0364535415Am J Community PsycholAm J Community PsycholAmerican journal of community psychology0091-05621573-277021800228410746110.1007/s10464-011-9461-yNIHMS324488ArticlePreliminary Examination of a Cartoon-Based Hostile Attributional Bias Measure
for Urban African American BoysLeffStephen S.123LeflerElizabeth K.14KheraGagan S.15PaskewichBrooke1JawadAbbas F.12The Children’s Hospital of PhiladelphiaThe University of Pennsylvania School of MedicineThe Philadelphia Collaborative Violence Prevention CenterUniversity of Northern IowaPeace First, IncCorrespondence should be mailed to: Stephen S. Leff at The
Children’s Hospital of Philadelphia, Dept. of Pediatric Psychology, Rm. 1480 at
CHOP North; 3405 Civic Center Blvd., Philadelphia, PA 19104.
leff@email.chop.edu662014620122372014490332346
The current study illustrates how researchers developed and validated a
cartoon-based adaptation of a written hostile attributional bias measure for a sample of
urban, low-income, African American boys. A series of studies were conducted to develop
cartoon illustrations to accompany a standard written hostile attributional bias vignette
measure (Study 1), to determine initial psychometric properties (Study 2) and
acceptability (Study 3), and to conduct a test-retest reliability trial of the adapted
measure in a separate sample (Study 4). These studies utilize a participatory action
research approach to measurement design and adaptation, and suggest that collaborations
between researchers and key school stakeholders can lead to measures that are
psychometrically strong, developmentally appropriate, and culturally sensitive. In
addition, the cartoon-based hostile attributional bias measure appears to have promise as
an assessment and/or outcome measure for aggression and bullying prevention programs
conducted with urban African American boys.
measurement developmentAfrican American boyshostile attributional biasrelational aggressionparticipatory action research
For many years researchers have documented both the frequency and impact of low
level physically aggressive actions, such as hitting, pushing, and threatening others. For
instance, research has demonstrated that many school-age children exhibit these behaviors
daily (Nansel et al., 2001), and that the
perpetrators of these actions are at heightened risk for experiencing peer relationship
difficulties, problem-solving deficits, internalizing problems, and academic challenges (see
Leff, Power, Manz, Costigan, & Nabors, 2001).
Further, repeated involvement as the perpetrator of low level physical aggression is
associated with higher levels of adjustment difficulties in adolescence and young adulthood
(Coie, Terry, Lenox, Lochman, & Hyman, 1995;
Loeber, Green, Lahey, & Kalb, 2000). Because
boys are more likely than girls to be involved in these types of physical conflicts with
their peers, studies have suggested that boys are therefore much more aggressive than girls.
However, in the mid-1990’s, researchers began to hypothesize that physical
aggression may not fully capture the scope of female aggressive behavior (Crick, 1995; Crick & Grotpeter,
1995) and that the primary focus on physical aggression in research may lead to
other types of aggression being overlooked and underestimated (McNeilly-Choque, Hart, Robinson, Nelson, & Olsen, 1996). Thus,
there has been a recent movement to examine the multitude of ways in which boys and girls
tend to express aggression towards others (e.g., Card,
Stucky, Sawalani, & Little, 2008; Crick,
Ostrov, & Kawabata, 2007).
Boys’ primary use of physical aggression is well documented in the
literature; however, it was only recently that aggression in girls was thoroughly examined.
Recent research suggests that girls tend to display aggression through the manipulation of
one’s social standing or reputation (see Card et al.,
2008). Crick and Grotpeter (1995) termed
this form of behavior relational aggression. Relational aggression includes
behaviors in which relationships serve as the vehicle of harm (Crick et al., 2007). Some examples include purposely excluding a
peer from the social group, gossiping, saying mean things or spreading rumors about a peer
in order to encourage rejection by others, and withdrawing friendship or acceptance in order
to control another child (Crick & Grotpeter,
1995). Relational aggression has been found to be relatively stable over time (Crick, 1996) and is associated with a range of current
and future psychosocial difficulties including problematic peer difficulties, internalizing
symptoms, and social problem-solving deficits (Card et al.,
2008; Cillesen & Mayeux, 2004; Crick, 1996; Leff,
Kupersmidt, & Power, 2003).
Pioneering studies in relational aggression demonstrated that the gender gap in
overall levels of aggression was smaller than previously believed when one considers
relational as well as physical aggression (e.g., Crick &
Grotpeter, 1995; McNeilly-Choque et al.,
1996; Ostrov & Keating, 2004). However,
the presumption that only girls display relationally aggressive behavior has been found to
be too simplistic (Swearer, 2008). While it holds
true that boys are more physically than relationally aggressive and that girls are more
relationally than physically aggressive (see Crick et al.,
2007), recent findings suggest that boys do indeed utilize relational forms of
aggression (Card et al., 2008; Leff, Waasdorp et al., 2010; Swearer, 2008). Thus, it may be more appropriate to consider that boys and girls
participate in and experience both forms of aggression, with varying levels of importance
placed on the type of aggressive act (Henington, Hughes,
Cavell, & Johnson, 1998; Rys & Bear,
1997; Tomada & Schneider, 1997; Underwood, Galen, & Paquette, 2001). Specifically,
research suggests that girls experience relational aggression as more hurtful or distressful
than boys (e.g., Crick, 1995; Crick, Bigbee, & Howes, 1996; Paquette & Underwood, 1999); however, recent research indicates that boys are
also distressed by relationally aggressive behaviors. For example, adolescents of both sexes
who experience high levels of relational aggression in school perceive their school to be
unsafe (Goldstein, Young, & Boyd, 2008). Further,
some researchers have found that for boys, being involved in relational aggression leads to
high levels of social-psychological maladjustment, perhaps because relational behaviors
(i.e., gossiping and social exclusion) may be considered a gender non-normative expression
of aggression (Crick, 1997). Thus, it follows that
boys’ experience of and distress over relational aggression is a very real
phenomenon that warrants additional research.
Paradoxically, it seems that just as girls were excluded from much of the early
research on physical aggression, boys are now being understudied in the relational
aggression literature. This is unfortunate given the aforementioned findings. Several
factors may impact the limited understanding of relational aggression in boys. First,
boys’ use of relational aggression may be overshadowed by the field’s focus
upon physically aggressive actions among boys. A second possible explanation may be that
boys are less likely than girls to talk about their experience of relational aggression, and
therefore tend to either ignore the aggressor or respond with a physically aggressive act,
thereby causing the relational aggression to go unnoticed as the overt behavior demands
attention (Paquette & Underwood, 1999).
Regardless, given the prevalence and negative implications of relational aggression in boys,
it is clear that more research is needed on the perception, processing, and experience of
relational aggression in boys.
The Urban School Context and Additional Challenges Faced by African American
Males
Youth living in urban, high poverty communities face a number of risk factors
that may be associated with increased levels of aggression and violence (Johnson, 2009, Stoddard, Henly,
Sieving, & Bolland, 2011; Tolan,
Gorman-Smith, & Henry, 2003). For instance, urban youth often must cope with
disorganized, unstructured schools, high levels of family unemployment, single-parent
homes, crime, and feelings of hopelessness about the future (Stoddard et al., 2011; Teitelman
et al., 2010; Tolan et al., 2003; Wandersman & Nation, 1998). In fact, sixty-two
percent of African-American children ages 6 through 11 (2.1 million individuals) live in
low-income families (Wight & Chau, 2009), and
homicide is currently the leading cause of death for African Americans 10 to 24 years old)
and rates of homicide for African-American males in this age group exceed those of
Hispanics and White males (Centers for Disease Control and
Prevention, 2010).
