,‘EWMQ

This is to certify that the
dissertation entitled

The Social Influences of School and Their Effect on the
Changes of Aggression Over Time: A Multilevel Model

presented by
Michael David Nicholson

has been accepted towards fulfillment
of the requirements for the

Doctoral degree in School Psychology

)L. %/{LA/M Av,
« Major Professor’s Signature
5/; [/0 5"

Date

MSU is an Affirmative Action/Equal Opportunity Institution

u” ————— -—-——

LIB TE EilIVERSITY
ICHIGAN STA
EXST LANSING, MICH 48824-1048

PLACE IN RETURN Box to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2/05" m/c'ifiE/o'a——teoue.irm-p.'1s‘

 

_, -—~.———

THE SOCIAL INFLUENCES OF SCHOOL AND THEIR EFFECT ON THE
CHANGES OF AGGRESSION OVER TIME: A MULTILEVEL MODEL

By

Michael David Nicholson

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Department of Counseling, Educational Psychology and Special Education

2005

ABSTRACT

THE SOCIAL INFLUENCES OF SCHOOL AND THEIR EFFECT ON THE
CHANGES OF AGGRESSION OVER TIME: A MULTILEVEL MODEL

By

Michael David Nicholson
Aggression, a complex phenomenon ranging from disruptiveness to physical and verbal
abuse that relies on a social label from those involved or those who witnessed the event,
resulted in nearly two million incidences of violence against students ages twelve to
eighteen in the year 2000 alone. Due to the seeming stability of aggression, it is
important to target these behaviors early. The school seems to be in a position to help
counter behavioral problems. Through the social interactions of the school in which
children participate five days a week for the better part of each year, the chance for
positive models of behavior are ever present. The present study examined relationships
with teachers and peers and the potential for those to influence changes in aggression
over time. Specific interest in teacher efficacy and its relationship to these changes are
explored. Results suggest that teacher efficacy is moderately related to changes in
aggression along with the previous aggression of the child and the overall climate of the
classroom. Examinations of current intervention and prevention programs and

implications on practice are discussed.

TABLE OF CONTENTS

LIST OF TABLES .................................................................................... v
CHAPTER 1

INTRODUCTION .................................................................................... 1
CHAPTER 2

LITERATURE REVIEW ........................................................................... 5

Aggression: Its Meaning and Origin ........................................................... 5

Definitions .................................................................................... 5

Symptoms and Risk Factors ................................................................ 7

The Stability of Aggression ............................................................... 12

Developmental Psychopathology ............................................................. 14

The Pathways of Aggression ............................................................. 14

Resiliency as an Agent of Change ....................................................... 19

The School as an Environment of Change .............................................. 22

Social Development ............................................................................ 24

Social Learning Theory .................................................................... 24

Social Information Processing Theory ................................................... 28

Social Development Theory .............................................................. 30

The Socializing Agents in the School ........................................................ 35

The Teacher Effect ......................................................................... 35

The Peer Effect ............................................................................. 38

Other Factors of School Failure and School Success ...................................... 41

iii

CHAPTER 3

METHODS .......................................................................................... 45
Participants ....................................................................................... 45
Measures ......................................................................................... 45

Aggression Scale ........................................................................... 45
Reading Grades ............................................................................. 46
Teacher Efficacy Scale .................................................................... 46
Teacher and Peer Support ................................................................. 47
Classroom Climate ......................................................................... 48
Data Analyses and Rationale .................................................................. 48

CHAPTER 4

RESULTS ............................................................................................ 50
Level 1 Variables: The Student Level ....................................................... 53
Level 2 Variables: The Classroom Level ................................................... 56

CHAPTER 5

DISCUSSION ....................................................................................... 58
Factors Within the Child Associated with Changes in Aggression ...................... 58
The Teacher’s Influence on Changes in Aggression ....................................... 60
The Classroom as a Whole .................................................................... 62
Implications and Future Directions ........................................................... 64
Limitations ....................................................................................... 68
Conclusions ...................................................................................... 69

REFERENCES ...................................................................................... 71

iv

LIST OF TABLES

Table 1: Means and Standard Deviations by Year .............................................. 51

Table 2: Estimates and Standard Errors by Time ............................................... 54

LIST OF FIGURES

Figure 1: Distributions of the outcome variable, Changes in Aggression ................... 52

vi

Chapter 1
Introduction

Nearly a decade ago, The National Education Goals Panel ambitiously set as a
goal the following: “by the year 2000, every school in America will be free of drugs and
violence and will offer a disciplined environment conducive to learning” (Batsche &
Knoff, 1994, p. 165). Now five years into the twenty-first century, schools still struggle
with the problem of violence and aggression, the United States having fallen well short of
its lofty goal. This is not to understate the progress that has been made. America has
seen a steady decline in the levels of violent acts within our schools since the creation of
that 1993 Goals Panel (Phillips & Greenfeld, 2002). However, school violence remains a
serious problem and threat to the safety of our nation’s children, a statement illuminated
by a brief examination of current data.

Among juveniles, serious violent crimes, which include rape, sexual assault,
robbery, and aggravated assault (Phillips & Greenfeld, 2002), are fairly rare occurrences,
with approximately 90 percent of schools reporting no such incidences during the 1996-
97 school year (Bear, Webster-Stratton, Furlong, & Rhee, 2000). In fact, since 1992,
there has been a 46% decrease in the number of violent crimes reported at school
(Phillips & Greenfeld, 2002). Still, according to the US. Departments of Education and
Justice in their 2002 Indicators of School Crime and Safety, Phillips & Greenfeld report
that this percentage translated into approximately 700,000 violent crime victims between
the ages of twelve and eighteen in the year 2000 alone. Of those, 128,000 were classified
as serious violent crimes and 47 resulted in deaths, 38 of which were homicides. The

homicide rate at school has remained relatively constant since 1992, its lowest point

1

being in 1997 when it was 21% above the average of the 1980’s (Snyder & Sickmund,
1999). Those committing these crimes were almost entirely male and between the ages
of fourteen and seventeen. In fact, between 35 and 45 percent of all violent and serious
violent crimes are committed by juveniles. Also strikingis the number of juveniles
participating in deviant acts. Twenty-eight percent of juveniles (and 37% of males) have
purposely destroyed property and 18 percent (23 for males) have committed assault
(Snyder & Sickmund, 1999).

The data above show that a large number of children and adolescents are
participating in violent acts. However, these numbers only tell part of the story. Violent
and serious violent crimes aside, other acts of aggression, such as bullying, teasing, and
rejecting peers are considerably more common among youth (Bear, Webster-Stratton,
Furlong, & Rhee, 2000). Indeed, over half of all schools reported that physical fights
escalated to the point of police involvement (Phillips & Greenfeld, 2002).

These data suggest that many school environments continue to be unsafe for the
students that attend them. In fact, in 2001, about six percent of students ages twelve to
eighteen carried a weapon to school, over twelve percent had been in a fight on school
property, nearly eight percent had been bullied (up nearly three percent from 1999) and
almost twenty percent reported that gangs were present at their school (Phillips &
ZGreenfeld, 2002). The result of this data manifests in the fact that six percent of these
students experience fear while at school. While these numbers may sound small, these
percentages translate into nearly two million incidences of violence against students ages
twelve to eighteen in the year 2000 alone (Phillips & Greenfeld, 2002).

Who are these juvenile offenders? Usually, there is no secret as to who they are.

2

These children have been getting into trouble their entire lives, and a quarter of them will
be officially referred to the juvenile justice system by the time they are fourteen years old
(Snyder & Sickmund, 1999). Over half of the violent and serious violent crimes will be
committed by these early-onset offenders. It is for this reason that we have seen an
increase of zero tolerance policies, violence prevention programs, security measures, and
onsite police officers in our nation’s schools (Phillips & Greenfeld, 2002). Although
these efforts have produced some positive change, the real threat and subsequent fear of
violence continues to torment millions of children nationwide.

This major threat to public health (Herrenkohl, Maguin, Hill, Hawkins, Abbott, &
Catalano, 2000) necessitates the need to understand aggression and to explore avenues
within the school that may circumvent the problem before more serious forms of violence
can ensue. For at least a couple reasons, this effort needs to start in the early years of
schooling, if not before. First, minor forms of aggression such as tardiness, breaking
rules, and fighting have been found to predict later, more serious acts of violence
(Kellam, Rebok, Ialongo, & Mayer, 1994). Rarely does an adolescent one day decide to
use violence. Instead, it seems more likely that acts of aggression result through both the
repeated occurrences of such acts and an inability to understand and utilize more
prosocial behavior. Second, it is important to target deviant behavior early because
children who demonstrate such behavior are nearly three times more likely to commit
serious violent crimes in the future during what is most likely a long, chronic career as a
criminal (Loeber & Farrington, 2000). Discovering avenues of success for these children
and tomorrow’s children is becoming increasingly important. With an expected increase

of over 20% in the number of people under the age of eighteen in the next thirty years

3

(Snyder & Sickmund, 1999), the problem of youth aggression can only increase.

The goal set by the National Education Goals Panel (Batsche & Knoff, 1994) is
still the desire of this nation. While unable to reach it by the year 2000 as planned, it is
still important to continue to strive for that plateau. Through an understanding of the
mechanisms of aggression and the ways those mechanisms interact with the dynamic
environment of school, we can begin to formulate solutions and continue to make strides
toward creating schools that are both physically and psychologically safe places to be.
Specifically, this study will examine the effects of the major social influences within the

school (teachers, peers) on the changes of childhood aggression over time.

Chapter 2
Literature Review
The story of how the mechanisms of aggression can interact with the dynamic
environment of school in either positive (prosocial) or negative (antisocial) ways can only
begin with an exploration of aggression itself. An understanding of what does and does
not constitute aggression, along with an examination of risk and protective factors, lays
the foundation for understanding how antisocial trajectories may be altered toward more
positive pathways. Thus the discussion would not be complete without looking at how
aggression develops over time, and in this particular context, how the school may aid in
such a development. The central premise of this chapter is that the school may be able to
provide an environment that could stimulate more prosocial behaviors in children.
Therefore, the remainder of this chapter will explore social theories as well as specific
influences within the child and school environment that will support this postulate.
Aggression: Its Meaning and Origin
Definitions
When asked to describe an aggressive child, many may conjure an image of a
foul—mouthed, child-hitting, dog-kicking, fire-starting boy who has no desire or perhaps
even ability to do the right thing. Popular movies have exploited these ideas for
terrifying effects in films such as The Bad Seed, The Omen, and The Good Son, the last of
which of course implying that there is a bad son. These depictions of severe conduct
disorders, from which between one and ten percent of children suffer (Conduct Problems
Prevention Research Group,1992; Hinshaw & Anderson, 1996) certainly represent

aggressive acts on the wrong end of a behavioral continuum. However, these depictions

5

provide a narrow conceptualization of aggression, a term that has been defined and
classified in many different ways.

Aggression has been defined as “any sequence of behavior, the goal response to
which is the injury of the person toward whom it is directed” (Bandura, 1973, pg. 2;
Parke & Slaby, 1983, pg, 549). This particular definition seems to depict an
oversimplified explanation of the phenomenon of aggression. The implication therein is
that there is only one purpose to an aggressive act, that being to cause injury. There
would seemingly be other goals that are possible for a child to resort to aggression. For
example, a child who wants a toy with which another child is playing may turn toward
violent solutions to remedy the situation, not necessarily to hurt the other child, but rather
to obtain the object.

Alternatively, aggression has been defined as “behavior that results in injury of
another individual” (Parke & Slaby, 1983, pg. 549). While this definition addresses some
of the shortcomings of the previous one, it does not seem to distinguish between
purposeful acts of aggression and those “injuries” that may be caused by a routine trip to
the dentist. While root canals may at times seem to be performed in malicious manners,
few would actually describe the dentist’s behavior as being aggressive.

Bandura (1973) suggests that for an act to be aggressive, it not only must result in
injury to another person or thing (“destruction of property”), but it also must be judged to
be aggressive by onlookers. Judges of aggression look at characteristics of the behavior,
the intensity of the responses, expressions of pain and injury, and the intentions attributed
to the performer. Characteristics of the labeler and of the aggressor further alter the

labels of the judges of aggression. What this conceptualization offers is the fact that an

6

act of aggression is a social act. It involves at least two people, each a labeler and an
actor, and in the schools often involves a host of other actors and labelers as well. Thus
in the social context of schools, this definition of aggression is appealing. Aggression,
then, is a “complex phenomenon ranging from disruptiveness to physical and verbal
abuse” (Gorsk & Pilotto, 1993, p. 36) that relies on a social label from those involved or
those who witnessed the event. In the specific example of this study, the social label
comes from the teacher by way of rating forms and it is these ratings or perceptions of
aggression levels that are being substituted for aggression itself. Research suggests that
teachers do supply reliable and valid measures of externalizing problems (Merydith,
2001).

Hinshaw and Anderson (1996) outline a number of forms which aggression can
take. Aggression can be verbal when a student threatens another or calls him/her a name
or it can be physical when the ensuing fight occurs. Aggression can be instrumental
(goal-directed) when a toddler uses a violent tantrum to play his favorite game or hostile
when that toddler’s favorite game is beating up his sister. Aggression can be direct as in
any boxing match on the school playground or indirect when rumors are started to hurt
another (Atlas & Pepler, 1998; Hinshaw & Anderson, 1996). Aggression can be
proactive when a student picks a fight or reactive when the victim fights back (Hinshaw
& Anderson, 1996). Aggression can be covert when someone steals when only turned
heads are seen or overt when the act is on display for others to see. Finally, aggression
can be destructive or nondestructive. For the current paper, only direct, overt aggressive
acts are considered because of the difficulty of measuring more covert behaviors.

Symptoms and Risk Factors

A risk factor is anything that increases a person’s chance to “succumb to
adversity” (Doll, & Lyon, 1998, p. 349). It seems unlikely that there is one cause or
factor that determines a child’s likelihood to become aggressive. Instead, there are
probably multiple factors that make up the aggressive child (Loeber, 1991), and the more
risk factors that a child possesses, the more likely there will be negative outcomes. In a
related area, Loeber and F arrington (2000) found that children with three more risk
factors than protective factors were eight times more likely to become seriously
delinquent than those children with a better balance of risk and protection. An
understanding then of the risk factors of aggression is important to comprehending the
construct.

There are three broad areas of risk factors that have been widely studied. First,
there are a number of factors within the child that can increase his/her chances to be
aggressive in the fiiture. The child’s temperament may have a small effect on later
aggression (Hinshaw, 2002; Loeber, & Farrington, 2000; Webster-Stratton, 1993). High
activity levels, poor adaptability, and general crankiness have some relation to
aggression, but unfavorable temperament does not in and of itself lead to
psychopathology. Secondly, gender is related to aggression in that males are
significantly more likely to partake in physically aggressive actions (Bear, Webster-
Stratton, Furlong, & Rhee, 2000; D011, & Lyon, 1998; Herrenkohl, Guo et al., 2001,
Herrenkohl et al., 2000). Children with cognitive and academic deficits such as reading
disabilities, language delays, and attention problems have all been related to more
negative outcomes (Bear et al., 2000; Conduct Problems Prevention Research Group,

1999; , Herrenkohl, Guo et al., 2001; Herrenkohl et al., 2000; Hinshaw & Anderson,

8

1996; Loeber, & F arrington, 2000; Loeber, F arrington, Stouthamer-Loeber, Moffitt, &
Caspi, 1998; Loeber, Green, Lahey, Christ, & Frick, 1992; Webster-Stratton, 1993).
Third, children with deficits in social skills such as distortions of social cues, poor social
problem solving skills, and a tendency to define problems in hostile ways leads to more
aggressive outcomes later in life (Bear et al., 2000; Webster-Stratton, 1993). Fourth,
hyperactivity and impulsivity puts a child at risk of becoming aggressive (Bear et al.,
2000; Conduct Problems Prevention Research Group, 1999; Herrenkohl, Guo et al.,
2001; Herrenkohl et al., 2000; Loeber, & Farrington, 2000; Loeber et al., 1998). F ifih,
involvement with antisocial peers as well as early antisocial behaviors are good
predictors of future antisocial acts (Bear et al., 2000; Herrenkohl, Guo et al., 2001;
Herrenkohl et al., 2000).

