University of Southern California Dissertations and Theses

Implicit theory ideology about human attributes and aggression

(USC Thesis Other)

Implicit theory ideology about human attributes and aggression

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

IMPLICIT THEORY IDEOLOGY ABOUT HUMAN ATTRIBUTES AND
AGGRESSION

by

Douglas Malcolm Stenstrom

A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(PSYCHOLOGY)

December 2008

Copyright 2008 Douglas Malcolm Stenstrom
ii
Table of Contents
List of Tables iv

List of Figures v

Abstract vi

Chapter 1: Introduction 1
Overview 1
Implicit Theories and Entity/Incremental Viewpoints 2
Aggressive Behavior and Implicit Person Theory 7

Chapter 2: Study 1 14
Study 1a 19
Method 19
Results and Discussion 22
Study 1b 27
Method 27
Results and Discussion 29

Chapter 3: Study 2 31
Study 2 32
Method 32
Results and Discussion 34

Chapter 4: Study 3 38
Study 3a 39
Method 39
Results and Discussion 40
Study 3b 41
Method 41
Results and Discussion 42
Study 3c 44
Method 44
Results and Discussion 45

Chapter 5: Study 4 48
Study 4a 49
Method 49
Results and Discussion 51
Study 4b 52
Method 52
Results and Discussion 58

iii
Chapter 6: General Discussion 71

Bibliography 77

Appendix 82

iv
List of Tables
Table 1: Study 1a - Correlation Between Implicit Theories 23
and Aggression Questionnaire

Table 2: Study 1a - Correlation Between Implicit Theories 24
and Measures of Aggression

Table 3: Study 1a - Correlation Between Implicit Theories 26
and Aggression Questionnaire

Table 4: Study 1b - Correlation Between Implicit Theories 30
and Measures of Aggression

Table 5: Study 2 - Correlation Between Implicit Theories 36
and State Version of Aggression Questionnaire

v
List of Figures
Figure 1: Study 4b - 2 x 2 analysis for Hiring Decision aggression measure 63

