Handbook of psychology volume 7 educational psychology

210

Relationships Between Teachers and Children

Other child characteristics that may be linked to the rela-

tionship that children develop with teachers include their own

social and academic competencies and problems (Ladd et al.,

1999; Murray & Greenberg, 2000). In a large sample of ele-

mentary school children, Murray and Greenberg reported that

children’s own reports of feeling a close emotional bond with

their teacher were related to their own and their teachers’ re-

ports of problem behavior and competence in the classroom.

Similarly, Pianta (1992) reported that teachers’ descriptions

of their relationships with students in kindergarten were re-

lated to their reports of the child’s classroom adjustment and, in

turn, related to first-grade teacher reports. A cycle of child be-

havior and interactions with the teacher appeared to influence

(in part) the teachers’ relationship with the child, which in turn

was independently related (along with reports of classroom be-

havior) to teacher reports of classroom adjustment in the next

grade. Ladd et al. (1999) suggested that relational style of the

child is a prominent feature of classroom behavior.

Also important to the formation of the child-teacher rela-

tionship, though less visible to teachers, are the thoughts and

feelings of their students, including their general feelings about

the school environment and about using adults as a source of

support. Third through fifth graders from urban, at-risk schools

who reported the highest levels of dissatisfaction with the

school environment reported less social support at school

and a more negative classroom social environment than did

their more satisfied peers (Baker, 1999). Similarly, elementary

school children who report an emotionally close and warm re-

lationship with their teacher view the school environment and

climate more positively (Murray & Greenberg, 2000). Clearly,

one issue in sorting out associations (or lack thereof) of

children’s judgments of school climate and the quality of

child-teacher relationships is the experiential source of these

judgments. Given the much greater weight on classroom expe-

rience in young children’s judgment of school climate, it is

likely that these findings demonstrate the influence of child-

teacher relationships on children’s judgments of climate and

social support in the broader school environment. Whether and

how this relation changes with development, such that school

climate plays a relatively greater role in judgments of the rela-

tional quality of classroom experiences over time, is at this

time unknown.

Just as teachers make judgments about when to invest

or not invest in a relationship, children, especially as they

grow older, calculate their relational investment based on

the expected returns (Muller, Katz, & Dance, 1999). Urban

youths (largely African American males) enrolled in two sup-

plemental programs near Boston reported investing more with

teachers who show that they care yet are also able to provide

structure and have high expectations for students progress. 

Clearly, these child characteristics, behaviors, and per-

ceptions are not independent of one another. As suggested

earlier, boys tend to act out more in primary grades, and this

behavior, rather than simply gender itself, may account for

the conflict they have with teachers. Similarly, there is

evidence linking child factors such as socioeconomic status,

ethnicity, and disability classification, and the associations

between these factors and the quality of the child-teacher

relationship are likely to be complex. Nevertheless, what the

child brings into the classroom each day is an important piece

of the child-teacher relationship.

Representational Models

An individual’s representational model of relationships

(Bowlby, 1969; Zeanah et al., 1993) is the set of feelings and

beliefs that has been stored about a relationship that guides

feelings, perceptions, and behavior in that relationship. These

models are open systems: The information stored in them,

while fairly stable, is open to change based on new experi-

ence. Also, representational models reflect two sides of a

relationship (Sroufe & Fleeson, 1988). A teacher’s represen-

tational model of how children relate to teachers is both the

teacher’s experience of being taught (and parented) and his or

her own experience as a teacher.

Representational models can have an effect on the forma-

tion and quality of a relationship through brief, often subtle

qualities of moment-to-moment interaction with children

such as the adult’s tone of voice, eye contact, or emotional

cues (Katz, Cohn, & Moore, 1996), and in terms of the toler-

ances that individuals have for certain kinds of interac-

tive behaviors. Therefore, adults with a history of avoidant

attachment, who tend to dismiss or diminish the negative

emotional aspects of interactions, will behave differently in

a situation that calls for a response to an emotionally needy

child than will adults whose history of secure attachment

provides support for perceiving such needs as legitimate and

responding to them sensitively.

In Pedersen et al.’s (1978) case study, adults were asked

to recall experiences with a particular teacher who had a

reputation as exceptional. This was an attempt to examine

(retrospectively) the features of experience associated with

an influential teacher. These recollections describe the impact

of a teacher who formed relationships with students that,

according to their reports, made them feel worthwhile, sup-

ported their independence, motivated them to achieve, and

provided them with support to interpret and cope with envi-

ronmental demands. Students’ representations of their rela-

tionships with a specific teacher appeared to be a key feature

of their experience in relationship with that teacher.

