Teaching and Teacher Education 58 (2016) 28e34

Contents lists available at ScienceDirect

Teaching and Teacher Education

journal homepage: www.elsevier.com/locate/tate

Teachers treat aggressive children: An outcome study

Zipora Shechtman*, Rony Tutian
Haifa University, Israel

h i g h l i g h t s

Teacher and student outcomes were compared in an experimental/control design.

Treatment students reduced aggression and increased perception of teacher empathy.

Teacher empathy was related to decrease in students' aggression.

Trained teachers seem to be effective in treating their aggressive students.

a r t i c l e i n f o a b s t r a c t

Article history: School aggression has become a major challenge to educators. Under the assumption that teacher
Received 31 October 2015 empathy is related to student aggression, we offer a group training program, employing bibliotherapy, for
Received in revised form teachers to help their aggressive students. Participants were 44 female teachers and 165 aggressive
31 March 2016
students. Teachers in the intervention group led the program with a small group of children; in the
Accepted 19 April 2016
contrast group such children were assisted academically. Results indicated more favorable outcomes for
children in the intervention group on aggression and on perceived teacher empathy, and gains on scores
on aggression were related to change in perceived teacher empathy.
Key words:
Child aggression
2016 Elsevier Ltd. All rights reserved.
Teacher training
Teacher empathy

Child and adolescent aggression is prevalent in schools, chal-
lenging the ability of helping professionals to provide services to all
children at risk (Reese, Norsworthy, & Rowlands, 2009). We suggest
that teachers can be trained to treat aggressive students in small
groups, and we offer a unique program based on bibliotherapy.
Bibliotherapy refers to the use of books for therapeutic purposes.
This program has been tested in previous studies, providing an
evidence base (for a review, see Author, 2010); however, it was
always conducted by professionals in mental health. The current
study measures the impact of treatment provided by teachers on
student aggression.
Teacher empathy is a major construct in education (Good &
Brophy, 2008), particularly in regard to challenging students, yet
its application in teacher education is rare (Barr, 2011). Such
empathy may be enhanced through training; the more teachers
understand child aggression and are willing to listen to their stu-
dents on an intimate level, the more they may be willing to support
* Corresponding author. Counseling and Human Developing, Haifa University,
Haifa, Israel.
E-mail address: ziporas@edu.haifa.ac.il (Z. Shechtman).
them. In turn, the more children sense teachers' care and empathy,
the more they may be willing to cooperate and relinquish some of
their disruptive behavior. The training process in the bibliotherapy
group provides teachers with the necessary knowledge, experi-
ences, and skills to convey empathy and assist their aggressive
students. Employing books in treatment is a natural pedagogy to
teachers and students; in addition, it provides a structured method
and a practical tool for them to cope with aggression in school.
1. Aggressive children
Aggressive behavior is dened by most researchers as an
intentional act to hurt others physically or psychologically (for
example, Moeller, 2001). It is frequently present in schools, with
rates ranging from 30% to over 50%, including verbal, relational,
physical, and cyber aggression, and for some children it is a daily
threat (Raczynski & Horne, 2014). Treatment of school aggression
usually takes a path of primary prevention (Conyne, Horne, &
Raczynski, 2013), but this is not sufcient for highly aggressive
children or youth. Aggression is a multi-dimensional phenomenon,
including social, cultural, familial, neurological, and/or biological
variables; therefore, these children and youth need a more rigorous

