies as a basis to know what courage
is, and the present studies’ convergence with the
existing literature, we might best describe or con-
ceptualize courage as: (a) a willful, intentional act,
(b) executed after mindful deliberation, (c) involving
objective substantial risk to the actor, (d) primarily
motivated to bring about a noble good or worthy
end, (e) despite, perhaps, the presence of the emotion
of fear. In general this description of courage is
comprised of situational (external circumstances),
cognitive, volitional, affective, and motivational
components. Perhaps there are additional compo-
nents, but those components may be peripheral to
the meaning of courage. However, if a core dimen-
sion of courage is absent (understanding that fear
may not be a necessary core component), then one
could not appropriately claim a complete definition
or description of courage. Additional studies are
required to address this particular issue.
As with any study, this one comes with its
limitations. This study was not developmental in its
approach to implicit theories of courage. Szagun
(1992) noted differing notions of courage in a cross-
sectional study of children in three age groups: 5–6,
8–9, and 11–12 years old. Older children appear to
understand courage in increasingly psychological and
nonphysical terms. Longitudinal studies, however,
are required to show whether or not these implicit
theories of courage change over the life span.
Szagun’s study, as with this one, was fixed at
a moment in time. We cannot provide evidence
to support or refute the notion that implicit theories
of courage change over time. These studies can only
be interpreted as accurately reflecting the implicit
theories of the participants at the time it was
conducted. Further, our study was limited to one
particular culture, rather than being cross-cultural.
While our results may generalize broadly within our
culture, it is reasonable to assume, based on the vast
number of definitions of courage alone, that concep-
tions of courage may differ from culture to culture.
Another possible limitation of this study was in terms
of the skewed sample of graduate versus under-
graduate participants (Yale sample), and in terms of
gender (USAFA sample). Researchers may also view
our samples as too ‘‘rarified.’’ Future studies could
employ the use of stratified community samples with
ample participants to further gauge the ‘‘every-
person’’ view of courage.
Future directions
Given the limitations and results of the present
study, several paths for future research should
Implicit theories of courage 95
be considered. First, it is important, whether con-
ducting explicit or implicit research, that appropriate
samples are used. The results of the present study
could be validated by first using stratified community
samples, then moving to cross-cultural samples to
garner broader views of the construct of courage.
Although the present study is concerned with finding
a more global conceptualization of courage, specia-
lized samples could target particular and specific
nuances.
In terms of methodologies, ecologically valid
studies might be conducted by developing vignettes
of courage using the methodology proposed by
McFall and Lillesand (1971) for their Conflict
Resolution Inventory. This procedural variation
could accomplish several things: (a) increase the
relevance to the participant’s daily lives; (b) increase
the ecological validity with respect to everyday acts of
courage experienced or witnessed by the sample
participants; (c) increase the potential for interven-
tion applications of the findings; and (d) improve the
assessment of realistic courageous behaviors. In
addition, Frisch and Higgins (1986) describe efforts
to behaviorally and unobtrusively assess assertiveness
that could be applied to the assessment of courage.
As Deutsch (1961) mentioned, it is difficult, how-
ever, to assess the multi-dimensional complexities of
courage in a laboratory setting.
Given the present study’s inconsistent results with
regards to the role of fear (however consistent with
the long-standing major debate) researchers should
continue to investigate fear, and emotion in general,
as it relates to courage. Investigations should focus
on courage and (a) not feeling fear at all, (b) feeling
fear but overcoming it in order to act, and (c) acting
with but despite of fear.
Finally, given the implicit–explicit dichotomy
debate, Rate is conducting an analysis that seeks to
integrate the two views and the myriad conceptua-
lizations of courage into a broad, umbrella concep-
tualization. This conceptualization will initially be
investigated in future vignette-based studies. The
integration of scholarly implicit- and explicit-theory
definitions should provide a solid springboard to
then singularly investigate the components of cour-
age and peripheral influences that have encouraged
branding several types of courage.
Conclusion
What do people think courage is? This study has
provided partial answers to this question. Those
seeking to behave in a courageous way may find it
useful to know what is considered courageous, but
also, what is not. One problem this research will not
resolve is whether any particular act is or is not
courageous because there will always be disagree-
ment as to the ‘‘necessary and sufficient’’ threshold
of its components, such as what is a noble or worthy
purpose. Such disagreements, regrettably, are the
basis of much of the conflict in the world; past,
present, and, in all likelihood, future.
Acknowledgements
The authors would like to express their gratitude to
the anonymous reviewers for their insightful com-
ments on a previous version of this manuscript. Their
feedback was extremely helpful in preparing this
article in its current form. This research was
supported in part by the Department of
Psychology, Yale University, and by a grant under
the Javits Act Program (Grant No. R206R00001) as
administered by the Institute for Educational
Sciences, US Department of Education, awarded to
the Yale Center for the Psychology of Abilities,
Competencies, and Expertise. Grantees undertaking
such projects are encouraged to express freely their
professional judgment. This article, therefore, does
not necessarily represent the position or policies of
the Institute for Educational Sciences, or the US
Department of Education, and no official endorse-
ment should be inferred. The views expressed in
this article also do not reflect the official policy or
position of the United States Air Force, Department
of Defense, or the US Government.