Figure 2: Study 4b - 2 x 2 analysis for “Intentionality” measure 64

vi
Abstract
Eight studies investigated the relationship between implicit theorist ideology of human
nature and aggression. Implicit theories of human attributes are dynamic meaning
systems used by people to understand, interpret, and predict their social world. A person
who believes human nature is fixed is called an entity theorist because s/he perceives
personality as a stable entity that does not change. A person who believes human nature
is more flexible and malleable is called an incremental theorist because s/he believes
personality can change across time or across different situations. The present research
extends the implicit theory research into the domain of aggression by identifying how
entity theorists perceive and interpret the same events or stimuli differently than
incremental theorists based upon the perceiver’s underlying implicit theories. Study 1a
and 1b investigated trait-based aggression using a wide variety of established measures of
aggression and aggression-related constructs. Study 2 then investigated state-based
aggression using imagined vignettes as the provocation manipulation. Study 3a through
3c extend this line of research by having the participants experience a real immediate
provocation, and also extend this line of research into utilizing a wider range of
aggression measures. Study 4a and 4b then directly manipulated implicit theories to test
the causal role of implicit theories in aggressive responding. In total, the eight studies
tested both trait and state aggression using different paradigms, methodologies, and
subject populations. The combined result was consistent support for trait-based
aggression, such that entity theorists reported increased anger and hostility after being
provoked, increased revenge planning, and increased physical and verbal aggression, but
the aggression was only directed at the actual provocateur, not third parties who were
vii
innocent or shared a group relationship with the provocateur. In terms of state-based
aggression, the combined result was mixed support for implicit theory ideology
influencing aggressive responding, such that using imagined vignettes and a real
immediate provocation produced the hypothesized relationship between implicit theory
ideology and aggression, but the effect did not replicate with new measures of aggression
and did not replicate when directly manipulating implicit theory ideology.
1
CHAPTER 1: INTRODUCTION
Overview
Lay theories of human nature are used by people to understand, interpret, and
predict their social world (Hong, Levy, & Chiu, 2001; Levy, Plaks, & Dweck, 1999;
Levy, Plaks, Hong, Chiu, & Dweck, 2001). Even though people may not be aware of
their own implicit lay beliefs, much research has identified how lay theories influence
self-perception (e.g., Dweck & Leggett, 1988; Ross, 1989; Sternberg, 1985), perception
of others (e.g., Fletcher & Thomas, 1996; Wright & Murphy, 1984), and perception of
groups (e.g., Chiu & Hong, 1999; Levy, Stroessner, & Dweck, 1998; Lickel, et al., 2000;
Pratto, Sidanius, Stallworth, & Malle, 1994; Wittenbrink, Hilton, & Gist, 1998). Lay
theories are able to influence this type of cognitive and behavioral functioning through
establishing a stable meaning system that organizes and directs perceptions and behavior
in accordance with the lay theory (Dweck, 1996; Hong, Levy, & Chiu, 2001; Levy, Plaks,
Hong, Chiu, & Dweck, 2001). In other words, the same experience is perceived and
interpreted differently based on the perceiver’s lay theory. One key underlying lay belief
that shapes subsequent cognitions and behavior is the determination people make about
the fixedness or malleability of human attributes (e.g. Chiu, Hong, & Dweck, 1997; Chiu,
Dweck, Tong, & Fu, 1997; Dweck, Chiu, Hong, 1995a, 1995b; Dweck & Leggett, 1988;
Erdley & Dweck, 1993; Levy et al, 2001; Hong, Chiu, Dweck, Lin, & Wan, 1999; Levy
& Dweck, 1998). For example, a person who believes that an attribute, such as
personality or morality, is a fixed and static entity (referred to as an entity theorist) holds
many different expectations and inferences about that attribute than a person who
believes that the same attribute is malleable and can be changed (referred to as an
2
incremental theorist). Using a variety of measures and manipulations, Dweck and
colleagues have consistently shown that the same experience can be perceived and
interpreted differently based on either an entity or incremental perspective (Dweck, Chiu,
Hong, 1995a, 1995b; Levy & Dweck, 1998; Levy, Plaks, & Dweck, 1999).
The purpose of the present research is to examine how the distinction between
entity and incremental orientations influences the perceptions, inferences, and judgments
related to aggressive responding. Below, I first describe the theoretical reasoning behind
entity and incremental ideology, and show how applying this distinction to aggressive
behavior provides a useful explanatory role in understanding aggressive responding.
Finally, I examine how implicit theories form a framework for influencing cognitive and
behavioral functioning associated with aggression.
Implicit Theories and Entity/Incremental Viewpoints
Originally involving implicit beliefs about intelligence (e.g., Dweck, Chiu, &
Hong, 1995; Dweck & Leggett, 1988; Henderson & Dweck, 1990; Hong, Chiu, Dweck,
Lin, & Wan, 1999; Sternberg, 1985), the distinction between entity and incremental
perspectives has also been investigated in the social domain (e.g., Chiu, Hong, & Dweck,
1997; Erdley, Cain, Loomis, Dumas-Hines, & Dweck, 1997; Erdley & Dweck, 1993;
Levy, Plaks, & Dweck, 1999; Chiu, Hong & Dweck, 1994), in the moral domain (Chiu,
Dweck, Tong, & Fu, 1997; Erdley & Dweck, 1993; Gervey et al., 1999), and in terms of
cognitive processing and stereotyping (Hong, Chiu, Dweck, & Sacks, 1997; Levy &
Dweck, 1999; Levy, Stroessner, & Dweck, 1998; Plaks, Stroessner, Dweck, & Sherman,
2001; Rydell, Hugenberg, Ray, & Mackie, 2007; Tong & Chiu, 2002). Not only is this
implicit theory of human attributes applicable across different content domains, but also
3
across different levels of attribute specificity (e.g., specific attributes of the person,
personal attributes as a whole, or attributes of the social world) and across different levels
of the self-other distinction (e.g., self-perception, perception of individual others, or
perceptions of groups) (Levy & Dweck, 1998; Levy et al., 1999). Even though entity and
incremental theories have been reliably and consistently found in this robust research
record, it is interesting to note that these implicit theories are domain specific. In other
words, an implicit belief in one attribute (i.e. believing that intelligence is fixed) is
independent of an implicit belief in a different domain (i.e. believing that moral character
is malleable) (Dweck, Chiu, & Hong, 1995a; Levy et al., 2001). If the belief in the degree
of malleability changes based on the specific attribute, then how did each implicit belief
develop? Dweck and colleagues (1995a, 1995b) argue that knowing whether something is
fixed or malleable is an essential component of human cognition that children must
develop before they can understand the physical world and social world. This distinction
then serves as one of the most fundamental and underlying characteristics that people
implicitly hold regarding the objects, processes and attributes in their life. This static
versus dynamic view of the world has similarly been conceptualized by philosophers (e.g.
Whitehead, 1929, 1938; Pepper, 1942), historians (Heller, 1967, 1981; Koyre, 1957),
linguists (Bloom, 1981; Whorf, 1956), and other psychologists (Piaget & Garcia, 1983).
Thus, converging viewpoints have postulated that the implicit theory of fixedness versus
malleability is a basic element of cognition that exists in all domains. It is subsequently
refined by socialization, inherent personality differences, and environmental forces
(Dweck, Chiu, & Hong, 1995) in order to form the particular implicit beliefs in each
domain of a person’s life.
4
Whether a person has a fixed or malleable implicit ideology in each domain also
depends on the person’s particular meaning system or knowledge structure (Chiu,
Dweck, Tong, & Fu, 1997; Dweck, Chiu, & Hong, 1995; Levy et al., 2001). For example,
a distinct knowledge structure exists for each domain or attribute and comprises a
network of associated goals, plans, and beliefs relating to either the entity theory or
incremental theory. This knowledge structure is designed to provide a framework for
constructing a narrative of social and moral events (Gervey et al., 1999; Levy et al.,
1999). The knowledge structure has many functions. As an illustration, consider the
moral domain. The first function of the knowledge structure is to organize existing
knowledge and beliefs about the moral domain around either an entity or incremental
viewpoint (Chiu, Dweck, Tong, & Fu, 1997; Dweck, Chiu, & Hong, 1995a; Levy &
Dweck, 1998) and provide a particular set of goals for the individual (Dweck, 1996;
Plaks et al., 2001). Entity theorists view moral character as consisting of fixed traits, and
hence the goal is to identify and measure people’s fixed traits. Traits become the primary
unit of analysis that entity theorists use to understand social events (Chiu, Hong, &
Dweck, 1997; Gervey et al., 1999; Levy et al., 1999; Levy & Dweck, 1999). In contrast,
incremental theorist views moral character as malleable, so the goal is to understand the
dynamics of behavior across different social events. Thus, incremental theorists use
factors other than traits to understand social events, such as external situational forces or
psychological processes within individual (e.g., goals, needs, current mood state, etc.).
The second function of a knowledge structure is to provide a framework for
processing information. Because entity theorists use traits for understanding human
action, they encode and categorize information about events and people in terms of its
5
trait-relevant evaluative meaning (Dweck, Chiu, & Hong, 1995a). Hence, they seek out
trait-relevant information and may disregard information that is inconsistent with their
trait-based construct of the social event (Gervey et al., 1999; Levy & Dweck, 1999). On
the other hand, because incremental theorists use situational and psychological factors to
understand human action, they pay more attention to these mediating factors instead of
trait-relevant information (Dweck, 1996; Levy et al., 2001). And, for both the entity and
incremental theorists, the more coherent their particular representation of the social event,
the less likely inconsistent information will be integrated into this narrative (Gervey et
al., 1999). Thus, the same social event can be perceived in different ways because of the
knowledge structure imposing either an entity or incremental framework.
The third function of a knowledge structure is to provide a framework for making
inferences or judgments. Because entity theorists perceive trait-relevant information more
readily, they use this information as the primary basis for making judgments or inferences
about people or social events (Gervey, et al, 1999; Levy & Dweck, 1999; Levy, Plaks, &
Dweck, 1999). Not only do entity theorist use this information, but they also more readily
believe that trait-relevant information explains behavior. For example, entity theorists
believe that behavior in a single situation implies a dispositional trait, so they are more
likely to draw trait inferences from social information, even when that information is very
limited, ambiguous, or non-trait interpretations for the behavior are provided (Levy et al.,
2001; Levy & Dweck, 1998). Moreover, once an entity theorist believes a person holds a
certain trait, they expect that the person’s behavior will be consistent with that
dispositional trait across time and across different situations (Chiu, Dweck, Tong, & Fu,
1997; Levy & Dweck, 1999). To an entity theorist, traits can predict future behavior
6
because if a trait is fixed, then there will be consistency in behavior that can be
ascertained from even a very limited amount of personal information.
Conversely, because incremental theorists perceive mediating factors more
readily, they use this information as their primary basis for making judgments or
inferences about people or social events (Gervey, et al, 1999; Levy & Dweck, 1998).
They also more readily believe that mediating factors, such as situational forces or
psychological processes (goals, needs, current mood state) explains behavior. For
example, incremental theorists do not believe that behavior in one situation implies
dispositional traits, or that traits can predict behavior in other situations (Chiu, Dweck,
Tong, & Fu, 1997; Levy et al., 2001). In other words, incremental theorists see persons
as malleable, so behavior can be inconsistent across situations, and not the result of an
underlying trait. Thus, the type of knowledge structure (either entity or incremental)
determines the type of inferences that are made about people and events in the moral
domain.
In summary, the distinction between entity and incremental perspectives is one
type of lay theory that people use to understand, interpret, and predict their social world.
An entity perspective provides a framework for constructing a narrative of social events
around dispositional traits. An entity theorist seeks trait-relevant information and tends to
form strong, stable, dispositional inferences in which to explain the causes of behavior
across different situations. On the other hand, an incremental perspective provides a
framework for constructing a narrative of social events around situational and
psychological mediating factors. An incremental theorist, therefore, employs information
about these mediating factors in order to explain the causes of behavior. In short, the
7
distinction between the fixedness or malleability of personality predicts important
cognitive and behavioral consequences.
Aggressive Behavior and Implicit Person Theory
The distinction between entity and incremental perspectives has practical and
theoretical implications for the literature on aggression. Aggressive behavior is a
pervasive interpersonal and societal ill that has engaged psychological researchers since
the seminal 1939 work Frustration and Aggression (Dollard, Doob, Miller, Mower, &
Sears, 1939). Given the serious physical and psychological impact of aggressive
behavior, the literature on aggression has sought to investigate why humans behave
aggressively and how to diminish its harmful effects (Anderson & Bushman, 2002). The
objective of the current research is to further the investigation by examining the role of
implicit theories of human nature in aggressive responding. This section of the current
manuscript will describe past research on implicit theories that are relevant to aggression-
related constructs, and then describe the predictions for how implicit theories of human
nature relate to aggression.
Negative evaluations of others have been a focus of prior research into the effects
of implicit ideology upon judgments and decision making. While not directly pertaining
to aggression, this prior research does provide a baseline from which to understand how
negative evaluations could extend to more severe forms of responding like aggression. In
Chiu, Parker, Hong, & Dweck (1994), for example, participants exposed to both positive
and negative hypothetical behaviors made judgments of whether or not the behaviors
evidenced underlying moral traits such as moral goodness or badness of the actors.
Compared to incremental participants, the entity theorists interpreted the behavior as
8
indicative of the person’s traits irrespective of whether the hypothetical behavior was
positive (risking one’s life for another) or negative (stealing a car). In Erdley & Dweck
(1993) the same finding also extended to negative actions that did not harm others
(plagiarism, telling a lie). The spontaneous trait judgments from minimal negative
information also extended to situations where the participants were not asked to make
trait judgments. In Hong (1994) participants were shown pictures of individuals with
positive and negative emotions that were paired with either a trait word or a state word.
Compared to incremental theorists, the entity participants recognized and labeled the
traits words more quickly than the state words.
Given this finding that entity theorists believe that negative behavior in single
situation implies a negative dispositional trait, research on implicit theories tested
whether entity theorists would then believe that the stable dispositional character of the
individual would persist beyond the immediate context. Consistent with prior work into
the stability of trait judgments (Kunda & Nisbett, 1986; Mischel, 1990; Nisbett, 1980),
Dweck and colleagues found that school-aged children watching a slide show of a
classmate who engaged in negative behaviors (lying, cheating) believed that the
classmate would be “bad” in the short-term and the long-term only if the participant was
a entity theorist, not an incremental theorist. (Erdley & Dweck, 1993). More directly
related to the current research on aggression, Chiu et al. (1994) found that being told a
person is aggressive prompted only entity theorist to indicate that the person would have
similarly aggressive traits in new and different situations.
If a person is expected to have the same negative traits in new and different
situations, then is that information used to explain the behavior in the new and different
9
situations? Hong (1994) presented participants with negative behaviors and asked
participants to explain why certain negative behaviors occurred. Using an open-ended
question format that asked participants to complete the following sentence, “This
probably occurred because…”, entity theorists explained the negative behavior in trait-
based terms. The causal attributions for the negative acts stemmed from trait-based
explanations for entity theorists but not for incremental theorists.
In the final step of this chain of events, how does the trait based inferences and
attributional judgments impact reactions to the target? In Chiu & Dweck (1994) school-
aged children presented with a troublemaking classmate were asked how they would
respond if they were the teacher. Entity theorists recommended more punishment for the
offending classmate. Similarly with adult-aged populations, Gervey et al. (1999)
presented participants with hypothetical scenarios of criminal acts (murder) and then
asked the participants what they felt was the primary function of imprisonment. As
predicted, entity theorists saw punishment as the primary function of imprisonment, and
incremental theorists saw rehabilitation as the primary function of imprisonment. Gervey
and colleagues postulated that believing moral character is fixed implies that
rehabilitation would be useless, so the only function of imprisonment is to punish and
keep this “bad” person out of society. Conversely, incremental theorists would see
rehabilitation as useful because they define moral character as something that can be
changed. While not directly relevant to aggressive behavior, the punitive tendencies of
entity theorists points to likely predictions of how implicit theories will relate to
aggressive behavior.
10
The different information processing approaches of the entity/incremental
orientations has important implications for the literature on aggressive behavior. As
described above, entity theorists and incremental theorists have distinct knowledge
structures that provide a framework for understanding and constructing social events, for
processing information, and for making inferences and judgments. To frame it in terms of
these distinct knowledge structures, entity theorists construct social events using traits as
the primary unit of analysis, so a provocation directed at the individual provides evidence
for the aggressive dispositional character of the provocateur. The behavior from a single
encounter defines the provocateur, from the entity theorist point of view, and allows an
accurate prediction of expected future behavior even when the initial act is limited or
non-trait based interpretation of the act exist. Incremental theorists, on the other hand
actively construct an organization of the events around mediating factors. For the
incremental theorist those mediating forces or psychological processes within individual
(e.g., goals, long term needs, short term needs, current mood state, etc.) then influence
how information is processed about the provocateur, just as the aggressive trait-based
attributions of the entity theorists dictate how they perceive the provocateur. The
resulting judgments and behaviors against the provocateur are a reflection of these
distinct knowledge structures. The prediction is that entity theorists would make a
dispositional inference from the limited incidental character information because they are
more willing to believe that behavior in a single situation implies a dispositional trait.
Entity theorists will more readily use that trait-based information when weighing their
decision about the appropriate response. Conversely, incremental theorists are predicted
11
to exhibit the opposite pattern of responding because incremental theorists perceive the
person as malleable and influenced by other factors.
Notice that the mediating factors within the incremental viewpoint are not the
same as the traditional attributional approach (Weiner, 1971; 1974; 1979) of
internal/external locus of causality. The mediating factors can be both an internal cause
(e.g., stemming from the individual, such as current mood state, goals, needs, etc) and
also an external cause (e.g., stemming from the situation or environment). In the same