http://dx.doi.org/10.1016/j.tate.2016.04.005
0742-051X/ 2016 Elsevier Ltd. All rights reserved.
 Z. Shechtman, R. Tutian / Teaching and Teacher Education 58 (2016) 28e34
treatment (Dodge, 2011).
Cognitive behavioral theories provide a convincing explanation
of child aggression. They use distorted social information process-
ing, by which they perceive limited cues in a given situation,
interpret them as intentionally negative, and select an aggressive
response because they perceive power as an advantage, believing
this will serve them best (Dodge, 2011). But, in fact, aggression does
not serve them well, as the consequences are usually not in their
favor; thus, they are anxious, angry, and helpless, and consequently
resistant to change. Most of the treatments offered to these chil-
dren are cognitive behavioral, aimed at changing their distorted
information processing and behavior (Dodge, Godwin, & Conduct
Problems Prevention Research Group, 2013; Kazdin, 2007). How-
ever, aggression seems to be more than a cognitive response.
Aggressive children are angry, anxious and lonely, nding it
difcult to express their feelings (Deffenbacher, Oetting, &
DiGuiseppe, 2002; Potter-Efron, 2005). In fact, Gurian (1997) ar-
gues that anger is the only emotion that aggressive boys feel
comfortable with, as they tend to act out their feelings instead of
expressing them. Aggressive children have been found to suffer
from verbal decits (Connor, 2002) and to have difculty express-
ing needs (Moeller, 2001). Some researchers even suggest that
denial of feelings is a survival strategy for them (Garbarino, 1999),
helping them overcome harsh life circumstances, stress, and failure.
Studies have shown aggression to be negatively correlated with
empathy (Author, 2003; Hrdy, 2009). As aggressive children are
disconnected from their own feelings, they cannot be highly
empathic to the suffering of others (Loudin, Loukas, & Robinson,
2003; Pollack, 2000). They appear to be indifferent and insensi-
tive to such suffering, especially of their victims. While they may
feel shame, guilt, and fear, they mask this with a false image of
strength and arrogance (Garbarino, 1999).
Aggressive children also endorse power. This is part of the dis-
torted information processing attributed to them (Dodge, 2011). As
most meaningful persons in their lives have caused them to
experience humiliation and despair, they nd it virtually impos-
sible to trust strangers. A hostile attribution bias leads them to
conclude that others harbor hostile intent towards them and
therefore deserve to be attacked. Thus, in a situation of social
conict, aggressive children feel they must be on guard and
respond with force; aggression seems their only viable option
(Calvete & Orue, 2012; Toth, Harris, Goodman, & Cicchetti, 2011).
Much of this aggression is a spontaneous reaction. These are the
scripts such children have developed through poor socialization
processes and insecure attachment (Ein-Dor, Mikulincer, & Shaver,
2011), and they act on them impulsively. In addition, many have
difculties regulating their own behavior (Ro ll, Koglin, &
Petermann, 2012). Self-regulation is developed at an early stage
of child development through relationships with parents (Dodge,
2011), the absence of such skills as inhibition of disruptive
behavior, cooperation with others, and self-assertion, it is difcult
for them to control themselves (Bohart & Stipek, 2001).
These characteristics cannot be addressed solely by skills
training. They must involve an emotional focus along with a
cognitive one. Moreover, because aggression is an interpersonal
behavior, treating aggressive children in small groups seems
reasonable, as interpersonal learning is a major therapeutic factor
in groups. Although some researchers raise concerns about the
potential adverse mutual inuence in such groups (Dishion, Dodge,
& Lansford, 2008; Dishion, McCord, & Poulin, 1999), other studies
showed that small groups are efcacious (Author, 2000, 2003,
2006; Author & Birani-Nasaraladin, 2006), and they are certainly
cost effective. The inconsistency in results may be attributed to the
level of child aggression; the later studies relate to aggression in
school children, which is milder than the aggression of the violent
29
youth in the Dishion studies. It may also be related to the effec-
tiveness of the unique bibliotherapy program.
1.1. Bibliotherapy
Bibliotherapy entails the use of books in the treatment of people
(Baker, 1987. It can be used as a cognitive intervention with minimal
therapist intervention (also called self-help) or an emotional
intervention, using ction literature, with intensive therapist
intervention. Most of the bibliotherapy treatment with children is
of an affective type (Gladding, 2005), relying on high quality liter-
ature. Stories are helpful in offering insight into personal problems
(Forgan, 2002) because they create a safe distance, bringing people
indirectly to the edge of sensitive and threatening issues. Biblio-
therapy is also helpful due to processes of identication; by iden-
tifying with literary characters, individuals are exposed to a range
of emotions which they can recognize in themselves, and thus
reconnect to their own emotions (Greenberg, 2002). Finally, high
quality literature presents a wide range of human thoughts that
readers can learn from and apply to their own lives (Kottler, 1986).
Stories, poetry, and lms all comprise wisdom from which in-
dividuals can learn.
Due to the high resistance to therapy, aggressive children need
an indirect method of treatment. In addition, children love stories,
identify with the characters, and understand the aggressive
behavior and its consequences from a distance, which in turn helps
them to express their own feelings and develop insight into their
own behavior. In a study that investigated the contribution of
bibliotherapy to the reduction of aggression (Author, 2006), chil-
dren in bibliotherapy treatment were compared to treatment
children using the same theoretical method but without biblio-
therapy. Results indicated more favorable outcomes on aggression
and empathy in the bibliotherapy. Moreover, those counselors who
used bibliotherapy were more satised with the treatment,
attributing it to the use of books and structure.
The literature we use in bibliotherapy is carefully selected to
cover the major issues of aggressive behavior, adjusted to age,
gender, and culture. Based on a review of aggressive behavior, we
include four content components: anger and its expression, the
endorsement of power, empathy to the suffering of others, and self-
regulation. Each session starts with a new poem, story, or lm. First
the literature is discussed (the indirect phase). Then, after exploring
the behavior, its reasons and consequences, the children are asked
to share their own experiences, understand their own behavior, and
make a commitment to change. Similar processes are used with
teachers in training. Teachers also identify with literary characters,
understand conictual situations, become aware of the difculties
aggressive children face, and develop empathy towards them.
1.2. Teacher empathy
To cope with students' deviant behavior, teachers need to deeply
care about them (Craig, Bell, & Leschied, 2011; Warren & Lessner,
2014). Indeed, empathy appears in the literature as a major
construct for effective teaching (Good & Brophy, 2008), yet little
research has examined empathy in the school context (Baron-