Notes
1. The addition of a volitional, cognitive dimension,
intentionality/deliberation, was included as it
had a strong presence in the Study 1 response-
generation task. It was determined that we
erroneously dropped this component as a modi-
fier (e.g., willingly) early on in this sequence of
studies. We later determined that this dimension
should be included as a potential ‘‘necessary’’
dimension of courage. For these same reasons,
we added the modifier ‘‘known’’ to personal risk
explicitly qualifying a cognitive connection of
‘‘awareness’’ to risk.
References
American Heritage Dictionary (1985). Boston, MA: Houghton
Mifflin Company.
Ardelt, M. (2005). Foreword. In R. J. Sternberg &
J. Jordan (Eds.), A handbook of wisdom: Psychological perspectives
(pp. 84–109). New York: Cambridge University Press.
96 C. R. Rate et al.
Aristotle. (1987). The Nicomachean ethics (J. E. C. Welldon,
Trans.). Amherst, NY: Prometheus Books.
Barnes, M. L., & Sternberg, R. J. (1997). A hierarchical model of
love and its prediction of satisfaction in close relationships.
In R. J. Sternberg & M. Hojjat (Eds.), Satisfaction in close
relationships (pp. 79–101). New York: Guilford Press.
Bluck, S., & Glu
¨ck, J. (2005). From the inside out: People’s
implicit theories of wisdom. In R. J. Sternberg &
J. Jordan (Eds.), A handbook of wisdom: Psychological perspectives
(pp. 84–109). New York: Cambridge University Press.
Cavanagh, G. F., & Moberg, D. J. (1999). The virtue of courage
within the organization. In M. L. Pava & P. Primeaux (Eds.),
Research in ethical issues in organizations (pp. 1–25). Stamford,
CT: JAI Press, Inc.
Clancy, T. R. (2003). Courage and today’s nurse leader. Nursing
Administration Quarterly,27, 128–132.
Comrey, A. L., & Lee, H. B. (1992). A first course in factor analysis
(2nd ed.). Hillsdale, NJ: Lawrence Erlbaum Associates.
Cox, D., Hallam, R., O’Conner, K., & Rachman, S. (1983). An
experimental analysis of fearlessness and courage. British
Journal of Psychology,74, 107–117.
Denver boy honored for aiding his brother (2005, November 29).
Des Moines Register. Retrieved November 30, 2005, from
http://desmoinesregister.com
Evans, P. D., & White, D. G. (1981). Towards an empirical
definition of courage. Behavioral Research and Therapy,19,
419–424.
Fehr, B. (1988). Prototype analysis of the concepts of love and
commitment. Journal of Personality and Social Psychology,55,
557–579.
Finfgeld, D. L. (1995). Becoming and being courageous in the
chronically ill elderly. Issues in Mental Health Nursing,16, 1–11.
Finfgeld, D. L. (1998). Courage in middle-aged adults with long-
term health concerns. Canadian Journal of Nursing Research,30,
153–169.
Frisch, M. B., & Higgins, R. L. (1986). Instructional
demand effects and the correspondence among self-report,
naturalistic, and role-play measures of social skill as
influenced byinstructional demand. Behavioral Assessment,8,
221–236.
Gould, N. H. (2005). Courage: Its nature and development.
Journal of Humanistic Counseling, Education and Development,33,
102–116.
Haase, J. E. (1987). Components of courage in chronically ill
adolescents: A phenomenological study. Advances in Nursing
Science,9, 64–80.
Hair Jr, J.F., Anderson, R.E., Tatham, R.L., & Black, W.C.
(1998). Multivariate data analysis (5th ed.). Upper Saddle River,
NJ: Prentice Hall.
Holliday, S. G., & Chandler, M. J. (1986). Wisdom: Explorations in
adult competence. Basel: Karger.
Ivanhoe, P. J. (2002). The virtue of courage in the Mencius.
In B. Darling-Smith (Ed.), Courage (pp. 65–79). Notre Dame,
IN: University of Notre Dame Press.
Kilmann, R. H., O’Hara, L. A., & Strauss, J. P. (2005). Developing
and validating a quantitative measure of organizational courage.
Manuscript submitted for publication.
Klein, M., & Napier, R. (2003). The courage to act: 5 factors of
courage to transform business. Palo Alto, CA: Davies-Black
Publishing.
Kruskal, J. B., & Wish, M. (1978). Multidimensional scaling.
Beverly Hills: Sage.
Lopez, S. J., O’Byrne, K. K., & Peterson, S. (2003). Profiling
courage. In S. J. Lopez & C. R. Snyder (Eds.), Positive
psychological assessment: A handbook of models and measures
(pp. 185–197). Washington, DC: American Psychological
Association.