In urban schools, 7.6% of students report feeling too unsafe to attend
class (Centers for Disease Control and Prevention,
2003). Additionally, African American boys are overrepresented in special
education classes (Jackson, 2008) and are more
likely to be suspended and expelled from school than boys from other race or ethnicities
(Noguera, 2003) and are underrepresented in
college preparatory and honors classes. Furthermore, African-American boys from low-income
families are at great risk for poor educational outcomes (Thomas & Stevenson, 2009; Wood, Kaplan,
& McLoyd, 2007), and are often at heightened risk for violence, drug use, and
school drop-out. The effect of this is evident in college enrollment numbers: in 2000,
12.9% of the U.S. population was African American; yet, African American males
represented less than 4% of the young people enrolled in colleges and universities
in the U.S (Harvey & Eugene, 2005). The risks
associated with the inner-city neighborhoods and schools where urban African American
youth spend their time make it critical to better understand relational aggression and
hostile attributional biases.
Social Information Processing and Hostile Attributional Biases
The social information processing (SIP) model (i.e., the way in which individuals
process and interpret the meaning of social cues) has been extensively investigated with
youth who are physically aggressive, and suggests that physical aggressors differ
fundamentally from non-physically aggressive youth at each stage of the SIP model (e.g.,
Crick & Dodge, 1994). These stages include
the encoding of social cues, interpretation of social cues, generation of possible
solutions, selection of a response, and enactment of a response. Similar research over the
past 10 to 15 years suggests that relational aggressors also experience some social
processing deficits (Crick, 1995; Crick, Grotpeter, & Bigbee, 2002; Leff, Kupersmidt et al., 2003).
The second step in this SIP model, the interpretation of social cues, has been
studied most extensively (see Orbio de Castro, Veerman,
Koops, Bosch, & Monshouwer, 2002). For instance, considerable research has
demonstrated that the interpretation of a neutral action (e.g., someone bumping into you
from behind or not being involved in a group activity) as hostile (Hostile Attributional
Bias; HAB) has been found in both physical and relational aggressors. In fact, HAB has
been studied more frequently than any of the other SIP steps because it has been shown to
be related to the development and maintenance of aggressive behavior in children (Orobio de Castro et al., 2002). Therefore, researchers
have attempted to better measure (Hughes, Meehan, &
Cavell, 2004; Leff, Kupersmidt et al.,
2003) and intervene (Conduct Problems Research
Group, 1999; Hudley, Britsch, Wakefield, Smith,
DeMorat, & Cho, 1998; Leff et al.,
2009; Leff, Waasdorp, et al., 2010; Lochman & Wells; 2003; Lochman & Wells, 2004) at the level of social information
processing.
Cartoon Illustrations as a Modality for Assessment
Following the lead of other researchers, cartoon illustrations were used by the
authors of the current study as a modality for exploring the second step in social
information processing. Cartoon illustrations have been proven useful in many
psychological measures, especially in the assessment of pain (Creemeens, Eiser, & Blades, 2006; Chambers, Giesbrecht, Craig, Bennett, & Huntsman, 1999), but
also to facilitate comprehension and interest in the measure (Bidaut-Russell, Valla, Thomas, Bergeron, & Lawson, 1998;
Praver, DiGiuseppe, Pelcovtiz, Mandel, & Gaines,
2000; Sadowski, Gulgoz, & LoBello,
1994; Shea et al., 2008).
The current assessment tool utilized cartoon illustrations to aid in the
acceptability and comprehension of an existing HAB measure for boys (Crick, 1995). The current research team previously developed a
cartoon-based version of this measure for girls (Leff et al., 2006). This adapted measure was novel in that cartoon illustrations
were designed specifically for an underserved population (urban African American girls)
and accompanied the standard written vignettes (Crick,
1995; Crick et al., 2002). Notably, this
version of the measure was created through participatory action research (PAR) with urban
youth and community members in order to ensure that it was developmental-appropriate and
culturally-responsive. Results from a subsequent study found that the adapted measure was
psychometrically strong, and was acceptable, interesting and meaningful to the targeted
population. In fact, there are relatively few measures that are culturally and
developmentally appropriate for urban youth (Fantuzzo,
Coolahan, & Weiss, 1997; Leff, Cassano et
al., 2010), a focus of the current investigation. A limitation of the original
cartoon-based measure was that it was developed solely for girls. Therefore, the
development of an attributional bias measure depicting relationally and physically
provocative social situations for urban African-American boys appeared to be a logical
next step.
Using a Participatory Action Research Approach in the Design of Measurement
Tools
Traditionally, psychological assessment tools have been developed by researchers
who define primary constructs and then develop and assess items based on extensive
literature reviews and psychometric analyses. While this approach has served the field
well and has resulted in theoretically and psychometrically sound instruments, the process
tends to be expert-driven (see Leff, Cassano et al.,
2010). Thus, the important role participants and community stakeholders can play
in the development of culturally sensitive assessment tools has often been ignored (Nastasi et al., 2000). This expert-driven development
may also be disenfranchising to ethnic minority populations and may (1) undermine these
individuals’ belief in the importance of the research being conducted (Fantuzzo et al., 1997), (2) cause these individuals to
conceal their true feelings thereby weakening the ecological validity of the measure
(Nastasi et al., 2000), or (3) allow these
individuals to put forth only minimal effort when completing the indices (Leff et al., 2006).
Because of this, some researchers have advocated for the use of a participatory
action research (PAR) paradigm in which the strengths of traditional assessment techniques
(prior empirical support, psychological theory, psychometric soundness) are combined with
important key stakeholder and/or participant feedback (Leff, Cassano et al., 2010; Leff, Power,
Costigan, & Manz, 2003; Nastasi et al.,
2000). Input from research participants and key stakeholders in the development
process helps ensure that the resulting measure is not only psychometrically sound, but is
also culturally-sensitive to the particular community from which the stakeholders were
drawn. Further, a PAR process enhances the meaningfulness and ecological validity of the
resulting research (Nastasi et al., 2000).
Therefore, the current measure was designed within a PAR framework. Specifically, the
research team recruited and worked with many key stakeholders from the local urban
community (students, teachers, and parents) in the development and refinement of this HAB
assessment tool.
It is noteworthy that the research team applied the PAR model to a relatively
under-studied and vulnerable population. Specifically, the partnering stakeholders and
youth for which the cartoon-based HAB measure was designed were low income, African
American males living in a large urban city. Because prior research has suggested that
some traditional written vignette measures, though valid, have been viewed as
inappropriate for and/or of low interest to urban African American youth (Leff et al., 2006), it seemed especially important to
engage youth and key stakeholders in adapting a measure to be of interest to this at-risk
group. Further, a number of studies have demonstrated that although physical aggression
tends to be the focus of research and media attention within urban African American
communities, recent studies suggest that relational aggression is also of considerable
concern (Leff et al., 2009; Talbott, Celinksa, Simpson, & Coe, 2002; Xie, Farmer, & Cairns, 2003). Moreover, the fact that some
relational aggressors also exhibit high levels of physical aggression (Card et al., 2008; Cillessen &
Mayeux, 2004) suggests that it is important to address both forms of aggression
among urban African American youth.