A sixth risk factor within the child may be the age of onset (Herrenkohl, Huang et
al., 2001; Hinshaw & Anderson, 1996; Loeber, & Farrington, 2000). There appears to be
a difference between early-onset and late-onset offenders, with the former having a
longer, even chronic difficulty with aggression. While late-onset offenders typically have
a prosocial orientation but succumb to peer pressures, early-onset offenders have a more
negative orientation right from the beginning. These orientations become crystallized
with subsequent exposures to harmful environmental stimuli (Herrenkohl, Huang et al.,
2001).

The second broad area of risk lies within the home or family. Parent skills
deficits such as being more violent and critical in their discipline (Batsche, & Knoff,
1994; Bear et al., 2000; Herrenkohl et al., 2000; Loeber, & Farrington, 2000; Loeber et

al., 1998; Webster—Stratton, 1993; Sines, 1987); being more permissive, erratic, and

9

inconsistent (Herrenkohl, Guo et al., 2001; Herrenkohl et al., 2000; Hinshaw &
Anderson, 1996; Parke & Slaby, 1983; Webster-Stratton, 1993); failing to monitor
children’s behaviors (Webster-Stratton, 1993); providing poor supervision (Hinshaw &
Anderson, 1996; Kolvin, Miller, Fleeting, & Kolvin, 1988; Loeber, & Farrington, 2000;
Loeber et al., 1998); having low involvement in children’s activities (Hinshaw &
Anderson, 1996); having poor problem-solving skills (Batsche, & Knoff, 1994); failing to
monitor children’s social interactions (Herrenkohl et al., 2000); and rewarding
inappropriate behaviors while ignoring or punishing prosocial behaviors are related to
more negative outcomes for the children (Webster-Stratton, 1993). Parental
psychological factors such as depression in the mother (Bear et al., 2000; Loeber, &

F arrington, 2000; Loeber et al., 1998, Webster-Stratton, 1993), alcoholism in the father
(Webster-Stratton, 1993), and substance abuse or antisocial behavior in either parent
influence the aggression of children (Hinshaw & Anderson, 1996; Loeber, & Farrington,
2000; Sines, 1987). Thirdly, parental attitudes in that they model violent or antisocial
values lead to more aggressive acts in the children (Bear et al., 2000; Herrenkohl et al.,
2000; Webster-Stratton, 1993). Low family income and crowded living conditions have
also been shown to lead to aggression in children, as has parental conflict, more so than
divorce (Bear et al., 2000; Conduct Problems Prevention Research Group, 1999; Doll, &
Lyon, 1998; Gorsk & Pilotto, 1993; Herrenkohl, Guo etal., 2001; Kolvin et al., 1988;
Loeber, & Farrington, 2000; Sines, 1987; Webster-Stratton, 1993). Furthermore, highly
coercive families tend to produce more aggressive children (Doll, & Lyon, 1998;
Patterson, Capaldi, & Bank, 1991).

There is some evidence that there is a genetic component involved in the

10

development of aggression. Generally speaking, it is difficult to ascertain the extent to
which aggression is inherited because children with aggressive parents are usually in a
negative environment (Brennan, Mednick, & Kandel, 1991). Better estimates may come
from studies of adoption and identical twins raised apart. That data suggest that about a
quarter of sons whose biological parents are chronic offenders later become offenders
themselves (Brennan, Mednick, & Kandel, 1991). Heritability estimates of
approximately .40 have been obtained through twin studies (Brennan, Mednick, &
Kandel, I991; Jacobson, Prescott, & Kendler, 2002). While this provides evidence of a
genetic component to aggression, it also reveals that the environment in which the child
lives and grows is an important factor that could either increase or decrease the risk of
future aggression (Aguilar, Sroufe, Egeland, & Carlson, 2000; Hinshaw & Anderson,
1996; Jacobson, Prescott, & Kendler, 2002).

The third broad area of risk lies within the school. Those factors that have been
shown to lead to later aggression in children are an involvement with antisocial peers
(Herrenkohl, Guo et al., 2001; Herrenkohl et al., 2000; Loeber, & Farrington, 2000; Parke
& Slaby, 1983), weak bonding to school (Loeber, & Farrington, 2000), peer rejection
(Conduct Problems Prevention Research Group, 1999; Webster-Stratton, 1993), poor or
negative teacher-child interactions (Bear et al., 2000), too strict or too permissive
classroom management style (Gorsk & Pilotto, 1993), a low frequency of praise
(Webster-Stratton, 1993), a low emphasis on academic work (Webster-Stratton, 1993),
noisy and disruptive classrooms (Conduct Problems Prevention Research Group, 1999),
and a high teacher-student ratio (Webster-Stratton, 1993).

Other areas of risk such as the influences of the neighborhood and the community

11

(Herrenkohl et al., 2000; Loeber et al., 1998) are seen to be important but are not the
focus of the present study. Of most concern to this paper are the child and school
domains, as well as the interaction between the two.

The Stability of Aggression

Aggression is generally seen to be a fairly stable construct throughout
development, with stability correlations as high as .81 (Hinshaw & Anderson, 1996;
Parke & Slaby, 1983). It would appear, however, that the stability of later-onset
aggression is significantly lower (Loeber et al., 1998). In fact, in addition to age of onset,
other factors that increase stability of aggression include the frequency of aggressive
behaviors (the more aggressive acts early in life predict greater stability over time), acts
occurring in more than one setting, a wider variety of aggressive acts displayed, and the
presence of hyperactivity and attention problems (Loeber, 1991). Loeber (1991) also
contends that these behaviors crystallize with age, making them more difficult to treat
later in life.

Despite this apparent stability, not all children who display aggressive tendencies
early in life continue to display them later. Two examples from clinical psychology can
exemplify this point. First, there seems to be a link between the development of
oppositional defiant disorder (ODD) and the later development of conduct disorder (CD;
Hinshaw & Anderson, 1996). Over 90% of children diagnosed with CD have previously
met the criterion for ODD, making the developmental progression from the earlier
disorder to the later one seem readily apparent. However, only a portion (about a quarter)
of the children with ODD later meets the criterion for CD. Half of these children

maintain an ODD diagnosis but do not meet criterion for CD, and the remaining quarter

12

no longer meet the criterion for ODD later in life.

A similar pattern exists with the association between conduct disorder (CD) and
antisocial personality disorder (APD; Hinshaw & Anderson, 1996). Nearly all of those
suffering from APD met the criterion for CD in the past, but only about a quarter of those
with CD will later develop APD. This is not to imply that most adolescents with CD will
have positive outcomes, since CD is associated with other negative outcomes besides
APD, such as substance abuse and internalizing disorders. Instead, this summary is only
to show that not all children with these difficulties maintain their aggression later in life,
or at least not to the extent from which they previously suffered. In fact, based upon the
relation between ODD and CD and then CD and APD, only about ten percent of children
with ODD will later meet the criterion of APD.

In short, aggression appears to be relatively stable when it begins early in life.
The early starter model often begins with the onset of ODD and continues throughout
life, eventually leading to the most serious of symptoms, interpersonal violence and
property violations, by the end of adolescence (Webster—Stratton, 1993). This would lead
one to believe then that early intervention is paramount to the successful futures of these
children. Despite the enduring quality of aggression and associated problems, “it is also
changeable, however, especially in the first 12 or so years of life. Many of the initial
causes of the disorder operate during childhood when children’s behavior tends to be
most malleable” (Loeber, 1991, pg. 396). Therefore the context of the school and
specifically the elementary school can be an important environment to attempt to
circumvent some of the negative pathways of aggressive children before the problem

becomes unmanageable. Through an understanding of the pathways of aggression and

13

the mechanisms of change, one can begin to see how the school may be able to lead
children toward more positive futures.
Developmental Psychopathology

The Pathways of Aggression

The development of aggression usually begins with more minor, less severe forms
such as arguing and begins to take on more serious forms such as physical aggression and
eventually violence (Loeber et al., 1998). In fact, less serious forms of aggression seem
to be a necessary step on the way to more serious modes of aggression (Loeber, 1991), at
least for boys (Kazdin & Kagan, 1994), making early intervention extremely important.

How do children begin down a path of violence? It seems unreasonable to
assume that there is only one way for a child to become aggressive. In other words, the
same risk factor is probably not going to lead to the same outcome for every child, due to
the unique characteristics of the child and that child’s environment. Therefore, it seems
more likely that the destination of aggression has multiple pathways (Kazdin & Kagan,
1994). Loeber et al. (1998) suggests that there are three pathways to violent behavior: the
overt, covert, and authority conflict pathways. The overt pathway sees children, mainly
boys (Hinshaw & Anderson, 1996), begin by bullying and annoying others, which, if left
unchecked, leads to physical fighting and eventually to violent outcomes such as rape or
attack. The covert pathway, which sees a higher proportion of girls (Hinshaw &
Anderson, 1996), begins with shoplifting, frequent lying, or other minor forms of covert
behavior and leads to more serious forms such as property damage and fraud. Finally the
authority conflict pathway begins with stubborn behavior at the mild end and defiance

and truancy on the more serious end. Some children progress down two or even all three

14

of the pathways, suggesting that over time, problems become varied and more complex.
For the purpose of this study, only more overt acts are studied.

The idea of different pathways is fundamental to the study of developmental
psychopathology. Not all aggressive adults have had similar pasts or similar risk factors.
That children can come from vastly different backgrounds and be exposed to very
different risk factors but have the same or similar outcomes is the concept of equifinality,
a fundamental concept in this theoretical orientation. Conversely, and an equally
important concept, children whose initial conditions are similar may have very different
outcomes, which is known as multifinality (Cicchetti & Richters, 1993; Hinshaw, 2002).
These two terms together pinpoint the complexity of psychopathological origin. Each
child is unique in that he/she has different cognitions, perceptions, and environments, and
all of these factors have positive and negative aspects, working congruently and at times
oppositionally, leading the child down different paths. The balance of the risk factors
against protective and other environmental influences comprises, in simplistic terms, the
raw materials in the construction of the different possible roads a child may follow. The
model of developmental psychopathology gives increased attention (in comparison to a
more medical model) to social and psychological factors in its attempt to reframe the
notion of how disorders arise (Sameroff, 2000).

The hundreds of competing influences in a child’s life make discovering the
origin for psychopathology difficult. To make it more complex however, as children
grow older, their cognitive and social capacities become more advanced and so they
interpret different social cues differently at different ages (Cicchetti & Richters, 1993).

Thus it is possible that the same factors that were influential at one age may be less

15

significant at another. Only with a side-by-side study of normal and abnormal
development can one begin to draw conclusions about the important factors in
development in general and in the development of psychopathology specifically. This
reciprocal relationship—knowledge of normal development can inform the study of
psychopathology and vice versa—has as its focus the full understanding of development
by studying the extremes because they diversify the range of possible outcomes.
Deviations may signify the beginning of a disorder, and by following the progression of
these disorders and their personal and environmental factors, it will broaden our
understanding of developmental processes (Sameroff, 2000).

Indeed, while pathways involving significant change are rare, they are potentially
more informative about the effects of contextual factors on development (Compas,
Hinden, & Gerhardt, 1995). From a very general standpoint, there are only five
developmental pathways down which a child/adolescent may travel. A child with many
protective factors and few risk factors throughout development are likely to be well
adjusted later in life (stable adaptive functioning). Conversely, the child who consistently
is faced with many risk factors without the balancing protective factors are most likely to
be poorly adjusted later in life (stable maladaptive functioning). The other three
pathways exist between those two extremes. For example, a child could be on a
maladaptive path and due to positive self and environmental impacts could change course
and head toward a positive future (adolescent turnaround or recovery). Conversely, a
child could be heading down an adaptive path, run into trouble, and change course toward
less favorable outcomes (adolescent decline). Finally, a child could temporarily alter the

current course, only to return to the original destination (temporary deviation or

16

maladaptation).

More specifically, considerable time has been devoted to the development of
conduct disorder (Conduct Problems Prevention Research Group,1992). Signs can be
seen as early as preschool with irritability, discipline problems, and impulsivity. This has
led to modeling different progressions of the disorder in an attempt to not only
understand how conduct disorders develop but also the different ways that children in
general can develop.

One such model of aggression is the early starter versus late starter (Aguilar,
Sroufe, Egeland, & Carlson, 2000; Compas, Hinden, & Gerhardt, 1995; Miller, Brehm, &
Whitehouse, 1998; Webster-Stratton, 1993). Early onset aggressive behavior (stable
maladaptive functioning) is associated with much worse outcomes than later onset
aggression. Adolescent onset (temporary deviation or maladaptation) is often limited to
the adolescent years. Early starters begin with oppositional qualities as early as preschool
and advance through more serious aggressive and nonaggressive delinquent behaviors.
Early onset may also have the fastest progression from mild to severe forms of aggression
(Loeber et al., 1992). Not surprisingly then, the prognosis for late starters is much more
favorable than for those adolescents with a chronic history of aggressive behavior.

This discussion is not meant to imply that all children on the same pathway reach
identical destinations. Not everyone makes it to Rome; there are many stops along the
way. To return to Loeber’s (1998) three pathways as an example, only a minority of boys
advances to the most serious levels within a pathway. However, a good portion of these
children (those who do not experience an adolescent turnaround or recovery) will be

faced with a variety of negative consequences such as a stability of negative behaviors

l7

due to repeated practice during formative periods of life; few opportunities to learn
prosocial skills due to years of delinquency; poor relationships with peers, relatives, and
employers; low interest in educational matters leading to classroom disruptions,
remediation, and truancy; fatherhood, despite an inability or unwillingness to fulfill the
father role; a higher risk of depression and suicide; a higher likelihood of being victims of
crime; and a higher likelihood of using illegal substances early in life (Loeber, &

F arrington, 2000).

The question becomes, what starts a child on a maladaptive pathway? Patterson,
DeBarysge, and Ramsey (1989) might contend that parents are a big factor in
determining the starting pathway for their children. Parents who demonstrate poor
discipline and monitoring may produce children who exhibit some conduct problems.
The conduct problems in turn lead to later peer rejection and academic failure, leading
these children toward deviant peer groups (the only people who seem to accept them)
which almost inevitably leads to delinquency. While the impact of the home and family
are not denied, it is not the focus of the current study.

Others cannot ignore the personal characteristics the child may bring, such as
hyperactivity, impulsivity, poor social skills, poor academic skills, aggression, and
oppositional behavior (Loeber, 1990, Mesman, Bongers, & Koot, 2001). Other child
characteristics that may influence the development of aggression are the thought
processes of the child. Children who are hypervigilant to hostile cues or see positive
outcomes to aggression, making aggressive choices primary in their decision-making are
more likely to have conduct problems (Dodge, 1993).