vein, the stability/malleability dimension of the implicit theory framework is
conceptually distinct from the traditional attributional approach of three dimensions:
stable/unstable, controllable/uncontrollable, internal/external. While there is some
conceptual overlap (Graham, 1995), the dimensions of the traditional attributional
approach do not fit interchangeably across the domains in which implicit theories are
investigated (Dweck & Legget, 1998) and the stability/unstability dimension is not
always separable from the controllable/uncontrollable dimension in reference to implicit
theories. Given the similarities between the different concepts, Dweck & Legget (1998)
have argued that from a conceptual standpoint the implicit theory framework may be a
source or guide for the traditional attributional dimensions.
One accord between the implicit theory framework and the traditional
attributional approach that relates to the current paper is the emphasis on the relationship
between the individual and their perception of the situational context. Framed within an
overarching theme of Person x Situation interactions (see Funder, 2001; Mischel, 2004;
Shoda, 2004) but more precisely applied to the individuals’ perceptions of the social
event, an entity theorist and incremental theorist respond differently based upon their
12
different informational processing approach. Given a provocation or attack an entity
theorist responds differently to that situation than an incremental theorist because of the
fixed/malleable viewpoint. In other words, just as prior research in domains other than
aggression has found that the same events can be perceived and interpreted differently
based on holding either an entity versus incremental approach (Dweck, Chiu, Hong,
1995), the current research predicts that features of the individual (the implicit theory
framework) interact with situational features (provocations or negative acts upon the
participant) to produce differential responding. A related prediction concerns whether or
not the target of the aggression is the one who evidenced prior bad acts. According to the
theorizing of this manuscript, the entity theorist will respond with aggression because of
the way in which they perceive the negative actions of the provocateur. Others targets
that did not evidence bad acts are not imbued with the same inferences or judgments.
Thus, entity theorists are not more aggressive in general, but more aggressive specifically
based upon perceiving the stable negative character of the provocateur. While this second
hypothesis is not the purpose of the present research, it is an interesting addendum to the
current theorizing that is tested in Study 1.
In summary, the current research investigated the relationship between
entity/incremental orientations and aggression. Eight studies were conducted using
correlational and experimental designs to test state aggression and trait aggression across
different paradigms, manipulations, and samples. The objective was to extend the
research on implicit theories into a new domain, aggression, in order to better understand
how and why aggression occurs following instances of provocation. By investigating the
role of implicit beliefs in aggressive responding the current research provides a better
13
understanding of how cognition influences aggression, and provides an avenue for
investigating new cognitive strategies to reduce aggression through inducing incremental
beliefs. Study 1a and 1b used a correlational design using established measures of direct
aggression, displaced aggression, retaliation, and other aggression-related constructs.
Study 2 manipulated provocations using imagined vignettes similar to previous research
on implicit theories. Study 3a through 3c utilized an ostensibly real provocation to more
accurately investigate real-world instances of aggression, and also utilized a variety of
different measures of aggression for external validity. Study 4a and 4b then extended the
research into a manipulation of implicit theories to test the causal role of implicit theories
in aggressive responding.
14
CHAPTER 2: STUDY 1
The initial test of the relationship amongst implicit theories and aggression
involved the association between well-established measures of implicit theories and
various components of aggression in order to assess the role of implicit theories in trait-
based instances of aggression. The Aggression Questionnaire by Buss & Perry (1992) is
the most widely used measure of aggression. The Aggression Questionnaire measures
four components of aggression: physical aggression, verbal aggression, hostility and
anger. The advantage of using the Aggression Questionnaire is that it is designed to
assess the behavioral (physical, verbal), cognitive (hostility) and emotional components
(anger) that comprise aggression in order to provide convergent validity and external
validity of the relationship between implicit theories of human attributes and aggression.
Since some have argued that aggression is multifaceted with conceptually overlapping
related concepts, multiple measures of aggression are useful for fully capturing the
construct (Ramirez & Andreu, 2006; Suris, Lind, Emmett, Borman, Kashner, & Barratt,
2003). The prediction was that entity theorists will evidence higher trait responses on all
four components of the Aggression Questionnaire.
The primary purpose of Study 1 was to identify the relationship between implicit
theories and the Aggression Questionnaire because the Aggression Questionnaire is the
most prevalent and psychometrically tested measure of aggression in the literature. Then,
beyond establishing that implicit theories correlate to the various components of
aggression measured in the Aggression Questionnaire, other well-established measures of
aggression-related constructs were assessed in order to identify the breadth and scope of
the relationship between implicit person theory and trait-based aggression. For example,
15
convergent validity of the relationship between implicit theories and aggression was
investigated by assessing constructs that confirm the underlying rationale for why
implicit theories of human traits predict aggressive responding. The Norm of Negative
Reciprocity (Eisenberger, Lynch, Aselage, & Rohdieck, 2004) is a measure of the
endorsement of reciprocating against unfavorable treatment by others. Responding higher
on the scale indicates endorsement of a norm of negative reciprocity to others via tit-for-
tat retribution. The Culture of Honor Scale (Cohen & Nisbett, 1994) was originally
develop to tap into reactions of White males from the southern United States which
display a culture of acceptable levels of violence in the face of perceived affronts. The
measure was used in the current manuscript because the content of the items pertain to
more severe forms of aggression, such as murder. In other words, both measures provide
additional value beyond the Aggression Questionnaire, with the Norm of Negative
Reciprocity used to measure the larger concept of tit-for-tat reciprocity, and the Culture
of Honor scale used to measure more extreme forms of responding than is assessed by the
Aggression Questionnaire. The prediction was that the same relationship found between
implicit theories and the Aggression Questionnaire also extends to these other
aggression-related constructs.
Discriminant validity of the relationship between implicit theories and aggression
was also investigated by assessing aggression-related constructs that diverge from the
underlying rationale for why implicit theories predict aggressive responding. For
example, the Displaced Aggression Questionnaire (Denson, Pedersen, & Miller, 2006) is
a measure of trait aggression displayed toward someone other than the original
provocateur. The three components measured in the questionnaire are angry rumination,
16
revenge planning, and displaced aggression. The prediction was that only the “revenge
planning” construct relates to implicit theories because only that component is consistent
with the underlying theorizing that entity theorists are more aggressive because of their
stable (negative) perceptions and interpretations of the actual provocateur. The revenge
planning component falls within the scope of the hypothesis by virtue of the fact the
items are designed to assess thoughts of planning direct revenge against the actual
provocateur. On the other hand, absent the negative information about the target, entity
theorist will not seek retribution, which is why implicit theories are not predicted to relate
to the “displaced aggression” component which pertains to displacing the aggression onto
an innocent third party. In other words, entity theorists are not more aggressive in
general, but only more aggressive toward an offending provocateur.
Another way that the Displaced Aggression Questionnaire provides discriminant
validity of the relationship between implicit person theory and aggression was by
involving an individual difference construct called “angry rumination” that is related to
the construct of aggression, but is also conceptually separable. For example, rumination
is an individual difference tendency to think about the provoking event long after it has
occurred as an emotional strategy to cope with the event (Denson et al., 2006). The
reason why angry rumination as a construct is included in the Displaced Aggression
Questionnaire is because those who ruminate will be more likely to emotionally overreact
and displace aggression onto an innocent third party. Angry rumination, however, is a
personality trait that does not pertain to the theorizing in the current manuscript. Thus, by
showing that angry rumination has no relation to implicit theories provides discriminant
evidence that entity theorists are not more aggressive in general, and are not associated
17
with an aggressive personality, but instead only pertain to the limited situation in which
they perceive negative dispositional traits from the actions of the provocateur.
Thus, the Displaced Aggression Questionnaire provided the ability to test
discriminant validity in terms of behavior that is beyond the scope of tit-for-tat
responding (the “displaced aggression” component”) and also in terms of personality
traits (the “angry rumination” component). In order to provide more confirmatory
evidence of both aspects of discriminant validity, an additional measure was included that
pertains to each aspect. For example, the Vicarious Retribution Scale (Stenstrom,
Denson, & Lickel) measures aggression against the outgroup members of the provocateur
and is designed to tap into a norm of group-based aggression against an outgroup for the
action of one of its members. The prediction was that the Vicarious Retribution construct
correlates with the other measures of aggression included in Study 1, but does not
correlate with implicit theories, thus validating the theorizing of the current manuscript
that it is the actions of the provocateur that produces aggression in entity theorists, and
not that entity theorists are simply more aggressive in general. The Impulsive Aggression
scale (Caprara, Cinanni, D’Imperio, Passerini, Renzi, & Travaglia, 1985) was used to
provide additional support that entity theorists do not have an aggressive personality, but
instead are reacting to the trait-based information provided by the actions of the
provocateur. Impulsive aggression is the tendency to react impulsively at the slightest
provocation, and is argued to be a component of aggressive personalities (Ramirez &
Andreu, 2006; Suris et al., 2003). By showing that the Impulsive Aggression scale
correlates with other measures of aggression, but not implicit theories, Study 1 provides
further discriminant validity of the relationship between entity orientation and aggression.
18
Another way to test both convergent validity and discriminant validity
concurrently was to utilize the Aggression Questionnaire in a new and interesting way.
The twenty-nine items of the Aggression Questionnaire can be categorized by whether or
not each item involves a provocation. For example, the content of some of the items
reference a provocation, such as “Given enough provocation, I may hit another person”
and “If somebody hits me, I hit back.” Conversely, the content of some of the items do
not reference a provocation, such as “I have threatened people I know” and “I wonder
why sometimes I feel so bitter about things.” The positive relationship between implicit
theories and aggression is predicated on an existing provocation. Thus, the prediction was
that implicit theories correlate only with the items that reference a provocation, but not
non-provocation items. More information about how the items of the Aggression
Questionnaire were categorized is provided in the Method section below.
In summary, the relationship between well-established measures of implicit
theories and aggression-related constructs were investigated to show convergent and
discriminant validity with respect to their hypothesized associations. The study was run
in two waves, Study 1a and Study 1b, because having participants respond to all the
measures in the same setting may produce response fatigue. By conducting the study in
two waves (Study 1a, Study 1b), the measures could be split into two groups. Another
advantage of separating into two separate studies was that measures of the same/similar
constructs that have same/similar items can be separated from each other. Since the
purpose of using the measures was to examine how implicit theories relate to the
measures, as long as the measure of implicit theories was included in both measurement
19
waves, splitting data collection into two separate studies would not impact the purpose of
this initial test of the relation between implicit theories and aggression
Study 1a
Method
Participants
Participants were 239 University of Southern California (USC) undergraduates
(174 females, 65 males. All participants were members of the psychology department
subject pool and volunteered for extra course credit.
Materials
Implicit theories were measured using an 8-item scale (Dweck, Chiu, Hong,
1995a). Items representative of entity viewpoints are: “People can do things differently,
but the important parts of who they are can't really be changed.” and “Everyone is a
certain kind of person, and there is not much that can be done to really change that.”
Representative items of the incremental viewpoint are: “People can always substantially
change the kind of person they are.” and “No matter what kind of person someone is,
they can always change very much.” All items were measured on a 6-point scale from
“Strongly Disagree” to “Strongly Agree”. The 8-item scale has been used in prior implicit
theory research and has shown strong psychometric properties of high internal validity
(>.93) and high test-retest reliabilities over 1-week intervals (>.82) and over 4-week
intervals (>.71) (Plaks, Stroessner, Dweck, & Sherman, 2001; Levy, Stroessner, &
Dweck, 1998). The 8-item scale in the present study formed a reliable composite (α =
.93).
20
Aggression and aggression-related constructs were measured in Study 1a with the
Aggression-Questionnaire (Buss and Perry, 1982), the Norm of Negative Reciprocity
(Eisenberger, et al., 2004), Culture of Honor (Cohen & Nisbett, 1994), and the Vicarious
Retribution Scale (Stenstrom et al., 2008). The Aggression Questionnaire is a twenty-nine
item measure of trait-based aggression. There are four components: Physical Aggression,
Verbal Aggression, Hostility, and Anger. Sample items perhaps will provide better
understanding of the four constructs. Representative items of Physical Aggression are
“There are people who pushed me so far that we came to blows.” and “I have become so
mad that I have broken things.” Representative items of Verbal Aggression are “I can’t
help getting into arguments when people disagree with me.” and “When people annoy
me, I may tell them what I think of them.” Representative items of Hostility are “At times
I feel I have gotten a raw deal out of life.” and “Other people always seem to get the
breaks.” Representative items of Anger are “Some of my friends think I’m a hothead.”
and “I have trouble controlling my temper.”. All items were measured on a 7-point scale
from “Strongly Disagree” to “Strongly Agree”. The entire scale in the present study
formed a reliable composite (α = .91), and the four components also formed reliable
composites: Physical Aggression (α = .84), Verbal Aggression (α = .72), Hostility (α =
.79), and Anger (α = .83).
The Norm of Negative Reciprocity (Eisenberger, et al., 2004) is a fourteen item
scale with no sub-factors. The items pertain to emotional, cognitive, and behavioral
responses to a prior provocation. Representative items of the scale are “If someone
dislikes you, you should dislike them”, “If a person wants to be your enemy, you should
treat them like an enemy”, and “If someone distrusts you, you should distrust them.” All
21
items were measured on a 7-point scale ranging from “Strongly Disagree” to “Strongly
Agree”. The scale formed a reliable composite (α = .93).
The Culture of Honor scale (Cohen & Nisbett, 1994) is a six-item scale with no
sub-factors. The items assess intense behavioral responding to perceived transgressions.
Representative items are “A man has the right to kill another man in a case of self-
defense.”, “A man has the right to kill a person to defend his house.”, and “It would be
justified for a man to shoot and kill a neighbor if that neighbor has sexually assaulted the
man’s daughter.” All items were measured on a 7-point scale ranging from “Strongly
Disagree” to “Strongly Agree”. The scale formed a fairly reliable composite (α = .73).
The Vicarious Retribution scale (Stenstrom et al., 2008) measures the response to
an attack on a person’s group or group member. The twenty-item scale involves support
for retaliation and blame against the offending group. Representative items are “When a
member of my group is harmed, other members of the offending group should be
punished.”, When one of my group members is harmed, I am unlikely to support revenge
on members of the other group.”, and “Others who share a group membership with those
who attack my group should be punished for the actions of their fellow group member.”
All items were measured on a 7-point ranging scale from “Strongly Disagree” to
“Strongly Agree”. The scale formed a reliable composite (α = .93).
All measures in Study 1a are included in the Appendix.
Procedure
Participants were instructed to complete a survey of their perceptions of different
issues and situations. Participants were presented with the questionnaire packet online
and asked to work through it at their own pace.
22
Results and Discussion
The primary hypothesis was that a positive relationship exists between implicit
theories and aggressive responding. A positive relationship indicates that agreement on
the implicit theory measure, which represents the entity theorist viewpoint, is associated
with the aggressive responses on the measures of aggression. To test this hypothesis, a
correlational analysis was conducted to analyze the relationship between the measure of
implicit theories and the other measures. Although technically the hypothesized
relationship is linear regression designed to identify if implicit theories predict
aggression, the data is presented as correlations in order to use a correlation matrix that
provides information about the intercorrelations amongst the measures that assesses
evidence of the discriminant validity described in detail below.
Table 1 provides the correlations among the measures from Study 1a. As can be
see in Table 1, implicit theories evidenced a positive and moderate correlation to the
Aggression Questionnaire, as predicted. How are effect sizes from correlations
interpreted? Cohen (1988) provided the most widely known guidelines, suggesting that
correlation coefficients in the order of .1, .3, and .5 are small, medium, and large,
respectively. Since those guidelines were “offered as a convention… for use when no
others suggest themselves” (Cohen, 1988, p. 79), some have argued that Cohen did not
intend his suggestions to be treated with rigidity (Trusty, Thompson, & Petrocelli, 2004).
To provide empirical guidelines to interpret effect sizes, Hemphill (2003) analyzed 380
meta analyses for the lower, middle, and upper third distributions of correlation-based
effect sizes, and found that under .2, between .2 and .3, and above .3, to be small,
medium, and large, respectively.
23
Table 1: Study 1a - Correlation Between Implicit Theories and Aggression Questionnaire
1 2 3 4 5 6

1. Implicit Theories
2. Hostility .19**
3. Anger .18** .50***
4. Verbal Aggression .17** .32*** .54***
5. Physical Aggression .16** .38*** .59*** .52***
6. Overall (composite) .23*** .72*** .84*** .71*** .84***
_______________________________________________________________________
*** = p < .001, ** = p < .01, * = p < .05

How does this apply to the current data in Table1? Table 1 shows that the
correlations are mostly small to medium in effect size. Since the current study involves
measures of prior aggression and/or memory of past occurrences, which are expected to
be weaker than a study involving direct contemporaneous experiences of
provocation/aggression (as tested in Study 2, Study 3, and Study 4), the small to medium
effect sizes are not unexpected.
The measures utilized in Study 1a were designed to show the breadth and depth of
the relationship between implicit theories and aggressive constructs. As predicted,
implicit theories correlated with the Aggression Questionnaire as a whole, as well as to
its four sub-components. In other words, entity theorists exhibited aggressive cognitions
(Hostility component), aggressive emotions (Anger component) and aggressive behaviors
(Verbal Aggression and Physical Aggression components).