The Current Study
The current study illustrates how researchers partnered with key stakeholders to
develop and preliminarily validate a cartoon-based adaptation of a written hostile
attributional bias measure for urban African American boys. A series of studies are
described in which the researchers attempted to better understand the cultural
sensitivity, developmental appropriateness, and psychometric properties of the newly
adapted measure that we hope will have applicability as an outcome measure for aggression
prevention programs conducted with urban African American boys. We expected that student
and teacher reported acceptability of the cartoon-based measure would be higher than the
traditional written vignette measure, and that the cartoon measure would demonstrate
adequate psychometric properties across relevant subscales. In addition, we hoped that our
study would serve as an illustration for how researchers can work collaboratively with
community stakeholders to ensure the developmental and cultural appropriateness of
measures for the needs of particular at-risk populations.
MethodsDescription of Standard Hostile Attributional Bias (HAB) Measure and Overview of a
Series of Adaptation Studies Undertaken
All activities in the current manuscript were undertaken to adapt a recognized
HAB measure to include cartoon illustrations specifically tailored for use with African
American boys that would accompany the standard written vignettes. The Measure of
Hostile Attributional Bias (HAB) and Feelings of Distress (Crick, 1995; Crick et al.,
2002; Leff et al., 2006)
includes ten written vignettes depicting common social conflict situations of ambiguous
intent that occur at school. Five of the vignettes depict a relationally provocative
social situations (i.e., someone is not included in an activity though it is unclear why
this occurred) and five depict a physically provocative social situations (i.e., someone
is bumped from behind though it is not clear why this has occurred). On two questions per
vignette, the participant is asked to indicate whether they think the action depicted was
intentional (score of 1 point) or unintentional (score of 0). Children’s
intentionality responses are then summed across the five relationally provocative or
physically provocative vignettes; scores range from 0 to 10, with higher scores indicating
higher levels of HAB. This score can also be represented as a percentage of the time that
the child makes an HAB on the measure (e.g., a 0.50 would indicate that 50% of the
time the child made a hostile attribution). Children’s level of distress is
measured by their response to how upset (scored on a 1 to 3 likert scale where 1 =
Not Upset at All and 3 = Very Upset) and how mad (scored on the same likert scale)
they would be if the depicted situation happened to them. Children’s feelings of
distress are then summed across the relational or physical vignettes and can range from 10
to 30.
A number of studies have demonstrated that the traditional measure has strong
reliability, validity, and acceptability (Crick,
1995; Crick et al., 2002). Despite this,
our pilot studies with urban African American youth suggest that this group of children
and adolescents may approach the standard written vignettes in a different manner than the
predominately Caucasian middle-class samples involved in the original Crick studies. Urban
African American youth reported that they were bored and not particularly interested in
completing the traditional written measure (Leff
et al., 2006). Given this, the research team developed a cartoon-based adaptation
of this measure designed for urban African American girls that has demonstrated strong
psychometric properties and higher acceptability levels than the traditional written
measure (Leff et al., 2006). Therefore,
developing a cartoon-based adaptation for urban African American boys appeared
warranted.
A series of four studies using a partnership-based framework were conducted for
the current manuscript. Specifically, Study 1 describes how the standard measure (Crick, 1995) was adapted and finalized so that
appropriate cartoons depicting urban African American youth accompanied each of the ten
standard written vignettes. In Study 2, the research team conducted a psychometric study
with 3rd and 4th graders and their teachers within the context of a
large urban elementary school. In Study 3, the acceptability of the cartoon-based
adaptation was directly compared to the standard written vignette measure in an
independent sample of boys attending a different urban elementary school. Finally, Study 4
was conducted to determine whether certain psychometric properties (especially the
test-retest reliability) of the primary subscales of the new measure were adequate for
another separate sample of urban boys.
Study 1: Measurement AdaptationParticipants
Forty boys ranging in age from 8 to 12 years participated in the measurement
adaptation phase of the current study. Participants were attending either a summer day
camp program or a summer school enrichment program serving a large public urban school
district. The majority of participants were African American (92.5%) with the
remaining youth being Asian American (7.5%). Youth were entering either the
3rd (17.5%), 4th (25%), 5th
(42.5%), or 6th (15%) grades. Ninety percent of the youth
attended public school with the remaining youth attending either private (5%) or
charter (5%) schools in the area.
Stakeholders
A team of community partners, including one teacher and two school support staff
from three different schools, helped the researchers interpret findings obtained from the
youth. All three individuals were African American and spent the majority of their lives
living in the community served by participating schools. They were chosen because of their
experience working with youth and teachers in the local urban schools and because they had
worked with the research team previously.
Procedure
The goal of Study 1 was to create cartoon illustrations to accompany each of the
ten written social conflict vignettes included in the original HAB measure (Crick, 1995; Crick et
al., 2002). The research team followed a series of steps in order to accomplish
this. First, an experienced cartoonist drafted cartoon illustrations of urban African
American boys to accompany each vignette. Participating boys were then individually shown
the cartoon illustration and asked to describe what they thought was happening in the
cartoons without knowledge of the vignette content. Subsequently, the corresponding
written vignette was read aloud to them while they looked further at the cartoon. Next two
standardized questions related to the intentionality of the provocateur and two
standardized questions related to feelings of distress (being mad or upset) were asked.
Each boy was also asked whether this type of situation had ever happened to him or to a
friend in order to determine the relevance of the social situation depicted. Finally,
participants were asked to provide specific feedback related to the depiction of the
cartoon characters and background context of each illustration so that they would be most
appropriate for use with urban males of their age. After following this process for all
ten of the stories, boys were asked several general questions about what could make the
experience more enjoyable and understandable.
The research process progressed in an iterative manner that is characteristic of
using a participatory action research model (PAR; see Leff, Costigan, & Power, 2004; Leff, Cassano
et al., 2010; Nastasi et al., 2000).
Approximately five or six youth at a time worked with the research team. Their feedback
was tabulated and modifications to the depiction of the cartoon characters and setting
characteristics were made, and then the updated versions of the cartoon vignettes were
used with the next group of boys.
Results
In general, as more youth participated in the measurement adaptation phase and
the cartoons were updated, it was clear that the majority of youth felt that the cartoons
provided appropriate illustrations of the written vignettes. It was determined that there
was an adequate heterogeneity in responding about the intentionality of the provocateur in
9 out of the 10 vignettes over the course of the 40 participant administrations. The one
vignette which was still largely viewed as depicting an unintentional action was further
developed through extensive feedback from the community partners prior to the finalization
of the measure. Suggestions for cartoon depiction modifications included providing longer
and baggier shorts for the characters, making closer cropped hair and changing the way
braids were drawn on the boys, giving characters a smaller nose and considerably darker
complexion, and trying to make several characters look younger and more like children. In
addition, a number of suggestions were made to change the spatial relations between
characters so that the intentionality of the situations was more ambiguous. Finally,
responses to the relevance of the depicted situations revealed that approximately
76% of the boys thought that each physical social situation had happened to
themselves or a friend (range from 67.5% to 87.5% per vignette), and
approximately 66% reported that each relational social scenario had happened to
themselves or a friend (range from 55% to 80% per vignette). Figure 1 contains the final cartoons used for one
relationally provocative social situation and one physically provocative social
situation.
Study 1 Conclusions and Implications for Psychometric Study
Feedback from the participants and the community stakeholders indicated that the
resulting cartoon illustrations were viewed as credible and engaging, and the situations
frequently occur for urban male youth in the 3rd to 6th grades.
