Temperament and peer relationships.

Peer relationships play an important role in socialization during childhood and young adolescence. Negative relationships may pose multiple difficulties for children related to their development that can have lifelong influences. Research indicates that the quality of peer relationships is associated with a variety of outcomes, such as social adjustment in school (Diehl, Lemerise, Caverly, Ramsay, & Roberts, 1998; Ladd, Kochenderfer, & Coleman, 1997); academic performance (Risi, Gerhardstein, & Kistner, 2003); and general self-regard (Hymel, Rubin, Rowden, & LeMare, 1990). Among a number of influential factors, temperament was found to be a major factor in building peer relationships (Szewczyk-Sokolowski, Bost, & Wainwright, 2005). Positive temperament patterns account for the frequency of peer interactions (Keogh & Burstein, 1988). Temperament has been operationalized as a composite conception about individual characteristics reflecting on emotional and behavioral display. Some temperamental dimensions affect the ways children interact with others in their social relationships. Therefore, it is important for teachers and educators to understand the influences of temperament on children's peer relationships.

What Is Temperament?

Two debates about temperament are ongoing: one concerning definition and one concerning measurement. For decades, researchers have defined temperament from different theoretical perspectives. For example, Rothbart and Derryberry (1981) stressed reactivity and self-regulation in defining temperament, while Buss and Plomin (1984) argued for the inherited origins of temperament. A widely accepted definition of individual differences in temperament suggests that it is biological--that is, it develops early in life and is stable throughout life (Bates, 1989; Buss & Plomin, 1984; Keogh, 2003a; Rothbart & Derryberry, 1981; Thomas & Chess, 1977). Thus, there is no definitive definition for temperament, and clouding this issue is the controversy on how individual differences in temperament are measured.

Numerous questionnaires have been developed by researchers that focus on different perspectives on temperament. For example, Thomas and Chess (1977) created a nine-category Child Temperament Questionnaire (CTQ) to evaluate temperament that contains the following subscales: activity, rhythmicity, approach/withdrawal, adaptability, threshold of responsiveness, intensity, quality of mood, distractibility, and persistence. In the 20-item Emotionality, Activity, Sociability, Impulsivity (EASI) survey, Buss and Plomin (1975) emphasized the inherited dimension of temperament through the examination of activity, emotionality, sociability, and impulsivity. Meanwhile, Rothbart and Derryberry (1981) argued that the scale of Thomas and Chess (1977) examined limited dimensions of temperament because it was based only on infant data. For example, the Thomas and Chess questionnaire did not acknowledge some behaviors displayed in later childhood. Therefore, Rothbart and her associates (1981) developed several questionnaires for different age groups to identify temperamental characteristics that are influenced by maturation and experience. For example, the Infant Behavior Questionnaire (IBQ) was established particularly for evaluating 3- to 12-month-old infants with regard to six categories: activity, distress to limits, fear, duration of orienting, smiling and laughter, and soothability. According to Rothbart and her colleagues (2001), these measurements provide age-appropriate items that better assess the temperamental characteristics for each individual age group. Additional concerns about temperament scales surround the issue of validity and reliability.

In summary, similarities and overlap of items exist in the temperamental patterns among the scales developed by these three groups of researchers. For example, activity level is a prevailing factor that can be seen in the CTQ (Thomas & Chess, 1977), the EASI (Buss & Plomin, 1975), and the IBQ (Rothbart & Derryberry, 2001). Buss and Plomin's sociability theory shares a similar view of temperament with Thomas and Chess's trait of approach/withdrawal and with Rothbart's shyness. Anger/frustration in Rothbart and Derryberry's scale contains qualities of mood similar to that of Thomas and Chess's questionnaire and to emotionality in Buss and Plomin's survey. Although not exactly comparable, Thomas and Chess's idea on persistence trait and Rothbart and Derryberry's idea of attentional focusing share certain aspects. The idea of sensitivity threshold in Thomas and Chess's theory is close to Rothbart's perceptual sensitivity. Therefore, these researchers approached temperament theory from different perspectives and, yet, they extend and expand on each other rather than contradict one another.