Cohen, 2011; Swan & Riley, 2012). Moreover, not all teachers are
empathic; training programs do not focus on enhancement of
empathy and do not provide teachers with tools to deal with
difcult children (Barr, 2011).
Empathy is dened as experiencing the psychological life of
another person by projecting one's self into the other, to under-
stand what he/she is thinking or feeling (Swan & Riley, 2012). In
common language, it is walking in someone else's shoes.
Empathy includes a cognitive component of perspective taking and
30 Z. Shechtman, R. Tutian / Teaching and Teacher Education 58 (2016) 28e34
an affective component of relating to the other's feelings (Engelen
& Rottger, 2012; Walter, 2012; Zahavi & Overgaard, 2012). When
someone senses how another person feels, he/she is not just
thinking about the other's needs, but also experiencing a shared
state that makes both parties feel better, increases the bond be-
tween them, and shifts attention away from the suffering and onto
the excitement of sharing (Swan & Riley, 2012). Decety and Jackson
(2004) add a behavioral component e an action taken to demon-
strate empathy.
Empathy is the cornerstone of effective human relationships. It
helps to regulate social interaction (Batson, 2012), enhances
cooperation regarding shared goals (de Waal, 2008), increases
satisfaction in intimate relationships (Long, Angera, & Hakoyama,
2008), and is negatively related to aggression (Hrdy, 2009).
Empathy is crucial in education. When teachers are able to
empathize, this creates a bridge with students. It makes students
feel comfortable with and trusting of teachers and improves the
quality of the teacher-student interaction (Warren & Lessner, 2014).
This also leads to higher motivation and functioning and increases
academic achievements (Cornelius-White, 2007; Hattie, 2009).
Studies have shown empathy to be a strong predictor of how
teachers react to aggressive behavior (Craig et al., 2011). A recent
review (Dedousis-Wallace, Shute, Varlow, Murrihy, & Kidman,
2014) found empathic teachers to be sensitive to the needs of the
victims and ready to respond to aggressive behavior. Several
studies showed that teacher empathy was related to classroom
climate and students' positive behavior (Pianta & Allen, 2008;
Pianta, 2006). Evaluating the effect of the My Teaching Partner
intervention, which focuses on teacher-student relationships, on
teacher trainees, Mikami and colleagues (Mikami, Gregory, Allen,
Pianta, & Lum, 2011) found a positive effect of the intervention
on teacher behavior and on student peer relationships, particularly
among students with highly disruptive behavior. Hence, teachers'
expressed empathy may also serve as a model for their students
(Barr, 2011). Furthermore, teachers may be more likely to intervene
when encountering aggressive behavior if they understand the
problem, care about the children, and are equipped with tools to
help (Barr, 2011).
Teacher training usually does not include empathy development
or provide tools to respond effectively to aggression. Thus, teachers
nd it difcult to deal with such behavior. Recently, researchers
proposed that teacher preparation programs should include
emotional intelligence training, in which empathy is a major
component (Barr, 2011; Dedousis-Wallace & Shute, 2009; Ergur,
2009). However, empathy cannot be taught in an informative way;
a greater focus on feelings and practical methods is needed
(Dedousis-Wallace & Shute, 2009).
The current study proposes teacher training that focuses on
empathy towards aggressive children and practical methods to
work with them, providing support throughout the intervention.
The program is expected to enhance teacher empathy and students'
reduction of aggressive behavior. Specically, we hypothesized
that: (1) children's scores on perceived teacher empathy would
increase following the intervention; (2) children scores on
aggression would decrease following treatment by their teachers;
and (3) gains in perceived teacher's empathy would correlate with a
reduction in children's aggression.
2. Method
2.1. Participants
The study was comprised of 165 aggressive children (124 boys
and 41 girls) from 25 schools in Israel: 19 elementary schools
(n 122 adolescents) and 6 junior high schools (n 43
adolescents). That is, 122 children (74%) were 10e12 years old, and
43 children (26%) were 13e15 years old. Schools varied in size from
35 students (special education schools) to 1050 students
(M 447.88, SD 228.71), and the number of classes per school
ranged from 4 to 36 (M 16.60, SD 7.28). The majority of the
children were living with married parents (72.1%), and about half of
the participants were diagnosed with ADD and/or ADHD.
Forty four female teachers from these schools participated in the
study. Their classrooms varied in size from 5 to 36 children per class
(M 23.25, SD 9.73), from which they selected 3e4 children to
participate in a small group intervention (M 3.75, SD 1.35).
These were the most aggressive children based on teacher evalu-
ation. Thirty nine teachers held a bachelor's degree and ve held a
master's degree. Seniority ranged from 3 to 31 years (M 17.27,
SD 6.65). Most had a background in special education and were
enrolled in a master's program titled Inclusive Classrooms. These
educators were highly skilled in treating problem behavior chil-
dren, but it was the rst time they were involved in the treatment
of aggression.
2.2. Measures
2.2.1. Screening tool
Aggressive children were identied by their teachers based on
the Peer Nomination Instrument (Crick, 1996). The original scale
includes 21 items relating to verbal, physical, and relational
aggression (e.g., The child pushes or hits other children). Internal
consistency ranged from 0.87 to 0.95; test-retest reliability after
four weeks ranged from 0.82 to 0.90. As ethical considerations did
not allow us to use peer nomination, we had the teachers complete
a short version of 10 items (1, 3, 5, 7, 9, 11, 13, 15, 18, 20) on a 4-point
scale, for each child in her classroom. Those children who scored 3
or 4 on the aggression scale were invited to participate in the group.
2.2.2. Self-reported aggression
Aggression was measured through a short version of the Child
Behavior Checklist (CBCL; Achenbach & Adelbrock, 1991a), including
19 items. (e.g., I tend to get into ghts frequently) that are parallel
to the TRF (see below). Possible responses range from 0 to 2
(0 not at all, 1 sometimes, 2 often), with a higher score
indicating greater aggression. Validity and reliability were reported
by the authors (Achenbach, Howell, McConaughy, & Stranger,
1995). In addition, the scale offers three clinical categories:
Normal, Marginal, and Clinical. The Hebrew version is commonly
used in Israel (e.g., Zilber, Auerbach, & Lerner, 1994). Alpha in the
current study was 0.88.
2.2.3. Teacher-reported aggression
Child aggression based on adjustment symptoms was measured
via the Teacher Report Form (TRF; Achenbach & Adelbrock, 1991b). A
short version of 19 items pertaining to aggressive behavior was
used (e.g., He tends to get into ghts frequently), with responses
on a 3-point scale, ranging from 0 to 2 (0 not at all, 1 some-
times, 2 often). For the analyses, scores were converted to T
scores (0e100) according to the guidelines provided by Achenbach.
Validity was based on a comparison with Conners' (1990) revised
Teacher Rating Scale, yielding a correlation of 0.80. Agreement
between raters (family, friends) ranged from 0.40 to 0.60. Reported