24
Table 2: Study 1a - Correlation Between Implicit Theories and Measures of Aggression
1 2 3 4

1. Implicit Theories
2. Culture of Honor .18**
3. Norm of Negative Reciprocity .29*** .58***
4. Norm of Vicarious Retribution .10 .43*** .54***
5. Aggression Questionnaire .23*** .43*** .50*** .48***
_______________________________________________________________________
*** = p < .001, ** = p < .01, * = p < .05

More evidence of the scope of the association is presented in Table 2, which
shows the relationship between implicit theories and the other measures in Study 1a. The
positive and moderate correlation with the Norm of Negative Reciprocity (Eisenberger, et
al., 2004) shows how implicit theories related to broader concepts that assess the
emotional, cognitive, and behavior reactions associated with reciprocity and revenge. The
positive and moderate correlation with Culture of Honor (Cohen & Nisbett, 1994) shows
how implicit theories also related to more intense and extreme responding, such as
killing/murder, in the face of provocations.
The measures in Study 1a were also designed to provide discriminant validity. For
example, as can be see from Table 2, the predicted relationship to the Vicarious
Retribution scale (Stenstrom et al., 2008) is non-significant because that scale assesses
retaliation not toward the provocateur, but other group members, making it exceed the
scope of the hypothesized relationships between implicit theories and aggression. Thus, it
is interesting to note that even if the targets of aggression have a definite relationship to
25
the provocateur, as long as the targets themselves displayed no evidence of initial
provocative aggressive behavior, an entity orientation showed no relationship with
aggressive retaliation.
Importantly, notice that the Vicarious Retribution scale correlated with the Norm
of Negative Reciprocity. The desire to retaliate evidenced in the Normal of Negative
Reciprocity is correlated with the desire to retaliate against other group members in the
Vicarious Retribution scale. Thus, the two scales measure a common construct of desire
to retaliate. However, they diverge from each other in terms of who is the target of the
retaliation, and it is that critical difference between them that results in implicit theories
correlating to the Norm of Negative Reciprocity but not the Vicarious Retribution scale.
Moreover, notice from Table 2 how all the measures of aggression exhibit reliable
correlations with each other, except for the non-significant correlation between implicit
theories and Vicarious Retribution. In other words, the only correlation that is not
significant within Table 2 is the only correlation that was hypothesized to be non-
significant. Thus, the results provide strong evidence of discriminant validity.
Another way to test both convergent validity and discriminant validity was to
utilize the Aggression Questionnaire in a new and interesting way to categorize the items
as either involving a provocation or not involving a provocation. The Aggression
Questionnaire listed in the Appendix has asterisks next to each item that involves a prior
provocation. Table 3 lists the correlations between implicit theories and the items
categorized as involving a provocation versus not involving a provocation. The items are
categorized within the sub-components of the Aggression Questionnaire, such as physical
aggression, verbal aggression, hostility and anger.
26
Table 3: Study 1a - Correlation Between Implicit Theories and Aggression Questionnaire
Implicit Theories

1. Implicit Theories
2. Hostility: Provocation .22***
3. No Provocation .12
4. Anger: Provocation n/a
5. No Provocation n/a
6. Verbal Aggression: Provocation .12*
7. No Provocation .18**
8. Physical Aggression: Provocation .18**
9. No Provocation .11
________________________________________________________________________
*** = p < .001, ** = p < .01, * = p < .05

The hypothesis was that implicit theories related to instances when a prior
provocation has occurred that provides entity theorists with information from which to
make stable dispositional inferences about the provocateur. In other words, the
hypothesis was that entity theorists are not more aggressive in general (in which case the
implicit theory measure should correlate with all the sub-scales in Table 3), but instead,
entity theorists are only aggressive following a provocation (in which case the implicit
theory measure should only correlate with the sub-scales in Table 3 that involve a
provocation). As can be seen from Table 3, the hypothesized relationship exists for the
Hostility component and the Physical Aggression component, but not the Verbal
Aggression component. Since all the items in the “Anger” component fail to refer to a
27
provocation, they are represented by a “n/a”. In other words, the Aggression
Questionnaire is a useful diagnostic tool for testing both convergent and discriminant
validity by showing that only those items that pertain to a prior provocation produced the
hypothesized relationship between entity orientation and aggression. In sum, splitting the
items of the Aggression Questionnaire into items involving a provocation versus no
provocation showed that implicit theories related more strongly to the items involving a
provocation, thus providing more support for the hypothesized relationship between
implicit theories and aggression.
Study 1b
Method
Participants
Data was collected online from 117 participants who responded to an
advertisement posted on a general use website and who indicated they wanted to be
contacted about taking part in future studies.
Materials
Implicit theories were measured using the same 8-item scale (Plaks et al., 2001;
Levy et al., 1998) as in Study 1a. The scale formed a reliable composite (α = .91). The
measures of aggression in Study 1b was the Displaced Aggression Questionnaire
(Denson, Pedersen, and Miller, 2006) and the Impulsive Aggression Scale (Caprara et al.,
1985). All measures in Study 1a are included in the Appendix.
The Displaced Aggression Questionnaire (Denson et al., 2006) is a thirty-one item
measure of aggression with three components: Angry Rumination, Revenge Planning,
and Displaced Aggression. The presentation of representative items from the measure
28
will provide a better understanding of the three constructs. Representative items of Angry
Rumination are: “I keep thinking about events that angered me a for a long time.”, “I
often find myself thinking over and over about things that have made me angry.”, and
“When angry, I tend to focus on my thoughts and feelings for a long period of time.”.
Representative items of Revenge Planning are: “I often daydream about situations where
I’m getting my own back at people.”, and “I think about ways of getting back at people
who had made me angry long after the event has happened.” Representative items of
Displaced Aggression are: “When someone or something makes me angry I am likely to
take it out on another person.” and “Sometimes I get upset with a friend of family
member even though that person is not the cause of my angry or frustration.” All items
were measured on a 7-point scale ranging from “Extremely uncharacteristic of me” to
“Extremely characteristic of me.” The entire scale in the present study formed a reliable
composite (α = .96), and the three components also formed reliable composites: Angry
Rumination (α = .94), Revenge Planning (α = .95), Displaced Aggression (α = .94).
The Impulsive Aggression Scale (Caprara et al., 1985) is a thirty-item scale that
consists of ten control items and twenty content items of impulsive aggression. The ten
control items are indicated by asterisks in the Appendix and are not included in the data
analysis of the Impulsive Aggression Scale. Representative items of impulsive aggression
are: “When I am tired I easily lose control”, “I often feel like a powder keg ready to
explode”, and “Some people irritate me if they just open their mouth.” All items were
measured on a 6-point scale ranging from “Strongly Disagree” to “Strongly Agree”. The
scale formed a reliable composite (α = .88).
Procedure
29
Participants were instructed to complete a survey of their perceptions of different
issues and situations. They were presented with the questionnaire packet online and asked
to work through it at their own pace.
Results and Discussion
Once again, the primary hypothesis was that a positive relationship exists between
implicit theories and aggressive responding. To test this hypothesis, a correlational
analysis was conducted to analyze the relationship between the measure of implicit
theories and the other measures. Table 4 shows the correlation matrix for the measures in
Study 1b. In terms of the Displaced Aggression Questionnaire, the prediction was that
implicit theories correlated with only the Revenge Planning component of the Displaced
Aggression Questionnaire because that component deals with the source of the
provocation, whereas the other components are beyond the theorizing behind why
implicit theories relate to aggression. In other words, entity theorists seek revenge after
experiencing a provocation, but they do not show the personality trait of rumination
(Rumination component), and they do not seek revenge against anyone other than the
provocateur (Displaced Aggression component). As can be seen from Table 4, all
hypotheses were confirmed.
Also as predicted, implicit theories showed no relationship to Impulsive
Aggression. Notice that as a personality trait associated with aggression, the Impulsive
Aggression scale was predicted to correlate with all three facets of the Displaced
Aggression Questionnaire. As seen from Table 4, as expected, the personality trait of
impulsive aggression is associated with angry rumination, revenge planning, and
displaced aggression. At the same time, the impulsive aggression construct was not
30
related to implicit theories, which bolsters the findings from Study 1a that entity theorists
are neither more aggressive in general nor have a more aggressive personality, but instead
are aggressive due to experiencing the provocation. Finally, similarly to Study 1a, notice
that the only non-significant correlations are the relationships that were hypothesized to
be non-significant because they fall outside the scope of why implicit theorists are
hypothesized to display aggressive responding.
To summarize Study 1a and Study 1b, entity theorists reported increased anger
and hostility after being provoked, increased revenge planning, and increased physical
and verbal aggression, but their aggression was only directed at the actual provocateur,
not third parties who are innocent or share a group relationship with the provocateur.
Thus Study 1a and Study 1b provide strong convergent and discriminant validity for the
hypothesized role of implicit theories laid out in the Introduction to the current
manuscript.
Table 4: Study 1b - Correlation Between Implicit Theories and Measures of Aggression
1 2 3 4 5 6

1. Implicit Theories
2. DAQ: Angry Rumination: .14
3. DAQ: Revenge Planning: .22* .51***
4. DAQ: Displaced Aggression: .10 .49*** .33***
5. DAQ: Total .20* .86*** .79*** .70***
6. Impulsive Aggression: .12 .60*** .48*** .73*** .75***
________________________________________________________________________
*** = p < .001, ** = p < .01, * = p < .05
31
CHAPTER 3: STUDY 2
As seen from Study 1, the initial test of the relationship between implicit theories
and aggression investigated psychometrically valid measures of both implicit theories and
aggression-related constructs. The advantage of this type of design is that it assessed the
accumulated trait-based experiences of one’s life, and thereby measuring a wide range of
conceptually overlapping aggression-related concepts. The disadvantage of this design,
however, is the retrospective nature of asking people to recall their own reconstructed
experiences. Study 2 was designed to overcome this limitation by testing the relationship
between implicit theories and aggression in a contemporaneous instance of
provocation/aggression. The expectation was that the weak to moderate relationships in
Study 1 are stronger in Study 2 because Study 2 investigated a present instance rather
than past aggression as measured in Study 1.
Similar to previous research on implicit theories presented in the Introduction of
the current manuscript, participants were asked to imagine scenarios after assessing their
level of implicit theory. As the present research concerns aggression, the imagined
scenarios were provocations directed at the participant, and the measures at the end of the
study were aggression constructs. When moving to investigate a present instance of
aggression, rather than recalling past instances of (trait) aggression as was measured in
Study 1, the measure of aggression needs to be a state-version of aggression. A state
version of the Buss-Perry Aggression Questionnaire has been developed for research on
state aggression (Karrar & Krcmar, 2006). This state version of the Aggression
Questionnaire is described in more detail in the Method section below.
32
As an overview, Study 2 was a between-subjects design in which participants
were asked to read one of four imagined provoking situations. Data was collected on the
USC campus as a field-study where potential participants were approached on campus to
take part in the psychology survey. One of the disadvantages of a field-study is that
outside the confines of a controlled laboratory environment there is the possibility that
participants are distracted by a variety of environment factors. To control for distraction
in Study 2, the experimenters coded the level of distraction.
Study 2
Method
Participants
Data was collected from 229 participants on University of Southern California
(USC) campus who were approached by the investigator and agreed to take part in the
study. The data from 84 participants were excluded due to not paying continuous
attention to the study.
Materials
Implicit theorizing was measured using a 3-item variation of the 8-item scale to
be used in Study 1. The 3-item scale is the more widely used version of the two scales
(Dweck, Chiu, & Hong, 1995a). It has high internal reliability (>.90), high test-retest
reliability (>.82), and is highly correlated with the original 8-item scale (>.83) (Dweck,
Chiu, & Hong, 1995a; Levy et al., 1998). Previous psychometric testing also showed that
agreement on the items was not a result of social desirability concerns, self-monitoring,
cognitive abilities, political attitudes, sex or age. The three items are: “The kind of person
someone is, is something very basic about them, and it can’t be changed very much.”,
33
“People can do things differently, but the important parts of who they are can’t really be
changed.”, and “Everyone is a certain kind of person, and there is not much that can be
done to really change that.” All items were measured on a 6-point scale ranging from
“Strongly Disagree” to “Strongly Agree”. The scale in the present study formed a reliable
composite (α = .86).
Study 2 used four imagined vignettes: (1) Later today you are walking home from
school when a mugger angrily pushes you to the ground and forcefully rips your money
from your wallet/purse before running away, (2) Later today you are on campus and
someone bumps into you, spilling your drink and the contents of your bag. They then
begin to curse and step toward you aggressively, (3) Later today you are driving on the
highway. The driver in the car in front of you throws some garbage out the car window,
and it smashes against your windshield, (4) Later today you are driving on the highway.
Your off ramp is quickly approaching. The driver next to you is purposefully driving in a
manner that is preventing you from changing lanes. You miss your exit.
The state measure of aggression (Farrar & Krcmar, 2006) consists of 11 items that
transform the Physical Aggression and Verbal Aggression components of the Buss-Perry
Aggression Questionnaire into a situation specific measure. For example, a representative
item from the original trait Buss-Perry Aggression Questionnaire for the Physical
Aggression component is “Once in a while I cannot control the urge to strike another
person”. That question is transformed into “I could not control my urge to strike this
person” in the state version of the scale. Similarly, a representative item from the Buss-
Perry Aggression Questionnaire for the Verbal Aggression component is “I often find
myself disagreeing with people”. That question is transformed into “I would find myself
34
disagreeing with this person” for the state version of the scale. The last item from the
state version was dropped from investigation because I deemed it confusingly
approaching a double-negative (e.g., I can think of no good reason for ever hitting a
person.). Of the remaining 10 items, four pertain to Verbal Aggression and six pertain to
Physical Aggression. The items were measured on a 6-point scale ranging from
“extremely uncharacteristic of me” to “extremely characteristic of me.” The items are
presented in the Appendix. The entire scale in the present study formed a reliable
composite (α = .8), and the two components also formed reliable composites: Physical
Aggression (α = .89) and Verbal Aggression (α = .79).
Procedures
The study was conducted as a field study on the USC campus. Potential
participants were approached and asked to take part in a psychology survey. The
experimenter obtained informed consent and then presented the survey, starting first with
the measure of implicit theory, then an imagined scenario, and finally the measure of
aggression. For the imagined scenario, the instructions read: “Please read the following
statement. Try to imagine yourself in this situation and that the incident is actually
happening to you.” Participants were randomly assigned to read one of the four imagined
vignettes.
Results and Discussion
To examine the hypothesis that there was a positive relationship between implicit
person theories and aggression, two types of analyses were conducted. First, the
correlation between the independent variable and the dependent variable was examined.
A positive correlation indicates that there is a positive relationship between implicit
35
theories (independent variable) and aggression (dependent variable). Second, I examined
whether the type of vignette that each participate read affected the correlation between
the independent variable and the dependent variable. For example, four vignettes were
used in the present research. Each participant read one of the four vignettes. If there was a
positive correlation between the independent variable and the dependent variable, it is
possible that the positive correlation was due to only one or two of the vignettes, but not
all the vignettes. The hypothesis was that there will be no difference between the
vignettes because the predicted relationship between implicit theories of personality and
aggression should be evident across all four vignettes, and not dependent upon a limited
instance of only one or few of the vignettes.
As can be seen from Table 5, implicit theories had a positive and medium
relationship with both verbal aggression and physical aggression subscales of the
aggression measure used in the study. In other words, implicit theories related to the level
of physical and verbal aggression from the state version of the Aggression Questionnaire.
The combined composite of both verbal and physical aggression also showed a positive
and medium relationship to implicit theory ideology. As expected, the relationship
between implicit theories and aggression in Study 2 was shown to be stronger than the
similar relationship in Study 1. If participants previously excluded due to distraction were
included into data analysis, the results remain the same with moderate effect sizes that are
highly significant for all three constructs, verbal aggression (r = .20, p < . 01), physical
aggression (r = .24, p < .001), and the overall scale (r = .25, p < .001). Thus, the results
from Study 2 testing state-based aggression replicate the findings from Study 1a and
Study 1b that tested trait-based aggression.
36
Table 5: Study 2 - Correlation Between Implicit Theories and State Version of
Aggression Questionnaire
1 2 3

1. Implicit Theories
2. Verbal Aggression .24***
3. Physical Aggression .28*** .55***
4. Overall (composite) .30*** .80*** .93***
________________________________________________________________________
*** = p < .001, ** = p < .01, * = p < .05