Thus, results from the measurement adaptation study suggested that the measure would be
appropriate for a systematic psychometric study, which was the goal of Study 2.
Study 2: Psychometric StudyParticipants
All 3rd and 4th grade boys from one large urban public
elementary school were offered the opportunity to be involved in the psychometric study.
One hundred and sixteen 3rd and 4th grade boys gave assent for the
study, had parental permission, and were present on the day of testing (82.3% of
the sample). Ninety-six of the boys were African American (83%), with the rest
being Caucasian (6%), Asian American (7%), or of another ethnic minority
background (4%). Given that this manuscript is specifically focused upon African
American youth, data were analyzed only for these 96 children. Most participants were
between the ages of 8 and 10 years of age with a range of 7 to 11 years of age. The
majority of youth who attend this school receive free or reduced lunch, and therefore this
group is considered a low socioeconomic status sample. No further information from the
families of these youth was available. Each of the 3rd and 4th grade
classroom teachers (n =10) agreed to be administered the
cartoon-based version of the measure in order to assess the acceptability of the measure
for their students. All of the teachers were female, 70% were African American and
30% were Caucasian, and they had an average of 13 years of teaching experience
(range from 1–24 years).
Procedures
Participating boys were randomized on a 1:2 ratio to the standard written
vignette only version (n = 34) or to the written plus cartoon
vignette version (n = 62). Participants were individually
administered this measure along with a brief acceptability measure immediately following
in order to better understand their perception of the measure. It took youth approximately
15 minutes to complete both forms. Approximately 20 boys from each condition were randomly
selected to participate in a 2 week test-retest reliability phase.
The ten teachers were then administered the ten vignettes with cartoon
illustrations. They were asked several questions about how a typical male youth in their
class would respond and/or feel in each situation.1 Following this, teachers were asked to complete a brief 5-item
acceptability questionnaire indicating how acceptable, engaging, and enjoyable they
thought that the written vignette plus accompanying cartoon illustrations would be for
their male students (see Figure 2). In addition, they
were asked to provide ratings for an additional six questions in order to help the
research team better understand their thoughts about the new measure.
MeasuresMeasure of Hostile Attributional Bias (HAB) and Feelings of Distress (<xref rid="R10" ref-type="bibr">Crick, 1995</xref>)
As described on pages 8 and 9, this measure includes 10 written vignettes
depicting common social conflict situations of ambiguous intent, and a series of studies
have demonstrated strong reliability and validity of the measure for use with a diverse
group of youth (Crick, 1995; Crick et al., 2002; Leff et al., 2006). The written vignettes of the standard measure were
administered to participants along with the cartoon illustrations described in Study
1.
Student Acceptability Questionnaire (SAQ)
The SAQ is a brief five-item acceptability questionnaire that has been used in
a previous study to determine perception of acceptability for the measure. The SAQ has
previously demonstrated adequate internal consistency and test-retest reliability
(Leff et al., 2006).
Teacher Acceptability Questionnaire (TAQ)
The TAQ is a brief five-item acceptability questionnaire that was adapted for
use in the current study based upon similar measures used in previous research (Leff et al., 2009). In addition, teachers were asked
to provide ratings for an additional six questions in order to help the research team
better understand their thoughts about the new measure and how they anticipated their
students would experience the new measure.
Results
Correlational analyses suggested that attributions of intentionality in
physically and relationally provocative social situations are moderately correlated with
one another in both the standard written (r = 0.43,
p = .01) and written plus cartoon (r =
0.51, p < .001) conditions. There was not a significant difference
between these two correlations (z = 0.46, p
= 0.64). In addition, feelings of distress were moderately correlated between
physically and relationally provocative social situations for the standard written
(r = 0.55, p = .001) and the written
plus cartoon (r = 0.35, p < .01) conditions.
There was also not a significant difference in correlations between the written versus the
written plus cartoon conditions for feelings of distress (z =
1.13, p = 0.26).
Two analyses of variance (ANOVA) were conducted to compare boys’
responses of intentionality and feelings of distress in physical situations across
conditions. These analyses indicated that boys exhibited similar levels of hostile
attributional bias in physical situations for the written condition (M
= 3.38, SD = 2.45) as compared to the written plus
cartoon condition (M =3.48, SD = 2.51)
and similar levels of distress in physical situations for the written condition
(M = 22.85, SD = 4.03) as compared to
the written plus cartoon condition (M =23.11, SD
= 4.43). Similar ANOVAS were conducted to compare intentionality responses and
feelings of distress in relational situations across conditions. These analyses indicated
that boys exhibited similar levels of hostile attributional bias in relationally
provocative situations for the written condition (M = 5.50,
SD = 2.08) as compared to the written plus cartoon condition
(M =5.08, SD = 2.37) and similar
levels of distress in relationally provocative situations for the written condition
(M = 19.0, SD = 4.18) as compared to
the written plus cartoon condition (M =18.03, SD
= 3.78).
Two paired sample t-tests were conducted in order to compare
boys’ hostile attributional biases in physical versus relational social
situations, collapsing across vignette condition (e.g., standard written and written plus
cartoon version) given the non-significant results reported above. Results indicated that
boys displayed higher levels of a hostile attributional bias in relational situations
(M = 5.23, SD = 2.27) as compared to
physical situations (M = 3.46, SD =
2.49); t = 7.09, p < .001; d
= 0.75). In contrast, results suggested that boys’ displayed higher levels
of distress in physical (M = 23.01, SD =
4.29) as opposed to relational situations (M =18.37,
SD = 3.93), t = 10.09,
p < .001; d = 1.13).
Psychometric Analyses
Adequate internal consistency was obtained for attributions of intentionality
and feelings of distress in both the standard written version and the written plus
cartoon version across both physical and relational situations (see Table 1). Test-retest reliability was conducted across both
the standard written version and written plus cartoon version for relational and
physical situations. As can be seen in Table 1,
test-retest reliability for attributions of intentionality and feelings of distress was
marginal in the physical vignettes across the written and written plus cartoon versions
and generally quite poor in relational situations across both versions.
Student Acceptability Analyses
The internal consistency of the acceptability measure was poor for the
standard written version (α = 0.45) and marginal for the written plus
cartoon version (α = 0.62). As such, student acceptability on this
measure is not reported.
Teacher Acceptability Analyses
Adequate internal consistency was found for the teacher acceptability measure
(α = 0.84). Teachers reported the written plus cartoon version of the
measure to be extremely acceptable. Specifically, teachers responded with a mean score
of 1.52 (SD = 0.60) on a 1 to 5 point scale anchored by 1 = Strongly
Agree (indicating high acceptability) to 5 = Strongly Disagree (indicating high
unacceptability).
Teachers also provided ratings on a series of descriptive questions designed
to better understand their thoughts about the measure and the experience they felt their
students had in completing the measure. As can be seen in Figure 2, teachers clearly felt that the average student in their class would
feel that the cartoons and stories made sense and were easy to follow, the length of the
measure was appropriate, and in comparison to the traditional written measure, students
would find the written plus cartoon version to be a better way to discuss friendships
and conflicts and much more enjoyable and fun to complete.