reliability was 0.96; test-retest reliability was 0.92 after 15 days
(Achenbach et al., 1995). Three clinical categories are offered in
addition: Normal, Marginal, and Clinical. In a previous study in
Israel, internal consistency was 0.83. In the current study, it was
0.93.
Perceived teacher empathy was measured through the
Empathy Scale (Persons & Burns, 1985). The questionnaire includes
 Z. Shechtman, R. Tutian / Teaching and Teacher Education 58 (2016) 28e34
10 items (e.g., My teacher understands me well) on a scale of 1e4
(1 doesn't understand me, 2 sometimes understands me,
3 often understands me, 4 always understands me), with a
higher score representing higher levels of perceived empathy. Re-
ported validity and reliability were good, and internal consistency
in the current study was 0.87.
2.3. Intervention
The intervention involved two phases: teacher training and
teacher practice with aggressive children. First, teachers were
trained in the program. Training included a didactic component, in
which teachers learned about the unique characteristics of
aggressive children (high levels of anger and a need for power and
low levels of empathy and self-control), principles of group lead-
ership, and bibliotherapy. An experiential component included the
teachers themselves in a group process using relevant literature.
This phase took about twelve 4-h sessions in which teachers sensed
the power of a small group and intimate relationships, as well as
the power of literature. Teachers use literature texts routinely, but
in these sessions they also learned how to employ literature in a
therapeutic way.
Second, teachers practiced the program with their students.
They learned to help aggressive children become aware of their
aggression, understand what triggers it, develop motivation to
change behavior, and learn to control it.
At the initial stage of the intervention with children, a thera-
peutic alliance was established in the group through therapeutic
activities and the creation of group norms. For example, following
ice-breaking activities (getting acquainted), the norm of self-
expressiveness was established using the book The Soul Bird
(Snunit, 2010). According to this story, the soul containing our
feelings is made up of drawers. After reading the story, children are
invited to open the drawer that represents their feelings in the
group (here and now focus) or the drawer of feelings that char-
acterizes their current situation in general (out of group focus).
This helps children express their emotional state, legitimizing
expression of feelings in the group.
In the working stage, children discussed issues that are perti-
nent to aggressive behavior: anger, what triggers it and ways to
control it; the need for power and force e coping with threats,
putdowns and disrespect; letting go and forgiveness; developing
empathy towards victims of aggression; and enhancing self-
control. For example, the story When Sophie Gets Angry (Bang,
1999) helps children express feelings of uncontrolled anger; the
poem Joey (Silverstein, 1974) helps them understand the conse-
quences of uncontrolled anger; the movie Madi (1987) helps chil-
dren develop empathy towards victims of aggression; and the
poem I am my Own Commander encourages them to take charge
of their aggressive behavior.
At the last stage of treatment participants summarized their
gains and made a commitment to increase self-control. (For a
detailed description of the intervention, see Shechtman, 2010.)
2.4. Procedure
All the educators participated in a master's program at one
teacher training college, in one of two courses: (1) Treatment of
Child Aggression (the intervention group; n 30) or (2) Learning
Disabilities (the contrast group; n 14). The difference in group
size is a result of course enrollment. The contrast group was initially
particularly small and, in addition, six teachers in the course opted
out of participating in the study.
Each teacher (intervention and contrast groups alike) selected a
few children in her class, based on scores of 3e4 on the adjusted
31
Peer Nomination Instrument (Crick, 1996). In the experimental
group, each teacher worked with the children for twelve 45-min
sessions. Conducting a small group was part of the course re-
quirements; hence, all teachers in the experimental group con-
sented to participate. In the contrast group, teachers assisted the
selected children with their academic work, also in twelve 45-min
sessions.
All children (intervention and contrast groups alike) completed
the CBCL (Child Behavior Checklist) and the PTE (Perceived Teacher
Empathy) questionnaires before the intervention started (time 1)
and upon termination (time 2). Because teachers worked with
these children, no drop out was observed. Educators also
completed the TRF (Teacher Report Form) twice, before the inter-
vention started and upon termination. Permission to conduct the
study was obtained from the Research Unit of the Israel Ministry of
Education and from parents. Adherence was achieved in three
ways. First, all intervention teachers used a written manual
comprised of the four content components and a list of literature
recommendations, to allow them to adjust the literature to the
children's age. Second, all intervention teachers received weekly
group supervision, so the instructor could monitor their work.
Third, because this was part of the intervention teachers' nal
project, the instructor read all reports, including a description of
each session.
2.5. Data analysis
Data analysis was conducted with SPSS (ver.22). Due to the
nested design (students within small groups and schools), and thus
the assumption of correlated data, data were analyzed with mixed
hierarchical models to account for the dependency of observations.
The students are the unit of analysis, nested within classes, which
in turn are nested within schools. Each small group within a school
was dened as the random variable. Analyses of change included
four levels: time within the individual (repeated measures), the
individual level, the class, and the school. Change was examined by
group (2) (intervention and contrast) and time (2) (pre and post)
which were dened as the xed effects of each mode. Post hoc
comparisons for signicant effects were based on estimated mar-
ginal means, comparing pairs of sub-groups. Gender (0-girls, 1-
boys) and age level (0-elementary school, 1-junior high school)
were dened as dummy variables and served as covariates.
3. Results
3.1. Preliminary results: pre-study differences
As most children were boys (about 75%) and most were
elementary school children (about 75%), gender and age level could
not serve as independent variables and were controlled for
throughout the analyses. Pre-study differences on children's vari-
ables (CBCL, TRF and perceived teacher empathy) were examined
by group (intervention, contrast). No signicant differences were
detected.
3.2. Pre-post differences in perceived empathy
The rst hypothesis related to teacher perceived empathy. It was
suggested that, following the program conducted by the teachers,
an improvement would be observed in the intervention group with
regard to teacher perceived empathy.
Group differences in pre-post change in teachers' perceived
empathy were examined with mixed model analyses of pre and
post scores by group (2) and time (2). Time was dened as a within
variable, group as a xed effect between variable, and the small
32 Z. Shechtman, R. Tutian / Teaching and Teacher Education 58 (2016) 28e34