Temperament and Peer Relationships

A variety of terms have been created to denote peer relationships that depend on the individual intragroup status. Peer relationships can be seen as a broad construct regarding the status a child achieves in connectedness with his/her peers.

To examine how temperament is associated with peer relationships, some researchers have adopted temperament composite as a whole concept (Szewczyk-Sokolowski, Bost, & Wainwright, 2005) or constellation groups (Parker-Cohen & Bell, 1988; Walker, Berthelsen, & Irving, 2001), while others have used specific dimensions of individual temperamental traits to pursue more precise results (e.g., Gleason, Gower, Hohmann, & Gleason, 2005; Parker-Cohen & Bell, 1988). The work of Szewczyk-Sokolowski, Bost, and Wainwright (2005) examines the relations among attachment, temperament, and peer acceptance. The results revealed that mothers' reports of having a child with a difficult temperament was not only related to peer acceptance but also presented as a stronger predictor in children's negative peer nominations than attachment. In their longitudinal study, Parker-Cohen and Bell (1988) investigated the relations between temperament and preschoolers' initial and later social behavior by considering temperament patterns individually and in constellation groups. For children's initial reactions toward peers when coming to a new preschool setting, the findings suggested that children with high activity and high approach/withdrawal, but a low sensitivity level, would be more socially responsive toward their peers. For later social behaviors, only activity and approach/withdrawal were found to be associated with preschoolers' social behavior. In addition, significant relationships emerged in the connection between the clusters of temperament and preschoolers' social behavior. As expected, the "easy" children were found to be more socially responsive to their peers, the "difficult" ones were in the middle, and the "slow-to-warm-up" children displayed the least socially responsive behavior. The differences found among these groups of children were not significantly sustained. Although Parker-Cohen and Bell (1988) provided more specific information about the relationship between temperament and children's social responsiveness toward peers, one weakness in their study was that they only used a teacher-rated temperament scale.

Walker, Berthelsen, and Irving (2001) investigated the linkage between difficult children and their peer relationships in preschool-age boys and girls. Defining difficult children as high in activity level, high in distractibility, and negative in quality of mood, the results indicated that children with difficult temperament tend to be rejected or neglected by their peers. Furthermore, difficult children tend to have a higher rate of negative quality of mood and lower level of adaptability than popular children. In contrast to girls, boys were rated by teachers as more active, more distractible, and less persistent. Different role expectation in socialization was advanced as the possible reason for different temperamental characteristics for the genders. According to the investigators, it is important for teachers and educators to be aware of the early appearance of social role stereotypes.

For those researchers who studied the relationship between specific dimensions of temperamental traits and peer relationships, activity level was the most frequently found significant factor for the connection between temperament and peer relationships (Dunn & Cutting, 1999; Gleason et al., 2005; Hess & Atkins, 1998; Houck, 1999; Keogh & Burstein, 1988; Skarpness & Carson, 1987). Similar to Walker and her associates, Dunn and Cutting (1999) were interested in exploring the association between temperament and interactive situations in friendship. Based on the data reported by teachers and mothers about their 4-year-olds, the authors suggested that highly active children engaged less frequently in joint pretend play with their friends. The less frequent interactions between highly active children and their friends were factors that prevent the highly active children from developing close relationships with others. Later, Gleason et al. (2005) examined the influence of temperament on children's choice of friends. Their findings suggested that children prefer to nominate those high on impulsivity and soothability as friends. The researchers also found that girls tended to nominate friends with low activity levels, whereas boys preferred to choose children with high activity levels. Skarpness and Carson (1987) investigated children as young as 2 years old and examined correlations between temperament, communicative competence, and adjustment in school. They used the Child and Adolescent Adjustment Profile, which contained five subscale areas of adjustment, such as peer relations, in their study. Based on the regression analysis model empirical evidence supported that activity level not only predicted communicative competence but also correlated to positive peer relationships in school. In addition to activity, other temperamental characteristics, including attention span/distractibility, rhythmicity, and mood, were found to be significantly predictive of children's adjustment.