Because Study 2 was a between-subjects design in which each participant
received only one of the four vignettes, the correlation coefficients for each vignette are
independent of each other. Thus, the way to test the impact of the vignettes is to treat the
correlation (between implicit theories and aggression) within each vignette as a separate
unit of analysis, and determine if there are significant differences between the
correlations. A statistical test exists to examine the difference between two independent
correlation coefficients. The procedure involves testing each pairwise comparison for a
total of six pairwise comparisons while controlling for familywise error. The statistical
test involves calculating a z-score using a Fisher’s r-to-z which is compared to a unit
normal distribution to determine the significant value (Cohen and Cohen, 1983). Since
this statistical test is not part of SPSS, a website is available with an imbedded script that
calculates the necessary information (Preacher, 2002). The prediction was that the six
pairwise comparisons are non-significant, indicating the type of vignette had no impact
upon the relationship between implicit theories and aggression. As predicted, all pairwise
37
comparisons were non-significant (ps > .59) irrespective of whether the participants who
were distracted are included or excluded from analysis.
38
CHAPTER 4: STUDY 3
Study 2 was the first test of state aggression and implicit theories. One limitation
of Study 2, however, was that the provocation is imagined by the participants. The
purpose of Study 3 was to extend this line of research into an ostensibly “real”
provocation that the participants experience. Participants were provided with an
ostensibly real article from the Daily Trojan that insulted USC students for whining about
tuition increases. The manipulation is presented in the Appendix. The article was taken
from a field study on Triggered Displaced Aggression (Vasquez, Lamoreaux, & Miller,
2008).
Another limitation of Study 2 was that it only used one type of measure – the state
version of the Aggression Questionnaire. Thus, a second purpose of Study 3 was to
utilize a wider range of measures of aggression. For example, Study 3b employed an
open-ended response format instead of the closed-ended response format from Study 1
and Study 2, and Study 3c used an ostensibly “real” measure of aggression in which
participants were lead to believe that they had the opportunity to aggress against the
author of the article in a way that had real-life negative consequences for the provocateur.
The third purpose of Study 3 was to employ a control condition. Notice that the
theorizing about the relationship between implicit theories and aggression speaks only to
a situation in which a provocation has occurred. In other words, the absence of a
provocation is beyond the scope of the underlying hypothesis. Study 3c involved a
control condition as well as a provocation manipulation. The control condition was
another ostensibly real Daily Trojan article, but instead of an insulting content, the
content of the control article was neutral. The article was taken from a field study on
39
Triggered Displaced Aggression (Vasquez, Lamoreaux, & Miller, 2008). The article is
presented in the Appendix.
In summary, three separate studies were conducted that replicate the underlying
hypothesis and then extend the findings into new areas. All four studies used the new
ostensibly “real” provocation in which USC students are insulted by the Daily Trojan
article, with the differences between the three studies were: (a) Study 3a utilized the state
version of the Aggression Questionnaire, (b) Study 3b used an open-ended response
format, and (c) Study 3c employed a control condition, as well as employed an ostensibly
real measure of aggression in which participants were lead to believe that they had the
opportunity to aggress in a way that had real-life negative consequences for the
provocateur,
Study 3a
Method
Participants
Data was collected from 100 participants on University of Southern California
(USC) campus who were approached by the investigator and agreed to take part in the
study. The data from 25 participants were excluded due to not paying continuous
attention to the study.
Materials
The measure of implicit theories and measure of aggression were the same as in
Study 2. Implicit theories were measured using the 3-item scale (Dweck et al., 1995a)
and aggression was be measured using the state version of the Aggression Questionnaire
(Farrar & Krcmar, 2006). The scales formed reliable composites for both implicit theories
40
(α = .94) and the state version of the Aggression Questionnaire (α = .84). The two
components of the state version of the Aggression Questionnaire also formed reliable
composites: verbal aggression (α = .80) and physical aggression (α = .82). The
provocation was induced by reading a Daily Trojan article that insults students for
whining about tuition increases. The article is reproduced in the Appendix.
Procedure
The study was a field-study on the USC campus. Participants were approached by
the experimenter and asked if they are a USC student. Because the provocation within the
Daily Trojan article is directed at USC students, it was essential that the participants in
the study were USC students. The experimenter was introduced as a researcher for the
university and asked if the participant has a few minutes to fill out a survey. The survey
as introduced as an opinion survey about an article that appeared in the Daily Trojan.
Participants were told that their responses are completely confidential and anonymous.
No identifying information was collected from the participants. The participants were
then directed to the first page of the survey, which measures implicit theories. The second
page of the survey was the Daily Trojan article. Below the Daily Trojan article was the
measure of aggression that directed the participant to answer the questions in regard to
the author of the Daily Trojan article. At the conclusion of the study the completed
survey from the participant was inserted into a blank envelope to ensure confidentiality
and anonymity.
Results and Discussion
Similar to the data analysis of Study 2, the correlation between implicit theories
and the measure of aggression was examined. The prediction was that implicit theories
41
have a positive and medium relationship to the aggression measure used in the study
following the ostensibly real provocation. It is also expected that the relationship between
implicit theories and aggression in Study 3 was stronger than the same relationship in
Study 1 which investigated trait-based aggression measures. As predicted, implicit
theories showed a relationship with the state version of the Aggression Questionnaire (r =
.26, p = .03). There was also a relationship the physical aggression composite (r = .25, p
= .02) but not the verbal aggression composite (r = .17, p = .14). If the participants
previously excluded due to distraction were included into the analysis, the results become
slightly stronger with the same moderate effect sizes for all three constructs, overall
composite (r = .34, p < . 001), physical aggression (r = .36, p < .001), and verbal
aggression (r = .21, p < .05).
Study 3b
Method
Participants
Data was collected from 122 participants on University of Southern California
(USC) campus who were approached by the investigator and agreed to take part in the
study. The data from 49 participants were excluded due to not paying continuous
attention to the study.
Materials
The measure of implicit theories and the Daily Trojan article were the same as in
Study 3a. The implicit theory measure formed a reliable composite (α = .88). The
addition for Study 3b is a new measure of aggression using an open-ended format.
Participants are asked: “What is your opinion of the author? What do you think about the
42
author of the article? We are interested in your opinion.” The explanation of how the
open-ended responses will be coded is explained in the “Results” section below.
Procedure
The procedure for Study 3b was identical to the procedure for Study 3a.
Results and Discussion
The open-ended responses were coded via four questions. Three undergraduate
coders coded each response. Question 1 of the coding manual asked the coders whether
the open-ended response was about the Daily Trojan article, about the author of the
article, or both. It is important to distinguish the content of the open-ended response
because a response that said the author was abusive, for example, is conceptually distinct
from one that sees the article itself as abusive. The reliability analysis indicated an alpha
of .74.
Question 2 of the coding manual asked the coders to rate “To what extent is the
text negative or positive?” using an 11-point scale ranging from -5 to 5. The instructional
text for the coders states: “For this question, you are giving your gestalt impression of the
entire response. For example, there could be both positive and negative comments, so
take them all into account when deciding the overall positivity versus negativity of the
text. It is possible that some of the negative comments are cancelled out by the positive
comments, and vice versa. It is possible that some of the negative comments are partially
cancelled out by the positive comments, and vice versa. In other words, ask yourself
whether overall the text is positive or negative (or neutral), and then how much positive
or negative?” The reliability analysis indicated an alpha of .86.
43
Question 3 of the coding manual asked the coders to rate “To what extent is the
text negative?” using a 6-point scale from 0 to 5. In Question 2, the purpose of having a
response format the ranged from negative (-5) to positive (5) is to capture the entire
continuum from positive to negative. The purpose of Question 3 is to focus only on the
negative aspects of the response, so the 6-point scale captures no negativity (0) to strong
negativity of the response (5). The instructional text for the coders states: “For this
question, you are only looking at the negative comments (and not using the positive
comments). For example, if there are no negative comments, put a 0. If there are negative
comments, ask yourself how negative the text is overall.” The reliability analysis
indicated an alpha of .86.
Question 4 of the coding manual asked the coders to rate “To what extent is the
text mean/angry/aggressive/hostile/etc?” using a 7 point response format ranging from
“not at all” to “very much”. The purpose of Question 3 is to measure the extent to which
the response is “negative”. The purpose of Question 4 is more limited and asks the extent
to which the response falls within the concept of angry or aggressive. The instructional
text for Question 4 states: “For this question, we are looking at the generic concept of
aggression. For example, “aggression” in this case could refer to the text being mean,
snarky, hostile, rude, obnoxious, sarcastic, angry, aggressive, etc. The reliability analysis
indicated an alpha of .83.
The prediction was that the new measure of aggression will evidence the same
level of relationship between implicit theories and aggression as found in Study 3a. Data
analysis produced non-significant findings (ps > .42) for the relationship between implicit
44
theories and aggression for all the open-ended questions, irrespective of whether or not
the distracted participants were included.
Study 3c
Method
Participants
Data was collected from 163 participants on University of Southern California
(USC) campus who were approached by the investigator and agreed to take part in the
study. The data from 68 participants were excluded due to not paying continuous
attention to the study. Given the nature of the deception in the study, the participants
were given a funnel debriefing at the conclusion of the study to identify if any of the
participants were suspicious about the true nature of the study, and 19 participants
indicated suspicion.
Materials
The measure of implicit theories was the same as in Study 3a and 3b. The
measure formed a reliable composite (α = .88). The provocation from the Daily Trojan
article was also the same as in Study 3a and 3b, however, in Study 3c a control condition
was added containing a neutral Daily Trojan article. The article is reproduced in the
Appendix. The neutral Daily Trojan article will be taken from a field study on Triggered
Displaced Aggression (Vasquez et al., 2008).
In addition to employing a control condition, the other addition for Study 3c was a
new measure of aggression. Because the provocation concerns a Daily Trojan reporter
insulting the participants, the measure of aggression gave the participants the opportunity
to evaluate the author of the article for an editorial position at the Daily Trojan. For
45
example, the participants were told that the Daily Trojan initiated a new procedure for the
hiring of editorial positions, and that since those positions offer a reasonable salary and
look good on a resume there have been more applicants than the number of positions
available, so the Daily Trojan decided to gather input from the public to help them make
their hiring decisions. Participants were informed that the author of the article they just
read was an applicant and the Daily Trojan would like their feedback regarding this
individual’s suitability for the position. The advantage of conducting the study as a field
study on the USC campus was that it is reinforces the belief that the Daily Trojan is
interested in getting feedback from actual USC students about the author of the article.
The measure consisted of five items: “I recommend this author for the position.”,
“I think this author should be hired for the position.”, “I think the author is a competent
individual.”, “I like the author.”, and “I think the author is intelligent.” All items were
measured on a 7-point scale ranging from “Strongly Disagree” to “Strongly Agree”. The
measure formed a reliable composite (α = .90).
Procedure
The procedure for Study 3c paralleled that of Study 3a except that participants
were randomly assigned to either the provocation article or the control article.
Results and Discussion
In terms of the provocation condition, the hypothesis was similar to that of
previous studies in the current manuscript, namely, that there was a positive relationship
between implicit theories and aggressive responding. In terms of the control condition,
there was no prediction because it is outside the scope of the theorizing behind why
implicit theories relate to aggression. For example, one possibility is that in the neutral
46
condition entity theorists will view a stable neutral opinion of the author, in which case
the author will be evaluated more neutrally on the measure aggression than the author in
the provocation condition. However, this does not necessarily mean that incremental and
entity theorists will evaluate the author of the article differently, which is the central
hypothesis for the studies in the current manuscript. As described in the Introduction of
the current manuscript, in the absence of a provocation, there is no reason for the
incremental and entity theorists to differentially evaluate the author of the article. In order
to test the interaction between implicit theory and condition, type of Daily Trojan article
(neutral versus provocation) was treated as a categorical variable that moderates the
relationship between aggression as a continuous linear dependent variable and the
continuous independent variable (measure of implicit theories). Interaction analysis
between categorical and continuous variables was conducted using linear regression
techniques. The result was a main effect of condition such that the provocation condition
elicited higher aggression than the neutral condition (beta = .49, p < .001), but there was
no effect between condition and implicit theory (beta = -.41, p = .22). In other words,
entity and incremental theorists showed no distinction in level of aggression between the
provocation article and the control article.
To summarize Study 3a, 3b and 3c, the primary purpose was to use an ostensibly
real provocation that the participants experience, rather than the imagined vignettes from
Study 2. In Study 3a using the new provocation from the Daily Trojan article, entity
theorists reported higher levels of aggression, as predicted, replicating the results from
Study 1a and 1b, and Study 2. The second purpose was to use new measures of
aggression, such as open-ended response format in Study 3b, and an ostensibly “real”
47
measure of aggression in Study 3c in which participants were lead to believe that they
had the opportunity to aggress against the author of the article in a way that had real-life
negative consequences for the provocateur. The predicted results were not confirmed for
either measure. The third purpose was employing a control condition. In Study 3c, the
participants were randomly assigned to either the provocation condition or the control
condition. The predicted results were not confirmed. Discussion of possible
interpretations for the null results is presented in the General Discussion when discussing
the limitations of the eight studies in the present article.
48
CHAPTER 5: STUDY 4
Study 1 through Study 3 investigated the role of endorsing implicit theory
ideology in aggressive responding. Study 4 extended this line of research by directly
manipulating the propensity to endorse implicit theory ideology. By manipulating
implicit theories, it is possible to test its causal role in predicting aggression. In all prior
studies in the current manuscript, implicit theories were measured before the provocation
and before the measure of aggression, thus providing indirect support for the causal role
of implicit theory ideology. Study 4, however, was designed to more directly test that
assertion by manipulating implicit theories before the provocation and measure of
aggression. Two studies were conducted: (a) Study 4a was a lab-based study in which
participants were randomly assigned to either the entity manipulation or incremental
manipulation, then presented with the provoking Daily Trojan article from Study 3, and
finally asked to answer the state version of the Aggression Questionnaire, (b) Study 4b
was an online study similar to Study 4a in that the participants were randomly assigned to
either the entity manipulation or incremental manipulation, but in Study 4b the
participants were then also randomly assigned to the provocation manipulation or control
manipulation, and answered a larger range of dependent measures that included two
measures of aggression, attributions about the provocation, and trait judgments about the
author of the article. Given that Study 4b measured possible reasons why implicit theory
ideology may impact aggression, namely attributions about why the provocation occurred
and trait judgments about the perpetrator, it was necessary to counterbalance the order of
the dependent measures, such that for half of the participants the first set of measures
involved aggression followed by attributions measures, and for the other half of the
49
participants the first set of measures involved attributions followed by the aggression
measures. Thus, Study 4b was a 2 x 2 x 2 between-subjects design. The purpose of the
counterbalancing of the dependent measures was that measuring the mediating
attributional variables before the measure of aggression may dilute the aggressive
response, but conversely, measuring the mediators before aggression increases the
validity of a mediational analysis by having the mediators precede the main outcome
measures of aggression. Counterbalancing the ordering of the measures will thus
maximize the potential validity of the study by ensuring that both goals of the study have
the best chance of being obtained.
Study 4a
Method
Participants
Participants were 102 University of Southern California (USC) undergraduate
members of the psychology department subject pool who volunteered for extra course
credit. The data from 27 participants were removed due to suspicion of the manipulation
of implicit theory, suspicion of the manipulation of provocation versus control, or
suspicion of the hypotheses in the study.
Materials
The manipulation of implicit theorist ideology was implemented by using a
manipulation developed by Dweck and colleagues that has been used successfully in
prior studies to directly manipulate a belief in either entity or incremental viewpoints
(Chiu, Hong, & Dweck, 1997; Hong, Chiu, Dweck, Lin, & Wan, 1999; Levy, 1998;
Levy, Plaks, Hong, Chiu, & Dweck, 2001; Levy et al, 1998). The manipulation was an
50
ostensibly “real” New York Times article which cites evidence and support from several
reputable sources that scientists have consistently found that people are stable and do not
change (entity viewpoint) or that people are malleable and do change (incremental
viewpoint). Both articles are reproduced in full in the Appendix.
The manipulation check was “To what extent do you believe that a person’s traits
or characteristics are stable?” rated on a 9-point scale ranging from “Not at all” to “Very
much.” On the same page as the manipulation check was filler items to bolster the
ostensible cover story told to the participants that the experimenters are interested in
social comprehension and memory of various types of newspaper articles. The filler
items were: “To what degree was the article informative?”, “How would you rate the
interest level of the article?”, “How understandable was the article?”
The other materials from the study were the same as Study 3a in which the
participants read the provoking Daily Trojan article and then completed the state version
of the Aggression Questionnaire. The state version of the Aggression Questionnaire
formed a reliable composite (α = .77). The two components of the state version of the
Aggression Questionnaire also formed reliable composites: verbal aggression (α = .77)
and physical aggression (α = .73).
Procedure
Participants were greeted by the experimenter and given the consent form to read
and sign. Next, the experimenter explained the ostensible purpose of the study that the
experimenters were interested in social comprehension and memory of various types of
newspaper articles. Participants were told they would read some newspaper articles and
51
answer questions including their memory of what they read. Participants were told the
first article was a summary of scientific findings and the second article was an editorial.
The study contained two phases. In Phase 1 the participants read one of the
manipulation articles, and then completed a questionnaire containing the measurement
check and filler items. Next, during Phase 2 participants read the provocation
manipulation from the Daily Trojan article, and then completed the state measure of the
Aggression Questionnaire. The study was conducted over two semesters. During the
second semester the manipulations were stapled together with the relevant questionnaire,
while during the first semester they were not.
Results and Discussion
The prediction was that the manipulation of implicit theories would induce an
entity or incremental viewpoint that causes higher levels of state aggression in the entity
condition compared to the incremental condition. After analyzing the manipulation check
items, statistical analysis tested the effect of the manipulation upon the dependent
variable, and also tested use the manipulation check as the independent variable to see if
the reported level of endorsement of implicit theory ideology from the manipulation
check influences the dependent variable.
The manipulation of entity versus incremental viewpoint influenced the
manipulation check item in the predicted direction, such that participants reported that
they believed a person’s traits or characteristics to be more stable in the entity
manipulation condition (M = 6.7) than the incremental manipulation condition (M = 5.3),
t(73) = -3.99, p <.001. The manipulation, however, did not influence the participant’s
responses on the measure of aggression using the state version of the Aggression
52
Questionnaire. For example, there was no difference on the state version of the
Aggression Questionnaire as a whole for entity versus incremental manipulation
conditions (M = 3.0, M = 3.4, respectively, t(73) = 1.55, p = .13) or for the verbal
aggression components (M = 4.1, M = 4.4, respectively, t(73) = 1.27, p = .21), or the
physical aggression component (M = 2.4, M = 2.7, respectively, t(73) = 1.24, p = .22). In
fact, the direction of the non-significant effect was the reverse of the hypothesized
relationship between the manipulation of entity versus incremental viewpoints upon
aggressive responding.
In order to further explore the data, a correlation was conducted between the
manipulation check and the dependent measures of aggression to identify if the
participants’ responses to the manipulation check indicating the level of entity versus
incremental belief had a relationship to aggressive responding, but no relationship existed
to the construct as a whole (r = .07, p = .58) or the verbal aggression component (r = .05,
p = .69) or the physical aggression component (r = .06, p = .62). In other words, no only
did the manipulation itself not have a relationship with aggression, but the self-reported
implicit theory belief from the manipulation check had no relationship to aggressive
responding.
Study 4b
Method
Participants and Design
Participants were 308 University of Southern California (USC) undergraduate
students (98 males and 210 females) who were solicited from the Psychology Subject
Pool, online advertising from Facebook targeting USC undergraduate students, and flyers
53
posted around campus. Given that the provocation manipulation within the study
concerned insulting USC undergraduate students, it was necessary for the participants to
be USC undergraduate students, so a question within the study screened for those
participants who were undergraduate USC students. The data from 17 participants were
removed due to not being undergraduate students at USC. The data from an additional 40
participants were removed due to not completing the study in a timely fashion. One
disadvantage of online studies compared to laboratory-based studies is that experimenters
are not present to ensure participants maintain attention to the study and possibly
discontinue/restart the study over a period of time. In the current online study a time-
stamp at the beginning and conclusion of the study allowed the experimenter to
determine the length of time each person participated in the study, which allowed the
experimenter to identify those subjects who take an uncharacteristically short or long
amount of time to finish the study.
Materials
The manipulation of implicit theorist ideology was implemented by using the
manipulation developed by Dweck and colleagues used in Study 4a. Also, similar to
Study 4a, Study 4b employed the same manipulation check of implicit theory ideology as
described in Study 4a, and the same state measure of the Aggression Questionnaire as
described in Study 4a. The new additions to Study 4b were the following materials.
Following the presentation of the implicit theory ideology manipulation,
participants answered an additional set of manipulation check items that also served to
reinforce the manipulation. For example, the original manipulation check item from
Study 4a that has been used successfully in prior published studies with the ostensibly
54
real “New York Times” entity/incremental manipulation (Chiu, Hong, & Dweck, 1997;
Hong, Chiu, Dweck, Lin, & Wan, 1999; Levy, 1998; Levy, Plaks, Hong, Chiu, & Dweck,
2001; Levy et al, 1998) asks the participants to what extent they believe that a person’s
traits or characteristics are stable. Thus, this manipulation check item is a check of how
strongly the participants believe in the stability/malleability of personality after reading
the articles. An additional set of manipulation check items was included in Study 4b that
checked how thoroughly the participants read the ostensibly real “New York Times”
article. The purpose of including this additional set of items was to reinforce the
manipulation by inducing participants to read the article more carefully in order to
answer a set of questions that asked about details of the study. For example, participants
were asked, “What is the main message of the article?”, with three answer choices: (a)
Personality characteristics are fixed entities, (b) Personality characteristics change over
time, (c) both a and b. The second question asked, “Why does the author begin the article
with the two examples of Mary S. and Benjamin M.?, with three answer choices: (a)
Mary S. and Benjamin M. are representative examples that personality characteristics are
fixed, (b) Mary S. and Benjamin M. are representative examples that personality
characteristics are malleable, and (c) both a and b. As you can see, participants must
identify the content of the article that they read, thus allowing the experimenter to screen
out those participants who did not read the article carefully. Notice that the original
manipulation check item used in past published research can be answered by the
participant without having ever read the manipulation articles, or by having quickly
skimmed the manipulation article without grasping the message of the article that serves
as the actual manipulation of implicit theory ideology. In the present research it was
55
possible to screen out inattentive participants while also reinforcing the manipulation
itself by having participants actively answer questions about the stability/malleability of
personality.
An added benefit of these manipulation check items was that it bolstered the
explanation that the questions were designed to assess general reading comprehension.
Two additional questions were included in the set of manipulation check items that
inducted the participants to more thoroughly read the article. For example, an additional
question asked: “How many longitudinal studies did Dr. Paul Medin publish about the
stability [malleability] of personality?”, with three answer choices: (a) 0-3, (b) 4-6, and
(c) 7-9. An additional question asked: “Why does the article report multiple evidence
from three different professionals?”, with three answer choices (a) An article that is about
research findings should contain confirmation from multiple sources, (b) Multiple pieces
of evidence show that the research findings are consistent, (c) both a and b.
Following the manipulation of implicit theory ideology, participants were then
randomly assigned to either the provocation manipulation (the Daily Trojan article from
Study 4a) or the control condition, which was the neutral Daily Trojan article from Study
3c. Both articles are reproduced in full in the Appendix.
Participants then completed the sets of dependent measures. For example, in
addition to the state version of the Aggression Questionnaire from Study 4a, participants
also completed the measure of aggression from Study 3c that gave the participants the
opportunity to evaluate the author of the article for an editorial position at the Daily
Trojan. For example, participants were told that The Daily Trojan initiated a new
procedure for the hiring of editorial positions, and that since those positions offer a
56
reasonable salary and look good on a resume there have been more applicants than the
number of positions available, so the Daily Trojan decided to gather input from the public
to help them make their hiring decisions. Participants were informed that the author of
the article they just read was an applicant and the Daily Trojan would like their feedback
regarding this individual’s suitability for the position. The measure consisted of three
items: “I think this author should be hired for the position.”, “I think recommend the
author for this position.”, and “I believe the author is qualified for the position.” All items
were measured on a 7-point scale ranging from “Strongly Disagree” to “Strongly Agree”.
Participants also completed a set of attributional and trait-based measures. For
example, trait judgments of the author of the article were assessed by asking the
participants to indicate their impressions and opinions of the author of the Daily Trojan
article on the following negative and positive traits: hostile, arrogant, aggressive, closed-
minded, malicious, annoying, rude, foul person, warm, pleasant, honest, likeable, kind,
friendly, good, nice. The purpose of including this set of trait judgments is to identify if
the manipulation of implicit theory ideology influences trait-based judgments of the
author of the article which then influences the level of aggressive responding. Another set
of attributional questions were included to identify if the manipulation of implicit theory
ideology influences participant’s attributions of the perpetrators actions which then
influences the level of aggressive responding. For example, questions asked about the
level of internal factors (To what degree is the cause(s) an internal factor of the author
(i.e., due to his/her personality?), external factors (To what degree is the cause(s) due to
situational factors that are external to the author?), controllability (To what degree was
the author’s inclination to write the article something that s/he could control?), stability
57
(To what degree is the cause(s) something stable/permanent about the author (as
compared to changeable)?), intention (To what degree is the impact or effect that the
article has on readers one that is intended by the author?), inherent character (To what
extent did the author write an article expressing negative opinions because that is the kind
of person he is?),the possibility that the author of the article could change (To what extent
can the author change the kind of person he is?), and the possibility that the author of the
article would express regret for his statements (To what degree do you think the author
regretted his statements afterwards?). All items were measured on a 7-point scale ranging
from “Not at all” to “Very Much”. The order of trait-based measures and attributional
measures was counterbalanced across the study, and the items within each set of
measures was full randomized.
A final set of questions at the conclusion of the study ascertained the degree to
which the participants agreed or disagreed with the statements from the Daily Trojan
article. For example, for those prior studies in the current manuscript that utilized the
Daily Trojan manipulation of provocation but did not find evidence significant results, it
is possible the lack of significant findings may be due to the fact that the participants
agreed to some degree with the statements from the provoking Daily Trojan article that
USC students are self-absorbed children who are annoying everyone with their constant
whining about tuition increases. In other words, the provocation manipulation may be
stronger for those participants who disagree more with the statements from the article.
Thus, three questions asked the participants to indicate: (1) “To what extent do you agree
with the author’s opinions in the Daily Trojan article?”, (2), “To what extent do you think
the author was correct in his statements in the Daily Trojan article?”, (3) “To what extent
58
do you think the article was an accurate depiction of the topic?” All items were measured
on a 7-point scale ranging from “Strongly Disagree” to “Strongly Agree”.
Procedures
Participants began the study by reading the online consent form and clicking a
button to indicate they agreed to participate in the study. The manipulations and measures
occurred in the same order as described above in the “Materials” section of the current
study. The only caveat is that the order of dependent measures was counterbalanced such
that for half of the participants the first set of measures involved aggression followed by
mediational measures (trait-based judgments and attributional judgments), and for the
other half of the participants the first set of measures involved mediators followed by the
aggression measures. At the conclusion of the study the participants were debriefed and
thanked for participation.
Results and Discussion
The manipulation check of implicit theory ideology showed the expected
relationship that those participants in the entity manipulation condition evidenced higher
belief in the stability of personality (M = 4.82) than the participants in the incremental
manipulation condition (M = 3.21), F(1, 240) = 108.48, p <.001. Regarding the
reinforcing manipulation check items that were new to Study 4b, 8 participants were
removed from analysis for not identifying the correct answer to the questions.
To examine the trait-based judgments, a Maximum Likelihood factor analysis
using Direct Oblimin rotation identified a coherent set of the eight negative items, and a
coherent set of the eight positive items, with the caveat that the trait “honest” showed a
weaker relationship than the other positive items to the grouping of positive items.
59
However, reliability analysis of the positive set of items indicated that whether “honest”
was included in the set of positive items (alpha = .92) or exclude from the set of positive
items (alpha = .95), the reliability of the composite did not appreciably change. The
negative set of items formed an equally strong composite (alpha = .95). And combining
together the negative set of items with the positive set of items (by first reverse coding
the positive set of items so that higher on all items indicates a stronger negative
impression of the perpetrator) again produced an equally reliable composite (alpha = .95).
In terms of the measures of aggression, the 10 items of the state version of the
Aggression Questionnaire formed a reliable composite (alpha = .82) and the three items
of the ostensibly real opinion survey of the Daily Trojan regarding the employability of
the author of the article formed a reliable composite (alpha = .87). In the remainder of
this article, this second measure of aggression will be referred to as the “Hiring Decision”
aggression measure.
In summary, there are three factors in the study: 2 (Entity manipulation versus
Incremental manipulation) x 2 (Provocation versus Control) x 2 (ordering of dependent
measures as either Mediator first – Aggression second, versus Aggression first –
Mediators second). There are also two measures of aggression (state version of the
Aggression Questionnaire, and Hiring Decision aggression measure) and two measures of
mediators (trait-judgments, and attributional measures). Since the items of the
attributional set of items measure discrete constructs, each of those items will be treated
separately, resulting in 12 outcome variables for the following data analysis: (1)
composite of state version of Aggression Questionnaire, (2) composite of Hiring Decision
aggression measure, (3) composite of negative items of trait-based judgments, (4)
60
composite of positive items of trait-based judgments, (5) the attributional item concerning
internal factors, (6) the attributional item concerning external factors, (7) the attributional
item concerning controllability, (8) the attributional item concerning stability, (9) the
attributional item concerning intention, (10) the attributional item concerning regret
afterwards, (11) the attributional item concerning inherent character, and (12) the
attributional item concerning the ability of the author to change.
A 2 x 2 x 2 multivariate analysis was conducted on all 12 outcome measures.
None of the three-way interactions were significant, as predicted. In terms of the two-way
interactions amongst the 2 x 2 x 2 analysis, the central hypothesis concerns the
interaction between the manipulation of implicit theory ideology (entity versus
incremental) and the manipulation of provocation (provocation versus no provocation).
More precisely, the primary hypothesis concerns aggression measures, namely that the
manipulation of implicit theory ideology will interact with the Daily Trojan manipulation
such that participants induced into the entity theorist perspective compared to the
incremental perspective will evidence aggression after being provoked as compared to
the no provocation condition. The state measure of the Aggression Questionnaire
evidenced no interactive effect (F(1, 230) = .20, p = .66) and the Hiring Decision
aggression measure was marginal (F(1, 230) = 3.37, p = .068). The secondary hypothesis
concerns the same 2 x 2 interaction, but now concerns the trait-based judgments and
attributional measures. None of the trait-based judgments and attributional measures
evidenced a significant interaction except for the “intention” variable (F(1, 230) = 4.04, p
= .05). As will be explained in more depth below, however, the “intention” variable was
reverse of the hypothesized direction. Thus, before discussing further analysis, it is
61
important to re-state the preceding results that, of the 12 outcome variables, the only
variable approaching significance in the expected direction was the Hiring Decision
aggression measure.
What of the other two types of interaction amongst the 2 x 2 x 2 analysis? For the
interaction between the implicit theory ideology manipulation (entity versus incremental)
and the ordering of variables (Aggression-Mediators versus Mediators-Aggression), if an
interaction exists then it implies that the ordering of the variables in some way influences
the outcome. In other words, an ideal situation would be no interactions for the 12
outcome variables. Data analysis found two interactions, one for the “controllability”
item (F(1, 230) = 9.45, p < .001) and one for the “kind of person he is” (F(1, 230) = 5.22,
p = .02). For the final type of interaction, namely between the provocation manipulation
(Provocation versus No provocation) and the ordering of variables (Aggression-
Mediators versus Mediators-Aggression), if an interaction exists then it implies that the
ordering of the variables in some way influences the outcome. In other words, once
again, an ideal situation would be no interactions for the 12 outcome variables. Data
analysis found no interactions. Thus, to summarize, 2 of the 12 outcome variables
evidence influence by the ordering of the outcome variables within the study, but neither
relate to the primary hypothesis regarding aggression measures
Another way to analyze the data is by further refining the analysis by using the
composite about the degree to which the participants agreed or disagreed with the
statements from the Daily Trojan article as a covariate. The same type of 2 x 2 x 2
multivariate analysis on the 12 outcome variable was repeated, but this time including the
covariate. How does including the covariate change the findings from the study? The
62
same results emerge, except now the findings are somewhat stronger. For example, the
Hiring Decision aggression measure is now significant (F(1, 229) = 5.13, p = .03), and
the “intention” item is slightly more significant (F(1, 229) = 4.18, p = .05) when
analyzing the 2 (entity/incremental) x 2 (provocation/no provocation). And the two
variables in the 2 (entity/incremental) x 2 (Ordering) analysis are essentially the same,
namely the “controllability” item (F(1, 229) = 9.52, p < .001) and “kind of person he is”
(F(1, 229) = 4.24, p = .04). No other results are significant, similar to the findings when
conducting the 2 x 2 x 2 analysis without the covariate.
To summarize the data analysis conducted up to this point, the primary hypothesis
of Study 4b was confirmed for the Hiring Decision aggression measure but not the state
version of the aggression measure. Figure 1 shows the nature of the interaction between
the manipulation of implicit theory and the manipulation of provocation compared to no
provocation. Follow-up analysis revealed that for participants in the “Provocation”
condition, there was not a significant difference between the participants induced with the
entity orientation (M = 5.44) compared to incremental orientation (M = 5.28), F(1, 123) =
.64, p = .43; whereas for participants in the “No Provocation” condition, there was a
significant difference between the participants induced with the entity orientation (M =
3.28) compared to incremental orientation (M = 3.76), F(1, 113) = 5.14, p = .03. The
secondary hypothesis regarding the trait-based judgments and attributional measures was
not supported, with the only item showing the predicted interaction as “intentionality” but
in the opposite direction, as shown in Figure 2.
63
Figure 1: Study 4b - 2 x 2 analysis for Hiring Decision aggression measure