All of these characteristics of the child and the environment represent, as

18

mentioned earlier, risk factors for future deviant behavior. Naturally, fewer problems or
risk factors lead to better outcomes (Lahey, Loeber, Burke, & Rathouz, 2002). Of special
interest, however, to developmental psychopathology are those children who diverge onto
deviant pathways but then recover back to more adaptive pathways (adolescent
turnaround or recovery), as well as those children who never succumb to the stressors
that lead to negative outcomes in others (stable adaptive functioning). These resilient
children may hold the key to helping alter maladaptive behaviors in deviant children
(Cicchetti & Richters, 1993).
Resiliency as an Agent of Change

Although usually chronic stress leads to unhappy personal and societal outcomes,
some individuals adapt and manifest competence (Cowen & Work, 1988). As can be
inferred from the previous discussion, resiliency acts as a counterweight to risk, lessening
the likelihood of negative outcomes, or, stated another way, resilience is successfiil
functioning in the context of high risk (Fraser, Richman, & Galinsky, 1999). A child’s
resiliency is thought to be composed of the protective factors (“influences that modify,
ameliorate, or alter a person’s response to some environmental hazard that predisposes to
a maladaptive outcome,” Rutter, 1985, pg. 600) in his/her life. These factors appear to be
most influential under conditions of high risk (Compas, Hinden, & Gerhardt, 1995),
probably because their impact is tough to assess under conditions of low risk—is it the
protective factors, the low risk, or some combination that attributes to positive gains?
These factors lead to positive outcomes in high-risk situations at least in part because
they decrease the likelihood of engaging in problem behavior (Jessor, Van Den Bos,

Vanderryn, Costa, & Turbin, 1995). This decrease occurs through incompatible attitudes

l9

and activities that are manifested within the individual, that individual’s family, or
institutions/organizations (such as school) to which that individual belongs (Fraser,
Richman, & Galinsky, 1999).

As with risk factors, there are three areas (discussed here) where protective
factors may lie, within the individual, the family, and the school, with the emphasis of
this study on the individual and the school. While protective sources may be found
elsewhere, such as within the community, those factors are seen to be beyond the scope
of the present paper.

There are several individual qualities that are thought to protect the child from the
harmful effects of risk. These children tend to feel a greater sense of autonomy (Cowen
& Work, 1988; Murphy, 1987), are more empathic (Cowen & Work, 1988), have good
self-confidence and esteem (Doll, & Lyon, 1998), have a capacity to satisfy their own
needs (Murphy, 1987), have a resilient belief system (Doll, & Lyon, 1998), and are filled
with curiosity and problem-solving skills (Cowen & Work, 1988). These children also
tend to have strong school skills such as a task-orientation (Cowen & Work, 1988), good
intellectual ability (Doll, & Lyon, 1998), strong reading skills (Vance, Bowen,
Fernandez, & Thompson, 2002), high expectations (Doll, & Lyon, 1998), and language
competence (Doll, & Lyon, 1998). Finally, their strong interpersonal skills (Vance et al.,
2002), easygoing disposition, positive social orientation, and a high rate of engagement in
productive activities (Doll, & Lyon, 1998) allow these children to make positive friends,
work well in school, and stay out of trouble.

Protective factors within the family include effective parenting characterized by

warmth, structure, and high expectations (Doll, & Lyon, 1998; Cowen & Work, 1988),

20

parental competency (Cowen & Work, 1988; Vance et al., 2002), open communication
and exchange of feelings (Cowen & Work, 1988), emotional support (Cowen & Work,
1988; Vance et al., 2002), and parental agreement on values and moral issues (Cowen &
Work, 1988). While it is important for the child to have a close and affectionate
relationship with at least one parent, also important is that the child has access to warm
relationships and guidance from extended family members (Doll, & Lyon, 1998).
Because of this, interventions have targeted parent management training and functional
family therapy to try and help the child in the family environment, to build protective
factors in the home (Loeber, & F arrington, 2000), to help alter negative pathways.

Not surprisingly, protective factors in the school are normally conceived as
coming from peers and teachers. Positive relationships with peers (Cowen & Work,
1988) and teachers (Doll, & Lyon, 1998; O’Donnell, Hawkins, & Abbott, 1995) are
important in developing school bonding and achievement within the student, both of
which are protective factors as well and have been able to discriminate, along with family
bonding and norms against substance use, between those who did or did not avoid
involvement in serious delinquent behavior (O’Donnell, Hawkins, & Abbott, 1995).
Naturally, the relationship with the classroom teacher can either be a source of support or
of stress. Close teacher-student relationships, characterized by, much like important
parental relationships, warmth and open communication may increase a student’s ability
to engage and participate in classroom activities (Ladd & Burgess, 2001), fostering
bonding to and enjoyment of school. Conversely, conflictual relationships with a teacher
may have an Opposite effect, a desire to escape what is perceived as a negative

environment. Likewise, acceptance by positive peers can help decrease attention

21

problems and misconduct and help foster cooperative participation and school bonding
(Ladd & Burgess, 2001). All of this takes place within a building, a school, and thus it
makes sense that the quality of the school would have an impact on the child (Doll, &
Lyon, 1998). Higher quality schools in terms of teaching, resources, safety, etc. can
serve as protection from a variety of risks within a child’s life by fostering that bond
within the school, and between the school and the home, as well as increasing academic
competence (Miller, Brehm, & Whitehouse, 1998). Therefore, before leaving
developmental psychopathology, a closer look at how a school may help a child to adjust,
to be successful is warranted.
The School as an Environment of Change

In any system, there are characteristics of the individual, such as gender,
personality, and coping mechanisms and characteristics of the environment such as
people and rules that constantly interact with each other (Ladd, 1996). A child entering a
school system brings with him/her certain experiences and feelings that influence the
reaction to the new environment, either positively or negatively. Children with poor
relationships with previous teachers may treat new teachers with similarly negative
expectations. Children rejected by peers may wish to avoid school altogether.
Furthermore, children who are concerned about events transpiring at home may behave
and adjust poorly to school. All of these characteristics and many others affect each
individual child at every moment of the school day.

Any reasonable observer at this point can realize the difficulty of meeting the
needs of every individual child. This is the obstacle that faces the school, however, and

there are many factors involved in successful adjustment to school. First and most

22

evident is the fit between the person and the environment (Eccles et al., 1993). Children
react to different teachers differently. While one child may respond well to a strict, hard-
nosed teacher, another may drown in such an environment. Likewise, while one student
may respond well to the coddling of a softer teacher, others may quickly misbehave in
such an environment. Other important factors may be they styles of teaching versus the
styles of learning or the arrangement of the desks (Rosenfield, Lambert, & Black, 1985).
In short, many factors influence the adjustment of children to the school environment.

Perry and Weinstein (1998), for example, conceptualize these factors as falling
into three broad domains, academic, social, and behavioral fimctioning. Under each
domain there are indicators of successful adjustment. Academic success leads to
achievement in school, the creation of school values, competence-related beliefs, and
academic goals. Successful social functioning leads to a sense of belonging, acceptance
by peers and with that a quality of friendships. Finally, successful behavioral fimctioning
manifests itself in the following of rules, the delaying of gratification, the controlling of
impulsivity, and the directing of attention to the appropriate stimuli.

These behaviors of academic adjustment are brought forth from the child when
the different pieces of the school community act as aprotective factor for that child. The
school therefore can be a dynamic environment for change because of the many resilient
factors that can be (but unfortunately not always are) present for children. The school
can address many of the deficits that children of risk face. It can provide positive role
models. The school can provide opportunities to socialize positively with both adults and
peers. It can lead to the strengthening of cognitive and social functioning. Finally, the

school can help alleviate some of the stressors at home by collaborating as well as

23

providing the escape that some children need to succeed. It is when schools fail to create
positive communities that school violence erupts (Baker, 1998). The idea of community
is important because it implies a dynamic social environment, which is one way to
describe a school. Therefore, to help understand this social place and the power of the
interactions therein, an exploration of social theories is predicated.
Social Development

Because people do not develop in a vacuum, when one thinks about the
developing individual, it is necessary to consider the social influences that help shape that
person. For children, the majority of what they learn is done through some social agent,
meaning that learning takes place in the context of other people and that these new ideas
are taught to that child, either directly or indirectly. The aggressive child must learn at
some point that aggression is a viable solution to a problem. How does the child learn
this idea? Whether it is through an adult or child model (it works for Rambo, why not
me) or by trial and error (hitting Tommy makes him give up the toy), aggression against
another person necessarily, by definition, must occur in a social context. The assumption
in this discussion, of course, is that people are not born with a selection of aggressive
behaviors—the art of karate is not something a baby comes into this world knowing—but
rather they are learned at some point in some way (Bandura, 1973). How these
behaviors develop, what sets the behaviors into motion, and how these aggressive options
are maintained within the response set as viable solutions is what a theory of aggression,
according to Bandura ( 1973), must explain.
Social Learning Theory

Recognizing a de-emphasis on social factors within Behaviorism, Bandura sought

24

to expand current views to recognize the impact of other people on the behavior of the
individual. Like Behaviorism, the primary mechanism in learning social behavior is
operant conditioning, the shaping of behavior by stimuli that follow or are the
consequence of that behavior (Akers, Krohn, Lanza-Kaduce, & Radosevich, 1979). Also
consistent with the previous theory, behavior is strengthened through reward and
avoidance of punishment or weakened by aversive stimuli or loss of reward. What is
different and what will be further examined below is that behavior can be learned not
only through direct conditioning, which is largely governed by the rewarding and
punishing consequences that follow any given action (Bandura, 1973), but also through
imitation (engaging in behavior after observation of similar behaviors, Akers, 1977, pg
48) and through the observation of the consequences of the behaviors of others. In other
words, through the interactions with and observations of significant groups and people in
life, behavior is evaluated as either good or bad, reinforcing or punishing (Akers, Krohn,
Lanza-Kaduce, & Radosevich, 1979). Those groups that seem to control an individual’s
sources of reinforcement or punishment can, in theory, produce significant behavioral
effects while exposing that individual to behavioral models, either positive or negative.
Children who repeatedly turn toward aggression must therefore have first learned
it somewhere and somehow and must second find aggression to be reinforcing (Bandura,
1978). Why and how is aggression reinforced? First, aggression is often an effective
way of acquiring tangible rewards. Why bother with the social construct of, “Please may
I play with that toy you have there,” which, when dealing with the younger crowd has
limited effectiveness at best outside adult supervision, when forcing the toy away is so

easy? Second, because you can learn almost anything by watching others, observed,

25

reinforced aggression is abundant in our culture (Bandura, 1978). Whether it is in our
own home, at school, on television, or in video games, aggression is often seen as not
only being a solution but the solution to problems, and the immediate result of the
aggressive act is nearly always positive. Seeing this positive result, this success in others
can function as a motivator by arousing in observers expectations that they can gain
similar rewards for analogous performances. A third source of reinforcement for
aggression can be found in aversive experiences which lead to emotional arousal and
anticipated consequences, which may lead to a variety of outcomes, one of which is
aggression (Bandura, 1978), depending on the types of responses the person has learned
for coping with stress and their relative effectiveness (Bandura, 1973). The moment
aggression resolves or is perceived to resolve aversive experiences, it becomes a very
viable option next time.

What influences aggressive behavior? According to Bandura (1973) aggression is
under stimulus, reinforcement, and cognitive control. Stimulus control suggests that in
order to function effectively a person must be able to anticipate the probable
consequences of different events and courses of action and regulate his behavior
accordingly. In aggressive children, stimuli are more likely to produce emotional and
aggressive responses. These emotional responses are frequently acquired on the basis of
vicarious rather than direct experiences, meaning that emotional or aggressive responses
exhibited by others toward certain people tend to arouse in observers strong emotional
reactions that can become conditioned to the same target, setting up a bully-victim
relationship with the victim being the stimulus for continued acts of aggression by the

bully.

26

Reinforcement control suggests that responses that cause unrewarding or
punishing effects tend to be discarded, whereas those that cause rewarding outcomes are
retained and strengthened (Bandura, 1973). As mentioned earlier, modeling and
vicarious reinforcement and punishment are fundamental in learning new behaviors and
changing old ones. Human behavior is therefore largely socially transmitted, either
deliberately or inadvertently, through the examples of the various models in our lives.
Thus, modeling influences are key to the spread of aggression because they are an
indispensable aspect of learning. Even in instances in which it is possible to establish
new skills through other means, the process of acquisition can be considerably shortened
by providing appropriate models, and therefore much human behavior is developed
through modeling (Bandura, 1977). From observing others (primarily families early in
life and increasingly peers later, Loeber, 1991), one forms a conception of how new
behavior patterns are formed. When thinking of reinforcement control, then, vicarious
reinforcement can help explain the rapid acquisition of behavior (Conger, 1976).
Observed reinforcement influences behaviors in much the same way as outcomes that are
directly experienced. Observed rewards increase and observed punishments decrease
similar behaviors in observers. It can be easily imagined then, when both direct and
vicarious reinforcement are experienced, as when situated within an entire group of
deviant peers, that the reinforcement is powerful. Indeed, the average number of
delinquent acts increases by 1.52 as the number of delinquent friends (reinforcers for
deviance) increase while reinforcement of appropriate behavior from conventional
sources remains low.

Finally, cognitive control suggests that people’s cognitive capacities

27

tremendously increase the information they can derive from their experiences, and thus
partly determine how those experiences will affect them. Thus, while approving
responses, recognition, status, and acceptance from significant others are universal and
powerful social reinforcers for humans (Akers, 1977), the cognitions of the individual
greatly influence the effect and the subsequent actions that are taken. While Bandura
acknowledged this factor within his Social Learning Theory, it is perhaps better
developed within Social Information Processing Theory.
Social Information Processing Theory

Social Information Processing Theory emphasizes the active role of cognition in
social adjustment (Bear et al., 2000). As opposed to emphasizing the external modeling
and reinforcements of behavior as the primary cause of antisocial behavior, this theory
instead focuses upon the student’s self-directed social cognitions. Dodge (1993)
conceptualizes the process of responding to stimuli as occurring in a sequence of
cognitive steps. First, the information is taken in or encoded into the brain. A mental
representation or the application of meaning to what was encoded is then formed. Third,
in a step referred to as response accessing, possible responses that are associated with the
mental representation such as verbalizations, motor activities, endocrine secretions,
autonomic arousal, and experienced affect are drawn upon. Those possible options are
evaluated in a decision-making step, and then finally the enactment, the behavior occurs.

Deficits and biases can occur at any and all of these stages (Bear et al., 2000;
Dodge, 1993). During the encoding stage, aggressive children may respond to fewer cues
in the environment and thus decisions are less accurate. Also, they tend to be overly

sensitive to hostile cues. The mental representations that aggressive children form may

28

be lacking affective or social perspective taking, making it difficult for these children to
understand others’ intent. They may therefore attribute hostile intent to ambiguous
situations, which is known as the hostile attribution bias. The available responses to
which aggressive children have access may be overloaded with deviant options and they
may evaluate these choices more positively than other children might. Furthermore, they
expect these deviant options to lead to more positive outcomes and judge it as being less
morally bad (Dodge, 1993) and as more effective and appr0priate (Deater-Deckard,
2001). They anticipate feeling better about themselves after aggressing (Dodge, 1993).
Therefore, aggressive children are less competent at enacting positive peer group entry
behaviors because the option or the behavior that they enact is often antisocial in nature.
Conversely, preschoolers with a more advanced understanding of other children’s
emotions are more likely to have friendships that are more prosocial and less conflicted
(Deater—Deckard, 2001). Crick and Dodge (1994) envision these steps as being not linear
but cyclical in nature, indicating that the behavioral enactments and peer responses
influence the encoding of future interactions.

Aggressive children and children with conduct disorder (who have fundamentally
different early social experiences than do typically developing children, Baker, 1998)
have deficits in social cognitive skills and thus are more likely to behave aggressively
(DeRosier, Cilessen, Coie, & Dodge, 1994), underutilize pertinent social cues (Hinshaw
& Anderson, 1996), misattribute hostile intent to ambiguous peer provocations, generate
fewer assertive solutions to social problems, distort social cues during peer interactions,
and expect that aggressive responses will lead to reward (Hinshaw & Anderson, 1996;

Webster-Stratton, 1993). Although it is acknowledged that many aggressive children are

29

popular and have high self-concepts, these cognitive and social deficits can in some
children cause a failure to successfully associate with peers, which may (though not
necessarily) lead to negative self-perceptions and social isolation that in turn are closely
linked to the development of further internalizing problems in later childhood (Mesman,
Bongers, & Koot, 2001).