Emotionality also has been seen as an essential factor when studying social competence as well as the relationship between temperament and peer interactions (Denham, McKinley, Couchoud, & Holt, 1990; Eisenberg et al., 1993; Fabes et al., 1999; Stocker & Dunn, 1990). Stocker and Dunn (1990) examined the effect of sibling relationships and temperament on friendships and peer relationships, respectively. Using hierarchical regression analysis, they found that emotionality and sociability significantly linked to the quality of children's relationships with friends and peers. The results also showed that temperament explained 7 to 35 percent of the variance in predicting friendships and peer relationships. Consistent with Stocker and Dunn's (1990) findings, Eisenberg and her colleagues (1993) highlighted the significance of self-regulation and emotionality in contributing to preschoolers' social capacities in response to peer interactions. In addition, the effect of gender differences was investigated in the relationship between temperament and social competence. The results showed that high negative emotionality was correlated to low social capacities for both genders. Highly self-regulated children were more likely to have positive peer interactions. Girls were found to be more able to self-regulate, be more expressive, and be more socially competent than boys.

The work of Fabes and his colleagues (1999) was similar to the study of Eisenberg and her colleagues (1993), except for the former's interest in understanding the relationships among context, temperament, and social competence. Based on the data obtained from teacher and parent ratings of temperament and through naturalistic observation, the researchers found that children with negative emotionality and who were low in self-regulation tended to overreact in the course of intense peer interactions. However, most children exhibited socially competent behaviors in a relaxed situation, regardless of their temperamental characteristics. Houck (1999) investigated a large number of children from infancy through toddlerhood, finding that the relationship between temperament and social competence was not only clearly correlated but also exhibited an increasing magnitude. The researcher suggested this as evidence of an increasing stability between the two factors. In addition, research has documented that negative emotionality was positively related to shyness while negatively associated with social disinterest (Colpan, Prakash, O'Neil, & Armer, 2004). However, researchers have argued that social disinterest does not absolutely lead to solitude. On the contrary, socially disinterested children did not differ from their peers in terms of the frequency of peer interactions.

Shyness also was found to have salient linkage with peer relationships, although this aspect has been seldom studied. Chen, Rubin, and Li (1995) conducted longitudinal work showing that Chinese children with high shyness levels were nevertheless accepted by their peers. Contradictory to the general beliefs possessed by Western researchers that shyness may lead to negative peer relationships, this important finding suggests that cultural issues moderate the association between temperamental patterns and peer relationships.

To summarize, the emotional dimension of temperament was found to significantly influence social competence, peer interactions, and peer relationships. Children who have negative emotionality and who lack the capacity for self-regulation tend to have negative experiences during peer interactions. Still, we must cautiously interpret the relations between emotionality and peer interactions. In addition, children who were highly active, easily distractible, and less persistent tend to receive negative social responses from peers. Boys were found to be less able to exhibit socially competent behaviors than girls were. Research findings also suggest cultural differences emerging from the associations between temperament and peer relations. Inasmuch as temperament has an effect on the course of peer interactions, it is important for teachers and educators to recognize the significance of this connection on influencing small-group cooperative learning.