Study 2 Conclusions and Implications for Follow-up Studies
Results from Study 2 highlight several conclusions as well as additional areas
of future research. It appears that urban African American boys attribute hostile
intentions to others relatively frequently when interpreting relationally provocative
social situations of ambiguous intent (e.g., 50% of the time boys think others
were purposefully mean when in fact the intent of the provocateur is unclear). In fact,
boys in the sample were more likely to express a hostile attributional bias in relational
as compared to physical social situations (52% versus 35% of the time). At
the same time, however, boys are more distressed in physically as compared to relationally
provocative situations. Taken together, these findings suggest that relationally
aggressive situations occur frequently among urban boys, and while they are sometimes
interpreted as being intentional, boys appear more upset and distressed by physical
aggression. These findings may provide a bridge between discrepant findings in the
literature by suggesting that relational provocation occurs quite frequently among boys
(e.g., Swearer, 2008), but that boys do not show
the same emotional reaction or distress that is often experienced among girls in
relationally aggressive situations (Crick et al.,
1996; Murray-Close & Crick, 2007;
Paquette & Underwood, 1999).
The psychometric analyses suggest that all subscales of the standard written and
written plus cartoon version of the measure are internally consistent. However, the
test-retest reliability suggests that the way in which boys respond to a number of
subscales across both the traditional written measure and the cartoon-based version may
not be consistent over a short period of time. In discussing these findings with the
community stakeholders, it was pointed out that the test-retest reliability may have been
low due to the time of year in which the study was conducted. Specifically, low
test-retest reliability may be due to the fact that participants were often tested on the
first occasion just before the winter holidays and then retested just after the holidays.
The stakeholders felt that the testing may be unreliable due to the children’s
decreased attention and increased excitement over the holiday. Unfortunately, when
conducting research within busy urban schools the timing of assessments may not always be
ideal, as was the case in this phase of our study.
Another primary finding was that teachers found the cartoon-based measure to be
extremely acceptable and enjoyable for their students, and also judged this version of the
measure to be a much better way to learn about friendships and conflicts than the
traditional written measure. On the other hand, low internal consistency was found for the
student acceptability measure, especially with regard to thoughts about the standard
written measure. This made the research team reluctant to interpret student levels of
acceptability for both the standard and adapted versions of the measure.
The combination of findings from Study 2 led the research team to propose two
additional studies. In Study 3 the researchers enrolled youth from a different urban
elementary school to determine which version of the HAB measure (traditional written
versus written plus cartoon) they would prefer after being administered several examples
of both versions. In Study 4, the research team aimed to determine whether the poor
test-retest reliability of both versions of the measure (standard written and written plus
cartoon) for urban boys in Study 2 was a replicable finding, or whether this finding may
have been attributable to the time-of-year concerns as hypothesized by the
stakeholders.
Study 3: Student Acceptability StudyParticipants
Participants included twenty 3rd and 4th grade boys from another urban
elementary school within the same public school district as Studies 1 and 2. All
participants were African American and in 3rd grade (n=9) and
4th grade (n=11). The vast majority of students at this school
received free or reduced price lunches.
Procedures
In order to directly test the hypothesis that the cartoon version would be more
acceptable than the standard written version for a sample of urban African American boys,
students were individually administered vignettes from each of the two conditions. Each
student received five standard written and five written plus cartoon vignettes in a
counterbalanced order.
MeasuresMeasure of Hostile Attributional Bias and Feelings of Distress (<xref rid="R10" ref-type="bibr">Crick, 1995</xref>)
See previous description.
Student Acceptability Measure
An eight item forced choice acceptability measure that has been used in prior
research (Leff et al., 2006) was utilized in the
current study. Participants were asked to choose which condition (standard written
versus written plus cartoon) was more enjoyable, acceptable, and meaningful to them.
Following each question students were given an opportunity to write an open-ended
response as to why they felt that one condition was more acceptable than the other.
Results
A binomial probability analysis was conducted to compare student responses
across the eight acceptability questions. Results clearly indicated that the written plus
cartoon version was more acceptable and enjoyable than the standard written version
(p < .001). Some of the qualitative responses for why students
preferred the cartoon version included: “I like cartoons and I can see what is
going on;” “Because you can hear (when it is read) and see what happened
(from the cartoons);” “Because seeing them (the cartoons) gave me
ideas;” and “Because cartoons are fun.”
Study 3 Conclusions and Implications
Study 3 suggests that the written plus cartoon version of the attributional
measure is more enjoyable, understandable, and acceptable for urban African American boys
than the traditional written version of the measure. This finding suggests that the use of
community partnerships to adapt assessment tools may increase the cultural sensitivity,
meaningfulness, and utility of the resultant measures. This is supported by other research
that has used acceptability as a marker for the developmental and cultural appropriateness
of measurement tools (Leff et al.,
2006).
Study 4: Test-Retest Reliability Study
This test-retest study was undertaken to determine whether the relatively weak
test-retest reliability found in Study 2 was indicative of an actual finding (e.g., that
urban boys do not consistently respond to the traditional or cartoon-adapted measure across
short time periods) or an artifact of the timing of the re-testing (e.g., in study 2 the
re-testing occurred primarily just before or after the winter holidays). A secondary goal of
Study 4 was to determine whether Study 2 findings regarding boys being more likely to
exhibit higher levels of hostile attributions in relational as opposed to physical
situations while experiencing more distress in physical than relational situations could be
replicated.
Participants
Seventy boys who were attending one of three urban public summer camps
participated in Study 4. This represented the vast majority of boys who had just completed
the 2nd – 5th grades in attendance on the days in which the
testing was conducted2. Ninety-one percent
of the sample was African American, 4.3% were biracial including African American,
and the remaining participants were drawn approximately equally from other ethnic minority
groups. The participants included five boys who had just finished the 2nd
grade, 24 boys who had just finished the 3rd grade, 25 boys who had just
finished the 4th grade, and 16 boys who had just finished the 5th
grade. Most participants were between the ages of 9 and 11 years of age with a range of 7
to 12 years of age. Only data from African American and/or biracial youth were used for
all subsequent analyses (n = 67).
Procedure
The boys were randomly assigned to the standard written vignette measure
(n = 32) or to the written plus cartoon measure
(n = 35). Participants were individually administered the
standard questionnaire on two occasions, such that the second occasion occurred
approximately two weeks after the initial administration for a test-retest reliability
trial.
MeasuresMeasure of Hostile Attributional Bias and Feelings of Distress (<xref rid="R10" ref-type="bibr">Crick, 1995</xref>)
See previous description.
Results
Strong test-retest reliability for attributions of intent and feelings of
distress in relationally provocative and physically provocative social situations were
obtained on both versions of the measure and can be seen in Table 2. The one exception to this was for feelings of distress
in physically provocative social situations in the standard written version which was
marginally significant (r = 0.62).
Correlational analyses suggested that attributions of intentionality in
physically and relationally provocative social situations are moderately correlated with
one another in both the standard written (r = 0.38,
p < .05) and written plus cartoon (r =
0.42, p < .05) conditions. There was not a significant difference
between these two correlations (z = 0.18, p
= 0.85). In addition, feelings of distress were moderately correlated between
physically and relationally provocative social situations for the standard written
(r = 0.47, p < .01) and the written plus
cartoon (r = 0.44, p = .01) conditions.
There was also not a significant difference in correlations between the written versus the
written plus cartoon conditions for feelings of distress (z =
0.15, p = 0.88).
Similar to the findings in Study 2, internal consistency was generally strong
for attributions of intentionality and feelings of distress in relational and physical
social situations for both the standard written measure and the written plus cartoon
version.
In order to replicate findings from Study 2, a paired sample
t-test was conducted to determine whether boys displayed stronger hostile
attributions in relational as compared to physical social situations while experiencing
more distress in physical as compared to relational situations. For these analyses we
collapsed across vignette condition (standard written versus written plus cartoon).