Thus, while Social Learning Theory contends that behavior is almost a reflex to
the environmental stimuli, Social Information Processing Theory focuses instead on
cognitive processes as the primary mechanism of behavioral change. While the
environment continues to be important in this later model, it is certainly de-emphasized.
What is important to the present paper is that it is well established theoretically that both
the environment and the individual, as well as the social interaction between these two
entities, are important in the process of developing behavior and behavioral change.
What seems to be missing or is at least underdeveloped in the above theories is the notion
of group dynamics and the social bonding that occurs between individuals and other
people, groups of people, or even an entire organization, in this case a school. It is
therefore imperative to discuss the theoretical underpinnings of the role of social
bonding.

Social Development T heory

Social learning theory does not specify the role of bonding in the etiology of
behavior (Huang, Kostennan, Catalano, Hawkins, & Abbott, 2001). The concept of
bonding can be broken into three components (O’Donnell, Hawkins, & Abbott, 1995).
First, to be bonded to an individual, there is an implicit understanding of attachment. By

definition, one cannot be bonded to an individual without feelings of attachment.

30

Second, it is a committed relationship, meaning that there is an investment and a desire to
continue that relationship. Finally, there is a common belief about what is right and
wrong. Without a common fundamental philosophy, bonding would seem unlikely at
best. In the specific example of schools, teachers attempt to bond with students
regardless of common philosophies, but this theory would posit that without some
common ground between the two parties, a mutual bond will be difficult to form. The
person or organization to whom the child is bonded then is referred to as the socializing
agent. Examples of socializing agents from whom children learn patterns of behavior are
family members, peers at school, teachers, or, at the organizational level, the family or
the school itself (Huang, Kosterman, Catalano, Hawkins, & Abbott,2001).

Children are socialized through processes involving four constructs. First, it is
necessary that there be an opportunity for involvement in activities and interactions with
others. Simply stated, you cannot form a bond with and therefore be socialized by an
agent with whom you have no involvement. Second, the degree of involvement and
interaction is important. Naturally, the more involvement, the more interaction a child
has with an agent, the more likely a bond will form. Third, the child must bring the skills
necessary to participate in these involvements and interactions. For example, the child
who cannot catch, cannot throw, and cannot hit a baseball may find it difficult to form a
bond with the team. Finally, following the performance within these activities and
interactions, it is imperative that reinforcement is forthcoming (Huang, et al., 2001;
O’Donnell, Hawkins, & Abbott, 1995 ). Putting it all together, when the child is given the
opportunity to participate, demonstrates the skills necessary to belong, and is reinforced

for the efforts, a bond is formed between that child and the agent (Huang, et al., 2001).

31

This social bond, once it is strongly established, can affect behavior independently
because of the informal control it commands on future behavior. Because of this
attachment, this commitment to the social agent, the child will behave in a manner that is
consistent with the standards and norms of the agent so as to not threaten the established
bond. In the school, students often have the opportunity to participate and it is the goal of
the teacher to improve the skills necessary for successfirl participation using a steady
dose of reinforcement in an attempt to not only form a bond with the student but to
develop the skills needed to be successful in future grades and in life after school.

How would the Social Development Model contend that aggressive behavior
develops? The theory suggests that these behaviors develop due to weak bonds to
prosocial institutions and to agents who promote the values those institutions represent
(Herrenkohl, Guo et al., 2001). It is of primary importance that families and schools
emphasize prosocial beliefs and good behavior. Beyond those institutions, peer
influences, particularly in early adolescence, are particularly strong. Negative peer
influences, as is often the case with aggressive and antisocial children, affect behavior
most powerfully when bonds to prosocial institutions and individuals are weak and bonds
to peers are strong. Likewise, adolescent problem behavior is inhibited by high levels of
social bonding to prosocial others (O’Donnell, Hawkins, & Abbott, 1995), implying that
the family and the school may be paramount to battling these negative peer influences.

Within the broader social context, there are factors at work that affect not only
whether aggression between the members of the group will occur but also the quality of
the other group members’ reaction to that aggression (DeRosier, Cilessen, Coie, &

Dodge, 1994). A group context marked by higher levels of physical activity, group

32

aversive behavior, and competition appear to provide a setting more conducive to
aggressive interactions. A highly active, aversive, and competitive context may promote
aggression because misinterpretation of others’ behavior is more likely

There are three other variables that seem to influence the power or the strength of
the formed social bond (Catalano & Hawkins, 1996). First, the individual’s position in
the social structure, as defined by socioeconomic status, race, gender, and age has an
indirect impact on behavior due to perceived opportunities for prosocial and antisocial
involvements and interactions. Second, constitutional or physiological factors (such as
cognitive ability) may impact opportunity or skills necessary to bond to a certain group.
Finally, external constraints such as the explicit clarity of rules, laws, and norms, and the
degree of consistency and immediacy of the sanctions imposed can impact the social
bond.

It has been suggested in this discussion that these social bonds influence future
behaviors. For example, involvement with antisocial peers will likely lead to delinquent
behavior, especially in the absence of prosocial agents to balance the effects. An
antisocial path may therefore be conceived as the following (Huang, et al., 2001). A
child may perceive an opportunity for problem behavior and antisocial interaction. This
may lead to an involvement in the problem behavior and an interaction with antisocial
others. The child then perceives rewards for the problem behavior and the interaction
with antisocial others. This perceived gain may lead to an attachment and a commitment
to antisocial others and activities to continue gaining from these behaviors. This brings
about a belief in antisocial values, which will promote future antisocial behavior.

Patterson, DeBaryshe, and Ramsey (1989) give a related example from a family

33

perspective. In their model, poor parental discipline and monitoring can be seen as a
failure to provide prosocial modeling. This in itself can lead to child conduct problems.
The child therefore, due to poor parenting, does not develop the skills necessary to relate
to positive peers or to school as a whole. This can lead to peer rejection and academic
failure. These children find that they can relate better to antisocial peers and form a
commitment to that deviant peer group. This then leads to delinquency. Opposite
pathways exist for positive behavior outcomes with the socializing agents being prosocial
instead of antisocial (Huang, et al., 2001).

The Social Development Model involves four phases of social development:
preschool, elementary, middle, and high (Catalano & Hawkins, 1996). Prosocial and
antisocial influences from one developmental period affect variables at the beginning of
the causal chain of the next (Huang, et al., 2001). Conceptually then, violent behavior in
the preschool phase can have a significant and direct effect on violence in the subsequent
phases. However, these social developmental processes continue to be important aside
from the violent behavior. Violence does not simply beget violence. In the middle
appears be an increase in antisocial socialization experiences.

The implications of this model are two-fold (Herrenkohl, Huang et al., 2001).
First, the model hypothesizes that similar developmental processes lead to prosocial and
antisocial behavior, which is supported by research. Second, it therefore implies that the
same interventions in childhood may be effective in adolescence. Interventions should
seek to enhance youth’s skills for prosocial interactions, increase opportunities for
involvement in prosocial activities, and reward involvement in those activities to

strengthen prosocial bonds. In the context of the school, there are two main groups of

34

people who could potentially serve as prosocial agents and can therefore be the targets of
intervention, the teachers and the peers.
The Socializing Agents in the School

The Teacher Effect

There is some evidence that would suggest that a student’s emotional and
academic adjustment to school is related to the bonding that occurs between that student
and the school (Catalano & Hawkins, 1996). The primary factors involved in that school
bonding would be the relationships between the child and his peers and between the child
and his teachers. Murray and Greenberg (2000) report four subtypes of students in
relation to school bonding: Dysfunctional, F unctional/Average, Positively Involved, and
School Anxious. Not surprisingly, the Dysfunctional group reported the least
involvement in school and the lowest emotional support from their teachers. These
characteristics blossomed into anger and reciprocal negative feelings between the two
parties. Conversely, early positive teacher-child bonds have been shown to be good
predictors of later behavioral adjustment in school and better future relationships with
teachers (Birch, & Ladd, 1998; Hamre & Pianta, 2001). It is unclear from this literature
how much of the impact is due to a more pleasing temperament on the part of the
students and how much is due to the efforts of the teachers. What seems to have good
face validity, however, is that a bond to the teacher can help strengthen the bond to the
school and may lead to better behavioral outcomes in children. Therefore, it is important
to examine the characteristics of effective teachers.

One factor heavily implicated in student academic achievement is teacher

efficacy, which is often divided into two dimensions: personal, the belief that you as a

35

teacher can have a positive impact on the learning of a child, and general, the belief that
teaching in general can have such an impact (Woolfolk, Rosoff, & Hoy, 1990). Differing
levels of efficacy may lead to different behaviors and levels of effectiveness among
teachers (Anderson, Greene, & Loewen, 1988; Gibson & Dembo, 1984). It is commonly
believed that teachers with a high sense of efficacy will have higher achieving students
than those teachers with low efficacy (Anderson et al., 1988; Gibson & Dembo, 1984;
Guskey, 1982; Tschannen-Moran, Woolfolk Hoy, & Hoy, 1998; Woolfolk & Hoy, 1990;
Woolfolk et al., 1990). This statement is intuitive theoretically. If a teacher has an
efficacy expectation (Bandura, 1977), or in other words is convinced that he/she can
successfully teach a child how to do some activity, arithmetic for example, then it stands
to reason that that teacher will be more successfirl at teaching addition than a teacher with
less confidence. According to Bandura (1977), these expectations are derived from
performance accomplishments, vicarious experience, verbal persuasion, and
physiological states with performance accomplishments producing the strongest
expectations, followed by vicarious experience. Receiving the immediate feedback of a
Well-run, high achieving classroom will often lead to a higher sense of efficacy
(Woolfolk et al., 1990), which in turn leads to higher goal challenges and a strong
CO Immitment to reach them (Bandura, 1993). Intuitively, this leads to higher student
3C h ievement.
Knowing that efficacy has a positive effect on student achievement leads one to
e)‘laJnine the qualities of teachers with a high sense of efficacy. They tend to be female
(Evans & Tribble, 1986; Anderson et al., 1988) and work in elementary schools (Evans &

Tribble, 1986), though why this might be is only speculative. It seems plausible that once

36

students reach middle school, there is a feeling among teachers that less can be done for
them, which would naturally lower their sense of efficacy. Other qualities of teachers
with high efficacy compared to those with low efficacy are higher levels of effort to the
many tasks of teaching, better goal-setting, higher levels of aspiration, and better
organization (Tschannen-Moran et al., 1998). They are also more determined in that they
keep trying to succeed, spend more time with academics, provide the extra help that some
students need to be successful, and praise the accomplishments of their students
(Bandura, 1993). There is also some evidence that there are differences in classroom
management strategies in that higher efficacy teachers place more trust in their students
and share in the responsibility of solving classroom problems (Woolfolk et al., 1990).
Finally, teachers with high efficacy are less likely to refer children to special education
(Meijer & Foster, 1988; Podell & Soodak, 1993; Soodak & Podell, 1993).
All of the above characteristics are qualities of teachers with a high sense of
e fficacy, which leads to better student achievement, academically, a notion that has been
Well supported in the literature. However, little evidence exists to evaluate the role that
teacher efficacy plays? on children’s behavioral outcomes. What seems to be the case is
th at teachers experience significantly less confidence when dealing with behavioral
Pro blems than when they are teaching academically (Martin et al., 1999). This lack of
coI‘lfidence probably stems from a lack of training in behavior management and thus
leEl.\ling teachers trying to learn techniques as they go (Martin et al, 1999; Wheldall,
1 991). Teachers therefore become frustrated and begin treating “problem children”
differently than the “well-behaved” children. They give the difficult students less

encouragement and punish them more often for negative behavior (Webster-Stratton,

37

1993 ), which generally creates less cooperative students (Winett & Vachon, 1974).
Furthermore, the greater the concern that the teachers have about a particular student, the
less confident they feel to deal with that student, and the more likely they are to refer that
child to other school personnel (Martin et al., 1999).

It stands to reason, however, that not all teachers behave the same way and that
there is variation in the efficacy of teachers when dealing with behavior problems. If
teachers with high efficacy, for example, feel that they have better control over their
classrooms (Woolfolk & Hoy, 1990) and if that efficacy directly affects the teachers’
behaviors, classroom management techniques, and instructional strategies employed
(Brownell & Pajares, 1999), then the effect of these variations of teacher efficacy are
important to study. In other words, the link between the efficacy the teacher feels and the

social relationship that that may impact is important to the understanding of problem
behaviors within the classroom and to the informing of potential interventions to the
problem. This of course is only one half of the story. As mentioned earlier, there are two
main social forces in the school, the other being the student’s peers.
Tile Peer Effect
Few would argue that the social interactions between students in a school are an
iI111:30rtant aspect of that student’s life. Each interaction has the potential to lead to
as c: eptance, which ultimately manifests itself in the form of friendship, or to rejection,
Wh ich may resemble, among other things, a bully-victim relationship. In theory, there are
Six types of rejecting behavior (Deater-Deckard, 2001): the exclusion from social
interactions, the prevention of access to friends or information, aggression in its various

forIns toward the child, the domination or controlling of a child, the expression of moral

38

disapproval of behavior, and the utilization of a third-party for tale-telling or saying
hurtful things within earshot of victim. Overall, children who are aggressive-withdrawn
are perhaps most at risk for being rejected (Hymel, Bowker, & Woody, 1993), which may
lead to behavioral and academic problems.