Social Behavior

Empirical evidence suggests that temperament contributes to both social behaviors (Billman & McDevitt, 1980; Calkins, Gill, Johnson, & Smith, 1999; Parker-Cohen & Bell, 1988; Rothbart, Ahadi, & Hershey, 1994; Rubin, Burgess, & Hastings, 2002; Russell, Hart, Robinson, & Olsen, 2003) and aggressive behaviors (Hess & Atkins, 1998; Rubin, Burgess, Hastings, & Dwyer, 2003; Rubin, Hastings, Chen, Stewart, & McNichol, 1998; Sutton, Cowen, Crean, & Wyman, 1999; Tallandini, 2004). Billman and McDevitt (1980) conducted a study in a preschool setting about the effects of temperament on social behaviors. Activity, approach-withdrawal, intensity, distractibility, and sensory threshold were found as predictors of one or more styles of social behavior. For example, highly active children tended to become involved in conflict situations but also were more sociable than inactive children. Withdrawal children spent more time watching other children, spoke less, and showed more unfriendly behaviors to their peers. Calkins, Gill, Johnson, and Smith (1999) researched the impact of emotional reactivity and regulation (Rothbart & Derryberry, 1981) on social-behaviors with peers in a large group of 2-year-olds. In a laboratory setting, children were first assessed individually on their emotional reaction by frustrating them. During the second measurement, emotional regulation was tested by observing toddlers' individual free play and cooperative play. At the end of the measurement, mother-reported data of temperament were collected. Correlational analyses indicated that emotional reactivity and emotional regulation were predictive of peer-directed conflict behavior and cooperation. The findings, according to the authors, suggest important relationships between emotional regulation and social skills in early childhood development.

Rothbart, Ahadi, and Hershey (1994) also conducted a study using the laboratory setting. They explored the association between temperamental traits and five patterns of social behaviors: empathy, guilt/shame, aggression, help-seeking, and negativity. Children ages 6 and 7 were assessed based on the mother ratings of Children's Behavior Questionnaire (CBS), and were observed for their responses to the stimuli. Results documented that each category of temperamental characteristics was related to one or more of the social behaviors, but none of the temperamental characteristics were associated with all of the social behaviors as a whole.

Recently, another longitudinal study was designed to ascertain whether different patterns of behavioral inhibition can predict children's social behaviors in later childhood (Rubin, Burgess, & Hastings, 2002). A cohort of subjects was measured two times, with a two-year interval. For measurement one, children's data of behavioral inhibition (i.e., the amount of time the toddler spent in physical contact with his/her mom) was obtained among the 2-year-olds. For measurement two, children's behaviors in different sessions (e.g., free play and speeches) were observed and coded. The results supported the hypothesis that children's ability to inhibit undesired behavior was a predictor of social behaviors.

The aforementioned studies, which used laboratory settings and mother-report scales, demonstrate congruence in their findings. It is suggested that the combination of the laboratory assessment and questionnaires can provide valuable additional data (Prior, 1992). However, Bates (1989) argued that laboratory measures may not yield adequate normative data compared to a real-life setting. In addition, Rothbart, Ahadi, and Hershey (1994) indicated that the dependence on mother ratings regarding their perceptions of children's temperamental characteristics may limit the research findings.

A cross-cultural study conducted by Russell and his colleagues (2003) utilized both teacher-rated and parent-reported questionnaires for evaluating temperament, as well as teacher-rated scales for social behavior. Based on correlation analysis, teacher ratings of activity, shyness, and sociability were found to correlate with aggression, social, and prosocial behaviors. Furthermore, the results of regression analysis revealed that temperament was a significant predictor of children's social behavior, although the amount of explained variance was relatively small.

Regarding the relationship between temperament and aggressive behaviors, difficult early temperament was documented to be not only associated with, but also predictive of, aggression in children (Sutton et al., 1999; Tallandini, 2004) and in toddlers (Rubin et al., 2003; Rubin et al., 1998). Conducted either in a naturalistic setting (Tallandini, 2004) or in a laboratory (Rubin et al., 2003; Rubin et al., 1998), consistent findings among these studies emphasized that children's capabilities to regulate emotions and inhibit undesired behaviors are central to the presence of aggressive behaviors.

Overall, temperamental characteristics have been suggested to be effective predictors of both social behavior and aggressive behavior as early as toddlerhood. Research findings, in naturalistic and laboratory settings, elucidated children's abilities to regulate emotion; skills of inhibition are the key determinants of aggression toward peers.