group as the random effect; and students were nested within Table 2
classes and schools. Results of this analysis are presented in Table 1. Time differences in children's CBCL aggression levels, according to clinical cate-
gories, by group (N 165).
A signicant time-by-group interaction was found regarding the
children's perceptions of teachers' empathy (see Table 1). Post hoc Time Intervention Contrast
analyses revealed that perceived teacher empathy increased N N
(%) (%)
signicantly in the intervention group: F(1261.25) 19.99,
p < 0.001, h2 0.113; but no signicant change was observed in the Pre Post Pre Post

contrast group: F(1261.25) 0.24, p 0.628, h2 0.002. Thus, the Normative 67 95 31 37

rst hypothesis was supported. (61.5) (87.2) (55.4) (66.1)
Marginal 16 3 20 14
(14.6) (2.8) (35.7) (25.0)
3.3. Pre-post differences in children's aggression Clinical 26 11 5 5
(23.9) (10.1) (8.9) (8.9)
c2(2) 30.92*** c2(2) 2.96
The second hypothesis suggested that experimental children's
scores on aggression would decrease following treatment by their ***p < 0.001.

teachers. Group differences in pre-post change were examined

with mixed model analyses by group (2) and time (2), controlling
Table 3
for gender and age level and their interactions with time. Table 1 Time differences in children's TRF aggression levels, according to clinical categories,
presents means, SDs and F values for change in children's aggres- by group (N 165).
sion. Results show signicant time-by-group interactions. For the
Time Intervention Contrast
CBCL, a signicant decrease in aggression was found in the inter- N N
vention group: F(1275.01) 46.27, p < 0.001, h2 0.185, while no (%) (%)
change was observed in the contrast group: F(1, 275.01) 1.43, Pre Post Pre Post
p 0.234, h2 0.010. Similarly, for the TRF, a signicant decrease in
Normative 40 83 19 24
aggression was found in the intervention group: F(1278.73) 47.72,
(36.7) (76.1) (33.9) (42.9)
p < 0.001, h2 0.182, and no change was observed in the contrast Marginal 39 15 30 25
group: F(1278.73) 1.03, p 0.312, h2 0.007. That is, children in (35.8) (13.8) (53.6) (44.6)
the intervention group gained from the treatment more than Clinical 30 11 7 7
(27.5) (10.1) (12.5) (12.5)
children in the contrast group on both aggression measures.
c2(2) 73.03*** c2(2) 2.15
Clinical progress was also analyzed on the basis of the CBCL and
TRF. Tables 2 and 3 present change in children's aggression based ***p < 0.001.

on clinical categories. Results indicate a signicant improvement in

the intervention group on both scales, but no change in the contrast
aggression levels, as well as in their perceptions of teachers'
group: c2(2) 30.92*** and c2(2) 2.96, n.s., for intervention and
empathy, scores were dened as adjusted gains computing change
contrast group, respectively, on the CBCL, and c2(2) 73.03*** and
while controlling for initial scores.
c2(2) 2.15, n. s., for intervention and contrast group, respectively,
The correlation between change in CBCL and change in TRF was
on the TRF. More specically, on the CBCL, the increase in the
positive and high (r 0.85, p < 0.001), showing high teacher-child
normal range of aggression was about 26% in the intervention
agreement for change in aggression. Perceived teacher empathy
group and the decrease in the clinical category was about 14%,
was negatively related to the change in CBCL (r 0.22, p 0.022)
compared to 11% and no change in these two categories in the
and in TRF (r 0.28, p 0.003). That is, children's perception of an
contrast group. On the TRF, the outcomes are even more dramatic:
increase in their teachers' empathy was related to a reduction in
40% and 17% in the intervention group vs. less than 10% and no
their own aggression levels. While these correlations are moderate,
change in the contrast group (see Tables 2 and 3). These ndings
they show that gains in the three measures e children's assess-
lend further support to the second hypothesis.
ments of teacher empathy, children's assessment of their level of
aggression, and teacher assessment of student aggression e are
3.4. Relationships between perceived empathy and aggression interrelated.