Conceptual-Theoretical Issues in Research on Child-Teacher Relationships

211

Teachers’ representations have been examined only re-

cently. Based on interview techniques developed for assess-

ment of child-parent relationships, Stuhlman and Pianta (in

press) gathered information from 50 teachers of kindergarten

and first-grade children. Teachers of first graders had rela-

tively higher levels of negative emotion represented in rela-

tionships with students than did the kindergarten teachers. For

both groups of teachers, representations of negative emotion

were related to their discipline and negative affect in interac-

tions with the children about whom they were interviewed.

Teachers’ representations were only somewhat related to the

child’s competence in the classroom, and teachers’ represen-

tations were related to their behavior with the child indepen-

dent of the child’s competence. The Stuhlman and Pianta (in

press) study suggests that teacher representations, while re-

lated in predicted directions to their ratings of child compe-

tence and to their behavior with the child, are not redundant

with them, with indications that representations are unique

features of the child-teacher relationship.

Muller et al. (1999) argued that teachers’ expectations for

students are essential components of the development of pos-

itive or negative relationships with children. They suggested

that teachers calculate expected payoffs from investing in

their relationships with students. The authors provided an

example of one teacher’s differential expectations for two

students with very similar backgrounds. Both were Latino stu-

dents with excellent elementary school records who were get-

ting poor grades in middle school. Although both boys were

involved with a peer group that encouraged cutting class, one

student broke away from his friends to attend the teacher’s

class. Interviews with the teacher suggested that she invested

much more time and energy into this student, who she thought

was less susceptible to peer pressure, because she believed she

could have a greater influence on him than on the other.

Further work on representational aspects of child-teacher

relationships is needed to distinguish these processes from

general expectations and beliefs held by the child or teacher.

At present, there is sufficient evidence from research on child-

parent relationships to posit that representational models, al-

though assessed via individuals’ reports and behaviors, may

in fact reflect properties of the relationship (Main, 1996) and

therefore be better understood not as features of individuals

but as relational entities at a different level of organization.

Information Exchange Processes: Feedback Loops

Between Child and Teacher

Like any system, the components of the child-teacher rela-

tionship system interact in reciprocal exchanges, or loops

in which feedback is provided across components, allowing

calibration and integration of component function by way of

the information provided in these feedback loops. In one way,

dyadic relationships can be characterized by these feedback

processes. For example, the ways that mother and child

negotiate anxiety and physical proximity under conditions

of separation characterize the attachment qualities of that

mother-child relationship (Ainsworth et al., 1978). As in the

attachment assessment paradigm, feedback processes are

most easily observed in interactive behaviors but also in-

clude other means by which information is conveyed from

one person to another. What people do with, say or gesture to,

and perceive about one another can serve important roles in

these feedback mechanisms.

Furthermore, the qualities of information exchange, or

how it is exchanged (e.g., tone of voice, posture and prox-

imity, timing of behavior, contingency or reciprocity of be-

havior), may be even more important than what is actually

performed behaviorally, as it has been suggested that these

qualities convey more information in the context of a rela-

tionship that does actual behavioral content (Cohn, Campbell,

Matias, & Hopkins, 1990; Greenspan, 1989).

Teacher-Child Interactions

Although there is a large literature on interactions between

teachers and children (e.g., Brophy & Good, 1986; Zeichner,

1995), it is focused almost entirely on instruction. Recent

work has integrated a social component to understanding

instructional interactions (e.g., Rogoff, 1990), but in the

majority of studies of teachers’ behaviors toward children in

classrooms, the social, emotional, and relational quality of

these interactions is almost always neglected.

Teachers and children come together at the beginning of

the year, each with their own personality and beliefs, and

from the moment the children enter the classroom, they

begin interacting with one another. It is through these daily

interactions, from the teacher welcoming students in the

morning to the moment the children run out the door to catch

the bus, that relationships develop. Recently, investigators

have gained a better understanding of the specific types of in-

teractions that lead to the formation of relationships between

students and teachers. As with studies on individual charac-

teristics and perceptions, these relational interaction studies

are imbedded within a much larger field of research on class-

room interactions.