Translating Research to Practice

There is a critical need to develop useful strategies for education practioners to use in the classroom that would help children with various levels of temperamental dispositions. In 1999, Chess and Thomas proposed the goodness-of-fit model to explain an appropriate status between the individual and his or her surroundings. According to Chess and Thomas,

Goodness of fit results when the properties of the environment and its expectations and demands are in accord with the organism's own capacities, characteristics, and style of behaving. When consonance between organism and environment is present, optimal development in a progressive direction is possible. (p. 3)

Keogh (2003b) described goodness of fit as "the match between a child's characteristics and the characteristics of environment--including the values, expectations, demands, and temperaments of adults" (p. 3). In other words, the major components of goodness of fit are the interactions between children's characteristics, an adult's characteristics, and anticipations in a social context. Adaptive outcomes will take place when the expectations and demands of the environment and the child's characteristics reach a congruent status. On the other hand, poorness of fit occurs when there is a discrepancy between the child and the environment. Chess and Thomas (1999) indicated that poorness of fit "leads to pathological functions" (p. 8).

The goodness of fit concept has been recognized by researchers, who have translated its usefulness into direct applications for the education field. For example, Churchill (2003) applied this theoretical concept to examine whether a higher teacher-child fit on children's temperament and the teacher's expectations and a better teacher-parent fit on the expectations for children's behaviors would positively account for children's social and academic outcomes. The results supported the hypothesis and suggested that children's temperamental traits have an impact on the style of teacher-child interactions. At the same time, children were viewed as behaving appropriately when the teacher's and parents' expectations for children's behaviors were consistent. Based on these findings, the researchers stated that it was critical for teachers to not only understand the children's individual characteristics in temperament but also fully communicate with the parents as much as possible.

In a child care center setting, De Schipper, Tavecchio, IJzendoorn, and Zeijl (2004) studied a large cohort of children ranging in age from 6 to 30 months old. They examined the correlations among children's temperament characteristics and well-being. Interpreting goodness of fit in another way, they found that the goodness of fit hypothesis has a different impact on children with different temperamental dispositions. For example, difficult children who changed care arrangements more were associated with internalizing behaviors. In addition, easygoing children demonstrated greater well-being when a trusted caregiver was more accessible. The interpretation for this finding may be that through frequent interactions, the trusted caregivers were more likely to offer what the easygoing children needed. However, the prominent relationships between trusted givers and children's well-being were not found to be related to difficult children. One weakness in the study was that only parent-and teacher-rated questionnaires were adopted, which limited the richness and depth of data. Longitudinal research and multiple research methods are needed to explore how temperamental traits contribute to well-being in real-world settings.

Keogh (2003a) indicated that intense, highly active, and impulsive children might display problem behaviors in a rigid-ruled, strict-scheduled classroom setting that required persistence and concentration for a long period of time. Slow-to-warm-up children may not adapt themselves well in a fast-paced classroom, particularly when they are new to the social context. Therefore, she advocated for the necessity to arrange a learning environment that matches children's temperamental tendencies. As a summary of this section, some useful suggestions that emerged follow:

* It is critical for a teacher to understand her students' temperaments by using temperament scales or observations.

* Frequent reflections on consistency (as well as inconsistency) between the teacher's expectations and the child's temperament are encouraged.

* Communication with parents is helpful for facilitating children's development and adjustment to school.

* It is important to establish trusting relationships with students, particularly with easygoing children.

* Teachers need to be more patient with children whose temperaments are categorized as difficult and slow-to-warm-up, more so than with children who are categorized as easygoing.

Summary

Based on the research findings described earlier, the goodness-off-it model has proven to be useful in facilitating children's learning according to their different temperament characteristics. In the classroom setting, teachers encounter children with different levels of temperament. To be a guide by the side rather than a sage on the stage, teachers need to adopt effective strategies and provide an environment that matches each individual's ways of learning.