Finally, an attempt was made to assess the relationships, in the

intervention group, between change in aggression and teachers'
perceived empathy. As changes were observed in children's

Table 1
Means, standard deviations and F values for pre-post differences in children's perceptions of teacher empathy, CBCL and TRF, by group (N 44 teachers, 165 children).

Time Intervention Contrast Ftime Fgroup Ftime  group

M M (h2) (h2) (h2)
(SD) (SD)

Pre Post Pre Post

Children's perception of teacher empathy 2.39 2.63 2.13 2.26 F(1261.25) 1.79 F(1,36.03) 7.99** F(1261.25) 4.38*
(1e3) (0.48) (0.43) (0.59) (0.50) (0.019) (0.046) (0.030)
CBCL aggression 62.00 55.45 61.84 60.45 F(1275.01) 2.24 F(1,40.00) 0.49 F(1275.01) 8.63**
(9.63) (7.65) (7.65) (7.71) (0.014) (0.003) (0.054)
TRF aggression 66.28 61.61 65.95 65.02 F(1278.73) 2.73 F(1,40.15) 0.54 F(1278.73) 9.88**
(6.88) (6.13) (4.78) (4.91) (0.017) (0.003) (0.063)

*p < 0.05, **p < 0.01.

Note. Intervention group: 30 teachers and 109 children; contrast group: 14 teachers and 56 children.
 Z. Shechtman, R. Tutian / Teaching and Teacher Education 58 (2016) 28e34
4. Discussion
The purpose of this study was to assess the impact of teacher
training and intervention with aggressive students on children's
perceived teacher empathy and level of aggression. Results indi-
cated that, following the intervention, aggressive children indeed
reduced their level of aggression compared to a contrast group, and
also felt that teachers were more empathic to their needs. More-
over, change occurred only in the intervention condition.
The reduction in aggressive behavior following the type of
intervention under study is not surprising, yet important. In several
studies conducted in the past, children exhibited such a reduction
(Author, 2006; Author & Birani-Nasaraladin, 2006), despite caution
suggested by others (Dishion et al., 2008) regarding group in-
terventions with aggressive children. However, the uniqueness of
the current study is that trained teachers, rather than school
counselors, were effective in conducting such treatment. Effec-
tiveness was established on children's mean scores provided both
by self and teacher reports, as well as the clinical scores of the in-
struments, for pre-post differences and compared to a contrast
condition. The high correlation between self and teacher report of
aggression adds strength to the results.
Moreover, children felt that teachers' empathy towards them
had increased following the intervention. Teacher empathy is a
major construct in education; empathic teachers create a positive
climate and relationships (Good & Brophy, 2008; Swan & Riley,
2012). Troubled students in particular need teachers' empathy to
feel accepted and understood (Dedousis-Wallace et al., 2014;
Mikami et al., 2011). The child-teacher relationship is considered
as a major component in children's success in school (Pianta &
Allen, 2008; Pianta, 2006). A close relationship with the teacher
leads to greater cooperation and less resistance (Batson, 2012; Long
et al., 2008), and was found to be negatively related to aggressive
behavior (Hrdy, 2009).
Growth in teacher empathy may be attributed to the treatment
they provided, but also to the training they received. In the process
of the intervention, teachers listened to the children's expression of
anger and were exposed to the children's distorted information
processing and to their struggle to reduce defensive reactions and
gain self-control. The intimate climate established in the small
group led to a sense of feeling understood and accepted, which was
reected in the increase in perceived teacher empathy. In addition,
the training itself enhanced understanding of aggressive children,
as stated by one of the participants: I'm a math teacher. I never
thought I would have to be involved in intimate relationships with
my students. Now that I've heard them, I'm not angry at them
anymore, actually I've grown to like them.
The correlations among gains on the variables add support to
the outcomes; gains on all three child variables e self-reported
aggression, teacher-reported aggression and perceived teacher
empathy e were correlated. This suggests that children's reduction
in aggression is related to relationships with their teacher (Batson,
2012; Long et al., 2008; de Waal, 2008).
The study is not without limitations. First, child aggression was
measured by a single type of instrument, and no follow up was
conducted. Although the CBCL and TRF are valid instruments,
widely used and highly accepted, observational measures would
add strength to the results. Follow-up measurement could reveal
the lasting effect of the intervention. Second, it was those very
teachers who worked with the children that reported on their
aggression level, creating a potential social desirability bias in
scores. Nonetheless, children's scores of their own aggression were
highly correlated with teacher scores. Third, the broad variety of
children and teachers, and the fact that only female teachers were
involved in the study, limit the possibility of generalization. Fourth,
33
a clinical intervention may be difcult for teachers, as it requires
some extra time and effort. Nonetheless, most teachers do have a
few out-of-classroom hours. Rather than using all of them for ac-
ademic support, teachers with the needed skills may want to apply
them in the provision of emotional support to their students Fifth,
although several measures were taken to ensure treatment delity