Macmillan, T., & Rachman, S. (1987). Fearlessness and courage
in paratroop veterans of the Falkland war. British Journal of
Psychology,78, 375–383.
McCain, J., & Salter, M. (2004). Why courage matters: The way to
a braver life. New York: Random House, Inc.
McFall, R. M., & Lillesand, D. B. (1971). Behavior rehearsal with
modeling and coaching in assertion training. Journal of
Abnormal Psychology,77, 313–323.
Miller, W. I. (2000). The mystery of courage. Cambridge, MA:
Harvard University Press.
Norus
ˇis, M. J. (2005). SPSS 13.0 advanced statistics procedures
companion. New Jersey: Prentice Hall.
O’Byrne, K. K., Lopez, S. J., & Peterson, S. (2000, August).
Building a theory of courage: A precursor to change? Paper
presented at the 108th Annual Convention of the American
Psychological Association, Washington, DC.
O’Connor, K., Hallam, R., & Rachman, S. (1985). Fearlessness
and courage: A replication experiment. British Journal of
Psychology,76, 187–197.
Peterson, C. (2006). A primer in positive psychology. New York:
Oxford University Press.
Peterson, C., & Seligman, M. E. P. (2004b). Character strengths
and virtues: A handbook and classification. New York: Oxford
University Press.
Philips, C. (2004). Six questions of Socrates. New York:
W.W. Norton & Company.
Plato. (1956). Protagoras and Meno (K. C. Guthrie Trans.). New
York: Penguin Books.
Plato. (1987). Laches (J. H. Nichols, Jr. Trans.). In T. L. Pangle
(Ed.). The roots of political philosophy: Ten forgotten Socratic
dialogues. Ithaca, NY: Cornell University Press.
Putman, D. (1997). Psychological courage. Philosophy, Psychiatry
and Psychology,4, 1–11.
Putman, D. (2001). The emotions of courage. Journal of Social
Philosophy,32, 463–470.
Rachman, S. J. (1990). Fear and courage (2nd ed.). New York:
W. H. Feeman and Company.
Rate, C. R. (2007). What is courage? A search for
meaning. Manuscript in preparation. New Haven, CT: Yale
University.
Rosch, E. (1978). Principles of categorization. In E. Rosch &
B. B. Lloyd (Eds.), Cognition and categorization (pp. 27–48).
Hillsdale, NJ: Erlbaum.
Seligman, M. E. P., & Csikszentmihalyi, M. (2000).
Positive psychology: An introduction. American Psychologist,
55, 5–14.
Shelp, E. E. (1984). Courage: A neglected virtue in the
patient–physician relationship. Social Science and Medicine,18,
351–360.
Shepela, S. T., Cook, J., Horlitz, E., Leal, R., Luciano, S., Lutfy,
E., et al. (1999). Courageous resistance: A special case of
altruism. Theory & Psychology,9, 787–805.
Snyder, C. R., & Lopez, S. J. (2007). Positive psychology: The
scientific and practical explorations of human strengths. Thousand
Oaks, CA: Sage.
Sternberg, R. J. (1985). Implicit theories of intelligence, creativity,
and wisdom. Journal of Personality and Social Psychology,49,
607–627.
Sternberg, R. J. (1987). Implicit theories: An alternative to
modeling cognition and its development. In J. Bisanz,
C. Brainerd & R. Kail (Eds.), Formal methods in developmental
psychology: Progress in cognitive development research (pp. 155–
192). New York: Springer-Verlag.
Sternberg, R. J., Conway, B. E., Ketron, J. L., & Bernstein, M.
(1981). People’s conceptions of intelligence. Journal of
Personality and Social Psychology,41, 37–55.
Implicit theories of courage 97
Szagun, G. (1992). Age-related changes in children’s under-
standing of courage. Journal of Genetic Psychology,153, 405–420.
Szagun, G., & Schauble, M. (1997). Children’s and adults’
understanding of the feeling experience of courage. Cognition
and Emotion,11, 291–306.
Tullis, T., & Wood, L. (2004). How many users are enough for
a card-sorting study? In Proceedings UPA 2004 (Minneapolis,
MN, June 7–11, 2004).
Walker, L. J., & Hennig, K. H. (2004). Differing conceptions of
moral exemplarity: Just, brave and caring. Journal of Personality
and Social Psychology,86, 629–647.
Walton, D. N. (1986). Courage: A philosophical investigation.
Los Angeles: University of California Press.
Wegner, D. M., & Vallacher, R. R. (1977). Implicit psychology:
An introduction to social cognition. New York: Oxford University
Press.
Wisconsin father, daughter die in icy pond (2005, November 25).
The Seattle Post-Intelligencer. Retrieved November 20, 2005,
from http://seattlepi.nwsource.com
Woodard, C. R. (2004). Hardiness and the concept of courage.
Consulting Psychology Journal: Practice and Research,56,
173–185.
98 C. R. Rate et al.