It is not surprising that teachers’ interactions with children

are related to characteristics of the children themselves. Peer-

rejected children tend to be more frequent targets of correc-

tive teacher feedback than nonrejected classmates (e.g.,

Dodge, Coie, & Brakke, 1982; Rubin & Clarke, 1983), and it

212

Relationships Between Teachers and Children

has been repeatedly demonstrated that teachers direct more

of their attention to children with behavior or learning prob-

lems and to boys (e.g., Brophy & Good, 1974). Similarly,

children rated as more competent in the classroom are more

frequent recipients of sensitive child-teacher interactions and

teachers’ positive affect (Pianta et al., 2002). Once again,

when considering child-teacher interactions in the context of

this dyadic relationship system, it is important to recognize

that there are strong bidirectional relations between child

characteristics and teacher behavior.

Research on teacher-child interactions as they relate to

student motivation provides some insight into associations

between interactions and relationships. Skinner and Belmont

(1993) suggested that although motivation is internal to a

child, it requires the social surrounding of the classroom to

flourish. They suggested three major components to this

social environment: involvement, autonomy support, and

structure. Involvement is defined as “the quality of inter-

personal relationships with teachers and peers. . . . [T]eachers

are involved to the extent that they take time for, express

affection toward, enjoy interaction with, are attuned to, and

dedicate resources to their students” (p. 573). This definition

closely resembles definitions of a positive child-teacher

relationship.

Skinner and Belmont (1993) found that upper elementary

teachers’ reports of greater involvement with students were

the feature of the social environment most closely associated

with children’s positive perceptions of the teacher. Further-

more, they found a reciprocal association between teacher

and student behavior such that teacher involvement facili-

tated children’s classroom engagement and that this engage-

ment, in turn, led teachers to become more involved.

Students who are able to form strong relationships with

teachers are at an advantage that may grow exponentially as

the year progresses. Similar research conducted with adoles-

cents suggests that student engagement with teachers is

dependent not only on their feelings of personal competence

and relevance of course material but also on students’ per-

ceptions of feelings of safety and caring in the school envi-

ronment (Roeser, Eccles, & Sameroff, 2000).

Teachers’ Interactions With Other Students

as Observed by Peers

An interesting line of recent research has focused on teachers

as social agents of information and the role that their interac-

tions with a given student serve as sources of information about

child-teacher relationships for the other students in the class-

room (Hughes, Cavell, & Willson, in press; White & Kistner,

1992). Hughes et al. (in press) reported that classmates’

perceptions of the quality of the relationship between their

teacher and a selected child in the classroom were related

to their own perceptions of the quality of their relationship

with the teacher. It is important to note that these relations were

observed independent of the characteristics of the child, sug-

gesting that this is a unique source of social information in the

classroom setting that has consequences for the impressions

that children form of their teacher (and vice versa), which in

turn could relate to help seeking and other motivational and

learning behaviors (Newman, 2000). In a related study, White

and Kistner (1992) examined relations between teacher feed-

back and children’s peer preferences in early elementary stu-

dents, finding that teachers’ negatively toned feedback toward

selected children was related negatively to classmates’ prefer-

ences for these children.

With regard to understanding the role that interactive be-

haviors play in the context of the entire teacher-child rela-

tionship, patterns of behavior appear to be more important

indicators of the quality of a relationship than do single in-

stances of behavior. It is not the single one-time instance of

child defiance (or compliance) or adult rejection (or affec-

tion) that defines a relationship. Rather, it is the pattern

of child and adult responses to one another—and the quality

of these responses. Pianta (1994) argued that these qualities

can be captured in the combination of degree of interactive

involvement between the adult and child and the emotional

tone (positive or negative) of those interactions. Birch and

Ladd (1996) pointed out that relationship patterns can be

observed in global tendencies of the child in relation to

the adult—a tendency to move toward, move away, or move

against.

Also involved in the exchange of information between

adult and child are processes related to communication, per-

ception, and attention (Pianta, 1999). For example, how a

child communicates about needs and desires (whiny and

petulant or direct and calm), how the teacher selectively at-

tends to different cues, or how these two individuals interpret

one another’s behavior toward each other (e.g. “This child is

needy and demanding” vs. “This child seems vulnerable and

needs my support”) are all aspects of how information is

shaped and exchanged between child and teacher. Percep-

tions and selective attending (often related to the individuals’

representations of the relationship; see Zeanah et al., 1993)

act as filters for information about the other’s behavior. These

filters can place constraints on the nature and form of infor-

mation present in feedback loops and can be influential in

guiding interactive behavior because they tend to be self-

fulfilling (Pianta, 1999). Over time, these feedback and in-

formation exchange processes form a structure for the

interactions between the adult and child.