it was not measured formally. Finally, the study would be stronger if
compared to another type of treatment; this was impossible in the
current study, whose purpose was to measure the efcacy of
teachers in treating aggression, but had been done previously (e.g.
Author, 2006). Future studies might examine the ingredients that
make this program effective.
Despite these limitations, the study is important in several re-
spects. First, it adds strength to group interventions with aggressive
children and to the particular intervention proposed in the current
study. More important is the unique outcome suggesting that
teachers can help aggressive students in their classroom to reduce
their aggression and to feel closer to their teachers. The teacher-
child relationship is a major construct in education, particularly
with children who display socio-economic (Pianta, 2006) and/or
socio-emotional difculties (Dedousis-Wallace et al., 2014).
Training teachers to create an emotional bond with their students is
therefore an avenue worth further exploration and effort.
References
Achenbach, T. M., & Adelbrock, C. (1991a). Manual for the child behavior checklist
(CBCL). Burlington: University of Vermont, Department of Psychology.
Achenbach, T. M., & Adelbrock, C. (1991b). Manual for the teacher's report form (TRF).
Burlington: University of Vermont, Department of Psychology.
Achenbach, T. M., Howell, C. T., McConaughy, S. H., & Stranger, C. (1995). Six-year
predictors of problems in a national sample of children and youth: I. Cross-
informant syndromes. Journal of the American Academy of Child and Adoles-
cent Psychiatry, 34, 336e347.
Author. (2000). An innovative intervention for treatment of child and adolescent
aggression: an outcome study. Psychology in the Schools, 37, 157e167.
Author. (2003). Therapeutic factors and outcomes in group and individual therapy
of aggressive boys. Group Dynamics: Theory, Research, and Practice, 7, 225e237.
Author. (2006). The contribution of bibliotherapy to the counseling of aggressive
boys. Psychotherapy Research, 16, 645e651.
Author. (2010). Treating child and adolescent aggression through bibliotherapy. New
York: Springer.
Author, & Birani-Nasaraladin, D. (2006). Treating mothers of aggressive children: a
research study. International Journal of Group Psychotherapy, 56, 93e112.
Baker, R. J. (1987). The social work dictionary. Silver Spring, MA: NASW.
Bang, M. (1999). When Sophie gets angry. New York: Blue Sky Press.
Baron-Cohen, S. (2011). Zero degrees of empathy: A new theory of human cruelty (1st
ed.). London: Allen Lane.
Barr, J. (2011). The relationship between teacher's empathy and perceptions of
school culture. Educational Studies, 37, 365e369.
Batson, C. D. (2012). The empathy-altruism hypothesis: issues and implications. In
J. Decety (Ed.), Empathy: From bench to bedside (pp. 41e54). Cambridge: MIT
Press.
Bohart, A., & Stipek, D. (2001). Constructive and destructive behavior: Implications for
family, school, and society. Washington, DC: American Psychological Association.
Calvete, E., & Orue, I. (2012). The role of emotion regulation in the predictive as-
sociation between social information processing and aggressive behavior in
adolescents. International Journal of Behavioral Development, 36, 338e347.
Conners, C. K. (1990). Conners' rating scales manual, Conners' teacher rating scales,
Conners' parent rating scales: Instruments for use with children and adolescents.
Toronto, North Tonawanda, NY: Multi-Health Systems.
Connor, D. F. (2002). Aggression and antisocial behavior in children and adolescents.
New York: Guilford.
Conyne, R., Horne, A., & Raczynski, K. (2013). Prevention in psychology: An intro-
duction to the prevention practice kit. Thousand Oaks, CA: Sage.
Cornelius-White, J. (2007). Learner-centered teacher-student relationships are
effective: a meta-analysis. Review of Educational Research, 77, 113e143.
Craig, K., Bell, D., & Leschied, A. (2011). Pre-service teachers' knowledge and atti-
tudes regarding school-based bullying. Canadian Journal of Education, 34,
21e33.
Crick, N. R. (1996). The role of overt aggression, relational aggression, and pro social
behavior in the prediction of children's future social adjustment. Child Devel-
opment, 67, 2317e2327.
Decety, J., & Jackson, P. (2004). The functional architecture of human empathy.
Behavioural and Cognitive Neurosciences Reviews, 3, 71e100.
Dedousis-Wallace, A., & Shute, R. H. (2009). Indirect bullying: predictors of teacher
intervention, and outcome of a pilot educational presentation about impact on
34 Z. Shechtman, R. Tutian / Teaching and Teacher Education 58 (2016) 28e34
adolescent mental health. Australian Journal of Educational & Developmental
Psychology, 9, 2e17.
Dedousis-Wallace, A., Shute, R., Varlow, M., Murrihy, R., & Kidman, T. (2014). Pre-
dictors of teacher intervention in indirect bullying at school and outcome of a
professional development presentation for teachers. Educational Psychology: An
International Journal of Experimental Educational Psychology, 37, 862e975.
Deffenbacher, J. L., Oetting, E. R., & DiGuiseppe, R. A. (2002). Principles of empiri-
cally supported interventions applied to anger management. The Counseling
Psychologist, 30, 262e280.
Dishion, T. J., Dodge, K. A., & Lansford, J. E. (2008). Deviant by design: risks asso-
ciated with aggregating deviant peers into group prevention and treatment
programs. The Prevention Researcher, 15, 8e11.
Dishion, T. J., McCord, J., & Poulin, F. (1999). When interventions harm: peer groups
and problem behavior. American Psychologist, 54, 755e764.
