Publications>
Research Bulletin

Phi Delta Kappa Center for Evaluation, Development, and Research
December 1999, No. 25

The Influence of the First-Grade Classroom on the Development of Aggressive Behavior
by Dr. Sheppard G. Kellam

The start of school is a major transition for children. The new context of the first-grade classroom presents great challenges. The teacher poses new demands, such as sitting still and paying attention. Getting along with classmates presents another set of challenges. When children respond to such demands aggressively or disruptively, even as early as first grade, such responses increase the risk of later aggressive behavior, conduct disorders, and drug abuse, particularly when coupled with shy or socially isolating behavior.1

The social context of the early classroom may play an important role in mediating and moderating the developmental course of children's aggressive behavior. This bulletin reports on a study that examined the influences of the first-grade classroom context on the course and malleability of children's aggressive behavior from entrance into first grade through the transition into middle school.2 The study also investigated the effects of a preventive intervention, implemented over first and second grades, on the course of children's aggressive behavior.

For purposes of this study, the conceptualization of normal and pathologic development and the choice of a preventive intervention were guided by life course/social field theory.3 This theory proposes that at each stage of life there are main social fields in which individuals are faced with social task demands. Individuals are formally or informally rated on the adequacy of their performance by natural raters, such as parents, teachers, and peers. While characteristics of the social field may enhance or inhibit adequate responses, each individual brings unique assets or liabilities in his or her adaptive capacity to respond adequately to social task demands. Early successful social adaptation tends to provide social, emotional, and cognitive competencies through reinforcement by natural raters of successful performance of social task demands, and these competencies remain accessible to the individual for later adaptation in new social fields. Early maladaptation results in a lack of competencies necessary for later social adaptation and psychological well being and reinforces one's sense of failure and/or alienation.

Based on the tenets of life course/social field theory, it is hypothesized that the classroom context markedly influences the developmental course of aggressive behavior. Aggressive, disruptive behavior has been shown to be influenced by other children in a peer group.4 In one study, aggressive youths in groups with other aggressive youths laughed at rule breaking and generally reinforced aggressive, disruptive behavior.5 In mixed groups of aggressive and nonaggressive youths, on the other hand, reinforcement was associated with prosocial behavior.

Early in elementary school, the social fields of the classroom and peer group overlap, and the social task demands of the two fields may be in conflict with each other, for example, if classmates are reinforcing aggressive, disruptive behavior while the teacher is trying to establish a classroom environment conducive to learning. A child who responds maladaptively to a teacher's demands for prosocial behavior may be at even greater risk for long-term aggressive behavior if he or she begins schooling in a classroom in which most other children are also aggressive compared to children with the same level of aggressive behavior who are in a classroom where most other children are less aggressive. The risk may be lower if he or she is in a classroom where most other children are adapting to teacher's social task demands.

The social field of the classroom and its social task demands by the teacher and classmates either enhance or inhibit the developmental paths of children with early aggressive responses to the teacher. In addition, high rates of aggressive behavior may cause teachers to react in ineffective ways, thereby reinforcing students' disruptive behavior. An intervention that improves children's social adaptation to teachers' and classmates' demands may diminish the risk of continuing aggression through early adolescence.

In this study, school and family influences were accounted for to evaluate the specific contributions of the classroom environment to a child's course of development. Also, because poverty may play an important role in the developmental course of aggressive behavior, the effects of school-level and individual poverty on aggressive behavior were considered.

DESIGN

The study investigated three questions: 1) What is the extent of variation in classroom aggressive, disruptive behaviors at the classroom and school levels? 2) What is the influence of the first-grade classroom on the course of aggressive behavior into middle school? and 3) What is the effect of a preventive intervention, the Good Behavior Game, on the course of aggressive behavior? The role of poverty was examined in relationship to the first two questions.

This was a longitudinal study that took place from the 1985-86 school year to the 1991-92 school year. At the beginning of the study, participants were 1,084 children enrolled in first grade in 18 Baltimore City Public Schools.6 By the end of first grade, 864 of these children were still participating in the study. At the end of the study six years later, 682 children were still enrolled in the city's middle schools. The population was 49% male, 64% African American, 29% white, and 7% other ethnic groups. At the beginning of the study, the average age of participants was 6 years, 4 months. All 40 first-grade teachers from the 18 schools agreed to participate in the study.

The 18 schools were in five urban areas representing a wide range of racial and ethnic make-up, income levels, types of housing, integrated and segregated neighborhoods, and extended and nuclear family living arrangements. In each of the five areas, three or four schools were matched, and each was randomly assigned to participate in one of two interventions -- a classroom management strategy called the Good Behavior Game or a reading curriculum -- or the regular school setting, that is, the standard school curriculum and the teacher's own classroom management strategy. Schools, teachers within schools, and children were randomly assigned to an intervention or to the regular school setting. Membership in the same classroom was maintained through the first two years of school, when the intervention programs were used.7

The Good Behavior Game is a team-based behavior management strategy that promotes prosocial behavior by rewarding teams of students whose members do not exhibit maladaptive behaviors during precisely defined periods when the game is "played."8 The goal of the strategy is to create classrooms with integrated social systems supportive of all children being able to learn. The teachers who were assigned to this intervention received 40 hours of training. Teachers assigned children to one of three teams after baseline measurements of behavior were established. Teachers determined team membership, making sure that teams had equal numbers of boys and girls, equal numbers of aggressive, disruptive children, and equal numbers of shy, socially isolated children.

Teachers clearly defined unacceptable behaviors and reinforced children for regulating their own and their teammates' behavior. Disruptive behaviors included shouting out, jumping out of one's seat, fighting, breaking things, teasing, and taking other children's possessions. (These behaviors were reflected in the measures of aggressive disruptive behavior used in the study.) The children decided with the teacher what the rewards were to be. At first rewards were such concrete items as colored pencils or erasers. Later rewards were more abstract, such as gold stars or marks on charts posted in the classroom. Winning teams were announced each week, and all teams could win during a particular week.

During the first weeks of the study, students played the game three times each week for 10 minutes. Over time, children played the game at different times of the day and during different activities. Over successive weeks the duration of the game increased until it was incorporated into the classroom environment at the end of the school year. At first, the teacher announced game periods and delivered the rewards immediately after the game. Later, the teacher initiated the game period without announcement and the rewards were delayed until the end of the school day or week.

Each student's level of aggression was rated using three types of measures: 1) teacher observations and ratings of each child's performance on the core tasks in the classroom; 2) peer nominations in which each child could nominate each of the other children in the classroom for a particular attribute, such as "fights," "gets into trouble," and "acts mean to others"; and 3) independent behavior observations based on time sampling in the classrooms, with independent observers noting the following behaviors: not complying, disrupting verbally, teasing, turning in one's seat, being out of one's seat without permission, taking others' property, damaging property, and attacking others. Teachers rated students in the fall and the spring of the first and second grades and in the spring of each year thereafter. Children completed peer nominations at the same time that teachers rated students, and independent observations were made before and after the other ratings in the fall and spring.

RESULTS

What is the extent of variation in classroom aggressive, disruptive behaviors at the classroom and school levels? The mean of each teacher's ratings for the children in her or his classroom was used to characterize a classroom in regard to level of aggressive behavior. Mixed model analyses were used to determine variation in aggressive behavior within and across classrooms and schools. Although there was some variation across schools, F = 1.91, p = .076, the more significant level of variation in aggressive behavior was among classrooms within schools, F = 4.98, p < .001.

Classrooms were then categorized as having either a lower or a higher aggressive level, with 24 classrooms identified as lower and 16 as higher. Results also indicated that there was a significant gender difference in the level of first-grade aggression, F = 71.72, p < .001, with boys more aggressive (M = 2.07) than girls (M = 1.65).

What is the influence of the first-grade classroom on the course of aggressive behavior into middle school? First-grade boys who were rated in the top quartile as aggressive and were in a higher aggressive first-grade classroom were far more likely -- as high as 59 times -- to be rated as highly aggressive in middle school compared to the average boys. (Boys who scored one standard deviation beyond the mean of English teachers' ratings compared to the total control population in sixth grade were considered to be highly aggressive in middle school.) First-grade boys who were rated in the top quartile as aggressive and were in a lower aggressive classroom were somewhat more likely to be aggressive in middle school, but much less at risk -- 2.7 times as likely to be aggressive.

For girls, only 18 were classified as highly aggressive among 201 girls in the analysis, making any inferences exploratory. The risk of being highly aggressive in middle school was not enhanced by high levels of first-grade classroom aggressive behavior.

What is the effect of the Good Behavior Game on the course of aggressive behavior? In other studies, the Good Behavior Game has been shown to decrease children's levels of aggressive behavior by middle school and curtail the initial use of tobacco. In the present study, there is suggestive evidence that the Good Behavior Game was effective in reducing the effect of the level of classroom aggression on the risk for aggressive boys in higher aggressive classrooms, though uncertainty remains in this inference. The game did not alter the course of aggressive behavior for girls.

The role of poverty. To determine the role of poverty on children's aggressive behavior, the percentage of children receiving free lunches in each classroom was used to measure classroom poverty levels. A child's individual measure of poverty was whether he or she was receiving free lunch. Whereas variation was found in levels of classroom aggressive behavior within schools, there was no variation among poverty levels of classrooms within schools, only across schools. Within each school there was consistency, with all classrooms within the school being either higher or lower in the percentage of children receiving free lunches.

First-grade boys' individual poverty level was associated with higher risks of being more aggressive and disruptive in the first grade. This increased their vulnerability to the level of aggression in the classroom as a whole. For both boys and girls, school/classroom poverty in first grade significantly predicted aggressiveness six years later, regardless of the children's first-grade baseline ratings of aggressive behavior. Individual poverty was found to affect aggressive behavior indirectly, in that the number of aggressive first-graders increased when individual poverty was taken into account.

DISCUSSION

The first-grade classroom environment and poverty appear to influence the course of the development of aggressive behavior, especially that of aggressive boys. Classroom aggression levels appear to affect aggressive boys but not girls, while classroom/school poverty affects all children.

Variation in aggressive behavior and in its impact was primarily at the classroom level and less so at the school level. Although children were randomly assigned to classrooms at the beginning of first grade, differences in levels of aggressive behavior at the classroom level emerged as early as the first quarter of first grade. This suggests that the origins of variation in classroom aggressive levels came from the classroom teacher or the mix of students and the teacher.

How might the classroom environment have this effect on more aggressive boys? It is possible that the skills of the teachers in highly aggressive, disruptive classrooms were not sufficient to promote an integrated social system. Effective classroom behavior management appears to be essential in socializing young children. For boys whose initial behavior is aggressive and disruptive, teacher training in effective classroom behavior management strategies is critical.

Unfortunately, teacher training typically does not provide effective methods and experience in classroom behavior management. This, coupled with inadequate staff support, places some children at greater risk of later conduct disorders and related academic problems. This inference should not be seen as placing blame on teachers. The teachers who participated in this study were dedicated and fully committed, but many did not have the resources to integrate the social system of their classroom. Thus the policy implications are that teacher's colleges and in-service training need to include specific training in classroom behavior management as an important part of the socialization role of the classroom.

The child's maladaptive, aggressive response to the teacher's demand to sit still, pay attention, and learn may depend on family processes that precede and parallel the process of social adaptation to the classroom. When the child is in a classroom with aggressive and disruptive classmates, other children are likely to reinforce this maladaptive behavioral response. This can escalate and result later in a higher risk of severe aggressive, disruptive behavior leading to high levels of conduct problems and a high risk of dropping out of school.

The classroom data were based on first-grade assessments, yet the findings persisted through the transition into middle school. Perhaps the experience of the aggressive child in an aggressive first-grade classroom sets the pattern for the child's behavioral responses, poor achievement, membership in similarly behaving peer groups, and detachment from school.

Another policy implication relates to the practice of tracking and ability grouping, which has been documented to result in aggressive, disruptive classrooms. The reasons often provided for such practices are that homogeneous classrooms are easier to teach and require fewer reading groups, especially where teachers have few resources and large classes. Yet, the findings of this study indicate that placing many aggressive boys together in one classroom may amount to socializing children directly toward antisocial outcomes, leading to very serious aggressive behavior later.

There is evidence that that the Good Behavior Game helped to reduce children's aggressive, disruptive behavior. This suggests that applying more precise classroom behavior management methods can reduce the impact of aggressive classrooms on the development of aggressive behavior. The classroom context is vitally important socially, and it is also malleable, giving us hope that severe aggressive behavior over the course of a child's development can be prevented if teachers are provided with effective methods of classroom behavior management.

ENDNOTES

1. J. Block, J.H. Block, and S. Keyes, "Longitudinally Foretelling Drug Usage in Adolesence: Early Childhood Personality and Environmental Precursors," Child Development 59 (1988): 336-55; M.E. Ensminger, S.G. Kellan, and B.R. Rubin, "School and Family Origins of Delinquency: Comparisons by Sex," in Prospective Studies of Crime and Delinquency, eds. K.T. Van Dusen and S.A. Mednick (Boston: Kluwer-Nijhoff, 1983), 73-97; D.P. Farrington et al., "Are There Successful Men from Criminogenic Backgrounds?" Psychiatry 51 (1988): 116-30; D.P. Farrington and J. Gunn, Aggression and Dangerousness (New York: Wiley, 1985); S.L. Hans et al., "Interpersonal Behavior of Children at Risk for Schizophrenia," Psychiatry: Interpersonal and Biological Processes 55 (1992): 314-35; S.G. Kellam et al., "Paths Leading to Teenage Psychiatric Symptoms and Substance Use: Developmental Epidemiological Studies in Woodlawn," in Childhood Psychopathology and Development, eds. S.B. Guze, F.J. Earls, and J.E. Barrett (New York: Raven Press, 1983), 17-51; J. McCord, "Parental Behavior in the Cycle of Aggression," Psychiatry 51 (1988): 14-23; L.N. Robins, "Sturdy Childhood Predictors of Adult Antisocial Behavior: Replications from Longitudinal Studies," Psychological Medicine 8 (1978): 611-22; A.E. Schwartzman, J.E. Ledingham, and L.A. Serbin, "Identification of Children at Risk for Adult Schizophrenia: A Longitudinal Study," International Review of Applied Psychology 34 (1985): 363-80; J. Shedler and J. Block, "Adolescent Drug Use and Psychological Health: A Longitudinal Study," Annual Progress in Child Psychiatry and Child Development (1991): 545-84.

2. See S.G. Kellam et al., "Effects of the Level of Aggression in the First-Grade Classroom on the Course and Malleability of Aggressive Behavior into Middle School," Development and Psychopathology 10 (1998): 165-85. This study was supported in part by a grant from the National Institute of Mental Health.

3. S.G. Kellam et al., Mental Health and Going to School: The Woodlawn Program of Assessment, Early Intervention and Evaluation (Chicago: University of Chicago Press, 1975); S.G. Kellam and M.E. Ensminger, "Theory and Methods in Child Psychiatric Epidemiology," in Studying Children Epidemiologically, ed. F. Earls, vol. 1 of International Monograph Series in Psycho-Social Epidemiology (New York: Neale Watson, 1980), 145-80; and S.G. Kellam and G.W. Rebok, "Building Developmental and Etiological Theory Through Epidemiologically Based Preventive Intervention Trials," in Preventing Antisocial Behavior: Interventions from Birth to Adolescence, eds. J. McCord and R. E. Tremblay (New York: Guilford, 1992), 162-95.

4. V. Battistich et al., "Schools as Communities, Poverty Levels of Student Populations, and Students' Attitudes, Motives, and Performance: A Multilevel Analysis," American Educational Research Journal 32 (1995): 627-58.

5. T. Dishion et al., "Deviancy Training in Male Adolescent Friendships," Behavior Therapy 27 (1996): 370-90.

6. Gratitude is expressed for the support and participation of the administrators and teachers of the Baltimore City Public Schools and the parents of the children who participated in the study.

7. This paper examines the effects of the Good Behavior Game on children's aggressive, disruptive behavior. Work continues on the effects of the reading curriculum on children's aggressive behavior.

8. For information about the Good Behavior Game, contact Dr. Sheppard Kellam, Center for Prevention Research, Mason Lord Building, Suite 500, 5200 Eastern Avenue, Baltimore, Maryland 21224 (410-550-3445).