Peer rejection is consistently linked with current behavioral problems and future
acts of violence (Deater-Deckard, 2001 ), as well as, more generally, to various acts of
aggression (Hinshaw & Anderson, 1996; Kazdin & Kagan, 1994; Parker & Asher, 1987;
Parkhurst & Asher, 1992). The relationship appears to be cyclical in nature in that the
rejection leads to externalizing behavioral problems which can in turn lead to the
maintenance and growth of that peer rejection. The problem itself may begin with a

generally poor social competence with peers (F arver, 1996). Indeed, the problem may
self-perpetuate given the idea that rejected peers may have less of an opportunity to
develop and master social skills (Skinner, Neddenriep, Robinson, Ervin, & Jones, 2002).
Regardless of the origin, this rejection can be a source of stress for the child, which may
I ead that child to seek out other rejected children as a possible way of alleviating some of
th at stress (Coie, Lochman, Terry, & Hyman, 1992). The problem is that, as mentioned
above, rejected children often exhibit behavioral difficulties and thus behaviorally
Ch allenged children often flock toward other behaviorally challenged children, forming in
a S ense antisocial friendships. These friendships form a base for deviancy training
(D eater-Deckard, 2001). They produce the atmosphere and the opportunity to perform
ant isocial behaviors that include, among others, aggression and substance abuse.
The assumption of the preceding paragraph is that aggressive youth always have a

rej ected status, which is simply not true. Something else, then, must be contributing to

39

the status of aggressive students. One reasonable theory is that rejection is based upon
the norms and contexts of the environment in which the behaviors take place (Farmer,
2000). A classroom of aggressive children may reject the prosocial, relatively well-
behaved child. That child, then, may feel stress and pressure to conform. Along these
lines, the social characteristics of the group setting play an influential role in the
expression of aggression within the group (DeRosier, Cilessen, Coie, & Dodge, 1994).
Aggressive behavior is more likely to occur when children are engaged in active rough
and tumble play than when they are engaged in parallel or cooperative play.
From an opposite perspective, the development of prosocial friendships (which is
easiest for children who have empathy and therefore understand the emotions of others,
Deater-Deckard, 2001) have numerous benefits. Research suggests that social support is
a potent protective factor (O’Grady & Metz, 1987). Friendships provide children with a
context for skill learning and development, emotional and cognitive resources, and
models for later friendships (Ladd, Kochenderfer, & Coleman, 1996). They can prompt
and reinforce desirable behaviors (Carden Smith & Fowler, 1984). These friendships are
as sociated with changes in several forms of maladjustment, including relative declines in
attention problems and misconduct, and relative gains in cooperative participation and
SC hool liking (Ladd & Burgess, 2001).
The implications of this research are fairly clear. Because the perceptions
Ch i ldren develop about friendship processes are associated with fiiendship outcomes such
as friendship satisfaction and stability (Ladd, Kochenderfer, & Coleman, 1996),
inte I’ventions such as peer monitoring with corrective feedback may be as effective as

teaCher monitoring in reducing disruptive behavior (Carden Smith & Fowler, 1984). In

40

addition, cooperative learning may foster self-esteem and altruistic behavior (F urrnan, &
Gavin, 1989).
Other Factors of School Failure and School Success
There are a number of child and environmental factors worth mentioning or
emphasizing within the discussion of school success and failure. As hinted at earlier,
physical aggression is the best predictor of later deviant behavior (Tremblay, Masse,
Pagani, & Vitaro, 1996). This is most often seen among boys during the preschool years.
(It should be noted, however, that while often downplayed, conduct disorder is one of the
most common psychiatric disorders among adolescent girls; Cote, Zoccolillo, Tremblay,
Nagin, & Vitaro, 2001). What all of this means within the school is that aggressive
children are often rated as being more hyperactive, more inattentive, more oppositional,
and less prosocial (Tremblay, Masse, Pagani, & Vitaro, 1996). These behaviors often
I ead to retentions, poor health, excessive drinking, drug use, and sexual intercourse all by
the age of 14. Other child characteristics that are associated with school success are:
Strong cognitive ability, strong reading skills, adequate social skills (Vance et al., 2002),
th e ability to self-regulate, and the possession of a positive temperament (Miller, Brehm,
85 Whitehouse, 1998). Conversely, school failure is often associated with being male,
eEll‘ly delinquency, prematurity, low birth weight, anoxia, other medical stresses, low
SC h 001 achievement (Yoshikawa, 1994), ADHD, deviant peers, substance abuse,
CO mmunication problems, negative cognitions, disobedience (which may include
tar(illness, breaking rules, and fighting; Webster-Stratton, 1993), and negative mood
(LOeber, 1990).

Although it is recognized then that characteristics of the child in fact impact the

41

outcomes, the central thesis of this paper is that the environment matters. It is clear from
the moment the child enters preschool that there will be demands imposed upon the
student by that environment. These demands are representative of key developmental
tasks such as making friends and learning required social skills (Mesman, Bongers, &
Koot, 2001). The better the child is able to bond to the school through the acceptance of
and adherence to the imposed rules, the better the outcome for that child will be
(O’Donnell, Hawkins, & Abbott, 1995). Schools then that focus upon the achievement,
both academic and social, of the individual student may have a positive impact (Miller,
Brehm, & Whitehouse, 1998).
Specifically within the classroom, it is important to remember that, as mentioned
above, aggressive behavior is influenced by peer group norms (Stormshak et al., 1999),
meaning that as classroom aggression increases, the negative effects of aggression on
peer preference decreases.
The focus of the current paper is upon the specific role of the teacher and the peer
as the major players in the maintenance or reduction of aggressive behavior over time.
S pecifically, there are several questions being explored. First, what characteristics within
“1 6: child are related to changes in aggression over time? Research suggests that
aggression is a relatively stable construct but yet is changeable. It is predicted that there
W i l 1 be significant variability in aggression scores from year to year in both directions,
me aning many children will either be more or less aggressive at the end of a time period
than at the beginning. Indeed research suggests that as many as 25% will see significant
imp rovement in scores. Obviously, these numbers leave a large majority of children who

do Iiot improve. This is partly because childhood aggression is one of the best predictors

42

of future aggression. In fact, it is believed that previous aggression strongly impacts
future aggression and that the more aggressive a child is, the more likely that child will
turn to aggression in the future. However, the amount of variation that is present will
allow a study of other factors that may influence these changes. Among these other
variables is the child’s reading ability, which has been consistently linked to aggression
in the literature. It is therefore predicted that a slight relationship may exist between the
child’s reading ability and changes in aggression.

Second, what are the social factors or variables (if any) related to the maintenance
or reduction of aggressive behavior? As mentioned earlier, the teacher and the peer are
the influences on the child being explored in this study. It is predicted that the quality of
the teacher-child relationship will be moderately related to the level of change of

aggression. In other words, the more positive personal and academic support that the
teacher gives, the less aggressive the child will become throughout the school year. A
related question being explored is what role does teacher efficacy with behavioral
p roblems play in the maintenance or reduction of aggressive behavior? Research
8 uggests that efficacy beliefs influence behaviors so that the better teachers feel equipped
to handle situations, the better they handle them. Therefore, it is predicted that teacher
e1“f‘icacy will have a small to moderate relationship with child aggression.

Third, how does peer support influence changes in aggression? The research

Cl e arly shows that peer support has a positive impact on children, and likewise peer
rej ection has a negative impact. It is believed that a lack of peer support will likely have
a r1“loderate and negative impact on the aggression of children. Additionally, children

who receive low support one year are likely to receive low support the next year and so

43

the effect will be consistently applied over time.

Finally, how do the behavioral norms of the classroom impact aggressive
behavior? Research suggests that if aggression is a normal occurrence within the
classroom, meaning that many children partake in aggressive acts regularly, then this will
increase the aggression of the whole classroom. It is believed that there will be a
moderate to strong relationship with aggression during the school year but would not

necessarily be carried over to the following year to a different classroom.

44

Chapter 3

Methods
Participants

Participants included 417 elementary school students and 68 elementary school
teachers from Southern elementary schools sampled over a two-year period. Among the
students there was a fairly even gender split with 50.6% being male. Over half of the
students were African American and approximately a third were White Caucasian.
Students ranged in age from six years, eight months to eleven years, eleven months. The
vast majority of the teachers were White/Caucasian females with three of them being
men and ten of them being African American. There was a wide range of experience
levels amongst the teachers with seven being in their first year of experience and twenty-
one having at least seventeen years of experience.
Measures
Aggression scale. The Behavior Assessment System for Children — Teacher

Rating Scale (BASC-TRS; Reynolds & Kamphaus, 1992) was completed for each child
near the beginning and end of each of the two school years of the study. This measure
asks teachers to rate children on a variety of behaviors that are coded on one of fourteen
scales, ten clinical (anxiety, aggression, attention problems, atypicality, conduct
problems, depression, hyperactivity, learning problems, somatization, and withdrawal)
and four adaptive (adaptability, leadership, social skills, and study skills). For the
purposes of this study, only the Aggression Scale was used. According to the manual,
this scale measures “the tendency to act in a hostile manner (either verbal or physical)

that is threatening to others” (Reynolds & Kamphaus, 1992, p. 48). Behaviors such as

45

name calling, arguing, and hitting are among those being assessed by this scale. The
scale uses T-scores with a mean of 50 and a standard deviation of 10. Aggression is
conceptualized not as a dichotomous variable (either you are or you are not aggressive),
but rather as a continuous variable of which all students display more or less aggression
than others. For the age range sampled in the present study, the TRS has internal
consistency alphas of greater than .90, test-retest reliabilities of .91, and interrater
reliability of .74.

Reading Grades. At four different points during the school year, students were
graded by their teachers in reading ability using their school’s grading policy. This data
was used as a general observation of student’s reading ability. More objective,
standardized scores were collected on some students but was not used during this study
due to the vast number of students who did not have this data available.

Teacher efficacy scale. The efficacy subscale of the Efficacy and Expectation

Measure (EZM) is a 5-item scale measuring teacher’s efficacy beliefs for each child.

 

Sample questions include: “I feel capable of helping this student behave appropriately in
my class,” and “I am certain I can manage this student’s behavior.” The five items were
significantly and moderately highly correlated with each other (Pearson product moment
correlations ranging from .44 to .84). The validity of the scale was explored using a
principal component factor analysis procedure. The results confirmed a strong unitary
factor with only one eigenvalue over one (3.63), which accounted for 73% of the
variance. All items loaded onto a single factor (all factor loadings above .82). The
internal consistency reliability (Cronbach’s alpha) for the subscale was .90. The scale

was used at the child level and aggregated at the classroom level to get a sense of the

46

overall efficacy of the teacher.

Teacher and peer support. Portions of the Student Information Questionnaire are
used to examine the quality of the teacher-child relationship. Teachers completed the
SE2 for each child, answering questions about how open, affectionate, and positive the
teacher-child relationship is seen to be. The validity of the scale was explored using a
principal component factor analysis procedure. The results confirmed a strong unitary
factor with only one eigenvalue over one (4.42), which accounted for 64% of the
variance. All items loaded onto a single factor (all factor loadings above .65). The
internal consistency reliability of this scale was .85.

Students completed portions of the Classroom Life Inventory to assess the
strength of teacher and peer academic and personal support. The scale explores the
child’s perceptions of how teachers and peers care about the child as both a friend and a
learner. The validity of the four scales extrapolated from this inventory (Teacher
Academic Support, Teacher personal Support, Student Academic Support, and Student
Personal Support) were explored using a principal component factor analysis procedure.
The results confirmed a strong unitary factor for each scale with only one eigenvalue over
one. The Classroom Life Inventory has an internal consistency reliability of .87.

To further explore another aspect of the child-teacher relationship, each student
completed the Child Report of Teacher Behavior. This is a five-item scale that identifies
the child’s perceptions of how well the teacher identifies expectations and meets the
educational needs of children. Two sample questions include “I know what my teacher
expects of me in class” and “My teacher makes sure I understand before she/he goes on.”

The validity of the scale was explored using a principal component factor analysis

47

procedure. The results confirmed a strong unitary factor with only one eigenvalue over
one (2.43). All items loaded onto a single factor (all factor loadings above .65). The
internal consistency reliability of this scale was .73. All of the above scales were used at
the child level and aggregated at the classroom level to get a picture of the overall support
in the classroom.

Classroom climate. The ways in which classroom climate may affect aggressive
tendencies in children were explored in two separate ways. First, the number of
aggressive students in each classroom is one way to measure classroom norms. For the
purposes of this study, all students with a BASC Aggression score of at least 60, one
standard deviation above the mean, are counted toward the total number of aggressive
students within a classroom. Second, the Vessels Classroom Climate Scale was
completed by each student and then aggregated at the classroom level. The scale
measures the overall atmosphere of the classroom with questions such as “The kids in my
class help each other” and “My teacher cares how I feel.” The internal consistency
reliability of the aggregated scale was .86.

Data Analyses and Rationale. Hierarchical linear modeling (HLM) is used in this
study for several reasons. First, the assumption that all the students in the sample are
independent (which is important in other forms of analyses such as multiple regression) is
not met because they all share with at least one other student a common teacher and thus
are commonly influenced by that teacher. HLM accounts for this problem and thereby
fixes other problems such as aggregation bias and misestimated standard errors.

The present study uses three two-level models, students within classrooms. These

models extend three different time periods, the fall to the spring of year 1, the spring of

48

year 1 to the fall of year 2, and the fall to the spring of year2. This was done to account
for the fact that students change teachers between years. Level one of each model
examines the child variables for each time period that are related to changes in
aggression, beginning with the unconditional model:

Yij = I301 + n,-
where

Yii is the change in aggression score for child i in school j;

Potential Level one variables that will be added to the model one at a time to
assess for significance include previous aggression, reading ability, behavior grades
teacher efficacy, teacher-child relationship, teacher academic and personal support, and
peer academic and personal support, along with demographic information such as age,
race, and gender.

Level two models examine the classroom variables related to changes in
aggression during each of the three time periods. Potential Level two variables that will
be added to the model one at a time to assess for significance include classroom climate,
the percentage of aggressive students in the classroom with a score of at least 60 on the
Aggression Scale of the BASC, the overall efficacy of the teacher when dealing with
behavior problems, the overall teacher and peer support in the classroom, the teacher-
child relationships (how positive or negative the teacher’s overall relationship with the

children) and the reading and behavior grades of the class as a whole.

49

Chapter 4
Results

Because students in the classroom are inherently not independent of each other
(each student is being acted upon by common variables such as the same teacher),
hierarchical linear modeling is used to tease out student level relationships from
classroom level ones with the outcome variable, in this case changes in aggression.
Because children change teachers from one year to the next, three separate analyses were
completed: from the fall to the spring of year one, from the spring of year one to the fall
of year two, and from the fall to the spring of year two. To assess the changes in
aggression, difference scores were taken for each of these time periods. Students with
missing data (N = 47) were generally deleted from the analysis. Since change in
aggression was the outcome of interest, students who demonstrated a change score of
zero, meaning they did not change during the time span, were deleted from analyses.
Some variables (Teacher Academic and Personal Support, Peer Academic and Personal
Support, and Teacher-Child Relationship) contained missing values less than five percent
of the time and were replaced with the mean for that variable.

Table 1 shows means and standard deviations of the variables studied, separated
by the level of analysis: student and classroom. As can be seen, years 1 and 2 generally
have similar averages and deviations with the summer in many cases being dramatically
different. This was especially true with the Support measures examined. These results
are not especially surprising given that these reports were given near the beginning of the
school year and much of that time period spanned the summer months when school was

not in session. Therefore, one would expect support to be less developed than at the ends

50

of the school year.

Table 1: Means and Standard Deviations by Year

Variable
Student

Aggression Difference
Previous Aggression

Age

Reading Grades
Behavior Grades
Behavioral Efficacy
Teacher Academic

Support

Teacher Personal Support
Teacher-Child

Relationship

Child Report of Teacher

Behavior

Classroom Climate
Student Personal Support
Student Academic Support

Classroom

Percentage of Aggressive

Children
Reading

Behavior Grades
Behavioral Efficacy
Teacher Academic

Support

Teacher Personal Support
Teacher-Child

Relationship

Teacher Behavior
Classroom climate

Student Academic Support
Student Personal Support

Range

-38-40
40-98
6:8-
11:11
1-5
1-5
3-15
0-8

0-8
3-38

0-10

6-30
0-8
0-8

0-75

2-5
2-5
6-15
0-8

0-8
20-30

0-10
15-27
0-8
0-8

Year 1

-2.36 (6.84)
52.17 (11.00)
7:6 (.88)

4.00 (.94)
3.98 (1.07)
13.42 (2.45)
7.49 (1.07)

6.86 (1.64)
24.76 (3.58)

8.43 (1.89)

21.90 (4.81)
5.55 (2.23)
5.82 (2.08)

21.88 (12.56)

3.97 (.43)
3.91 (.52)
13.69 (.96)
7.49 (.32)

6.89 (.43)
24.07 (1.38)

7.83 (.51)
22.01 (2.32)
5.17 (.61)
5.23 (.53)

Summer

.93 (11.92)
54.34 (11.92)
8:7 (1.41)

3.58 (1.04)
3.65 (1.01)

13.74 (2.16)
.67 (1.21)

1.27 (1.96)
24.70 (3.43)

1.52 (2.10)

22.85 (4.78)
2.27 (2.31)
2.34 (2.51)

29.26 (13.52)

3.54 (.34)
3.52 (.55)
13.53 (1.79)
.73 (.32)

1.24 (.45)
24.86 (2.01)

1.50 (.63)
22.79 (1.32)
2.28 (.57)
2.22 (.52)

Year 2

-l.69 (7.68)
49.87 (11.19)
8:8 (1.78)

3.93 (.97)
4.17 (.90)
14.43 (1.18)
7.51 (.97)

6.89 (1.58)
23.67 (2.78)

8.55 (1.71)

22.70 (4.62)
6.13 (1.90)
5.83 (2.19)

13.96(16.11)

3.92 (.31)
4.09 (.37)
13.94 (.86)
7.14 (.51)

6.44 (.52)
23.57 (1.71)

8.39 (.45)
20.90 (2.34)
5.40 (.80)
5.72 (.71)

In addition, while the unconditional models of the intercept for years one and two

were significant, this was not the case for the summer months. Due to this fact,

classroom level variables were not modeled for the time span from the Spring of year 1 to

51

the F all of year 2 and thus only student level variables were examined using regression.
Furthermore, hierarchical linear modeling assumes that the distribution of the outcome
variable is normal. Figure 1 shows that changes in aggression did follow a fairly normal
distribution during all three time spans and thus this was not seen as a limitation in the

present study.

 

 

 

 

 

 

 

Year 1 Summer

 

 

 

 

Year 2

Figure l: Distributions of the outcome variable, Changes in Aggression

Level 1 Variables: The Student Level

Student level variables were centered around the group mean and Table 2 lists

52

estimates and standard errors for both student and classroom variables. Results were
somewhat inconsistent across time periods in that no variable was significant at all three
time points. This is not all together surprising given that significant changes occur during
the summer months and into the fall, such as summer vacation and then likely new
teachers when the students return to school. Two variables were significant at two time
periods: previous aggression and behavioral efficacy. Due to the nature of difference
scores and to the fact that aggression is the object of change, interpreting the impact of
previous aggression is not advised. Modeling previous aggression simply assures that it
has been controlled and perhaps lends more weight to the other findings (Willett, 1998).

Behavioral efficacy seems to positively relate to aggression over time. Teachers
who feel that they can help students with behavioral problems be successful are more
likely to meet that goal. In other words, the stronger the efficacy belief, the greater the
decline in aggression over time.

Behavioral grades were significant during one of the three time points. As
behavioral grades decline, i.e. behavior gets worse, aggression increases over time.
There would seemingly be considerable overlap between this variable and aggression, a
prime reason why it was not significant at other time points. During the first year,
however, these grades help explain more of the variance than aggression does alone.
While the grades would surely be impacted by aggression, they do cover a broader range
of behaviors than purely aggression, which explains the result and the inconsistency to
some degree. Perhaps during year one there was a higher rate of students with non-
aggression difficulties, such as hyperactivity, that impacted grades and yielded the result.

Table 2: Estimates and Standard Errors by Time
Variable Year 1 Summer Year 2

53

Student Est SE Est SE Est SE

Previous Aggression .116 .06 .471** .08 .782" .08

Age .622 .61 .088 .98 .215 .78

Sex .110 .92 -1.37 1.69 -.038 .95

Reading Grades .858 .55 -2.44** .87 .791 .70

Behavior Grades 2.65“ .71 1.82 .99 1.1 1 .69

Behavioral Efficacy .561** .19 .248 .37 .603** .21

Teacher Academic Support .809* .33 .165 .66 .140 .61

Teacher Personal Support -.025 .36 -.006 .61 .319 .42

Teacher-Child Relationship -.038 .10 -.382** .14 -.053 .08

Child Report of Teacher -.022 .27 .185 .50 -.089 .37

Behavior

Classroom Climate -.l37 .12 .198 .22 .047 .18

Student Personal Support .05 .25 .014 .48 . 163 .38

Student Academic Support -.008 .23 -.252 .42 .137 .38

Classroom

Percentage of Aggressive .104* .04 -.062* .03

Children

Reading -.559 1.47 -.587 1.63
Behavior Grades .154 1.10 1.14 1.13
Behavioral Efficacy -.802 .48 -.31 .50

Teacher Academic Support 1.04 .76 .527 .65

Teacher Personal Support .684 1.16 .31 1.10
Teacher-Child Relationship -.053 .17 -.013 .19

Teacher Behavior 1.05 .63 1.78 1.56
Classroom climate .080 .21 1.82* .88

Student Academic Support .076 .63 -1.28 1.45
Student Personal Support .26 .65 1.25 1.41

* significant at .05 level
** significant at .01 level

The teacher-child relationship was significant during the summer time period
only. Students who had a better relationship with their teacher at the end of year one
evidenced lower aggression at the start of year 2. This result may be due to the
possibility that students who ended the year more positively had an easier time

transitioning into the next year.

One additional variable was significant at only one time point: Reading Grades.

54

From the spring of year one to the fall of year two, reading ability was a significant
predictor of changes in aggression. Children with better reading scores had positive
changes in aggression, i.e., aggression decreased. This may be due to the possibility that
lower ability readers had a higher loss of information over the summer months than
higher reading students, causing more frustration early in the school year. The fact that
reading was a less important factor during the school year despite stability in reading
grades throughout the year (average change in reading throughout the school year was
approximately .285) would seemingly support this suggestion. Lower level readers do
not get more aggressive as the school year progresses because they already started at a
higher level of aggression and the frustration is constant.

Also of interest were those variables that proved to not be significant. First, the
age of the child did not seem to matter in terms of his/her change in aggression. Older
children are just as likely to change (or not) as younger children and behavioral efficacy
continues to be a moderate predictor when age is controlled.

Two other insignificant variables in this study were the race and sex of the child.
While one might expect race and sex to impact aggression, it does not appear to have any
bearing upon the change of aggression over time. This finding, along with age discussed
above, may lend some credence to the idea that aggression is relatively stable over time.

The support variables measured also proved to be insignificant in changing
aggression. Both teacher and peer academic and personal support were not related to
changes in aggression in this study. This is counterintuitive to thought and theory and
may be a result of the scales used. All of the scales had limited ranges and questions and

thus may not have measured support completely.

55

Level 2 Variables: The Classroom Level

Because the unconditional models of the intercepts were only significant for years
one and two (not summer), only during these time spans were teacher variables modeled.
Only one of the classroom-level variables modeled in year one were significant at the .05
level. It is not surprising that variables failing to reach a level of significance at the
student level failed to reach significance when aggregated at the classroom level.
Somewhat more surprising is the fact that behavioral efficacy at the classroom level was
insignificant. The overall behavioral efficacy of the teacher when dealing with problems
did not relate to changes in aggression while the efficacy with the individual child did.
This is probably due to the fact that teachers feel more comfortable with some students
than others and thus the overall efficacy does not matter when faced with a student with
whom the teacher feels less efficacious.

Most surprising in year one was that classroom climate was not related to changes
in aggression, despite the fact that the percentage of aggressive students in the classroom
was significant. It would make sense that the worse the classroom climate, the more
aggression will rise in all students and vise versa. This held to be somewhat truer in year
two when both of these variables were significant at the .05 level. The results of year 2
are more consistent with theory and have greater face validity than those of year 1. This
may simply be due to the fact that year 2 had more students per teacher, was therefore
better distributed per classroom, and thus provided a better picture of the relationship

between classroom climate and changes in aggression over time.

56

Chapter 5
Discussion

This study evaluated the extent to which social agents (teachers and peers) within
the school setting related to changes in aggression over time. Several hypotheses were
examined including that the amount of variability in aggression scores over time would
be significant, that reading ability would be related to changes in aggression, that teacher
efficacy, teacher-child relationship, and personal and academic support from peers and
teachers would all be correlated with aggression change.

First, it was predicted that there would be significant variability in the changes of
aggression so that it would be appropriate to model level 2 (classroom) variables. This
hypothesis was supported by the data for two of the three time points. During the middle
term, which took place primarily in the summer, this variability was not found. Since the
majority of students are not in school during the summer and since they tend to switch
teachers the following term, this is not surprising. It takes time to build relationships
with teachers and peers and thus any relationship between classroom level variables and
changes in aggression will develop during the course of the school year. Other
hypotheses examining within-child, between teacher, and whole classroom characteristics
will be examined in turn, as well as implications for intervention.

Factors Within the Child Associated with Changes in Aggression

While previous aggression cannot be interpreted in the present model due to the
measurements of difference score, it can be noted that aggression changed enough to find
some predictors of that change. This idea speaks to the stability of aggression. As in the

example from clinical psychology where about 25% of children with Oppositional

57

Defiant Disorder no longer exhibit these symptoms in the future (Hinshaw & Anderson,
1996), the data may be showing similar fluctuations and improvements. On the other
hand, because the distribution of change scores followed a fairly normal curve, children
witnessing declines in behavior may be evidence for the progression of pathology such as
that suggested by Loeber and his colleagues (1998). They contended that aggression
would become more severe over time, which would produce higher scores on
standardized measures of aggression, like the BASC. The purpose of the present study
and future studies should be to attempt to determine those factors that help children move
into the 25% of those who improve and to avoid those factors that cause a behavioral
decline. Overall, this study seems to suggest that many teachers in many instances are
finding little success when dealing with aggressive students.

In addition, two other variables were significant during at least one time period.
Students’ behavioral grades were related to changes in aggression as expected, in that
better-behaved students evidence lowering aggression scores. The fact that behavioral
grades at one time point helped explain more variance than previous aggression alone
may be attributable to the grades themselves. Because these grades do not only measure
aggression, they may be identifying related problems. For example, Attention Deficit /
Hyperactivity Disorder and Anxiety Disorders are related to Conduct Disorder
specifically and Aggression more broadly (Hinshaw & Anderson, 1996). Students with
attention or impulsivity problems may receive lower behavioral grades. These
difficulties with attention and impulsivity may then correlate with future aggression
somewhat independently of previous aggression scores.

Reading ability was associated with changes in aggression during the summer

58

period only. While reading appears to be related to aggression levels (Tomblin, Zhang,
Buckwalter, & Catts, 2000), it is generally not related to changes in aggression. It is
hypothesized that children with lower reading ability may lose more information over the
summer, making school a more frustrating environment, which impacts behavior at the
beginning of the school year. Previous attempts to study reading and aggression did not
examine changes in ability or in aggression. Interventions for aggression have targeted
reading as one area to help students based upon this literature. This study offers little
support for this hypothesis in that reading appears resistant to change, at least under
normal instruction, and reading ability showed little relationship to aggression changes.
Further research is needed, however, to continue to assess this relationship.
The Teacher’s Influence on Changes in Aggression

While teacher academic and personal support was not related to changes in
aggression in this study, the hypothesis that the teacher-child relationship would be
related to changes in aggression received minimal support. The child’s relationship with
the teacher at the end of year one could predict aggression at the beginning of year 2. It
is hypothesized that children who end the year on a positive note with a good relationship
are more prepared to start the following school year. Why this relationship was not
present during the school year itself is unknown, but it should be noted that cumulative
effect of this variable may be more significant. Since at all three time points a similar
relationship exists, it may be that cumulatively over time, positive teacher relationships
can help mold prosocial students.

The hypothesis that teacher efficacy would be related to changes in aggression

scores was supported by the data at two of the three time points. It does not follow that

59

teacher efficacy on one student at the end of year one would lead to lower aggression in
the fall of year 2 because there is a change of teacher at that point and so it was not an
unexpected result to fail to find significance during the summer time period. Because
previous efficacy was used to predict future change, it would appear that the more
efficacious a teacher feels when dealing with behavior problems, the lower aggression
will become over time. This finding warrants a closer look at the construct of teacher
efficacy.

Teacher efficacy is the “belief in his or her capability to organize and execute
courses of action required to successfully accomplish a specific teaching task in a
particular context” (Tschannen-Moran, Hoy, & Hoy, 1998, pg. 233). When thinking
about efficacy with behavior problems, then, behavioral efficacy is the belief about one’s
capability to successfully correct the negative behavior of a student, with the negative
behavior in this specific case being aggression. Because most of the research on teacher
efficacy has been done on academic efficacy, one must extrapolate from the literature to
discover what factors are important in the creation of efficacy beliefs.

Efficacy is largely derived from experience, either positive or negative
(Tschannen-Moran, Hoy, & Hoy, 1998). Efficacy develops during teacher education
programs and especially during the fieldwork associated with them. It is during these
times that the student teacher draws from the coursework and the observations done in
the field to develop initial sets of skills and attitudes about teaching. These early
experiences likely impact the efficacy beliefs of that teacher throughout his or her career
since once the efficacy beliefs are established; it appears they are difficult to change

(Tschannen-Moran, Hoy, & Hoy, 1998). Therefore, attending to early or pre-service

60

teachers’ beliefs about their efficacy when working with children with aggressive
behaviors may be an important application of these findings.
The Classroom as a Whole

The classroom is a fairly social dynamic environment. Initially somewhat
surprisingly then, the level of peer support was not related to changes in aggression.
Upon fiirther reflection, the finding makes considerable sense. If it can be assumed as
research suggests that peer rejection is related to aggression (Hinshaw & Anderson, 1996;
Kazdin & Kagan, 1994; Parker & Asher, 1987; Parkhurst & Asher, 1992) and thus
conversely peer support is oppositely related, then it seems likely that a couple different
things are occurring. First, many of the students studied here had already been in school
for some time. Based upon the results of this paper, it can be inferred that students who
showed aggression during the time frames studied most likely showed aggression earlier
as well. Therefore, aggressive students may have already been rejected (or accepted
depending upon the norms discussed below) and friends and enemies had already been
more or less established. While the results showed that children on either end of the
aggression scale did change somewhat in that aggressive students became more
aggressive and vise versa, due to their position on the aggression scale (closer to the
basement or to the ceiling than other children), their change scores would by necessity be
closer to zero than those children who fell closer to the middle of the scale and thereby
showed greater fluctuation. Therefore, it seems reasonable that even if theory is correct
in that aggressive students are more likely to be rejected, then it follows that rejected
students would show little change over time. It is also assumed that these students,

unless they are exposed to some intervention, rarely have the opportunity to become

61

friends with prosocial peers. It seems unlikely that this happens naturally but rather
aggressive peers have more antisocial friends. Furthermore, in some instances it seems
reasonable that the aggressive student may even be preferred by other students than less
aggressive students and that aggression may even be socially appealing (Stormshak et al.,
1999). In other words, not all aggressive students are rejected. The convergence of these
factors could muddle the data enough to eliminate significant results.

The hypothesis that the percentage of aggressive students in the classroom, as
well as the overall climate of the classroom will impact the aggression of individual
students was fully supported in year 2 but not in year 1. As mentioned above, theory and
the makeup of the data suggest that perhaps year 2 was a better indicator than year 1.
Further study will be needed to fully appreciate the relationship between these variables
and changes in aggression. As of now, only conjectures are possible.

It seems reasonable that when a classroom reaches a certain level of aggression or
a certain number of aggressive students that aggression itself changes from being
dysfunctional to being functional for the child. No longer does following prosocial
values lead to being accepted and “surviving” in that classroom. Instead, just the
opposite becomes a reality. In order to fit in, to be become “one of the boys,” it becomes
necessary to act like the majority. In this case, aggression becomes the norm and the way
for assimilating in that environment.

This is an important idea to research further for several reasons. First, what are
the long-term effects of this deviation? If the deviation occurs early in schooling, it
seems possible that it could lead to a steadier maladaptive pathway, as compared to it

occurring later (such as early adolescence or beyond) in which case it may only be a

62

temporary deviation. Deviations left unchecked in early elementary school may carry
over to the following year as school becomes associated with negative attitudes and may
begin a cycle of negativity. Second, functional aggression within the schools seems
understudied in the literature. Often, aggression is approached as being antisocial,
negative, and dysfunctional. One could argue that aggression is all of these things, but
perhaps in some instances aggressive acts are started out of the necessity of surviving in
hostile environments. Few people want to be a victim and thus the response may be a
choice between being one or joining the majority.
Implications and Future Directions

There are several implications on current practice that can be derived from this
study. First, interventions targeting aggressive students have often focused upon the
cognitions of the aggressive child. Using a Social Information Processing approach,
some programs have found mild to moderate results in the changing of aggressive acts
over time. Two examples of violence prevention curriculums that focus on aggression
and related behaviors are Second Step and Promoting Alternative Thinking Strategies
(PATHS; Bear, Webster-Stratton, Furlong, & Rhee, 2000). While both programs have
shown to be effective in reducing aggression in the classroom during the program and at
six-month follow-ups, effect sizes have been generally small, probably due to the
complex nature of a child, an environment, and the interaction between the two. Future
research continues to be important within this area.

Second, in the area of reading ability, the Fast Track program is one example of a
comprehensive prevention program that targets reading as one of its components.

Students within the program receive reading remediation during the school year. While

63

these programs may maintain or even slightly improve a child’s reading ability, the
absence of the program during the summer months, an idea that seems critical from the
present data, may ensure the child falling behind. Keeping children learning all year
round is a difficult task that necessitates that parents help reinforce the behavior at home.

Third, believing that teacher efficacy has a strong impact on the aggression of
students leads one to desire strengthening the efficacy of teachers. If Martin, Linfoot,
and Stephenson (1999) are correct in that teachers may not be equipped to handle the vast
array of problems that arise, then better teacher education programs in classroom
management may be warranted, as well as more focused in-service programs. Giving
teachers the tools to manage their classroom may lead to higher efficacy, which in turn
may lead to better adjusted, higher achieving classrooms. Unfortunately, previous
research has not suggested that routine in-service programs influence teacher efficacy
(Ross, 1994). The mere exposure to knowledge does little without the active use of the
knowledge gained. Targeting teachers at the beginning of their careers before ideologies
become crystallized may have more influence over their overall efficacy than short,
routine programs near the end of their careers.

Because efficacy is related to experiences, it makes sense that exposure to
behavior management, especially early in a teaching career is vital to the teacher’s
preparation for dealing with behavior problems. However, of the top 10 ranked teacher
education programs in the United States, according to US. News & World Report (Morse
and Flanigan, 2005), none require students in their teacher education program to take a
course on behavior management and only one appears to cover such topics within the

context of a broader course, such as Educational Psychology. This statement is not

64

intended to reflect the author’s assessment of the top education programs in America,
only that this list may be perceived by some to be just that and that it is justifiable
therefore to scrutinize. In addition, it is believed, though not verified, that these ten
schools are in fact fairly representative of teacher education programs and that it is the
general rule of these programs to not require specific courses in behavior management.
Since behavioral efficacy is related to changes in aggression and since efficacy is related
to early experiences, it may be important for future research to examine these teacher
education practices.

In addition to this lack of a requirement in teacher education programs and in
addition to the problem of stability of efficacy, another problem lies in the prevention
programs that do exist. Most of these programs utilize some form of curriculum (a few
of which, as mentioned, have shown minimal results), but they offer little aid to teachers
dealing with specific behavioral problems. Since many of the problems teachers face on
a daily basis fall outside the realm of these programs, the program itself does little more
than a cough suppressant on lung cancer—it alleviates the symptoms for a time but in the
end it may not have helped the bigger problem. Project ACHIEVE, however, appears to
be one program that focuses on the teaching of behavior management techniques that can
be applied to many different problems (Bear, Webster-Stratton, Furlong, & Rhee, 2000).
It has been shown to be effective in reducing aggressive acts, at least on a short-term
basis. Again, more research is needed on longer-term outcomes and similar programs to
better understand the program ’5 efficacy on increasing teacher efficacy and decreasing
aggression.

All of the programs discussed (Fast Track, Second Step, Project ACHIEVE)

65

contain components of whole classroom (or school wide) interventions. While effect
sizes have been relatively small and generally the programs have been under researched,
especially longitudinally, these programs represent the first step toward the reduction of
aggression in the schools.

Attempts to correct aggression through peer mediation or peer-led interventions
have been shown to be largely ineffective (Bear, Webster-Stratton, Furlong, & Rhee,
2000). Generally, these interventions are based upon the idea that positive interactions
with prosocial peers will be beneficial. These ideas may be based upon the peer-rejection
literature that supports the idea that aggressive students tend to be rejected and thus
providing positive peer role models could be an effective intervention. As mentioned
throughout this paper, however, the basis of this line of thought may not be all together
true. Certainly some aggressive students are rejected, but just as certainly, some are very
much accepted. Either way, there are several factors that may impede the interventions
such as but not limited to apprehension of the peer mediator, dominance by the
aggressive student, or poor training of the mediator.

In addition to those areas specifically addressed above, future research should
explore teacher level variables that help further explain the variation among teachers on
measures of changes in aggression. If more of the variation can be explained than in this
study, more efforts can be made to help teachers deal with the many challenges that they
face, which in turn will help students become more successful in school.

Limitations
There are a few limitations to the current study. Some of the scales used may

not have been ideal at examining the construct they purported to measure. For example,

66

teacher academic support showed a general ceiling effect with the average score being
half a point below the maximum. Students in elementary school may have the belief that
all teachers want them to learn and do well (which may be true) and may hold onto that
belief even when there is a poor relationship with that teacher or even when they report
poor personal support from that teacher. In the child’s mind, the phrases “My teacher
really cares about me” and “My teacher likes to help me learn” are mutually exclusive
from each other. In other words, it may be difficult for children in elementary school to
not feel supported academically simply by way of what the idea of teacher stands for
even when they feel a disconnect personally with that teacher.

Second, the use of grades in elementary school is a difficult variable because of
the subjective nature of the grade itself. In other words, grades given in the elementary
years are not necessarily based upon percentages on assignments and tests (which can be
subjective as well) but rather on one teacher’s perceptions of the differences between a
grade of Excellent and Good and Satisfactory etcetera. The distances between Excellent
and Good and between Good and Satisfactory may not be the same within the same
teacher let alone between teachers. Thus it is a difficult prospect to interpret findings
based on letter grades. While access was granted to standardized test scores, the number
of missing data for those scores was high enough to make utilizing those scores
unjustifiable.

Third, especially in level two there was little power in the current study. In some
cases, as few as eighteen teachers were used to examine a multitude of variables that may
impact the changes in aggression. Because of this small power, some variables that may

in reality have a small relationship with the changes in aggression over time will fail to

67

reach significance. Further research may be necessary to truly determine these small
effects.

Finally, the same person rating her efficacy with a particular student is also the
one rating the adjustment of that student. This creates a potential for bias, even
unintentional bias, on the part of the rater. Thinking how well a student is doing may
lead to inflated efficacy scores and vice versa. It is not believed that this factor discounts
the findings, but it is acknowledged that the strength of the relationship may be somewhat
inflated.

Conclusions

In order to continue to strive for the goal set by the National Education Goals
Panel (Batsche & Knoff, 1994), it would be beneficial give teachers the help that many
are lacking. While it is important to continue researching prevention programs that have
the potential to significantly reduce aggressive acts, at least within the confines of the
school, it is ultimately the teacher who must interact with these children on a day-to-day
basis. While focusing on academic teaching may lead to controlled classrooms, the
reality appears to be that many teachers are simply not equipped to deal with the myriad
of behavior problems faced. Because teacher efficacy may stabilize somewhat over time,
waiting for Masters and In-service programs to address behavior management techniques
may simply be too late. Understanding different techniques to deal with behavioral
issues in teacher education programs and early in the teacher’s career should increase the
likelihood of successful interactions with students with behavior problems during the
critical early years of teaching. These successful interactions will be positive

reinforcement for the teacher and will lead to higher efficacy. It is no longer enough to

68

state that teachers can make a difference in the lives of children both academically and

behaviorally. Now, we need to believe it.

69

References

Aguilar, B., Sroufe, L. A., Egeland, B., & Carlson, E. (2000). Distinguishing the early-
onset/persistent and adolescent-onset antisocial behavior types: From birth to
sixteen years. Development and Psychopathology, 12, 109-132.

Akers, R. L. (1977). Deviant behvaior: A social learning approach (2nd edition).
Belmont, CA: Wadsworth Publishing Company.

Akers, R. L., Krohn, M. D., Lanza-Kaduce, L., & Radosevich, M. (1979). Social
learning and deviant behavior: A specific test of a general theory. American
Sociological Review, 44, 636-655.

Anderson, R. N., Greene, M. L., & Loewen, P. S. (1988). Relationships among teachers’
and students’ thinking skills, sense of efficacy, and student achievement. The
Alberta Journal of Educational Research, 34, 148-165.

Atlas, R. S., & Pepler, D. J. (1998). Observations of bullying in the classroom. The
Journal of Educational Research, 92(2), 86-97.

Baker, J. A. (1998). Are we missing the forest for the trees? Considering the social
context of school violence. Journal of School Psychologi, 36, 29-44.

Bandura, A. (1973). Aggression: A social learning analysis. New Jersey: Prentice Hall.

Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change.
Psychological Review, 84, 191-2 1 5.

Bandura, A. (1978). Social learning theory of aggression. Journal of Communication,
12-29.

Bandura, A. (1993). Perceived self-efficacy in cognitive development and functioning.
Educational Psychologist, 28(2), 117-148.

Batsche, G. M., & Knoff, H. M. (1994). Bullies and their victims: Understanding a
pervasive problem in the schools. School Psychology Review, 23, 165-174.

Bear, G. G., Webster-Stratton, C. Furlong, M. J ., & Rhee, S. (2000). Preventing
aggression and violence. In K. M. Minke & G. C. Bear (Eds) Preventing school
problems—Promoting school success: Strategies and programs that work (pp. 1-
70). Bethesda, MD: NASP Publications.

Birch, S. H., & Ladd, G. W. (1998). Children’s interpersonal behavior and the teacher-
child relationship. Developmental Psychology, 34, 934-946.

70

Brennan, P., Mednick, S., & Kandel, E. (1991). Congenital determinants of violent and
property offending. In D. J. Pepler & K. H. Rubin (Eds.), The development and
treatment of childhood aggression (pp. 81-92). Hillsdale, NJ: Erlbaum.

Bronfenbrenner, U. (1979). The ecology of human development: Experiments by nature
and design. Cambridge: Harvard University Press.

Brownell, M. T., & Pajares, F. (1999). Teacher efficacy and perceived success in
mainstreaming students with learning and behavior problems. Teacher Education
and Special Education, 22(3), 154-164.

Calkins, S. D., & Fox, N. A. (2002). Self-regulatory processes in early personality
development: A multilevel approach to the study of childhood social withdrawal
and aggression. Development and Psychopathology, 14, 477-498.

Carden Smith, L. K., & Fowler, S. A. (1984). Positive peer pressure: The effects of peer
monitoring on children’s disruptive behavior. Journal of Applied Behavior
Analysis, 17, 213-227.

Catalano, R. F., & Hawkins, J. D. (1996). The social development model: A theory of
antisocial behavior. In J. D. Hawkins (Ed.) Delinquency and Crime: Current
Theories (pp. 149-193). New York: Cambridge University Press.

Cicchetti, D. & Richters, J. E. (1993). Developmental considerations in the investigation
of conduct disorder. Development and Psychopathology, 5, 331-344.

Coie, J. D., Lochman, J. E., Terry, R., & Hyman, C. (1992). Predicting early adolescent
disorder from childhood aggression and peer rejection. Journal of Consulting and
Clinical Psychology, 5, 783-792.

Compas, B. E., Hinden, B. R., & Gerhardt, C. A. (1995). Adolescent development:
Pathways and processes of risk and resilience. Annual Review of Psychology, 4 6,
265-293.

Conduct Problems Prevention Research Group. (1992). A developmental and clinical
model for the prevention of conduct disorder: The FAST track program.
Development and Psychopathology, 4, 509-527.

Conduct Problems Prevention Research Group. (1999). Initial impact of the fast track
prevention trial for conduct problems: 11. classroom effects. Journal of Consulting
and Clinical Psychology, 67, 648-657.

Conduct Problems Prevention Research Group. (1999). Initial impact of the fast track
prevention trial for conduct problems: I. The high-risk sample. Journal of

Consulting and Clinical Psychology, 6 7, 631-647.

71

Conduct Problems Prevention Research Group. (2002a). The implementation of the fast
track program: An example of a large-scale prevention science efficacy trial.
Journal of Abnormal Child Psychology, 30, 1-17.

Conduct Problems Prevention Research Group. (2002b). Evaluation of the first three
years of the fast track prevention trial with children at high risk for adolescent
conduct problems. Journal of Abnormal Child Psychology, 30, 19-35.

Conduct Problems Prevention Research Group. (2002c). Predictor variables associated
with positive fast track outcomes at the end of the third grade. Journal of Abnormal
Child Psychology, 30, 37-52.

Conger, R. D. (1976). Social control and social learning models of delinquent behavior:
A synthesis. Criminology, 14, 17-40.

Costello, E. J ., & Arnold, A. (1993). Toward a developmental epidemiology of the
disruptive behavior disorders. Development and Psychopathology, 5, 91-101.

Cote, S., Zoccolillo, M., Tremblay, R. E., Nagin, D., & Vitaro, F. (2001). Predicting
girls’ conduct disorder in adolescence from childhood trajectories of disruptive
behaviors. Journal of the American Academy of Child and Adolescence Psychiatry,
40, 678-684.

Cowen, E. L., & Work, W. C. (1988). Resilient children, psychological wellness, and
primary prevention. American Journal of Community Psychology, 16, 591-607.

Crick, N. R., & Dodge, K. A. (1994). A review and reformulation of social information-
processing mechanisms in children’s social adjustment. Psychological Bulletin,
115, 74-101.

Deater—Deckard, K. (2001). Annotation: Recent research examining the role of peer
relationships in the development of psychopathology. Journal of Child Psychology
and Psychiatry, 42, 565-579.

DeRosier, M. E., Cilessen, A. H. N., Coie, J. D., & Dodge, K. A. (1994). Group social
context and children’s aggressive behavior. Child Development, 65, 1068-1079.

Dodge, K. A. (1993). Social-cognitive mechanisms in the development of conduct
disorder and depression. Annual Review of Psychology, 44, 559-584.

Doll, B., & Lyon, M. A. (1998). Risk and resilience: Implications for the delivery of
educational and mental health services in schools. School Psychology Review, 27,
348-363.

Eccles, J. S., Midgley, C., Wigfield, A., Buchanan, C. M., Reuman, D., Flanagan, C., &
Iver, D. M. (1993). Development during adolescence: The impact of stage-

72

environment fit on young adolescents’ experiences in schools and in families.
American Psychologist, 49, 90-101.

Elliott, D. S. (1994). Serious violent offenders: onset, developmental course, and
termination—The American Society of Criminology 1993 presidential address.
Criminology, 32, l- 21.

Evans, E. D., & Tribble, M. (1986). Perceived teaching problems, self-efficacy, and
commitment to teaching among preservice teachers. Journal of Educational
Research, 80(2), 81-85.

Farmer, T. (2000). Misconceptions of peer rejection and problem behavior:
Understanding aggression in students with mild disabilities. Remedial and Special
Education, 21, 194-208.

Farrington, D. P. (1992). Explaining the beginning, progress, and ending of antisocial
behavior from birth to adulthood. In F. Adler, W. Laufer (Series Eds.) & J. McCord
(Vol. Ed.) Advances in Criminological Theory: Vol. 3. Facts, frameworks, and
forecasts (pp. 253-286). New Brunswick: Transaction Publishers.

F arver, J. M. (1996). Aggressive behavior in preschoolers’ social networks: Do birds of a
feather flock together? Early Childhood Research Quarterly, 11, 333-350.

Forrnan, S. G., & Kalafat, J. (1998). Substance abuse and suicide: Promoting resilience
against self-destructive behavior in youth. School Psychology Review, 27, 398-406.

Fowler, S. A. (1986). Peer-monitoring and self-monitoring: Alternatives to traditional
teacher management. Exceptional Children, 6, 573-581.

Fraser, M. W., Richman, J. M., & Galinsky, M. J. (1999). Risk, protection, and
resilience: Toward a conceptual framework for social work practice. Social Work
Research, 23, 131-143.

Funnan, W., & Gavin, L. A. (1989). Peers’ influence on adjustment and development: A
view from the intervention literature. In T. J. Berndt & G. W. Ladd (Eds.) Peer
relationships in child development (pp. 319-340). New York: Wiley-Interscience
Publications.

Gibson, 8., & Dembo, M. H. (1984). Teacher efficacy: A construct validation. Journal
of Educational Psychology, 76, 569-582.

Gorski, J. D., & Pilotto, L. (1993). Interpersonal violence among youth: A challenge for
school personnel. Educational Psychology Review, 5, 35-61.

Greenberg, M. T., Kusche, C., & Milhalic, S. F. (1998). Blueprints for violence
prevention, book ten: Promoting Alternative Thinking Strategies (PA T HS).

73

Boulder, CO: Center for the Study of Prevention and Violence.

Guskey, T. R. (1982). Differences in teachers’ perceptions of personal control of positive
versus negative student learning outcomes. Contemporary Educational
Psychology, 7, 70-80.

Hamre, B. K., & Pianta, R. C. (2001). Early teacher-child relationships and trajectory of
children’s school outcomes through eighth grade. Child Development, 72, 625-638.

Hellman, D. A., & Beaton, S. (1986). The pattern of violence in urban public schools:
The influence of school and community. Journal of Research in Crime and
Delinquency, 23, 102-127.

Herrenkohl, T. 1., Guo, J., Kosterman, R., Hawkins, J. D., Catalano, R. F ., & Smith, B. H.
(2001). Early adolescent predictors of youth violence as mediators of childhood
risk. Journal of Early Adolescence, 21, 447-469.

Herrenkohl, T. 1., Huang, B., Kosterman, R., Hawkins, J. D., Catalano, R. F ., & Smith, B.
H. (2001). A comparison of social development processes leading to violent
behavior in late adolescence for childhood initiators and adolescent initiators of
violence. Journal of Research in Crime and Delinquency, 38, 45-63.

Herrenkohl, T. 1., Maguin, E., Hill, K. G., Hawkins, J. D., Abbott, R. D., & Catalano, R.
F. (2000). Developmental risk factors for youth violence. Journal of Adolescent
Health, 26, 176-186.

Hinshaw, S. H. (2002). Prevention/intervention trials and developmental theory:
Commentary on the fast track special section. Journal of Abnormal Child
Psychology, 30, 53-59.

Hinshaw, S. P. (2002). Process, mechanism, and explanation related to externalizing
behavior in developmental psychopathology. Journal of Abnormal Child
Psychology, 30, 431-446.

Hinshaw, S. P., & Anderson, C. A. (1996). Conduct and oppositional defiant disorders.
In E. J. Mask & R. A. Barkley (Eds.) Child psychopathology (pp. 113-149). New
York: The Guilford Press.

Huang, 3., Kosterman, R., Catalano, R. F ., Hawkins, J. D., & Abbott, R. D. (2001).
Modeling mediation in the etiology of violent behavior in adolescence: A test of the
social development model. Criminology, 39, 75-107.

Hughes, J. N., Cavell, T. A., & Jackson, T. (1999). Influence of the teacher-student
relationship on childhood conduct problems: A prospective study. Journal of
Clinical Child Psychology, 28, 173-184.

74

Hymel, S., Bowker, A., & Woody, E. (1993). Aggressive versus withdrawn unpopular
children: Variations in peer and self-perceptions in multiple domains. Child
Development, 64, 879-896.

Jacobson, K. C., Prescott, C. A., & Kendler, K. S. (2002). Sex differences in the genetic
and environmental influences on the development of antisocial behavior.
Development and Psychopathology, 14, 395-416.

Jessor, R., Van Den Bos, J., Vanderryn, J ., Costa, F. M., & Turbin, M. S. (1995).
Protective factors in adolescent problem behavior: Moderator effects and
developmental change. Developmental Psychology, 31, 923-933.

Kazdin, A. E., & Kagan, J. (1994). Models of dysfimction in developmental
psychopathology. Clinical Psychology: Science and Practice, 1, 35-52.

Kellam, S. G., Rebok, G. W., Ialongo, N., & Mayer, L. S. (1994). The course and
malleability of aggressive behavior from early first grade into middle school:
Results of a developmental epidemiologically-based preventive trial. Journal of
Child Psychology and Psychiatry, 35, 259-281.

Kolvin, 1., Miller, F. J. W., Fleeting, M., & Kolvin, P. A. (1988). Social and parenting
factors affecting criminal-offence rates: Findings from the Newcastle Thousand
Family Study (l947-l980_. British Journal of Psychiatry, 152, 80-90.

Ladd, G. W. (1996). Shifting ecologies during the 5 to 7 year period: Predicting
children’s adjustment during the transition to grade school. In A. Sameroff & M.
Haith (Eds), The five to seven year shift: The age of reason and responsibility (pp.
363-386). Chicago: University of Chicago Press.

Ladd, G. W., & Burgess, K. B. (2001). Do relational risks and protective factors
moderate the linkages between childhood aggression and early psychological and
school adjustment? Child Development, 72, 1579-1601.

Ladd, G. W., & Price, J. M. (1987). Predicting children’s social and school adjustment
following the transition from preschool to kindergarten. Child Development, 58,
1168-1189.

Ladd, G. W., Kochenderfer, B. J ., & Coleman, C. C. (1996). Friendship quality as
predictor of young children’s early school adjustment. Child Development, 6 7,
1103-1118.

Lago-Delello, E. (1998). Classroom dynamics and the development of serious emotional
disturbance. Exceptional Children, 64, 479-492.

Lahey, B. B., Loeber, R., Burke, J., & Rathouz, P. J. (2002). Adolescent outcomes of
75

childhood conduct disorder among clinic-referred boys: Predictors of improvement.
Journal of Abnormal Child Psychologi, 30, 333-348.

Larson, J. (1998). Managing student aggression in high schools: Implications for
practice. Psychology in the Schools, 35, 283-295.

Loeber, R. & Farrington, D. P. (2000). Young children who commit crime:
Epidemiology, developmental origins, risk factors, early interventions, and
implications. Development and Psychopathology, 12, 737-762.

Loeber, R. (1990). Development and risk factors of juvenile antisocial behavior and
delinquency. Clinical Psychology Review, 10, 1-41.

Loeber, R. (1991). Antisocial behavior: More enduring than changeable? Journal of the
American Academy of Child and Adolescent Psychiatry, 30, 393-397.

Loeber, R., Farrington, D. P., Stouthamer—Loeber, M., Moffitt, T. E., & Caspi, A. (1998).
The development of male offending: Key findings from the first decade of the
Pittsburgh Youth Study. Studies on Crime and Crime Prevention, 7(2), 141-171.

Loeber, R., Green, S. M., Lahey, B. B., Christ, M. G., & Frick, P. J. (1992).
Developmental sequences in the age of onset of disruptive child behaviors. Journal
of Child and Family Studies, 1, 21-41.

Loeber, R., Wung, P., Keenan, K., Giroux, B., stouthamer-Loeber, M., Van Kammen, W.
B., & Maughan, B. (1993). Developmental pathways in disruptive child behavior.
Development and Psychopathology, 5, 101-133.

Martin, A. J ., Linfoot, K., & Stephenson, J. (1999). How teachers respond to concerns
about misbehavior in their classroom. Psychology in the Schools, 36, 347-358.

Meijer, C. J ., & Foster, S. F. (1988). The effect of teacher efficacy on referral chance.
The Journal of Special Education, 22, 378-385.

Merydith, S. P. (2001). Temporal stability and convergent validity of the Behavior
Assessment System for Children. Journal of School Psychology, 39, 253-265.

Mesman, J ., Bongers, I. L., & Koot, H. M. (2001). Preschool developmental pathways to
preadolescent internalizing and externalizing problems. Journal of Child
Psychology and Psychiatry, 42, 678-689.

Miller, G. E., Brehm, K., & Whitehouse, S. (1998). Reconceptualizing school-based
prevention for antisocial behavior within a resiliency framework. School
Psychology Review, 27, 364-379.

Minuchin, P. P., & Shapiro, E. K. (1983). The school as a context for social
76

development. In P. H. Mussen (Series Ed.) & E. M. Hetherington (Vol. Ed.),
handbook of child psychology: Vol. 4. Socialization, personality, and social
development (4th ed., pp. 197-274). New York: Wiley.

Morse, R. J., & F lanigan, S. M. (2005). America’s best colleges 2005. US. News 7
World Report.

Murphy, A. E. (1987). John, a boy who acquired resilience. In E. J. Anthony & B. J.
Cohler (Eds) The invulnerable child (pp. 106-145). New York: The Guilford
Press.

Murray, C., & Greenberg, M. T. (2000). Children’s relationships with teachers and
bonds with school: An investigation of patterns and correlates in middle childhood.
Journal of School Psychology, 38, 423-445

O’Donnell, J ., Hawkins, J. D., & Abbott, R. D. (1995). Predicting serious delinquency
and substance abuse among aggressive boys. Journal of Consulting and Clinical
Psychology, 63, 529-537.

O’Grady, D., & Metz, J. R. (1987). Resilience in children at high risk for psychological
disorder. Journal of Pediatric Psychology, 12, 3-23.

Olweus, D. (1994). Annotation: Bullying at school: Basic facts and effects of a school
based intervention program. Journal of Child Psychology and Psychiatry, 35,
1171-1190.

Parke, R. D., & Slaby, R. G. (1983). The development of aggression. In P. H. Mussen
(Series Ed.) & E. M. Hetherington (Vol. Ed.), handbook of child psychology: Vol. 4.
Socialization, personality, and social development (4th ed., pp. 547-641). New
York: Wiley.

Parker, J. G., & Asher, S. R. (1987). Peer relations and later personal adjustment: Are
low-accepted children at risk? Psychological Bulletin, 102, 357-389.

Parkhurst, J. T., & Asher, S. R. (1992). Peer rejection in middle school: Subgroup
differences in behavior, loneliness, and interpersonal concerns. Developmental
Psychology, 28, 231-241.

Patterson, G. R., Capaldi, D., & Bank, L. (1991). An early starter model for predicting
delinquency. In D. J. Pepler & K. H. Rubin (Eds.), The development and treatment
of childhood aggression (pp. 139-168). Hillsdale, NJ: Erlbaum.

Patterson, G. R., DeBaryshe, B. D., & Ramsey, E. (1989). A developmental perspective
on antisocial behavior. American Psychologist, 44, 329-335.

Perry, K. E., & Weinstein, R. S. (1998). The social context of early schooling and
children’s school adjustment. Educational psychologist, 33, 177-194.

77

Phillips, G. W., & Greenfeld, L. A. (2002). Indicators of school crime and safety.
Retrieved on November 18, 2002, from http://www.ojp.usdoj.gov/bjs/pub/pdf/
isc502.pdf

Pianta, R. C., Steinberg, M. S., & Rollins, K. B. (1995). The first two years of school:
Teacher-child relationships and deflections in children’s classroom adjustment.
Development and Psychopathology, 7, 295-312.

Podell, D. M., & Soodak, L. C. (1993). Teacher efficacy and bias in special education
referrals. Journal of Educational Research, 86(4),247-253.

Prinz, R. J. (2002). The fast track project: A seminal intervention efficacy trial. Journal
of Abnormal Child Psychology, 30, 61-63.

Rosenfield, P., Lambert, N. M., & Black, A. (1985). Desk arrangement effects on pupil
classroom behavior. Journal of Educational Psychology, 77, 101-108.

Rutter, M. (1985). Resilience in the face of adversity: Protective factors and resistance to
psychiatric disorder. British Journal of Psychiatry, 147, 598-611.

Rutter, M., & Sroufe, L. A. (2000). Developmental Psychopathology: Concepts and
Challenges. Development and Psychopathology, 12, 265-296.

Sameroff, A. J. (2000). Developmental systems and psychopathology. Development and
Psychopathology, 12, 297-312.

Seifer, R. & Sameroff, A. J. (1987). Multiple determinants of risk and invulnerability. In
E. J. Anthony & B. J. Cohler (Eds) The invulnerable child (pp. 51-70). New York:
The Guilford Press.

Sines, J. O. (1987). Influence of the home and family environment on childhood
dysfunction. In B. B. Lahey & A. E. Kazdin (Eds) Advances in clinical child
psychology (Vol. 10, pp. 1-44). New York: Plenum Press.

Skinner, C. H., Neddenriep, C. E., Robinson, S. L., Ervin, R., & Jones, K. (2002).
Altering educational environments through positive peer reporting: Prevention and
remediation of social problems associated with behavior disorders. Psychology in
the Schools, 39, 191-202.

Snyder, H. N., & Sickmund, M. (1999). Juvenile offenders and victims: 1999 national
report. Retrieved November 18, 2002, from http://www.ncjrs.org/html/ojjdp/
nationalreport99/toc.html

Soodak, L. C., & Podell, D. M. (1993). Teacher efficacy and student problem as factors
in special education referral. The Journal of Special Education, 27, 66-81.

78

Storrnshak, E. A., Bierrnan, K. L., Bruschi, C., Dodge, K. A., Coie, J. D., & The Conduct
Problems Prevention Research Group (1999). The relation between problems and
peer preference in different classroom contexts. Child Development, 70, 169-182.

Tomblin, J. B., Zhang, X., Buckwalter, P, & Catts, H. (2000). The association of reading
disability, behavioral disorders, and language impairment among second-grade
children. Journal of Child Psychology & Psychiatry, 41, 473-482.

Tremblay, R. E., Masse, L. C., Pagani, L., & Vitaro, F. (1996). From childhood
aggression to adolescent maladjustment: The Montreal Prevention Experiment. In
Preventing childhood disorders, substance abuse and delinquency (pp. 268-298).
Thousand Oaks, CA: Sage.

Tschannen-Moran, M., Hoy, A. W., & Hoy, W. K. (1998). Teacher efficacy: Its meaning
and measure. Review of Educational Research, 68(2), 202-248.

Vance, J. E., Bowen, N. K., Fernandez, G. & Thompson, S. (2002). Risk and protective
factors as predictors of outcomes in adolescents with psychiatric disorder and
aggression. Journal of the American Academy of Child and Adolecent Psychiatry,
41, 36-43.

Webster-Stratton, C. (1993). Strategies for helping early school-aged children with
oppositional defiant and conduct disorders: The importance of home-school
partnerships. School Psychology Review, 22, 437-457.

Wheldall, K. (1991). Managing troublesome classroom behavior in regular schools: A
positive teaching perspective. International Journal of Disability, Development,
and Education, 38, 99-116.

Willett, J. (1988). Questions and answers in the measurement of change. In Rothkopf
(Ed.), Review of research in education (pp. 224-241). Washington, DC: American
Educational Research Association.

Woolfolk, A. E., & Hoy, W. K. (1990). Prospective teachers’ sense of efficacy and
beliefs about control. Journal of Educational Psychology, 82, 81-91.

Woolfolk, A. E., Rosoff, B., & Hoy, W. K. (1990). Teachers’ sense of efficacy and their
beliefs about managing students. Teaching and Teacher Education, 6(2), 137-148.

Yoshikawa, H. (1994). Prevention as cumulative protection: Effects of early family

support and education on chronic delinquency and its risks. Psychological Bulletin,
115, 28-54.

79

 

 

     

   

STAI} UNIVEHMH I

111111111Mill

3 1293 02736 1157