Reference

Bates, J. E. (1989). Concepts and measures of temperament. In G. A. Konstamm, J. E. Bates, & M. K. Rothbart (Eds.), Temperament in childhood (pp. 3-26). Chichester, England: John Wiley & Sons, Ltd.

Billman, J., & McDevitt, S. C. (1980). Congruence of parent and observer ratings of temperament with observations of peer interaction in nursery school. Child Development, 51, 395-400.

Buss, A. H., & Plomin, R. (1975). A temperament theory of personality development. New York: Wiley.

Buss, A. H., & Plomin, R. (1984). Temperament: Early developing personality traits. Hillsdale, NJ: Lawrence Erlbaum Associates.

Calkins, S. D., Gill, K. L., Johnson, M. C., & Smith, C. L. (1999). Emotional reactivity and emotional regulation strategies as predictors of social behavior with peers during toddlerhood. Social Development, 8(3), 310-334.

Chen, X., Rubin, K. H., & Li, Z. (1995). Social functioning and adjustment in Chinese children: A longitudinal study. Developmental Psychology, 31(4), 531-539.

Chess, S., & Thomas, A. (1999). Goodness of fit: Clinical applications from infancy through adult life. Ann Arbor, MI: Taylor & Francis.

Churchill, S.L. (2003). Goodness-of-fit in early childhood settings. Early Childhood Education Journal, 31(2), 113-118.

Colpan, R. J., Prakash, K., O'Neil, K., & Armer, M. (2004). Do you "want" to play? Distinguishing between conflicted shyness and social disinterest in early childhood. Developmental Psychology, 40(2), 244-258.

Denham, S. A., McKinley, M., Couchoud, E. A., & Holt, R. (1990). Emotional and behavioral predictors of peer status in young preschoolers. Child Development, 1, 1145-1152.

De Schipper, J. C., Tavecchio, L. W. C., IJzendoorn, M. H. V., & Zeijl, J. V. (2004). Goodness-of-fit in center day care: Relations of temperament, stability, and quality of care with the child's adjustment. Early Childhood Research Quarterly, 19, 257-272.

Diehl, D. S., Lemerise, E. A., Caverly, S. L., Ramsay, S., & Roberts, J. (1998). Peer relations and school adjustment in ungraded primary children. Journal of Educational Psychology, 90(3), 506-515.

Dunn, J., & Cutting, A. L. (1999). Understanding others, and individual differences in friendship interactions in young children. Social Development, 8(2), 201-219.

Eisenberg, N., Fabes, R., Bernzweig, J., Karbon, M., Poulin, R., & Hanish, L. (1993). The relations of emotionality and regulation to preschoolers' social skills and sociometric status. Child Development, 64, 1418-1438.

Fabes, R. A., Eisenberg, N., Jones, A., Smith, M., Guthrie, I., Poulin, R., Shepard, S., & Freidman, J. (1999). Regulation, emotionality, and preschoolers' socially competent peer interactions. Child Development, 70, 432-443.

Gleason, T. R., Gower, A. L., Hohmann, L. M., & Gleason, T. C. (2005). Temperament and friendship in preschool-aged children. International Journal of Behavior Development, 29(4), 336-344.

Hess, L. E., & Atkins, M. S. (1998). Victims and aggressors at school: Teacher, self, and peer perceptions of psychosocial functioning. Applied Developmental Science, 2(2), 75-89.

Houck, G. M. (1999). The measurement of child characteristics from infancy to toddlerhood: Temperament, developmental competence, self-concept, and social competence. Issues in Comprehensive Pediatric Nursing, 22(2), 101-127.

Hymel, S., Rubin, K. H., Rowden, L., & LeMare, L. (1990). Children's peer relationships: Longitudinal prediction of internalizing and externalizing problems from middle to late childhood. Child Development, 61, 2004-2021.

Keogh, B. K. (2003a). Temperament in the classroom: Understanding individual differences. Baltimore: Paul H. Brookes.