Dodge, K. A. (2011). Social information processing models of aggressive behavior. In
M. Mikulincer, & P. R. Shaver (Eds.), Understanding and reducing aggression,
violence, and their consequences (pp. 165e186). Washington, DC: American
Psychological Association.
Dodge, K. A., Godwin, J., & Conduct Problems Prevention Research Group. (2013).
Social-information-processing patterns mediate the impact of preventive
intervention on adolescent antisocial behavior. Psychological Science, 24,
456e465.
Ein-Dor, T., Mikulincer, M., & Shaver, P. R. (2011). Attachment insecurities and the
processing of threat-related information: studying the schemas involved in
insecure people's coping strategies. Journal of Personality and Social Psychology,
101, 78e93.
Engelen, E. M., & Rottger, B. (2012). Current disciplinary and interdisciplinary de-
bates on empathy. Emotional Review, 4, 3e8.
Ergur, D. O. (2009). How can education professionals become emotionally intelli-
gent? Procedia Social and Behavioral Sciences, 1, 1023e1028.
Forgan, J. (2002). Using bibliotherapy to teach problem-solving. Intervention in the
School and Clinic, 38, 75e82.
Garbarino, J. (1999). Lost boys: Why our sons turn violent and how we can save them.
New York: Free Press.
Gladding, S. T. (2005). Counseling as an art (4th ed.). Alexandria, VA: American
Counseling Association.
Good, T. L., & Brophy, J. E. (2008). Looking in classrooms. New York: Pearson.
Greenberg, L. S. (2002). Emotion-focused therapy. Washington, DC: American Psy-
chological Association.
Gurian, M. (1997). The wonder of boys: What parents, mentors and educators can do to
shape boys into exceptional men. New York: Penguin.
Hattie, J. (2009). Visible learning: A synthesis of over 800 meta-analyses relating to
achievement. London: Routledge.
Hrdy, S. B. (2009). Mothers and others. Cambridge, MA: The Belknap Press of Har-
vard University Press.
Kazdin, A. E. (2007). Mediators and mechanisms of change in psychotherapy
research. Annual Review of Clinical Psychology, 3, 1e27.
Long, E., Angera, J., & Hakoyama, M. (2008). Transferable principles from a formative
evaluation of a couples' empathy program. Journal of Couple & Relationship
Therapy: Innovations in Clinical and Educational Interventions, 7, 88e112.
Loudin, J. L., Loukas, A., & Robinson, S. (2003). Relational aggression in college
students: examining the roles of social anxiety and empathy. Aggressive
Behavior, 29, 430e439.
Mikami, A. Y., Gregory, A., Allen, J. P., Pianta, R. C., & Lum, J. (2011). School Psychology
Review, 3, 367e385.
Moeller, T. G. (2001). Youth aggression and violence: A psychological approach.
Mahwah, NJ: Erlbaum.
Persons, J. B., & Burns, D. D. (1985). Mechanism of action of cognitive therapy:
relative contribution of technical and interpersonal intervention. Cognitive
Therapy & Research, 9, 539e551.
Pianta, R. C. (2006). Schools, schooling, and developmental psychopathology. In
D. Cicchetti, & D. J. Cohen (Eds.), Developmental psychopathology (pp. 494e529)
(Hoboken, NJ :Wiley).
Pianta, R. C., & Allen, J. P. (2008). Building capacity for positive youth development
in secondary school classrooms: changing teacher interactions with students. In
M. Shinn, & H. Yshikama (Eds.), Toward positive youth development: Transforming
schools and community programs (pp. 21e39). New York: Oxford Press.
Pollack, W. S. (2000). Real boys' voices. New York: Penguin.
Potter-Efron, R. T. (2005). Handbook of anger management. New York: Haworth
Press.
Raczynski, K., & Horne, A. (2014). Psychoeducational and counselling groups for
bullying. In J. DeLucia-Waack, C. Kalodner, & M. Riva (Eds.), Handbook of group
counseling and psychotheraphy (pp. 495e505). Thousand Oaks, CA: Sage.
Reese, R., Norsworthy, L., & Rowlands, S. (2009). Does a continuous feedback model
improve psychotherapy outcome? Psychotherapy: Theory, Research, and Practice,
46, 418e431.
ll, J., Koglin, U., & Petermann, F. (2012). Emotion regulation and childhood
Ro
aggression: longitudinal associations. Child Psychiatry & Human Development,
43, 909e923.
Silverstein, S. (1974). Where the sidewalk ends. New York: Harper & Row.
Snunit, M. (2010). The soul bird. London: Constable.
Swan, P., & Riley, P. (2012). Mentalization: a tool to measure teacher empathy in
primary school teachers. In Paper presented at the joint AARE APERA international
conference, Sydney.
Toth, S., Harris, L., Goodman, G. S., & Cicchetti, D. (2011). Inuence of violence and
aggression on children's psychological development: trauma, attachment, and
memory. In P. R. Shaver, & M. Mikulincer (Eds.), Understanding and reducing
aggression, violence, and their consequences (pp. 351e366). Washington, DC: APA
Books.
de Waal, F. B. M. (2008). Putting the altruism back into altruism: the evolution of
empathy. Annual Review of Psychology, 59, 279e300.
Walter, H. (2012). Social cognitive neuroscience of empathy: concepts, circuits, and
genes. Emotion Review, 4, 9e17.
Warren, C. A., & Lessner, S. (2014). Who has family business? exploring the role of
empathy in student-teacher interactions. Perspectives on Urban Education, 11,
122e132.
Zahavi, D., & Overgaard, S. (2012). Empathy without isomorphism: a phenomeno-
logical account. In J. Decety (Ed.), Empathy: From bench to bedside (pp. 3e20).
Cambridge, MA: MIT Press.
Zilber, N., Auerbach, J., & Lerner, Y. (1994). Israeli norms for the Achenbach child
behavior checklist: comparison of clinically-referred and non-referred children.
Israeli Journal of Psychiatry and Related Sciences, 31, 5e12.