Another way to analyze the data is to screen out participants who did not
thoroughly read the article by using the two questions designed to assess whether the
participants thoroughly read the article: (1) “How many longitudinal studies did Dr. Paul
Medin publish about the stability [malleability] of personality?”, and (2): “Why does the
article report multiple evidence from three different professionals?”. The data from 140
participants were removed for not correctly identifying the answer to each question,
leaving 102 participants within the data analysis. Repeating the 2 x 2 x 2 multivariate
analysis revealed no three way interactions, a 2 (entity/incremental) x 2 (Provocation
versus no provocation) interaction for the “intended” question (F(1, 101) = 6.74, p = .01),
64
and a 2 (entity/incremental) x 2 (Ordering) interaction for “control” item (F(1, 101) =
7.17, p = .01) and Hiring Decision aggression measure (F(1, 101) = 4.30, p = .04). In
other words, the central hypothesis about the interaction between the implicit theory
manipulation and provocation manipulation revealed no interactive effects upon
aggression.
Figure 2: Study 4b – 2 x 2 analysis for “Intentionality” measure

To further explore the data, the manipulation check item of implicit theory
ideology was analyzed as a predictor of the 12 outcome variables. In other words, similar
to the data analysis conducted in Study 4a, instead of using the manipulation of implicit
theory ideology (Entity versus Incremental) as a predictor within the analysis, the
65
manipulation check indicating the level of entity versus incremental belief was entered as
a predictor of the outcome variables. An interaction analysis was conducted between the
continuous manipulation check item and categorical manipulation of Provocation (versus
No Provocation) while using the composite about the degree to which the participants
agreed or disagreed with the statements from the Daily Trojan article as a covariate. The
interaction analysis revealed no effect upon the 12 outcome variables except for the
Hiring Decision aggression measure which produced a significant interactive effect (beta
= .24, p = .02). In other words, the self-reported implicit theory belief from the
manipulation check produced the same outcome as the implicit theory manipulation
itself.
A final way in which to analyze the data involves examining the relationship
between the current dataset and a self-reported level of implicit theory ideology acquired
at the beginning of the semester via a prescreening survey for the participants sampled
through the USC subject pool. In other words, a prescreening survey contained the same
3-item implicit theory ideology measure used in Study 2 and 3. Thus, for a subset of the
participants in the current study who were acquired via the USC subject pool, it was
possible to identify their level of self-reported implicit theory ideology. A total of 138
participants answered both the prescreening 3-item measure of implicit theory ideology
and also completed the current study. The 3-item measure from the prescreening survey
formed a reliable composite (α = .84).
This self-reported level of implicit theory ideology can be used in a variety of
ways to better understand the current dataset. For example, the first type of analysis is to
examine the relationship between the 3-item implicit theory measure from the
66
prescreening survey and the manipulation check item of implicit theory ideology. The
correlation between the two measures was small and non-significant (r = .01, p = .89)
which indicates that the participants’ level of implicit theory ideology changed between
the prescreening survey and the current study, which were approximately 4 weeks apart.
Previous psychometric testing of the implicit theory measure found a high test-retest
reliability (>.82, Dweck, Chiu, & Hong, 1995a) which suggests that another factor,
beyond changes over time, affected the implicit theory ideology in the current study. One
possibility is that the manipulation of implicit theory ideology was, as intended, the
intervening cause of the change in implicit theory ideology. If so, then it may be possible
to use the 3-item prescreening measurement of implicit theory ideology to strengthen the
manipulation of implicit theory ideology in the current study by identifying the
participants whose prescreening level of implicit theory match the manipulation of
implicit theory in the current study. In other words, randomly assigning participants to
conditions in the current study results in entity theorists (as self-reported by the
prescreening survey) to fall in the incremental manipulation, and vice versa. In this case,
their preexisting level of implicit theory ideology may have muted the manipulation.
Ideally, the manipulation would have been the dominant force that overrode any
preexisting implicit theory ideology, which was the specific purpose of implementing the
manipulation. However, the results in this study are weak and inconsistent. Thus, one
way to re-analyze the data in a way that strengthens the manipulation is by identifying the
participants whose prescreening level of implicit theory match the manipulation of
implicit theory in the current study. Previous research into the 3-item implicit theory
ideology measure categorizes incremental theorists as those participants in the bottom
67
40% of the scale, and categorizes entity theorists as those participants in the top 40% of
the scale (see Dweck, Chiu, & Hong, 1995a for a review). After categorizing the
participants according to the prescreening measure of implicit theory ideology, the result
was 46 participants whose implicit theory orientation from the prescreening survey
matched the corresponding manipulation in the current study. Data analysis conducted on
those 46 participants evidenced no significant interactions for all 12 outcome variables on
the hypothesized interaction between implicit theory manipulation and the manipulation
of provocation versus no provocation.
Another way to use the 3-item prescreening measure of implicit theory ideology is
to treat that variable as a covariate while examining the hypothesis about the interaction
between implicit theory manipulation and the manipulation of provocation versus no
provocation. The multivariate analysis on the 12 outcome variables showed a marginal
interaction for the Hiring Decision aggression measure (F(1, 130) = 3.27, p = .073), a
marginal interaction for the “intention” measure (F(1, 130) = 3.82, p = .053), and a
marginal interaction for the “regret” measure (F(1, 130) = 3.52, p = .063). Regarding the
interaction for the Hiring Decision aggression measure, follow-up analysis revealed that
for participants in the “Provocation” condition, there was not a significant difference
between the participants induced with the entity orientation (M = 5.66) compared to
incremental orientation (M = 5.22), F(1, 74) = 1.20, p = .28; and for participants in the
“No Provocation” condition, there was also no significant difference between the
participants induced with the entity orientation (M = 3.22) compared to incremental
orientation (M = 3.64), F(1, 55) = 1.77, p = .19. Regarding the interaction for the
“intention” measure, follow-up analysis revealed that for participants in the
68
“Provocation” condition, there was not a significant difference between the participants
induced with the entity orientation (M = 4.06) compared to incremental orientation (M =
4.50), F(1, 74) = .80, p = .37; and for participants in the “No Provocation” condition,
there was a significant difference between the participants induced with the entity
orientation (M = 5.03) compared to incremental orientation (M = 4.41), F(1, 55) = 4.94, p
= .03. Finally, for the “regret” measure, follow-up analysis revealed that for participants
in the “Provocation” condition, there was not a significant difference between the
participants induced with the entity orientation (M = 2.74) compared to incremental
orientation (M = 2.52), F(1, 74) = .26, p = .61; and for participants in the “No
Provocation” condition, there was a significant difference between the participants
induced with the entity orientation (M = 1.68) compared to incremental orientation (M =
2.37), F(1, 55) = 7.28, p = .01. In summary, using the 3-item prescreening measure of
implicit theory ideology as a covariate in the analysis revealed the same general pattern
within all the analysis in Study 4b that the hypothesis about the interaction between
implicit theory ideology and provocation versus no provocation was marginally
supported for the Hiring Decision aggression measure, but not the other measure of state
aggression.
Beyond the purpose of the present research to examine the relationship between
implicit theory ideology and aggression, another way to utilize the self-reported level of
implicit theory ideology from the prescreening survey is to identify the degree to which
the manipulation of implicit theory ideology influenced the participants’ particular beliefs
in implicit theory ideology. In other words, to what degree did the manipulation of
implicit theory change the participant’s self-reported level of Entity/Incremental beliefs
69
between the prescreening survey and the manipulation check of Entity/Incremental
beliefs in the current study. The purpose of answering this question is to better
understand the weak and inconsistent results found in Study 4b. For example, a 2
(prescreening categorization as Entity versus Incremental) x 2 (manipulation of Entity
versus Incremental) univariate analysis on the manipulation check of implicit theory
belief evidenced no interaction, a main effect of manipulation (F(1, 107) = 49.69, p <
.001), and a marginal main effect of categorization from the prescreening survey (F(1,
107) = 3.00, p = .086). More specifically, for the main effect of manipulating implicit
theory beliefs, the mean level of belief in stability of personality from the manipulation
check was higher for the Entity manipulation (M = 4.86) than for the Incremental
manipulation (M = 3.07). For the marginal main effect of the prescreening categorization,
the mean level of belief in stability of personality from the manipulation check was
higher for the Entity manipulation (M = 4.18) than for the Incremental manipulation (M =
3.74). In other words, the manipulation of implicit theory ideology had a stronger impact
upon the manipulation check item than did the preexisting level of implicit theory
ideology before the study began. However, the preexisting level of implicit theory
ideology still had a (marginal) impact upon how participants viewed the stability of
personality (manipulation check item) following the manipulation. Thus, the weak and
inconsistent results from Study 4b may have been due to the manipulation of implicit
theory ideology not sufficiently overriding the participant’s preexisting ideologies,
either/both because the manipulation was too weak to mute the preexisting level of
implicit theory ideology, or the preexisting level of implicit theory ideology was too
strong to allow the manipulation to impact the participant’s ideology during the study.
70
However, the fact that previous studies in this manuscript found moderately strong
relationships between preexisting implicit theory ideology and aggression (Study 2 and
Study 3a), and yet little to no relationship between the manipulation of implicit theory
ideology and aggression in Study 4b, suggests that the preexisting beliefs may be more
enduring and robust than a manipulation of implicit theory ideology. If so, then how is
one to interpret the fact the manipulation of implicit theory ideology in the current
research has been used successfully in prior implicit theory research (Chiu, Hong, &
Dweck, 1997; Hong, Chiu, Dweck, Lin, & Wan, 1999; Levy, 1998; Levy, Plaks, Hong,
Chiu, & Dweck, 2001; Levy et al, 1998). Possible explanations and interpretations of the
data are discussed in the following General Discussion.
71
CHAPTER 6: GENERAL DISCUSSION
In summary, eight studies investigated the relationship between implicit theories
of human nature and aggression. Study 1a and 1b investigated trait-based aggression
using a variety of established measures of the emotional, cognitive, and behavioral
responding to both direct and displaced targets. As predicted by the hypothesized
relationship between implicit theories and aggression as laid out in the Introduction,
entity theorists reported increased anger and hostility after being provoked, increased
revenge planning, and increased physical and verbal aggression, but their aggression was
only directed at the actual provocateur, not third parties who are innocent or share a
group relationship with the provocateur. Thus Study 1a and Study 1b provide strong
convergent and discriminant validity for the hypothesized role of implicit theories in
aggressive responding. Study 2 used an imagined scenario paradigm common in prior
research on implicit theories that overcame the limitation of retrospectively asking people
to recall their own reconstructed experiences from Study 1a and Study 1b by instead
incorporating a contemporaneous instance of provocation/aggression within the study. As
predicted, the relationship between implicit theories and aggression in Study 2 replicated
the findings from Study 1 while also evidencing a stronger relationship due to no longer
assessing retrospective instances of aggression. Study 3a through 3c then extended the
analysis using an ostensibly “real” provocation and a larger range of aggression
measures. The result was that Study 3a evidenced the predicted relationship between
implicit theories and aggression using the new “real” provocation and state version of the
Aggression Questionnaire, while Study 3b and Study 3c found no relationship between
implicit theories and aggression using an open-ended response format of aggression
72
(Study 3b) or employing an ostensibly real measure of aggression in which participants
were lead to believe that they had the opportunity to aggress in a way that had real-life
negative consequences for the provocateur (Study 3c). Study 4a and 4b then directly
manipulated implicit theories to test the causal role of implicit theories in aggressive
responding. The result was that the laboratory-based study evidenced non-significant
results, while the online study was significant for one of the aggression measures.
Thus, in combination, the proposed studies tested both trait and state aggression
using different paradigm, methodologies, and subject populations. Taken as a whole,
Study 1a, Study 1b, Study 2, Study3a and Study 4b evidenced results consistent with the
hypothesized relationships between implicit theories and aggression, while Study 3b,
Study 3c, and Study 4a produced non-significant results. To state it another way, the
eight studies showed: (1) consistent results for trait aggression (Study 1a, Study 1b), (2)
mixed results for state aggression (Study 2, Study 3a, Study 4b only), (3) mixed results
when using a control condition (significant results for Study 4b, but not for Study 3c), (4)
mixed results when manipulating implicit theory ideology (significant results for Study
4a but not Study 4b), (5) mixed results for the state version of the Aggression
Questionnaire (significant results for Study 2 and Study 3a, but not for Study 4a or Study
4b), (6) mixed results for the ostensibly real measure of aggression involving the hiring
decisions at the Daily Trojan (significant results for Study 4b, but not for Study 4a or
Study 3c).
How are the results to be interpreted? Across the eight studies there is a general
pattern that emerges of a small to medium effect for implicit theory ideology to influence
aggressive responding, such that entity theorists are more aggressive following a
73
provocation. However, the general pattern exhibits mixed results across different types of
aggression measures and mixed results when manipulating implicit theory ideology. It is
not necessarily surprising that there are mixed results across different measurement
devices because measures have different levels of diagnosticity and applicability based
upon the topics/situations under investigation. It is also not surprising that there are
mixed results when trying to directly manipulate implicit theories since implicit theories
are an underlying belief systems outside conscious awareness that may not be easily
induced. Also, it is possible that the provocation manipulation overrides the preceding
manipulation of implicit theory because a provocation engenders strong emotional
reactions that overrides the more muted cognitively-based manipulation of implicit
theories. Nonetheless, the inconsistent nature of the hypothesized relationship between
implicit theory and aggression across the eight studies is some cause for concern
regarding the robustness of the findings.
Conversely, one issue that showed quite remarkable robustness of findings was
the convergent and discriminant evidence from Study 1a and Study 1b regarding the
conditions necessary for implicit theory ideology to impact aggressive responding. For
example, the breadth and depth of the relationship between implicit theories and
aggressive constructs was found with different types of aggression-related constructs,
such as hostility, anger, physical aggression, verbal aggression, severe forms of
aggression, norms of aggressive responding, and revenge planning. At the same time,
there was no relationship to the aggression-related constructs that fell outside the scope of
the relationship between implicit theories and aggression, such as aggression to innocent
third parties, aggression to group members of the perpetrator, aggression in the absence
74
of provocation, or with personality traits like rumination and impulsiveness that are
associated with aggression but not associated with the underlying rationale that entity
theorists are only more aggressive in the face of provocations that provide evidence of the
character of the perpetrator. In other words, entity theorists perceive and interpret the
same events or stimuli differently than incremental theorists based upon the perceiver’s
underlying implicit theories.
Another way to test the convergent and discriminant validity of the hypothesized
relationship between implicit theories and aggression is to employ a paradigm involving
displaced aggression and triggered displaced aggression. For example, the finding from
the current manuscript is that even if the targets of aggression have a definite relationship
to the provocateur, as long as the targets themselves displayed no evidence of initial
provocative aggressive behavior, an entity orientation showed no relationship with
aggressive retaliation. A displaced aggression paradigm involves a manipulation of
provocation but in situations in which the target is an innocent third party. The findings
from the present manuscript indicate that entity theorists would not display aggression
toward the innocent third party. However, a modification of the displaced aggression
paradigm includes a mildly provoking trigger from the target of aggression (who is not
the original provocateur), in which case the present findings indicate that the entity
theorists would display aggression because the target has provided negative character
evidence. Thus, a study in which there is a provocation manipulation with three
conditions -- target is perpetrator (direct aggression), target is innocent third party
(displaced aggression) and target is not perpetrator but displays triggering event
(triggered displaced aggression) -- would create a situation in which to test discriminant
75
validity by showing that entity theorists evidence more direct and triggered displaced
aggression, but not displaced aggression, because the former involve instance of prior
negative character information, while the latter does not.
Other forms of future research could investigate other conditions necessary for the
hypothesized effects to occur, such as the level of negative character information or the
ambiguity of the negative character information. Alternatively, more real-world instance
of aggression could be investigated such as gangs, violent crimes, and larger scale
conflicts like political, racial, or religious conflicts. One of the primary reasons for
conducting research on aggression is to understand the reasons behind aggressive
behavior in order to ameliorate the harmful physical and psychological effects of violence
in everyday life. Finally, another line of research could investigate conscious
understanding of how implicit theories impact our thoughts, behaviors, and feelings. For
example, research into implicit theories has shown how people are unaware of how their
beliefs about the stability/malleability of human influence their interpretations and
reactions to social events. If implicit theories are one of the most fundamental and
underlying characteristics that people implicitly hold regarding the objects, processes and
attributes in their life, as has been argued by many researchers over the years, then
making individuals aware of how these underlying beliefs impact so many aspects of
their lives may help mute the detrimental effects, such as aggression, that may occur.
In summary, the distinction between entity and incremental perspectives is one
type of lay theory that people use to understand, interpret, and predict their social world.
An entity perspective provides a framework for constructing a narrative of social events
around dispositional traits whereas an incremental perspective provides a framework for
76
constructing a narrative of social events around situational and psychological mediating
factors. As a result of the different information processing approaches, the
entity/incremental orientations has important implications for how we perceive, interpret,
and make inferences about ourselves and others. The purpose of this article was to show
how the distinction between entity and incremental orientations has important
implications for the perceptions, inferences, and judgments related to aggression by
breaking new ground in extending the research on implicit theory ideology into the
domain of aggressive responding.

77
Bibliography
Anderson, C. A., & Bushman, B. J. (2002). Human aggression. Annual Review of
Psychology, 53, 27-51.

Bloom, A. (1981). The linguistic shape of thoughts. Hillsdale, NJ: Lawrence Erlbaum
Associates, Inc.

Buss, A. H., & Perry, M. (1992). The aggression questionnaire. Journal of Personality
and Social Psychology, 63, 452-459.

Caprara, G. V., Cinanni, V., D’Imperio, G., Passerini, S., Renzi, P., & Travaglia, G.
(1985). Indicators of impulsive aggression: Present status of research on
irritability and emotional susceptibility scales. Personality and Individual
Differences, 6, 665-674.

Chiu, C., Dweck, C. S., Tong, J. Y., & Fu, J. H. (1997). Implicit theories and conceptions
of morality. Journal of Personality and Social Psychology, 73, 923-940.

Chiu, C., & Hong, Y. (1992). The effects of intentionality and validation on individual
and collective responsibility attribution among Hong Kong Chinese. Journal of
Psychology, 126, 291-300.

Chiu, C., Hong, Y., & Dweck, C. S. (1994). Toward an integrative model of personality
and intelligence: A general framework and some preliminary steps. In R. J.
Sternberg & P. Ruzgis (Eds.), Personality and intelligence (pp. 104-134).
Cambridge, England: Cambridge University Press.

Chiu, C., Hong, Y., & Dweck, C. S. (1997). Lay dispositionism and implicit theories of
personality. Journal of Personality and Social Psychology, 73, 19-30.

Cohen, J. (1988). Statistical power analysis for the behavioral sciences. (2nd ed.)
Hillsdale, NJ: Erlbaum.

Cohen, J., & Cohen, P. (1983). Applied multiple regression/correlation analysis for the
behavioral science. Hillsdale, NJ: Erlbaum.

Cohen, D., & Nisbett, R. E. (1994). Self-protection and the culture of honor: Explaining
southern violence. Personality and Social Psychology Bulletin, 20, 551-567.

Denson, T. F., Pedersen, W. C., & Miller, N. (2006). The displaced aggression
questionnaire. Journal of Personality and Social Psychology, 90, 1032-1051.

Dollard, J., Doob, L. W., Miller, N. E., Mowrer, O. H., & Sears, R. R. (1939).
Frustration and Aggression, New Haven, CT: Yale University Press.

78
Dweck, C. S. (1996). Implicit theories as organizers of goals and behavior. In R. M.
Gollwitzer & J. A. Bargh (Eds.), The psychology of action (pp. 69-90). London,
England: The Guilford Press.

Dweck, C. S., Chiu, C., & Hong, Y. (1995a). Implicit theories and their role in judgments
and reactions: A world from two perspective. Psychological Inquiry, 6, 267-285.

Dweck, C. S., Chiu, C., & Hong, Y. (1995b). Implicit theories: Elaboration and
Extension of the model. Psychological Inquiry, 6, 322-333.

Dweck, C. S. & Leggett E. L. (1988). A social-cognitive approach to motivation and
personality. Psychological Review, 95, 256-273.

Eisenberger, R., Lynch, P. Aselage, J., & Rohdieck, S. (2004). Who takes the most
revenge? Individual differences in negative reciprocity norm endorsement.
Personality and Social Psychology Bulletin, 30, 787-799.

Erdley, C. A., Cain, K. M., Loomis, C. C., Dumas-Hines, F., & Dweck, C. S. (1997). The
relations among children’s social goals, implicit personality theories, and
responses to social failure. Developmental Psychology, 33, 283-272.

Erdley, C. A., & Dweck, C. S. (1993). Children’s implicit personality theories as
predictors of their social judgments. Child Development, 64, 863-878.

Farrar, K., & Krcmar, M. (2006). Measuring state and trait aggression: A short,
cautionary tale. Media Psychology, 8, 127-138.