Results replicated Study 2 findings, such that boys displayed higher levels of a hostile
attributional bias in relational situations (M = 5.59,
SD = 2.15) as compared to in physical situations
(M = 3.98, SD = 2.98);
t = 4.52, p < .001; d
= 0. 62). Also similar to Study 2 findings, boys displayed higher levels of
distress in physical (M = 24.17, SD =
4.46) as opposed to relational situations (M =20.11,
SD = 4.85), t = −6.92, p <
.001; d = 0.87).
Study 4 Conclusions and Implications
Findings from Study 4 suggest that the new cartoon-based version of the Crick
attributional bias measure demonstrates adequate psychometric properties (internal
consistency and test-retest reliability) for urban predominately African American boys.
Further, these findings suggest that weak test-retest reliability findings from Study 2
may have, in fact, been attributable to causes separate from the measure itself (i.e.,
possibly time of year considerations or other factors). This highlights the importance of
implementation factors and the timing of test administrations when conducting research in
the public school system. It further suggests that the active involvement of community
stakeholders may lead researchers to better understand the meaning of initially
surprising results. Alternatively, the divergent findings between Study
2 and Study 4 may be due to other factors, such as age. Specifically, the sample for Study
2 included a slightly younger sample of youth than those participating in Study 4.
Study 4 also replicates an interesting finding from Study 2. Specifically, with
an independent sample, urban boys were again more likely to demonstrate a hostile
attributional bias in relational social situations as opposed to physical social
situations (56% to 40% of the time, respectively). At the same time, they
were more distressed in physical situations than relational situations. Future research is
needed to elucidate whether this finding is specific to urban predominately African
American boys or to the majority of boys regardless of their background and community
context.
Discussion
Given the high rates of community violence that plague many urban communities
across the nation (Tolan et al., 2003; Stoddard et al., 2011), it is extremely important to
develop accurate and culturally-responsive assessment tools that can help guide prevention
planning and also help to determine intervention success. Research suggests that not only do
urban African American youth witness acts of violence within their community, but that many
of these youth interpret these stressors as concerning and even traumatic (Farrell & Bruce, 1996; Teitelman et al., 2010). Developing measures that are responsive to the needs of
these youth is extremely important in helping to better understand their needs and the
success of prevention programming within the urban school context.
The series of studies outlined here highlight the importance of engaging key
stakeholders in a partnership-based process in the design and adaptation of measures,
especially measures for individuals in low-income, urban minority groups (Tucker & Herman, 2002; U.S. Public Health Service, 2001). This is particularly notable given the growing
recognition that assessment tools are only reliable and valid for use with the specific
samples of participants for which they have been designed and/or validated (Leff, Cassano et al., 2010). The fact that the cartoon-based
adaptation of the HAB measure demonstrated strong psychometric properties and was also
considerably more acceptable than the standard written measure for the sample of urban
African American boys suggests that it is possible to maintain psychometric soundness while
adapting measures to ensure their developmental and cultural sensitivity, acceptability, and
appropriateness. It also suggests that more researchers may want to consider following
procedures similar to those outlined in this paper in order to adapt other measures to
minority populations.
Moreover, the engagement of key community partners in measurement development and
adaptation can also be helpful in interpreting discrepant or unexpected findings. This was
evidenced in Study 2, in which test-retest reliability was lower than researchers
anticipated. Rather than assume these findings were attributable to the measure itself,
community partners helped the research team understand the potential relevance of other
factors in the study, such as the time of year when the students were retested. Because the
results of Study 4 indicated that the psychometric properties of both the standard written
measure and the cartoon-adapted measure were adequate, including strong test-retest
reliability across all subscales, the researchers have confidence in the use of the adapted
measure within the urban predominately African American male sample. Without input from the
community stakeholders, the researchers may have reached different conclusions and not fully
appreciated the usefulness of the measure.
In addition to the benefits of the participatory action research model, there are
also challenges to conducting this type of research. First, it can be difficult at times to
forge relationships with community members when they may not have a positive opinion of
research (see Leff, Thomas et al., 2010). This
barrier is not prohibitive, but it can take some time to build trusting relationships (Israel, Schulz, Parker, & Becker, 1998). Second,
although PAR produced a beneficial suggestion in our series of studies, it is not always the
case that key stakeholders will make suggestions that are in line with the team’s
research goals. For example, community stakeholders may provide suggestions that will add a
significant amount of work to the study, may not be feasible, or are not consistent with the
goals of the research. In situations like this it is important for researchers to balance
the goals of the research team with the needs of the community. However, having partners in
the community may increase the cultural sensitivity of the research being conducted, and
that contribution makes PAR extremely valuable, despite these possible challenges (Hughes, 2002; Leff,
Thomas, et al., 2010; Nastasi et al.,
2000).
The cartoon-based hostile attributional measure may be of use in understanding an
important mediator of violence prevention and treatment in urban ethnic minority communities
and schools. Specifically, this measure appears to be appropriate for 3rd
– 6th grade African American urban boys. Further, given that hostile
attributions have been shown to be a key social information processing deficit in aggressive
children (Crick & Dodge, 1994; Orbio de Castro et al., 2002), this measure may also be useful in
designing and evaluating secondary and tertiary prevention/intervention programs for urban
African American boys. That is, because HAB can lead to increased aggression, it is
extremely important to have measures that appropriately measure this construct, and can
identify any effects of treatment. It is also possible, that a times a hostile attributional
bias may serve a protective function given the neighborhoods and schools that African
American males find themselves in. Future research needs to examine hostile attributional
biases as a mediator between urban settings and life outcomes.
Future Directions and Considerations
Because this cartoon-based HAB measure may potentially be used to evaluate the
outcomes of prevention/intervention programs, its sensitivity to treatment effects is an
important area of future study. For instance, future research needs to examine whether the
measure reliably identifies children who are more or less likely to make hostile
attributions, and then to relate this to relevant outcome indices. If this outcome measure
is found to be sensitive to treatment effects for urban African American boys, then it
could be disseminated and used with this specific population. Additionally, given that
some samples included in the current study were relatively small, future research may also
be warranted with larger sample sizes. Although it is recommended that the cartoons be
adapted before use with different treatment populations (i.e., Caucasian boys; boys from
rural settings), the process by which key school and community stakeholders are able to
engage in the research in order to ensure its meaningfulness sheds light upon the
flexibility often required of researchers working in partnership within the urban schools
and the importance of adapting social cognitive measurement tools so that they are
maximally responsive to the developmental and cultural context.
This research was supported by two NIMH grants to the first author, R34MH072982 and
R01MH075787, and by cooperative agreement number 5 U49 CE001093 from The Centers for Disease
Control and Prevention. Its contents are the sole responsibility of the authors and do not
represent the official position of the Centers for Disease Control and Prevention. This
research was made possible, in part, by the School District of Philadelphia. Opinions
contained in this report reflect those of the authors and do not necessarily reflect those
of the School District of Philadelphia. The authors wish to thank Elizabeth Woodburn for her
work on this project.
Results from this portion of the study are not reported as they go beyond the purpose of
the current manuscript.
The Institutional Review Board approved all phases of the studies described in the
manuscript. While in Studies 1–3 of the research active parent permission was
obtained for participants, in Study 4 researchers and camp leaders collaborated such that
the completion of the measures were integrated within the typical camp day. Given this,
and that it is extremely challenging to obtain parent permission within the urban summer
camp format, parents were informed of the study through flyers and meetings but did not
have to sign permission forms for their child’s participation. Enrollment in the
study was stopped after 70 boys had participated during the first administration.