Conceptual-Theoretical Issues in Research on Child-Teacher Relationships

213

In sum, processes involving transmission of information

via behavioral, verbal, and nonverbal channels play a central

role in the functioning of the dyadic system of the child-

teacher relationship. For the most part, research has focused

on descriptions of instructional behaviors of teachers, on

teachers’ differential attentiveness to children, and on chil-

dren’s engagement and attention in learning situations (see

Brophy & Good, 1974). There is much less information

available on social and affective dimensions of child-teacher

interactions, nonverbal components of interaction, and the dy-

namics of multiple components of interaction in classroom

time or developmental time. Furthermore, how these interac-

tive processes are shaped by and shape school- and system-

level parameters (e.g. school climate, policies on productive

use of instructional time) is even less well described. Nonethe-

less the available data provide support for the developmental

systems perspective of child-teacher relationships and the

complex ways in which information is transmitted through

multiple channels between child and teacher and the fact that

this information plays an important role in children’s and

teachers’ perceptions and representations of one another.

External Influences

Systems external to the child-teacher relationship also exert

influence. Cultures can prescribe timetables for expectations

about students’ performance or the organization of schools

(Sameroff, 1995) that can shape how students and teacher

relate to one another. What other external influences shape

student-teacher relationships? State regulations mandate

standards for student performance that affect what teachers

must teach, and at times how they must teach it. School sys-

tems have codes for discipline and behavior, sometimes man-

dating how discipline will be conducted. States and localities

prescribe policies and regulations regarding student-teacher

ratios, the placement of children in classrooms, at what grade

students move to middle school, or the number of teachers a

child comes into contact with in a given day. Teachers also

have a family and personal lives of their own.

Structural Aspects of the School Environment

Structural variables in classrooms and schools play an impor-

tant role in constraining child-teacher relationships through

direct effects on the nature of interaction and indirectly via at-

tributes of the people involved. For example, observations of

child-teacher interactions in kindergarten and first grade are

observed to vary as a function of the ratio of children to adults

in the room, the activity setting (small or large group), and the

characteristics of the children in the classroom. In large

samples of students in kindergarten (Pianta et al., 2002) and

first grade (NICHD Early Child Care Research Network

[ECCRN], 2001), children in classrooms with a low ratio of

children to adults receive more frequent contacts with their

teacher and contacts that are more positive emotionally.

Teachers in these classrooms are observed to be more sensi-

tive. Similarly, in both of these samples, when the activity

setting was large-group or whole-class instruction, children

had many fewer contacts with the teacher than when in small

groups. In a sample of more than 900 first-grade classrooms,

children, on average, were engaged in individual contact with

their teachers on approximately 4 occasions during a 2-hr

morning observation (NICHD ECCRN, 2001).

It appears that attributes of the class as a whole are related

to the quality of interactions that teachers have with an indi-

vidual child (NICHD ECCRN, 2001). Therefore, when the

classroom is composed of a higher percentage of African

American children or children receiving free-reduced lunch,

teachers were observed to show less emotional sensitivity

and support and lower instructional quality in their intersec-

tions with an individual (unselected) student. These findings

suggest perhaps that the racial and poverty composition of

the classroom may represent demand features of the children,

which can result in a teacher’s behaving more negatively with

children (with attendant consequences for their relationships)

when high as a function of the concentration of children with

these characteristics in classroom. This suggestion is sup-

ported by survey data demonstrating that teachers with high

concentrations of ethnic minority or poor children in their

classrooms experience a greater degree of burden (Rimm-

Kaufman, Pianta, Cox, & Early 1999).

Finally, the level or organization of the school also affects

how child and teacher relationships form and function.

Eccles and Roeser (1998) summarized findings suggesting

that as children move through elementary school and into

middle school, there is an increasing mismatch between their

continuing needs for emotional support and the school’s

increasing departmentalization and impersonal climate.

School Climate and Culture

How the school values and supports the emotional-social

component of teacher-child interactions involves its view of

the role and importance of child-teacher relationships (e.g.,

Battstich et al., 1997; Haynes, 1998). As noted earlier, it is

difficult to disentangle the extent to which teacher-child rela-

tionships and school climate influence one another and the

extent to which the balance of influence shifts as children

grow older and their experiences are more widely distributed

within a school. Nonetheless, there is ample evidence that