Fletcher, G. J. O., & Thomas, G. (1996). Close relationship lay theories: Their structure
and function. In G. J. O. Fletcher & J. Fitness (Eds.). Knowledge structures in
close relationships: A social psychological approach (pp.3-24). Mahway, NJ:
Lawrence erlbaum Associates, Inc.

Funder, D. C. (2001). Personality. Annual Review of Psychology, 52, 197-221.

Gervey, B. M., Chiu, C., Hong, Y., Dweck, C. S. (1999). Differential use of person
information in decisions about guilt versus innocence: The role of implicit
theories. Personality and Social Psychology Bulletin, 25, 17-27.

Graham, S. (1995). Implicit theories as conceptualized by an attribution researcher.
Psychological Inquiry, 6, 294-297.

Heller, A. (1981). Renaissance man (R. Allen, Trans.). New York: Schocken. (Original
work published, 1967).

Hemphill, J. F. (2003). Interpreting the magnitudes of correlation coefficients. American
Psychologist, 58, 78-80.
79

Henderson, V. & Dweck, C. S. (1990). Motivation and achievement. In S. S. Feldman &
G. R. Elliot (Eds.). At the threshold, The developing adolescent (pp. 308-329).
Cambridge, MA: Harvard University Press.

Hong, Y., Chiu, C, Dweck, C. S., Lin, D. M., & Wan W. (1999). Implicit theories,
attributions, and coping: A meaning system approach. Journal of Personality and
Social Psychology, 77, 588-599.

Hong, Y., Chiu, C., Dweck, C. S., & Sacks, R. (1997). Implicit theories and evaluative
processes in person cognition. Journal of Experimental Social Psychology, 33,
296-323.

Hong, Y., Levy, S. R., & Chiu, C. (2001). The contribution of the lay theories approach
to the study of groups. Personality and Social Psychology Review, 5, 98-106.

Koyre, A. (1957). From the closed world to the infinite universe. Baltimore: John
Hopkins University Press.

Kunda, Z., & Nisbett, R. E. (1986). The psychometrics of everyday life. Cognitive
Psychology, 18, 195-224.

Levy, S. R. & Dweck, C. S. (1998). Trait- versus process-focused social judgment.
Social Cognition, 16, 151-172.

Levy, S. R., & Dweck, C. S. (1999). The impact of children’s static versus dynamic
conceptions of people on stereotype formation. Child Development, 70 1163-
1180.

Levy, S. R., Plaks, J. E., & Dweck, C. S. (1998). Modes of social thought: Person
theories and social understanding. In S. Chaiken & Y. Trope, (Eds.). Dual
process theories in social psychology. New York: Guilford.

Levy, S. R., Plaks, J. E., Hong, Y., Chiu, C., & Dweck, C. S. (2001). Static versus
dynamic theories and the perception of groups: Different routes to different
destinations. Personality and Social Psychology Review, 5, 156-168.

Levy, S. R., Stroessner, S. J., & Dweck, C. S. (1998). Stereotype formation and
endorsement: The role of implicit theories. Journal of Personality and Social
Psychology, 74, 1421-1436.

Mischel, W. (1990). Personality dispositions revisited and revised: A view after three
decades. In L. A. Pervin (Ed.), Handbook of personality: Theory and research
(pp. 111-134). New York: Guilford.

80
Mischel, W. (2004). Toward an integrative science of the person. Annual Review of
Psychology, 55, 1-22.

Nisbett, R. E. (1980). The trait construct in lay and professional psychology. In L.
Festinger (Ed.), Retrospections on social psychology (pp. 109-130). Oxford
England: Oxford University Press.

Pepper, S. (1942). World hypothesis. Berkely: University of California Press.

Piaget, J., & Garcia, R. (1989). Psychogensis and the history of the sciences. (H. Fielder,
Trans.). Paris: Flamarion. (Original work published 1983).

Plaks, J. E., Stroessner, S. J., Dweck, C. S., & Sherman, J. W. (2001). Person theories and
attention allocation: Preferences for stereotypic versus counterstereotypic
information. Journal of Personality and Social Psychology, 80, 876-893.

Pratto, F., Sidanius, J., Stallworth, L. M., Malle, B. F. (1994). Social dominance
orientation: A personality variable predicting social and political attitudes.
Journal of Personality and Social Psychology, 67, 741-763.

Preacher, K. J. (2002, May). Calculation for the test of the difference between two
independent correlation coefficients. Computer software available from
http://www.quantpsy.org.

Ramirez, J. M., & Andreu, J. M. (2006). Aggression, and some related psychological
constructs (anger, hostility, impulsivity): Some comments from a research project.
Neuroscience & Biobehavioral Reviews, 30, 276-291.

Ross, M. (1989). Relation of implicit theories to the construction of personal histories.
Psychological Review, 96, 341-357.

Rydell, R. J., Hugenberg, K., Ray, D., & Mackie, D. M. (2007). Implicit theories about
groups and stereotyping: The role of group entitativity. Personality and Social
Psychology Bulletin, 33, 549-558.

Schoda, Y (2004). Individual differences in social psychology: Understanding situations
to understand people, understanding people to understand situations. In Sanson,
C., Morf, C. C., & Panter, A. T. (Eds.), The sage handbook of methods in social
psychology (pp. 117-141). Thousand Oaks, CA: Sage Publications.

Stenstrom, D. M., Denson, T. F., & Lickel, B. (2008). Vicarious retribution scale.
Manuscript in preparation.

Sternberg, R. J. (1985). Implicit theories of intelligence, creativity, and wisdom. Journal
of Personality and Social Psychology, 49, 607-627.

81
Suris, A., Lind, L., Emmett, G., Borman, P. D., Kashner, M., & Barratt, E. S. (2003).
Measures of aggressive behavior: Overview of clinical and research instruments.
Aggression and Violent Behavior, 9, 165-227.

Tong, Y., & Chiu, C. (2002). Lay theories and evaluation-based organization of
impressions: An application of the memory search paradigm. Personality and
Social Psychology Bulletin, 28, 1518-1527.

Trusty, J., Thompson, B., & Petrocelli, J. V. (2004). Practical guide for reporting effect
size in quantitative research in Journal of Counseling and Development. Journal
of Counseling and Development, 82, 107-110.

Vasquez, E., Lamoreaux, J., & Miller, N. (2008). Non-personal frustration in the
triggered displaced aggression paradigm. Unpublished Manuscript.

Weiner, B. (1971). Perceiving the causes of success and failure. General Learning Press.

Weiner, B. (1974). Achievement motivation and attribution theory. Morristown, N. J.:
General Learning Press.

Weiner, B. (1979). A theory of motivation for some classroom experiences. Journal of
Educational Psychology, 71, 3-25.

Wittenbrink, B., Hilton, J. L., Gist, P. L., (1998). In search of similarity: Stereotypes as
naïve theories in social categorization. Social Cognition, 16, 31-55.

Whorf, B. L. (1956). Language, thoughts, and reality. Cambridge, MA: MIT Press.

Whitebread, A. N. (1929). Process and reality. New York: Free Press.

Whitebread, A. N. (1938). Modes of thought. New York: Free Press.

Wright, J. C., Murphy, G. L. (1984). The utility of theories in intuitive statistics: The
robustness of theory-based judgments. Journal of Experimental Psychology, 113,
301-322.

82
Appendix

Implicit Theory (8-item)
1. The kind of person someone is, is something very basic about them, and it can’t
be changed very much.
2. People can do things differently, but the important parts of who they are can’t
really be changed.
3.
Everyone, no matter who they are, can change their basic characteristics.
4. As much as I hate to admit it, you can’t teach an old dog new tricks. People
can’t really change their deepest attributes.
5.
People can always substantially change the kind of person they are.
6. Everyone is a certain kind of person, and there is not much that can be done to
really change that.
7.
No matter what kind of person someone is, they can always change very much.
8.
All people can change even their most basic qualities.

Aggression Questionnaire
Physical Aggression
1.
Once in a while I can’t control the urge to strike another person.
2.
* Given enough provocation, I may hit another person.
3.
* If somebody hits me, I hit back.
4.
I get into fights a little more than the average person.
5.
* If I have to resort to violence to protect my rights, I will.
6.
* There are people who pushed me so far that we came to blows.
7.
* I can think of no good reason for ever hitting a person.
8.
I have threatened people I know.
9.
I have become so mad that I have broken things.

Verbal Aggression
10.
My friends say that I’m somewhat argumentative.
11.
* I can’t help getting into arguments when people disagree with me.
12.
* When people annoy me, I may tell them what I think of them.
13.
I often find myself disagreeing with people.
14.
* I tell my friends openly when I disagree with them.

Hostility
15.
I am sometimes eaten up with jealousy.
16.
* At times I feel I have gotten a raw deal out of life.
17.
* Other people always seem to get the breaks.
18.
I wonder why sometimes I feel so bitter about things.
19.
I know that “friends” talk about me behind my back.
20.
* I am suspicious of overly friendly strangers.
21.
I sometimes feel that people are laughing at me behind my back.
83
22.
* When people are especially nice, I wonder what they want.

Anger
23.
I flare up quickly but get over it quickly.
24.
When frustrated, I let my irritation show.
25.
I sometimes feel like a powder keg ready to explode.
26.
I am an even-tempered person.
27.
Some of my friends think I’m a hothead.
28.
Sometimes I fly off the handle for no good reason.
29.
I have trouble controlling my temper.

Culture of Honor
1.
It is often necessary to use violence to prevent violence.
2.
When a person harms you, you should turn the other cheek and forgive them.
3.
A man has a right to kill another man in a case of self-defense.
4.
A man has the right to kill a person to defend his family.
5.
A man has the right to kill a person to defend his house.
6. It would be justified for a man to shoot and kill a neighbor if that neighbor has
sexually assaulted the man’s daughter.

Norm of Negative Reciprocity
1.
If someone dislikes you, you should dislike them.
2.
If a person despises you, you should despise them.
3.
If someone says something nasty to you, you should day something nasty back.
4.
If a person wants to be your enemy, you should treat them like an enemy.
5.
If someone treats me badly, I feel I should treat them even worse.
6.
If someone treats you badly, you should treat that person badly in return.
7.
If someone has treated you poorly, you should not return the poor treatment.
8. If someone important to you does something negative to you, you should do
something even more negative to them.
9.
A person who has contempt for you deserves your contempt.
10.
If someone treats you like an enemy, they deserve your resentment.
11.
When someone hurts you, you should find a way to get even.
12.
You should not give help to those who treat you badly.
13.
When someone treats me badly, I still act nice to them.
14.
If someone distrusts you, you should distrust them.

Vicarious Retribution
1. If a member of another group harms a member of my group, I would blame all
the members of that group.
2. When a member of my group is attacked, I feel like I must retaliate against the
other group or my group’s reputation would be harmed.
84
3. When a member of my group is attacked, I feel like I must retaliate against the
other group or else members of my group would think less of me.
4. When a member of my group is attacked, I would retaliate against other
members of the group to prevent the attack from ever occurring again.
5. When a member of my group is harmed, I would retaliate against other members
of the group to even the score.
6. If a member of my group were harmed, I would retaliate against other members
of the group to make the person who was hurt feel better.
7. When a member of my group is harmed, retaliating against other members of the
group is the wrong thing to do.
8. If a member of my group were attacked, I would feel like I must retaliate against
other members of the group or else the harmed member of my group would feel
betrayed by me.
9. When a member of my group is attacked, I feel like I will lose my status in my
group if I don’t retaliate against other members of the offending group.
10. When a member of my group is harmed, “an eye for an eye” applies to any
member of the offending group.
11. Retaliating on behalf of a harmed group member sends a message to the other
group to never attack my group again.
12. When one of my group members is harmed, I am unlikely to support revenge on
members of the other group.
13. My group will lose status and respect if we don’t retaliate against the other group
when someone attacks our group.
14. When a one of my group members is attacked, retaliation is justified against
anyone who shares a group membership with those who attacked my group.
15. Others who share a group membership with those who attack my group should
be punished for the actions of their fellow group member.
16. If a member of my group is attacked, it would be morally wrong to attack a
member of that person’s group.
17. If a member of my group were harmed by another, I would support revenge
toward other members of the offending group.
18.
When my group is harmed, I do NOT get angry at all members of the group.
19. When a member of my group is harmed, other members of the offending group
should be punished.
20. When a member of my group is harmed, other members of the offending group
should NOT be held responsible.

Displaced Aggression Questionnaire
Angry Rumination
1.
I keep thinking about events that angered me for a long time.
2.
I get “worked up” just thinking about things that have upset me in the past.
3.
I often find myself thinking over and over about things that have made me angry.
4.
Sometimes I can't help thinking about times when someone made me mad.
5.
Whenever I experience anger, I keep thinking about it for a while.

85
6. After an argument is over, I keep fighting with this person in my imagination.
7.
I re-enact the anger episode in my mind after it has happened.
8.
I feel angry about certain things in my life.
9.
I think about certain events from a long time ago and they still make me angry.
10. When angry, I tend to focus on my thoughts and feelings for a long period of
time.

Revenge Planning
11. When someone makes me angry I can't stop thinking about how to get back at
this person.
12.
If somebody harms me, I am not at peace until I can retaliate.
13.
I often daydream about situations where I’m getting my own back at people.
14. I would get frustrated if I could not think of a way to get even with someone who
deserves it.
15. I think about ways of getting back at people who have made my angry long after
the event has happened.
16.
If another person hurts you, it’s alright to get back at him or her.
17.
The more time that passes, the more satisfaction I get from revenge.
18.
I have long living fantasies of revenge after the conflict is over.
19.
When somebody offends me, sooner or later I retaliate.
20.
If a person hurts you on purpose, you deserve to get whatever revenge you can.
21.
I never help those who do me wrong.

Displaced Aggression
22. When someone or something makes me angry I am likely to take it out on
another person.
23.
When feeling bad, I take it out on others.
24.
When angry, I have taken it out on people close to me.
25. Sometimes I get upset with a friend or family member even though that person is
not the cause of my anger or frustration.
26.
I take my anger out on innocent others.
27.
When things don’t go my way, I take out my frustration on the first person I see.
28.
If someone makes me angry I would likely vent my anger on another person.
29. Sometimes I get so upset by work or school that I become hostile toward family
and friends.
30.
When I am angry, I don’t care who I lash out at.
31. If I have had a hard day at work or school, I’m likely to make sure everyone
knows about it.

Impulsive Aggression
1.
I easily fly off the handle with those who don’t listen or understand.
2.
I am often in a bad mood.
3.
* Usually when someone shows a lack of respect for me, I let it go by.
4.
* I have never been touchy.
86
5.
It makes my blood boil to have somebody make fun of me.
6.
* I think I have a lot of patience.
7.
When I am irritated I need to vent my feelings immediately.
8.
When I am tired I easily lose control.
9.
I think I am rather touchy.
10.
When I am irritated I can't tolerate discussions.
11.
* I could not put anyone in his place, even if it were necessary.
12.
* I can't think of any good reason for resorting to violence.
13.
I often feel like a powder keg ready to explode.
14.
* I seldom strike back even if someone hits me first.
15.
I can't help being a little rude to people I don’t like.
16.
Sometimes when I am angry I lose control over my actions.
17.
* I do not know of anyone who would wish to harm me.
18.
Sometimes I really want to pick a fight.
19.
* I do not like to make practical jokes.
20.
When I am right, I am right.
21.
* I never get mad enough to throw things.
22.
When someone raises his voice I raise mine higher.
23.
Sometimes people bother me just by being around.
24.
Some people irritate me if they just open their mouth.
25.
Sometimes I shout, hit and kick and let off steam.
26.
I don’t think I am a very tolerant person.
27.
* Even when I am very irritated I never swear.
28.
It is others who provoke my aggression.
29.
Whoever insults me or my family is looking for trouble.
30.
It takes very little for things to bug me.

Implicit Theory (3-item)
1. The kind of person someone is, is something very basic about them, and it can’t
be changed very much.
2. People can do things differently, but the important parts of who they are can’t
really be changed.
3. Everyone is a certain kind of person, and there is not much that can be done to
really change that.

Vignettes for Study 2
1. Later today you are walking home from school when a mugger angrily pushes
you to the ground and forcefully rips your money from your wallet/purse before
running away.
2. Later today you are on campus and someone bumps into you, spilling your drink
and the contents of your bag. They then begin to curse and step toward you
aggressively.
87
3. Later today you are driving on the highway. Your off ramp is quickly
approaching. The driver next to you is purposefully driving in a manner that is
preventing you from changing lanes. You miss your exit.
4. Later today you are driving on the highway. The driver in the car in front of you
throws some garbage out the car window, and it smashes against your
windshield.

Aggression Questionnaire – State Version
1.
I would tell this person openly that I disagreed with them.
2.
I would find myself disagreeing with this person.
3.
When this person annoyed me, I would tell them what I think of them.
4.
This person would say that I’m somewhat argumentative.
5.
I could not control my urge to strike this person.
6.
Given enough provocation, I may hit this person.
7.
If this person hit me, I’d hit back.
8.
I’d get into a fight with this person a little more than the average person.
9.
If I had to resort to violence to protect my rights, I would.
10.
This person would have pushed me so far that we came to blows.
11.
* I can think of no good reason for ever hitting this person.