Bidaut-RussellMVallaJPThomasJMBergeronLLawsonE1998Reliability of the Terry: A mental health cartoon-like screener for African
American childrenChild Psychiatry and Human Development2824926310.1023/A:10226361154859628057CardNAStuckyBDSawalaniGMLittleTD2008Direct and indirect aggression during childhood and adolescence: A
meta-analytic review of gender differences, intercorrelations, and relations to
maladjustmentChild Development791185122910.1111/j.1467-8624.2008.01184.x18826521Centers for Disease Control and Prevention2003Youth Risk Behavior SurveillanceRetrieved from http://www.cdc.gov/healthyyouth/yrbs/index.htmCenters for Disease Control and Prevention, National Center for Injury Prevention
and Control2010Youth violence: facts at a glanceAtlanta, GACenters for Disease Control and Prevention, Division of Violence
PreventionRetrieved from http://www.cdc.gov/ViolencePrevention/pdf/yv-datasheet-a.pdfChambersCTGiesbrechtKCraigKDBennettSMHuntsmanE1999A comparison of faces scales for the measurement of pediatric pain:
Children’s and parent’s ratingsPain83253510.1016/S0304-3959(99)00086-X10506669CillessenAHNMayeuxL2004From Censure to Reinforcement: Developmental Changes in the Association
Between Aggression and Social StatusChild Development7514716310.1111/j.1467-8624.2004.00660.x15015681CoieJDTerryRLenoxKLochmanJHymanC1995Childhood peer rejection and aggression as predictors of stable patterns of
adolescent disorderDevelopment and Psychopathology769771310.1017/S095457949800176XConduct Problems Prevention Research Group1999Initial impact of the Fast Track Prevention Trial of Conduct Problems: II.
Classroom effectJournal of Consulting and Clinical Psychology6764865710.1037/0022-006X.67.5.64810535231CremeensJEiserCBladesM2006Characteristics of health-related self-report measures for children aged
three to eight years: A review of the literatureQuality of Life Research1573975410.1007/s11136-005-4184-x16688506CrickNR1995Relational aggression: The role of intent attributions, feelings of
distress, and provocation typeDevelopment and Psychopathology31332210.1017/S0954579400006520CrickNR1996The role of overt aggression, relational aggression, and prosocial behavior
in the prediction of children’s future social adjustmentChild Development6752317232710.2307/11316259022243CrickNR1997Engagement in gender normative versus nonnormative forms of aggression:
Links to social-psychological adjustmentDevelopmental Psychology33461061710.1037/0012-1649.33.4.6109232376CrickNRBigbeeMAHowesC1996Gender differences in children’s normative beliefs about
aggression: How do I hurt thee?Child Development6731003101410.1111/j.1467-8624.1996.tb01779.x8706506CrickNRDodgeKA1994A review and reformulation of social information-processing mechanisms in
children’s social adjustmentPsychological Bulletin11517410110.1037/0033-2909.115.1.74CrickNRGrotpeterJK1995Relational aggression, gender, and social-psychological
adjustmentChild Development6671072210.2307/11319457789197CrickNRGrotpeterJKBigbeeMA2002Relationally and physically aggressive children’s intent
attributions and feelings of distress for relational and instrumental peer
provocationsChild Development7341134114210.1111/1467-8624.0046212146738CrickNROstrovJMKawabataYFlanneryDJVazsonyiATWaldmanID2007Relational aggression and gender: An overviewThe Cambridge handbook of violent behavior and aggression245259New York, NY USCambridge University PressFantuzzoJWCoolahanKWeissACicchettiDTothS1997Resiliency partnership-directed research: Enhancing the social competencies
of preschool victims of physical abuse by developing peer resources and community
strengthsDevelopmental perspective on trauma: Theory, research and intervention463514Rochester, NYUniversity of Rochester PressFarrellADBruceSE1996Impact of exposure to community violence on violent behavior and emotional
distress among urban adolescentsJournal of Clinical Child Psychology2621410.1207/s15374424jccp2601_19118172GoldsteinSEYoungABoydC2008Relational aggression at school: Associations with school safety and social
climateJournal of Youth and Adolescence37664165410.1007/s10964-007-9192-4HarveyWBEugeneLA2005Minorities in Higher Education: Twenty-First Annual Status Report:
2003–2004Washington DCAmerican Council on EducationHeningtonCHughesJNCavellTAThompsonB1998The role of relational aggression in identifying aggressive boys and
girlsJournal of School Psychology3645747710.1016/S0022-4405(98)00015-6HudleyCBritschBWakefieldWSmithTDeMoratMChoS1998An attribution retraining program to reduce aggression in elementary school
studentsPsychology in the Schools3527128210.1002/(SICI)1520-6807(199807)35:3<271::AID-PITS7>3.0.CO;2-QHughesJN2002Commentary: Participatory action research leads to sustainable school and
community improvementSchool Psychology Review323843HughesJNMeehanBTCavellTA2004Development and validation of a gender-balanced measure of
aggression-relevant social cognitionJournal of Clinical Child and adolescent Psychology33229230210.1207/s15374424jccp3302_1115136194IsraelBASchulzAJParkerEABeckerAB1998Review of community-based research: assessing partnership approaches to
improve public healthAnnual review of public health1917320210.1146/annurev.publhealth.19.1.173JacksonJH2008Given half a chance: The Schott 50-state report on public education and African
American malesCambridge, MASchott Foundation for Public EducationRetrieved from www.blackboysreport.org/files/schott50statereport-execsummary.pdfJohnsonH2009Vulnerable youth and the transition to adulthoodASPE Research BriefOffice of the Assistant Secretary for Planning and Evaluation, Office of
Human Services Policy, U.S. Department of Health and Human ServicesRetrieved from aspe.hhs.gov/hsp/09/vulnerableyouth/6/index.pdfLeffSSCassanoMMacEvoyJPCostiganT2010Initial validation of a knowledge-based measure of social information
processing and anger managementJournal of Abnormal Child Psychology381007102010.1007/s10802-010-9419-920449645LeffSSCostiganTEPowerTJ2004Using participatory-action research to develop a playground-based
prevention programJournal of School Psychology4232110.1016/j.jsp.2003.08.005LeffSSCrickNRAngelucciJHayeKJawadAFGrossmanMPowerTJ2006Social Cognition in Context: Validating a Cartoon‐Based
Attributional Measure for Urban GirlsChild development7751351135816999803LeffSSGullanRLPaskewichBSAbdul-KabirSJawadAFGrossmanMPowerTJ2009An initial evaluation of a culturally-adapted social problem solving and
relational aggression prevention program for urban African American relationally
aggressive girlsJournal of Prevention and Intervention in the Community3726027410.1080/1085235090319627419830622LeffSSKupersmidtJBPowerTJ2003An initial examination of girls’ cognitions of their rationally
aggressive peers as a function of their own social standingMerrill-Palmer Quarterly: Journal of Developmental Psychology491285410.1353/mpq.2003.0003LeffSSPowerTJCostiganTEManzPH2003Assessing the climate of the playground and lunchroom: Implications for
bullying prevention programmingSchool Psychology Review32418430LeffSSPowerTJManzPHCostiganTENaborsLA2001School-based aggression prevention programs for young children: Current
status and implications for violence preventionSchool Psychology Review30343360LeffSSWaasdorpTEPaskewichBGullanRLJawadAMacEvoyJPPowerTJ2010The Preventing Relational Aggression in Schools Everyday (PRAISE) Program:
A preliminary evaluation of acceptability and impactSchool Psychology Review3956958721686034LeffSSThomasDEVaughnNAThomasNAMacEvoyJPFreedman2010Using community-based participatory research to develop the PARTNERS youth
violence prevention programProgress in Community Health Partnerships: Research Education and Action420721610.1353/cpr.2010.0005LochmanJEWellsKC2003Effectiveness of the Coping Power Program and of classroom intervention
with aggressive children: Outcomes at a 1-year follow-upBehavior Therapy3449351510.1016/s0005-7894(03)80032-1LochmanJEWellsKC2004The Coping Power program for preadolescent aggressive boys and their
parents: Outcome effects at the 1-year follow-upJournal of Consulting and Clinical Psychology7257157810.1037/0022-006x.72.4.57115301641LoeberRGreenSMLaheyBBKalbL2000Physical fighting in childhood as a risk factor for later mental health
problemsJournal of American Academy of Child and Adolescent Psychiatry3942142810.1097/00004583-200004000-00010McNeilly-ChoqueMKHartCHRobinsonCCNelsonLJOlsenSF1996Overt and relational aggression on the playground: Correspondence among
different informantsJournal of Research in Childhood Education114767Murray-CloseDCrickNR2007Gender differences in the association between cardiovascular reactivity and
aggressive conductInternational Journal of Psychophysiology6510311310.1016/j.ijpsycho.2007.03.01117481760NanselTROverpeckMPillaRSRuanWJSimons-MortonBScheidtP2001Bullying behaviors among U.S. youth: Prevalence and association with
psychological adjustmentJournal of the American Medical Association2852094210010.1001/jama.285.16.209411311098NastasiBKVarjasKSchensulSLSilvaKTSchensulJJRatnayakeP2000The participatory intervention model: A framework for conceptualizing and
promoting intervention acceptabilitySchool Psychology Quarterly1520723210.1037/h0088785NogueraPA2003The trouble with Black boys: The role and influence of environmental and
cultural factors on the academic performance of African American malesUrban Education38443145910.1177/0042085903038004005Orbio de CastroBVeermanJWKoopsWBoschJDMonshouwerHJ2002Hostile attribution of intent and aggressive behavior: a
meta-analysisChild Development391693410.1111/1467-8624.00447OstrovJMKeatingCF2004Gender difference in preschool aggression during free play and structured
interactions: An observational studySocial Development13225527710.1111/j.1467-9507.2004.000266.xPaquetteJAUnderwoodMK1999Gender differences in young adolescents’ experiences of peer
victimization: Social and physical aggressionMerrill-Palmer Quarterly452242266PraverFDiGiuseppeRPelcovtizDMandelFSGainesR2000A preliminary study of a cartoon measure for children’s reaction to
traumaChild Maltreatment527328510.1177/107755950000500300711232273RysGSBearGG1997Relational aggression and peer relations: Gender and developmental
IssuesMerrill-Palmer Quarterly43187106SadowskiCJGulgozSLoBelloSG1994An evaluation of the use of content-relevant cartoons as a teaching
deviceJournal of Instructional Psychology21368370SheaJAGuerraCEWeinerJAguirreACRavenellKLAschDA2008Adapting a patient satisfaction instrument for low literate and
Spanish-speaking populations: Comparison of three formatsPatient Education and Counseling7313214010.1016/j.pec.2008.03.02618486414StoddardSAHenlySJSievingREBollandJ2011Social connections, trajectories of hopelessness, and serious violence in
impoverished urban youthJournal of Youth in Adolescence4027829510.1007/s10964-010-9580-zSwearerSM2008Relational aggression: Not just a female issueJournal of School Psychology4661161610.1016/j.jsp.2008.08.00119083375TalbottECelinksaDSimpsonJCoeM2002“Somebody else making somebody else fight”: Aggression and
the social context among urban adolescent girlsExceptionality1020322010.1207/S15327035EX1003_4TeitelmanAMcDonaldCWiebeDThomasNGuerraTKassam-AdamsNRichmondTS2010Youth’s strategies for staying safe and coping with the stress of
living in violence communitiesJournal of Community Psychology3887488510.1002/jcop.2040221765562ThomasDEStevensonW2009Gender risks and education: The particular classroom challenges for urban
low-income African American boysReview of Research in Education3316018010.3102/0091732X08327164TolanPHGorman-SmithDHenryDB2003The developmental ecology of urban males’ youth
violenceDevelopmental Psychology1927429110.1037/0012-1649.39.2.27412661886TomadaGSchneiderBH1997Relational aggression, gender, and peer acceptance: Invariance across
culture, stability over time, and concordance among informantsDevelopment Psychology33460160910.1037/0012-1649.33.4.601TuckerCMHermanKC2002Using culturally-sensitive theories and research to meet the academic needs
of low-income African American childrenAmerican Psychologist5776277310.1037//0003-066x.57.10.76212369499UnderwoodMKGalenBRPaquetteJA2001Top ten challenges for understanding gender and aggression in children: Why
can’t we all just get along?Social Development10224826610.1111/1467-9507.00162U.S. Public Health Service2001The surgeon general’s mental health supplement on culture, race and
ethnicityWashington, DCDepartment of Health and Human ServicesRetrieved from www.surgeongeneral.gov/library/mentalhealth/cre/sma-01-3613.pdfWandersmanANationM1998Urban neighborhoods and mental health: Psychological contributions to
understanding toxicity, resilience, and interventionsAmerican Psychologist5364765610.1037/0003-066X.53.6.6479633265WightVRChauM2009Basic facts about low-income children, 2008National Center for Children in PovertyRetrieved from www.nccp.org/publications/pub_892.htmlWoodDKaplanRMcLoydVC2007Gender differences in the educational expectations of urban, low-income
African American youth: The role of parents and the schoolJournal of Youth and Adolescence3641742710.1007/s10964-007-9186-2XieHFarmerTWCairnsBD2003Different forms of aggression among inner-city African-American children:
Gender, configurations and school social networksJournal of School Psychology4135537510.1016/s0022-4405(03)00086-4
Cartoon Illustration of a Relationally Provocative Vignette and a Physically
Provocative Vignette
Imagine that you are taking a walk in your neighborhood one day. After you walk a block
or two, you see two kids that you know from school. You walk over to the kids and say
“hi”. The kids act as if you are not there -- they don’t say
anything to you. Then they say something to each other that you can’t hear and
they walk the other way.
Imagine that you have just finished an art project for school. You’ve worked on
it a long time and you’re really proud of it. Another kid comes over to look at
your project. The kid is holding a jar of paint. You turn away for a minute and when you
look back the kid has spilled paint on your art project. You worked on the project for a
long time and now it’s ruined.
Teacher Perceptions of the Cartoon-Based Version of the HAB Measure
Internal consistency and test-retest reliability for Study 2
Internal Consistency Reliability
Test-Retest Reliability
Physical Situations
Relational Situations
Physical Situations
Relational Situations
α written
α cartoon
α written
α cartoon
r written
r cartoon
r written
r cartoon
Hostile Attribution Bias
.68
.73
.74
.73
.67
.58
.30
.39
Feelings of Distress
.75
.71
.77
.80
.50
.46
.58
.41
Internal consistency and test-retest reliability for Study 4