Keogh, B. K. (2003b). Understanding child temperament can have far-reaching impact on behavior and quality of life. The Brown University Child and Adolescent Behavior Letter, 19(10), 1-4.

Keogh, B. K., & Burstein, N. D. (1988). Relationship of temperament to preschoolers' interactions with peers and teachers. Exceptional Children, 54(5), 456-461.

Ladd, G. W., Kochenderfer, B. J., & Coleman, C. C. (1997). Classroom peer acceptance, friendship, and victimization: Distinct relational systems that contribute uniquely to children's school adjustment? Child Development, 68(6), 1181-1197.

Parker-Cohen, N. Y., & Bell, R.Q. (1988). The relationships between temperament and social adjustment to peers. Early Childhood Research Quarterly, 3, 179-192.

Prior, M. (1992). Childhood temperament. Journal of Child Psychology and Psychiatry, 33(1), 249-279.

Risi, S., Gerhardstein, R., & Kistner, J. (2003). Children's classroom peer relationships and subsequent educational outcomes. Journal of Clinical and Child and Adolescent Psychology, 32(3), 351-361.

Rothbart, M. K., Ahadi, S. A., Hershey, K. (1994). Temperament and social behavior in childhood. Merrill-Palmer Quarterly, 40(1), 21-39.

Rothbart, M. K., Ahadi, S. A., Hershey, K., & Fisher, P. (2001). Investigations of temperament at three to seven years: The Children's Behavior Questionnaire. Child Development, 72, 1394-1408.

Rothbart, M. K., & Derryberry, D. (1981). Development of individual differences in temperament. In M. E. Lamb & A. L. Brown (Eds.), Advances in developmental psychology (Vol. I, pp. 37-86). Mahwah, NJ: Erlbaum.

Rubin, K. H., Burgess, K. B., & Hastings, P.D. (2002). Stability and social-behavioral consequences of toddlers' inhibited temperament and parenting behaviors. Child Development, 73(2), 483-495.

Rubin, K. H., Burgess, K. B., Hastings, P. D., & Dwyer, K.M. (2003). Predicting preschoolers' externalizing behaviors from toddler temperament, conflict, and maternal negativity. Developmental Psychology, 39(1), 164-176.

Rubin, K. H., Hastings, P. D., Chen, X., Stewart, S., & McNichol, K. (1998). Intrapersonal and maternal correlates of aggression, conflicts, and externalizing problems in toddlers. Child Development, 69(6), 1614-1629.

Russell, A., Hart, C. H., Robinson, C. C., & Olsen, S. F. (2003). Children's sociable and aggressive behavior with peers: A comparison of the U.S. and Australia, and contributions of temperament and parenting styles. International Journal of Behavioral Development, 27(1), 74-86.

Skarpness, L. R., & Carson, D. K. (1987). Correlates of kindergarten adjustment: Temperament and communicative competence. Early Childhood Research Quarterly, 2, 367-376.

Stocker, C., & Dunn, J. (1990). Sibling relationships in childhood: Links with friendships and peer relationships. British Journal of Developmental Psychology, 8, 227-244.

Sutton, S. E., Cowen, E. L., Crean, H. & Wyman, P. A. (1999). Pathways to aggression in young, highly stressed urban children. Child Study Journal, 29(1), 49-67.

Szewczyk-Sokolowski, M., Bost, K. K., & Wainwright, A.B. (2005). Attachment, temperament, and preschool children's peer acceptance. Social Development, 14, 379-397.

Tallandini, M.A. (2004). Aggressive-behavior in children's dolls' house play. Aggressive Behavior, 30(6), 504-519.

Thomas, A., & Chess, S. (1977). Temperament and development. New York: Brunner/Mazel.

Walker, S., Berthelsen, D., & Irving, K. (2001). Temperament and peer acceptance in early childhood: Sex and social status differences. Child Study Journal, 31(3), 177-191.

Imei Ma is a doctoral candidate, Elementary and Social Studies Education, University of Georgia, Athens.