Daily Trojan Article as Provocation
It’s that time of year again: time for the annual announcement that – surprise! –
tuition is going up! It’s time to begin the typical chorus… of whining.
It seems like each year everybody is up in arms about the annual hike in tuition.
Stop complaining about it! No one likes to hear a cry-baby.
Yah, sure tuition has gone up. Yeah, every year the amount that tuition goes up
gets higher and higher. So what? It seems like a lot at first. But think about how much
you are paying for school already. Not even counting room and board, the tuition
increase comes to about 7 percent more money than you were paying. That’s barely
even a tuition hike when you take into account the fact that the inflation rate has been
around 3 percent this year.
Yet every year, there is the same report on the tuition increase in the paper,
without fail only barely managing to avoid veering out of the realm of journalism and
into the realm of opinion. This is unavoidably followed by a torrent of letters from you
complaining about the tuition increase, as though we are all shocked and appalled at an
event we all knew was coming anyway. It happens every year. You can handle. It. Here
is a thought – stop acting like self-absorbed children and get over it!
You pretty much sign on for tuition increases as part of the deal when you enroll,
didn’t you realize that? All whining does it annoy the rest of us who are able to
understand that this is a fact of life.
It is clear that the school administration have no plans to halt or lessen the
increases in tuition, no matter how much students whine, especially since you take no
action other than whining. Why should the administration listen to lazy and indolent
children? So stop the chorus of whining so we can all get on with our lives in peace.
88

Daily Trojan Article as Control (No Provocation)
You don’t need the glint of Egyptian treasure or the shock of value of plasticized
human remains to draw a massive crowd through museum doors. In fact, you may be
able to entice more than 200,000 visitors to an exhibit featuring a 66-million-year-old
Triceratops skull, a red 1970s Winnebago and a makeshift Mayan temple – all in close
range of each other – with a little ingenuity, and a challenge to viewerss to read
between the lines.
“Collapse?” a new exhibit at the Los Angeles Natural History Museum, provides
visitors with more than a walk through an object-littered marble hall. While it does host
an intriguing buffet of mixed-media eye candy, the exhibition dares views to think
about the potentially dark fate that modern society faces by taking a journey through
the past to see how man’s choices brought about the successes and decline of
civilizations in difference eras across the globe.
Inspired by Pulitzer Prize winning author Jared Diamond’s best-selling book,
“Collapse: How Societies Choose to Fail or Succeed”, the exhibit take museum visitors
on a journey that starts in modern-day Montana, and then travels backward in time
through Mayan civilization and Tokugawa, Japan before returning to the contemporary
scene in Australia and Southern California.

Measure about evaluating article author for position at Daily Trojan
The Daily Trojan has initiated a new procedure for the hiring of editorial
positions. Since these positions offer a reasonable salary and look good on a resume,
there have been more applicants than the number of positions available, so we have
decided to gather input from the public to help us make our hiring decisions. The author
of the article you just read is an applicant and we would like your feedback regarding
this individual’s suitability for the position. Please do not write your name or any other
identifying mark on this sheet since we want the answers to be completely anonymous.
1. I do not recommend this author for the position.
2. I do not think this author should be hired for the position.
3. I do not think that the author is a competent individual.
4. I do not like the author.
5. I do not think the author is intelligent.

Coding Manual for Study 3c to code open-ended responses
Question #1 Is the text about the article (put a 1), the author (put a 2), or about both
the article and the author (put a 3).
Question #2 “To what extent is the text negative or positive?” Remember, for this
question, you are giving your gestalt gut impression. For example,
there could be both positive and negative comments, so take them all
into account when deciding the overall positivity versus negativity of
the text. It is possible that some of the negative comments are cancelled
out by the positive comments, and vice versa. It is possible that some of
the negative comments are partially cancelled out by the positive

89
comments, and vice versa. In other words, ask yourself whether overall
the text is positive or negative (or neutral), and then how much positive
or negative?
Question #3 “To what extent is the text negative? Remember, for this question, you
are only looking at the negative comments (and not using the positive
comments). For example, if there are no negative comments, put a 0. If
there are negative comments, ask yourself how negative the text is
overall.
Question #4 “To what extent is the text mean/aggressive/hostile/etc.?” Remember,
for this question, we are looking at the generic concept of aggression.
For example, “aggression” in this case could refer to the text being
mean, hostile, rude, obnoxious, sarcastic, angry, aggressive, etc.

Implicit Theory Induction Articles – Fixed
As a preschooler, Mary S. would not leave her mother to make friends with other
children. Later, when she grew up, she had difficulty getting along with people. In her
late forties, she was still single and led a lonely life.
Benjamin M. exhibited a lot of self-discipline during his early childhood. When
he was seven-years old he didn’t need his parents to urge him to get dressed in the
morning or to go to bed at night. Later, in school, he always had a well-planned study
schedule and was better prepared for examinations than the other students.
These cases were among the eight hundred and twelve cases that researchers
have collected at the Personality and Development Unit at Stanford University, and
they are typical examples of personality development.
Does personality change? Researchers at the Personality and Development Unit
at Stanford University (PDU) are interested in the origins of personality characteristics
and personality patterns over an individual’s life. To collect cases for the data bank,
these researchers launched a large scale longitudinal (that is, long-term) study.
For more than twenty-five years, the PDU has been following eight hundred
individuals. The researchers identified them at birth and have been collecting elaborate
data on them since, including birth records, school records, extensive observations at
home and in the laboratory, and in-depth interviews with the individuals, their family
members, and close friends.
In a recent article published in the Journal of Personality Research, Dr.
Lawrence Rescoria, the director of PDU in the psychology department at Stanford,
reported the findings of their extensive case study research. As was observed
repeatedly, Dr. Rescoria concluded that “personality characteristics seem to be rather
fixed and to develop consistently along the same path over time.” In fact, people’s
personality characteristics can be conceived as fixed entities. “Even in the early years,
personality characteristics can be recognized as a cohesive personality profile,” he
wrote. He argued that “at no time in a person’s life are basic personality characteristics
likely to change much at all.”
Similar conclusions were drawn by Dr. Paul Medin, a psychologist at the
National Institute on Mental Health. In his speech at the American Psychological
Association’s annual convention held in Washington, D.C. in August, Dr. Medin

90
argued that “in most of us, our character is set like plaster and will never soften.” He
reported numerous large longitudinal studies that show that people “age and develop,
but they do so on the foundation of enduring dispositions.”
Dr. Medin’s conclusions about personality are based on six longitudinal studies
published between 1978 and 2000, including two of his own. All six had considerably
different samples and rationales, but “were nearly unanimous in their conclusions on
the stability of personality,” he said.
These studies, together with many others, have made clear the fact that people’s
personality is relatively stable.
Many historically significant figures possessed a stable personality, too.
Interestingly, stable personality has been found to be the rule for significant
figures in history. Dr. Marsha Schneider, a historian at the University of Chicago, has
done research on the personality of important historical figures. Her research is based
largely on biographies and published interviews with these individuals.
In her article, appearing in the last December issue of the American Historian,
she reported that “many significant figures in history displayed their key personality
characteristics at an early age. These characteristics then led them to behave in ways
that made them famous.”
She mentioned several examples, one being Jack Partlow, a entrepreneur of the
1920’s, featured in Fortune 500 magazine, who became known for his aggressive
method of doing business. According to the people who knew him when he was in
college, he was never a passive person. In fact, Jack was known to be rather pushy. He
sometimes bullied the other students to get what he wanted. Later, Jack maintained his
aggressive approach when dealing with people both in his company and in rival
companies. Like other business tycoons who were aggressively taking over smaller
companies and putting others out of business, Jack chose business solutions that never
allowed other companies to thrive along with his. “Consistent with his aggressive
beginnings, Jack turned out to be a forceful business leader,” Dr. Schneider concluded.
Dr. Schneider also discussed the example of Rebecca Mahoney, mayor of
Elkhart, Indiana in the 1970s, who earned a reputation as a fair and honest mayor,
trusted by many. As an adolescent, Rebecca was also known to be an honest person.
She never cheated on exams, and, reportedly, was trusted by the other students in her
school, even those that could be considered rivals for popularity. Later, Rebecca
maintained her reputation as a fair and honest politician, earning the trust of the people
of Elkhart, Indiana. She kept her campaign promises and won re-election several times.
“Rebecca became well known as an honest and trustworthy mayor, as would have been
predicted by her reputation in the past,” Dr. Schneider reported.
Resulting from her analysis of the personality development of seventy-two
historically significant figures, Dr. Schneider concluded that “overall, historically
significant figures are no different from common people in the sense that their
personality is relatively fixed and stable. Perhaps, the difference is that they had a
distinctive personality to begin with.”
To conclude, research findings from a wide range of studies, including large-
scale longitudinal studies, case studies, and historical analyses, converge to one major
conclusion: Personality seems to be fixed and stable.

91
Implicit Theory Induction Articles – Malleable
As a preschooler, Mary S. would not leave her mother to make friends with other
children. However, as a kindergartner, she had outstanding social skills. She always
had friends to play with both at school and at home.
Benjamin M. exhibited a lot of self-discipline during his early childhood. When
he was seven-years old he didn’t need his parents to urge him to get dressed in the
morning or to go to bed at night. However, later in school, he never had a well-planned
study schedule and was less prepared for examinations than the other students.
These cases were among the eight hundred and twelve cases that researchers
have collected at the Personality and Development Unit at Stanford University, and
they are typical examples of personality development.
Does personality change? Researchers at the Personality and Development Unit
at Stanford University (PDU) are interested in the origins of personality characteristics
and personality patterns over an individual’s life. To collect cases for the data bank,
these researchers launched a large scale longitudinal (that is, long-term) study.
For more than twenty-five years, the PDU has been following eight hundred
individuals. The researchers identified them at birth and have been collecting elaborate
data on them since, including birth records, school records, extensive observations at
home and in the laboratory, and in-depth interviews with the individuals, their family
members, and close friends.
In a recent article published in the Journal of Personality Research, Dr.
Lawrence Rescoria, the director of PDU in the psychology department at Stanford,
reported the findings of their extensive case study research. As was observed
repeatedly, Dr. Rescoria concluded that “personality characteristics seem to be
malleable and can be developed over time. In fact, personality characteristics are
basically a bundle of potentialities that wait to be developed and cultivated,” he wrote.
He argued that “at almost any time in a person’s life, his or her personality
characteristics can change.”
Similar conclusions were drawn by Dr. Paul Medin, a psychologist at the
National Institute on Mental Health. In his speech at the American Psychological
Association’s annual convention held in Washington, D.C. in August, Dr. Medin
argued that “no one’s character is hard like a rock that cannot be changed.” He
reported numerous large longitudinal studies that show that some people “change their
character and personality even late in life.”
Dr. Medin’s conclusions about personality are based on six longitudinal studies
published between 1978 and 2000, including two of his own. All six had considerably
different samples and rationales, but “were nearly unanimous in their conclusions on
the malleability of personality,” he said.
These studies, together with many others, have made clear the fact that people’s
personality can change throughout their lives.
Many historically significant figures possessed a malleable personality, too.
Interestingly, malleable personality has been found to be the rule for significant
figures in history. Dr. Marsha Schneider, a historian at the University of Chicago, has
done research on the personality of important historical figures. Her research is based
largely on biographies and published interviews with these individuals.

92
In her article, appearing in the last December issue of the American Historian,
she reported that “the key characteristics of many significant figures in history
developed and changed over time. These characteristics then led them to behave in
ways that made them famous.”
She mentioned several examples, one being Jack Partlow, a entrepreneur of the
1920’s, featured in Fortune 500 magazine, who became known for his nonaggressive
method of doing business. According to the people who knew him when he was in
college, he was never a passive person. In fact, Jack was known to be rather pushy. He
sometimes bullied the other students to get what he wanted. Later, though, Jack
developed a more peaceful approach to dealing with people both in his company and in
rival companies. While some business tycoons were taking over smaller companies and
putting others out of business, Jack found business solutions that allowed both his
company and others to thrive at the same time. “Despite his aggressive beginnings,
Jack turned out to be a peaceful business leader,” Dr. Schneider concluded.
Dr. Schneider also discussed the example of Rebecca Mahoney, mayor of
Elkhart, Indiana in the 1970s, who earned a reputation as an unfair and dishonest
mayor, mistrusted by many. However, as an adolescent, Rebecca was known to be an
honest person. She never cheated on exams, and, reportedly, was trusted by the other
students in her school, even those that could be considered rivals for popularity. Later,
though, Rebecca acquired a reputation as an unfair and dishonest politician, losing the
trust of the people of Elkhart, Indiana. She kept very few of her campaign promises
and was therefore not re-elected. “Rebecca became well known as a dishonest and
untrustworthy mayor, even though her reputation had been different in the past,” Dr.
Schneider reported.
Resulting from her analysis of the personality development of seventy-two
historically significant figures, Dr. Schneider concluded that “overall, historically
significant figures are no different from common people in the sense that their
personality is relatively changeable. Perhaps, the difference is that they developed a
distinctive personality.”
To conclude, research findings from a wide range of studies, including large-
scale longitudinal studies, case studies, and historical analyses, converge to one major
conclusion: Personality seems to be malleable and can change.

Study 4b - Manipulation Check, Reinforcing Questions, and Filler items
1. What is main message of the article?
a. Personality characteristics are fixed entities
b. Personality characteristics change over time
c. both a and b
2. Why does the author begin the article with the two examples of Mary S. and
Benjamin M.?
a. Mary S. and Benjamin M. are representative examples that personality
characteristics are fixed
b. Mary S. and Benjamin M. are representative examples that personality
characteristics are malleable
c. both a and b

93
3. Why does the article report multiple evidence from three different professionals?
a. An article that is about research findings should contain confirmation from
multiple sources
b. Multiple pieces of evidence show that the research findings are consistent
c. both a and b
4. How many longitudinal studies did Dr. Paul Medin publish about the stability
[malleability] of personality?
a. 0-3
b. 4-6
c 7-9
5.
To what extent was the article informative and well-written?
6.
To what extend do you believe that a person’s traits or characteristics are stable
7. If the New York Times was thinking of hiring the author for a prestigious
editorial position, to what extent do you think the author should be hired for the
position?

Study 4b – Attributions
1. To what degree is the cause(s) an internal factor of the author (i.e., due to his/her
personality)?
2. To what degree is the cause(s) due to situational factors that are external to the
author?
3. To what degree was the author’s inclination to write the article something that
s/he could control?
4. To what degree is the cause(s) something stable/permanent about the author (as
compared to changeable)?
5. To what degree is the impact or effect that the article has on readers one that is
intended by the author?
6.
To what degree do you think the author regretted his statements afterwards?
7. To what extent did the author write an article expressing negative opinions
because that is the kind of person he is?
8.
To what extent can the author change the kind of person he is?

Study 4b – Trait judgements

The items below concern your impressions and opinions of the author of the
Daily Trojan article. Please rate the author on the following traits.
_____ hostile _____ warm _____ arrogant _____ pleasant
_____ honest _____ aggressive _____ likeable _____ kind
_____ friendly _____ malicious _____ nice _____ closed-minded
_____ annoying _____ rude _____ foul person _____ good

94
Study 4b – Author evaluation

The Daily Trojan has initiated a new procedure for the hiring of editorial
positions. Since these positions offer a reasonable salary and look good on a
resume, there have been more applicants than the number of positions available,
so we have decided to gather input from the public to help us make our hiring
decisions. The author of the article you just read is an applicant and we would
like your feedback regarding this individual’s suitability for the position.
1.
I think the author should be hired for the position.
2.
I recommend the author for this position.
3.
I believe the author is qualified for the position.

Study 4b – Agreement with Article

Using the scale below, please indicate the extent to which you agree or disagree
with each of the following statements regarding the Daily Trojan article.
1. To what extent do you agree with the author’s opinions in the Daily Trojan
article?
2. To what extent do you think the author was correct in his statements in the Daily
Trojan article?
3.
To what extent do you think the article was an accurate depiction of the topic?

Abstract (if available)

Abstract Eight studies investigated the relationship between implicit theorist ideology of human nature and aggression. Implicit theories of human attributes are dynamic meaning systems used by people to understand, interpret, and predict their social world. A person who believes human nature is fixed is called an entity theorist because s/he perceives personality as a stable entity that does not change. A person who believes human nature is more flexible and malleable is called an incremental theorist because s/he believes personality can change across time or across different situations. The present research extends the implicit theory research into the domain of aggression by identifying how entity theorists perceive and interpret the same events or stimuli differently than incremental theorists based upon the perceiver's underlying implicit theories. Study 1a and 1b investigated trait-based aggression using a wide variety of established measures of aggression and aggression-related constructs. Study 2 then investigated state-based aggression using imagined vignettes as the provocation manipulation. Study 3a through 3c extend this line of research by having the participants experience a real immediate provocation, and also extend this line of research into utilizing a wider range of aggression measures. Study 4a and 4b then directly manipulated implicit theories to test the causal role of implicit theories in aggressive responding. In total, the eight studies tested both trait and state aggression using different paradigms, methodologies, and subject populations. The combined result was consistent support for trait-based aggression, such that entity theorists reported increased anger and hostility after being provoked, increased revenge planning, and increased physical and verbal aggression, but the aggression was only directed at the actual provocateur, not third parties who were innocent or shared a group relationship with the provocateur.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

The impact of naturally-occurring rumination on aggression

PDF

The angry brain: neural correlates of interpersonal provocation, directed rumination, trait direct aggression, and trait displaced aggression

PDF

Children's triggered displaced aggression

PDF

Projection on outgroups: complementary projection in aversive intergroup contexts

PDF

Personality and culture revisited: Integrating the social cognitive approach to personality and the dynamic constructivist approach to culture

PDF

Effects of racial exclusion from media

PDF

Living with the most humanlike nonhuman: understanding human-AI interactions in different social contexts

PDF

Decoding information about human-agent negotiations from brain patterns

PDF

Essays on improving human interactions with humans, algorithms, and technologies for better healthcare outcomes

PDF

Detecting joint interactions between sets of variables in the context of studies with a dichotomous phenotype, with applications to asthma susceptibility involving epigenetics and epistasis

Asset Metadata

Creator Stenstrom, Douglas Malcolm (author)

Core Title Implicit theory ideology about human attributes and aggression

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Psychology

Degree Conferral Date 2008-12

Publication Date 12/06/2008

Defense Date 10/27/2008

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag aggression,implicit theory ideology,OAI-PMH Harvest

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Miller, Norman (committee chair), Huey, Stanley J., Jr. (committee member), Kim, Peter H. (committee member), Read, Stephen J. (committee member)

Creator Email dstenstrom@email.com,stenstro@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m1884

Unique identifier UC1448361

Identifier etd-Stenstrom-2559 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-129682 (legacy record id),usctheses-m1884 (legacy record id)

Legacy Identifier etd-Stenstrom-2559.pdf

Dmrecord 129682

Document Type Dissertation

Rights Stenstrom, Douglas Malcolm

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu