Handbook of psychology volume 7 educational psychology

DEVELOPMENTAL SYSTEMS THEORY

In the last two decades, views that embrace the perspective

that the study of development is in large part the study of

living systems and is therefore informed by the study of sys-

tems have been adopted as the primary conceptual paradigm

in human development (see Lerner, 1998, for example). As

noted by Lerner (1998), “a developmental systems perspec-

tive is an overarching conceptual framework associated with

contemporary theoretical models in the field of human devel-

opment” (p. 2). General systems theory has a long history

in the understanding of biological, ecological, and other

complex living systems (e.g., Ford & Ford, 1987; Ford &

Lerner, 1992) and has been applied to child development by

Ford and Lerner (1992) and Sameroff (1995) in what is called

202

Relationships Between Teachers and Children

developmental systems theory (DST). DST can be applied

to the broad array of systems involved in the practice of

psychology with children and adolescents (Pianta, 1999).

The principles of DST help integrate analysis of the multiple

factors that influence young children, such as families, com-

munities, social processes, cognitive development, schools,

teachers, peers, or conditions such as poverty. This analysis

of child-teacher relationships draws heavily on develop-

mental systems perspectives for principles and constructs

that guide inquiry, understanding, and integration of diverse

knowledge sources.

For the purposes of this discussion, systems are defined

as units composed of sets of interrelated parts that act in

organized, interdependent ways to promote the adaptation

and survival of the whole. Families, classrooms, child-parent

and child-teacher relationships, self-regulatory behaviors, and

peer groups are systems of one form or another, as are various

biologic systems within the organism. These systems function

at a range of levels in relation to the child—some distal and

some more proximal (Bronfenbrenner & Morris, 1998). They

are involved in multiple forms of activity involving interac-

tions within levels and across levels (Gottlieb, 1991) that form

a pattern, or matrix, of reciprocal, bidirectional interac-

tions that varies with time. In the case of child-teacher rela-

tionships, this perspective is reflected in analysis of the

ways that school policies about child-teacher ratios affect

student-teacher interactions that in turn are related to students’

and teachers’ perceptions and affects toward one another. It is

important to note that one must recognize the vertical as well

as lateral interactions across and within levels and associated

systems. The concept of within- and across-level interactions

among systems is a key aspect of DST as applied to child-

teacher relationships; for example, just as these relationships

are influenced by the interactions of two individuals, they

are in turn affected (and affect) classroom organization and

climate.

Principles Influencing the Behavior and Analysis

of Developmental Systems

The behavior of developmental systems is best understood in

the context of a number of general principles. These princi-

ples apply across all forms of living systems (Magnusson &

Stattin, 1998).

Holism and Units of Analysis

Because of the preponderance of rich, cross-level interac-

tions, interpretation and study of the behavior of systems at

any level must take place in the context of activity at these

other levels. Behavior of a “smaller” system (e.g., children’s

self-regulation in a classroom) should be understood in rela-

tion to its function in the context of systems at more distal

levels (e.g., child-teacher relationships) as well as more prox-

imal or micro levels (e.g., biological systems regulating

temperament) and vice versa. The rich, reciprocal intercon-

nections among these units promote the idea that a relational

unit of analysis is required for analysis of development

(Lerner, 1998). From this perspective, noted Lerner (1998),

the causes of development are relationships among systems

and their components, not actions in isolation. This is highly

similar to the perspective advanced by Bronfenbrenner and

Morris (1998), who argued that the primary engine of devel-

opment is proximal process—interactions that take place be-

tween the child and contexts over extended periods of time.

Bronfenbrenner and Morris cited interactions with teachers

as one course of proximal process. For several developmen-

tal theorists, acknowledging the existence of multilevel inter-

actions leads directly to the need for research that has these

interactions and relationships as their foci.

Magnusson and Stattin (1998) approached the issue of

holism from a somewhat different perspective. They noted

that most psychological (and educational) research and the-

ory are variable-focused—that is, a construct of interest is the

sole focus of measurement and inquiry, inasmuch as variation

in that construct relates to other sources of variation. This ap-

proach, argued Magnusson and Stattin, yields a science that

examines selective aspects of the person but misses large sec-

tors of experience that may hold descriptive and explanatory

power. Behavior is better viewed in terms of higher order or-

ganized patterns of relations across different components of

the system.

The developing child is also a system. From this point of

view, motor, cognitive, social, and emotional development

are not independent entities on parallel paths but are inte-

grated within organized, dynamic processes. Psychological

practices (assessment or intervention) that focus solely on

one of these domains (e.g., cognition, personality, attention

span, aggression, or reading achievement) can reinforce the

notion that developmental domains can be isolated from one

another and from the context in which they are embedded.

Taking a developmental systems perspective, many argue

that child assessment should focus on broad indexes reflect-

ing integrated functions across a number of behavioral

domains as they are observed in context (e.g., Greenspan &

Greenspan, 1991; Sroufe, 1989b). Terms such as adaptation

have been used to capture these broad qualities of behavioral

organization, and although fairly abstract, they call attention

to a focus on how children use the range of resources avail-

able to them (including their own skills and the resources of

Developmental Systems Theory

203

peers, adults, and material resources) to respond to internal

and external demands.

In terms of this analysis of child-teacher relationships,

holism means that to understand the discipline-related behav-

ior of a teachers in their classrooms, one must know some-

thing about the school, school system, and community in

which the teachers are embedded, their experiences, and their

own internal systems of cognition and affect regulation in re-

lation to behavioral expectations in the classroom. From the

perspective of holism, the whole (i.e., the pattern or organi-

zation of interconnections) gives meaning to the activity of

the parts (Sameroff, 1995).

Reciprocal, Functional Relations Between Parts

and Wholes

Systems and their component entities are embedded within

other systems. Interactions take place within levels (e.g.,

beliefs about children affect a teachers’ beliefs about a partic-

ular child; Brophy, 1985) and across levels (e.g., teachers’ be-

liefs about children are related to their training as well as to

the school in which they work; Battstitch et al., 1997) over

time. It is a fundamental tenet of developmental systems the-

ories that these interactions are reciprocal and bidirectional.

Gottlieb (1991) refered to these interactions as coactivity

in part to call attention to the mutuality and reciprocity of

these relations. Similarly, Magnusson and Stattin (1998) and

Sameroff (1995) emphasized that in multilevel, dynamic, ac-

tive, moving systems, it is largely fictional to conceptualize

“cause” or “source” of interactions and activity. Again, this

view has consequences for considerations of child-teacher re-

lationships when examining the large number of components

of these relationships as well as the multilevel systems in

which they are embedded (Eccles & Roeser, 1998).

Motivation and Change

Systems theory offers alternative views of the locus of

motivation and change. Within behavioral perspectives,

change and motivation to change are often viewed as derived

extrinsically—from being acted on by positive (or negative)

reinforcement, or reinforcement history. Maturationist or

biological views of change posit that the locus of change

resides in the unfolding of genetic programs, or chronological

age. From both perspectives the child is a somewhat passive

participant in change—change is something that happens to

the child, whether from within or without.

In developmental systems theory the motivation to change

is an intrinsic property of a system, inherent in that sys-

tem’s activity. Developmental change follows naturally as a

consequence of the activity of interacting systems. That chil-

dren are active can be seen in the ways they continually con-

struct meaning, seek novelty and challenges, or practice

emergent capacities. Furthermore, the child acts within con-

texts that are dynamic and fluid. Motivation, or the desire to

change, is derived from the coaction of systems—of child

and context (Bronfenbrenner & Morris, 1998). That relation-

ships play a fundamental role as contexts for coaction be-

tween child and the world is supported by Csikszentmihalyi

and Rathunde’s (1998) proposition that relationships with

parents are foundational for establishing the rhythm of inter-

action between the child and the external world.

Maturationist or biological views of the motivation for

developmental change tend to rely on characteristics of

the child as triggers for developmental experience and can

result in practices and policies that neglect individual varia-

tion or notions of adaptation. Strongly behavioral views of

motivation focus solely on contingencies while failing to ac-

knowledge the meaning of target behaviors and contextual

responses to the child’s goals, leading to a disjunction be-

tween how the child perceives his or her fit in the world and

how helpers may be attempting to facilitate change. Views of

motivation informed by systems theory lead to a develop-

mental interactionism (Magnusson & Stattin, 1998) that

focuses attention on issues of goodness of fit, relationships,

and related relational constructs. As Lerner (1998) acknowl-

edged, because of relationism, an attribute of the organism

has meaning for psychological development only by virtue of

its timing of interaction with contexts or levels.

Developmental change occurs when systems reorganize

and transform under pressure to adapt. Development takes

place, according to Bronfenbrenner and Morris (1998),

through progressively more complex reciprocal interactions.

Change is not simply a function of acquiring skills but a reor-

ganization of skills and competencies in response to internal

and external challenges and demands that yields novelty in

emergent structures and processes (Magnusson & Stattin,

1998).

Competence as a Distributed Property

Children, as active systems, interact with contexts, exchang-

ing information, material, energy, and activity (Ford & Ford,

1987). Within schools, teacher and children engage within a

context of multilevel interactions involving culture, policy,

and biological processes (Pianta & Walsh, 1996). The

dynamic, multilevel interactionism embodied in the principle

of holism also suggests that children’s competence is so in-

tertwined with properties of contexts that properties residing

in the child (e.g., cognition, attention, social competence,

204

Relationships Between Teachers and Children

problem behaviors) are actually distributed across the child

and contexts (e.g., Campbell, 1994; Hofer, 1994; Resnick,

1994). Cognitive processes related to attending, compre-

hending, and reasoning (Resnick, 1994); emotion-related

processes such as emotion regulation and self-control; help-

seeking; and social processes such as cooperation are all

properties not of the child but of relations and interactions of

the child in the context of the classroom: They reflect a cer-

tain level of organization and function (Magnusson & Stattin,

1998).

The concept of affordance (see Pianta, 1999, for an expla-

nation of this construct as applied to classrooms) embodies

the idea that contexts contain resources for the child that can

be activated to sustain the child’s adaptation to the demands

of that setting. It is important to note that the affordance of a

context must be accessed by interactions with the child. From

the perspective of developmental systems theory, compe-

tence (and problems) in a classroom setting cannot be con-

ceptualized or assessed separately from attributes of the

setting and the interactions that features of the child have

with those setting attributes (and in turn how these interac-

tions are embedded in larger loops of interaction). At the

level of social and instructional behavior between teachers

and children, understanding these interactions may require a

moment-by-moment analysis of behavioral loops, whereas

understanding how the teacher’s behavior in these loops is a

function of her education and training may require a much

wider time frame (Pianta, La Paro, Payne, Cox, & Bradley,

2002). The coordination of these temporal cycles of interac-

tion within and across levels is critical to understanding be-

havior within the specific setting of interest.

Centrality of Relationships in Human Development

In the context of all these multilevel and multisystem interac-

tions, enduring patterns of interaction between children and

adults (i.e., relationships) are the primary conduit through

which the child gains access to developmental resources.

These interactions, as noted earlier, are the primary engine of

developmental change. Relationships with adults are like the

keystone or linchpin of development; they are in large part

responsible for developmental success under conditions of

risk and—more often than not—transmit those risk condi-

tions to the child (Pianta, 1999).

Our focus here is primarily on school-age children and

their relationships with teachers in classroom settings. There

is virtually no question that relationships between children

and adults (both teachers and parents) play a prominent role in

the development of competencies in the preschool, elemen-

tary, and middle-school years (Birch & Ladd, 1996; Pianta &

Walsh, 1996; Wentzel, 1996). They form the developmental

infrastructure on which later school experiences build. Child-

adult relationships also play an important role in adaptation

of the child within the context in which that relationship

resides—home or classroom (e.g., Howes, 2000b; Howes,

Hamilton, et al., 1994; Howes, Matheson, & Hamilton, 1994).

The key qualities of these relationships appear to be related to

the ability or skill of the adult to read the child’s emotional

and social signals accurately, respond contingently based on

these signals (e.g., to follow the child’s lead), convey accep-

tance and emotional warmth, offer assistance as necessary,

model regulated behavior, and enact appropriate structures

and limits for the child’s behavior. These qualities determine

that relationship’s affordance value.

Relationships with parents influence a range of competen-

cies in classroom contexts (Belsky & MacKinnon, 1994;

Elicker, Egeland, & Sroufe, 1992; LaFreniere & Sroufe,

1985; Pianta & Harbers, 1996; Sroufe, 1983). Research has

established the importance of child-parent (often child-

mother) relationships in the prediction and development of

behavior problems (Campbell, 1990; Egeland, Pianta, &

O’Brien, 1993), peer competencies (Elicker et al., 1992;

Howes, Hamilton, et al., 1994), academic achievement, and

classroom adjustment (Pianta & Harbers, 1996; Pianta,

Smith, & Reeve, 1991). Consistent with the developmental

systems model, various forms of adaptation in childhood are

in part a function of the quality of child-parent relationships.

For example, a large number of studies demonstrate the

importance of various parameters of child-parent interaction

in the prediction of a range of academic competencies in

the early school years (e.g., de Ruiter & van IJzendoorn,

1993; Pianta & Harbers, 1996; Pianta et al., 1991; Rogoff,

1990). These relations between mother-child interaction and

children’s competence in mastering classroom academic

tasks reflect the extent to which basic task-related skills

such as attention, conceptual development, communication

skills, and reasoning emerge from, and remain embedded

within, a matrix of interactions with caregivers and other

adults. Furthermore, in the context of these interactions chil-

dren acquire the capacity to approach tasks in an organized,

confident manner, to seek help when needed, and to use help

appropriately.

Qualities of the mother-child relationship also affect the

quality of the relationship that a child forms with a teacher.

In one study, teachers characterized children with ambiva-

lent attachments as needy and displayed high levels of nur-

turance and tolerance for immaturity toward them, whereas

their anger was directed almost exclusively at children with

histories of avoidant attachment (Motti, 1986). These find-

ings are consistent with results in which maltreated and

Conceptual-Theoretical Issues in Research on Child-Teacher Relationships

205

nonmaltreated children’s perceptions of their relationships

with mothers were related to their need for closeness with

their teacher (Lynch & Cicchetti, 1992), as well as to the

teachers’ ratings of child adjustment (Toth & Cicchetti,

1996). Cohn (1990) found that boys classified as insecurely

attached to their mothers were rated by teachers as less

competent and more of a behavior problem than were boys

classified as securely attached. In addition, teachers re-

ported that they liked these boys less. This link between the

quality of child-parent relationships and the relationship

that a child forms with a teacher confirms Bowlby’s (1969)

contention that the mother-child relationship establishes for

the child a set of internal guides for interacting with adults

that may be carried forward into subsequent relationships

and affect behavior in those relationships (Sroufe, 1983).

These representations can affect the child’s perceptions of

the teacher (Lynch & Cicchetti, 1992), the child’s behavior

toward the teacher and the teacher’s behavior toward the

child (Motti, 1986), and the teacher’s perceptions of the

child (Pianta, 1992; Toth & Cicchetti, 1996). On the other

hand, there are limits to concordance and stability in

mother-child and teacher-child relationships as children

move from preschool to school (Howes, Hamilton, &

Phillipsen, 1998).

In sum, there is no shortage of evidence to support the

view that—particularly for younger children, but also for

children in the middle and high school years—relationships

with adults are indeed involved centrally in the development

of increasingly complex levels of organizing one’s interac-

tions and relationships with the world. In this view, adult-

child relationships are a cornerstone of development, and

from a systems perspective intervention involves the inten-

tional structuring or harnessing of developmental resources

(such as adult-child relationships) or the skilled use of

this context to developmental advantage (Lieberman, 1992).

This is inherently a prevention-oriented view (Henggeler,

1994; Roberts, 1996) that depends on professionals’ under-

standing the mechanisms responsible for altering devel-

opmental pathways and emulating (or enhancing) these

influences in preventive interventions (e.g., Hughes, 1992;

Lieberman, 1992). 

In conclusion, a developmental systems perspective draws

attention to this child as an active, self-motivated organism

whose developmental progress depends in large part on qual-

ities of interactions established and maintained over time

with key adult figures. Such interactions, and their effects, are

best understood using child-adult relationships as the unit of

analysis and then embedding this focus on relationships

within the multilevel interactions that impinge on and are

affected by this relationship from various directions.

CONCEPTUAL-THEORETICAL

ISSUES IN RESEARCH ON

CHILD-TEACHER RELATIONSHIPS

This section updates and extends Pianta’s (1999) model of

relationships between children and teachers and reviews

research related to components of this model. The model is

offered as an integrative heuristic. It draws heavily on princi-

ples and concepts of systems theory, positing that by focusing

at the level of relationships as the unit of analysis, significant

advances can be made in understanding the development of

child-teacher relationships and in their significance in rela-

tion to child outcomes. Considering the wide-ranging and

diverse literatures that currently address, in some form or

another, the multiple systems that interact with and comprise

child-teacher relationships, a model of such relationships

needs to be integrative. As suggested by the examples used

in the previous discussion of developmental systems theory,

this model must incorporate aspects of children (e.g., age or

gender), teachers (experience, efficacy), teacher-child inter-

actions (discipline, instruction), activity settings, children’s

and teachers’ perceptions and beliefs about one another, and

school policy (ratios) and climate (Battistich et al., 1997;

Brophy, 1985; Brophy & Good, 1974; Eccles & Roeser,

1998).

It is our firm belief that greater understanding of the de-

velopmental significance of school settings can be achieved

by this focus on child-teacher relationships as a central, core

system involved in transmitting the influence of those set-

tings to children. Narrow-focused examinations of one or two

of the factors just noted, as they relate in bivariate fashion to

one another, are unlikely to yield a comprehensive under-

standing of the dynamic, multilevel interactions that take

place in schools, the complexities of which have frustrated

educational researchers and policy makers for years (Haynes,

1998). In many ways, a focus on the system of child-teacher

relationships as a key unit of analysis may provide the kind of

integrative conceptual tool for understanding development

in school settings that a similar focus on parent-child rela-

tionships provided in the understanding of development in

family settings (Bronfenbrenner & Morris, 1998; Pianta,

1999).

Hinde (1987) and others (e.g., Sameroff, 1995) describe

relationships as dyadic systems. As such, relationships are

subject to the principles of systems behavior described ear-

lier; they are dynamic, multicomponent entities involved in

reciprocal interactions across and within multiple levels of

organization and influence (Lerner, 1998). They are best

considered as abstractions that represent a level or form of

organization within a much larger matrix of systems and