Eisenberg, Pidada, Liew - Child Development 2001

Child Development, November/December 2001, Volume 72, Number 6, Pages 17471763

The Relations of Regulation and Negative Emotionalityto Indonesian Childrens Social Functioning

Nancy Eisenberg, Sri Pidada, and Jeffrey Liew

The purpose of this study was to examine the relations of individual differences in regulation and negativeemotionality to 127 third-grade Indonesian childrens social skills/low externalizing problem behavior, socio-metric status, and shyness. Parents and multiple teachers provided information on childrens regulation, nega-tive emotionality, and social functioning; peer sociometric information on liking and social behavior was ob-tained; and children reported on their self-regulation. In general, childrens low socially appropriate behavior/high problem behavior and rejected peer status were related to low dispositional regulation and high negativeemotionality (intense emotions and anger), and regulation and negative emotionality (especially teacher rated)sometimes accounted for unique (additive) variance in childrens social functioning. Adult-reported shynesswas related to low peer nominations of disliked/ghts (although shy children were not especially liked), lowadult-reported regulation, and (to a lesser degree) low teacher-rated negative emotionality. Findings are com-pared with work on regulation, negative emotionality, social competence, and shyness in other countries.

INTRODUCTION

In the last decade, there has been considerable inter-est in the role of regulation and emotionality in thequality of childrens social functioning (e.g., Saarni,1990). A growing body of research supports the viewthat emotionality and regulation are associated withchildrens concurrent and long-term social compe-tence and adjustment (Block & Block, 1980; Caspi,1998, 2000; Eisenberg, Fabes, Guthrie, & Reiser, 2000;Pulkkinen & Hamalainen, 1995; Rothbart, Posner, &Hershey, 1995). However, most of this work has beenconducted in the United States or other western, in-dustrialized countries. The primary purpose of thepresent study was to examine the relations of individ-ual differences in dispositional regulation and nega-tive emotionality to quality of childrens social func-tioning in a non-Western cultural environmentthatof Java, Indonesia. As is discussed shortly, Java is aninteresting place to study emotion regulation becauseit is highly prized in the culture.

Emotion-related regulation is the process of initiat-ing, maintaining, modulating, or changing the occur-rence, intensity, or duration of internal feeling states,emotion-related physiological processes, and the be-havioral concomitants of emotion. The concomitantsof emotion include facial and gestural reactions andother behaviors that stem from, or are associatedwith, internal emotion-related psychological or phys-iological states and goals (Eisenberg et al., 2000).Emotion-related regulation often can involve effortfulcontrol, a construct discussed by Rothbart and col-leagues (e.g., Ahadi & Rothbart, 1994; Rothbart,Ahadi, & Evans, 2000). Effortful control is dened asthe ability to inhibit a dominant response to perform

a subdominant response (Rothbart & Bates, 1998,p. 137). It involves both attentional regulation (e.g.,the ability to voluntarily focus attention as needed)and behavioral regulation (e.g., the ability to inhibitbehavior as appropriate). Effortful control (includingattentional and behavioral regulation) has been re-lated to low levels of, or better modulated, negativeemotionality (Derryberry & Rothbart, 1988; Kochan-ska, Coy, Tjebkes, & Husarek, 1998); high levels of em-pathy, prosocial behavior, and conscience (Eisenberg,Fabes, Karbon, et al., 1996; Kochanska, Murray, & Coy,1997; Kochanska, Murray, Jacques, Koenig, & Vande-geest, 1996; Rothbart, Ahadi, & Hershey, 1994); andsocial competence and low levels of externalizingproblems (e.g., Eisenberg, Fabes, Guthrie, et al., 1996;Eisenberg, Guthrie, et al., 1997; Rothbart et al., 1994;for a review, see Eisenberg et al., 2000). Moreover, lowlevels of attentional control sometimes have beenlinked to shyness (Eisenberg et al., 2001; Eisenberg,Fabes, & Murphy, 1995),

Individual differences in intensity and valence ofemotion also appear to play an important role in qual-ity of social functioning. These individual differencesgenerally are viewed as rooted in temperament (e.g.,Rothbart & Bates, 1998). If people experience strongnegative emotions and cannot adequately modulatetheir emotion and control their expression, they arerelatively likely to behave in inappropriate ways (e.g.,act out anger) or may be socially withdrawn andshy (Asendorpf, 1987; Caspi et al., 1995; Eisenberg etal., 2001; Gottman, Katz, & Hooven, 1996; Lerner,

2001 by the Society for Research in Child Development, Inc.All rights reserved. 0009-3920/2001/7206-0009

1748 Child Development

Hertzog, Hooker, Hassibi, & Thomas, 1988). For ex-ample, children who are moody and prone to nega-tive emotions such as anger are less liked by peers(Coie & Dodge, 1988; French, 1988; Maszk, Eisenberg,& Guthrie, 1999; Newcomb, Bukowski, & Pattee,1993; Stocker & Dunn, 1990). Moreover, shyness hasbeen associated with internalizing negative emotions(e.g., sadness, anxiety) in both children and adults(Asendorpf, 1987; Eisenberg, Shepard, et al., 1998; Izard,Libero, Putnam, & Haynes, 1993; Leary, 1986). In con-trast, relations between shyness and externalizingemotions such as anger generally have been non-signicant (e.g., Eisenberg et al., 2001; Eisenberg,Shepard, et al., 1998), although positive relations alsohave been obtained between adults reports of shy-ness and their reports of anger, contempt, and disgust(Izard et al., 1993). Thus, the relation between shynessand anger or other such displays of emotion is un-clear. Moreover, in a culture that values the control ofemotional experience and expression, shy individu-als may not display their negative emotions, such asanxiety, in an attempt to conform with norms.

Although individual differences in negative emo-tionality and regulation are likely to be correlated,Eisenberg and Fabes (Eisenberg et al., 2000) have ar-gued that they often provide some unique variance tothe prediction of social competence and problem be-havior. Consistent with this view, the prediction ofsocial competence and problem behaviors is higherwhen both negative emotionality and regulation areconsidered (e.g., Eisenberg et al., 2001; Eisenberg, Fabes,Guthrie, et al., 1996; Eisenberg, Fabes, et al., 1997).Moreover, individual differences in regulation andnegative emotionality sometimes interact when pre-dicting social functioning, such that the combination oflow effortful regulation and negative emotionality is es-pecially problematic. For example, in samples of Amer-ican children, children high in negative emotionalityand low in regulation are most prone to externalizingproblems or low social competence (Eisenberg, Fabes,Guthrie, et al., 1996; for a review, see Eisenberg et al.,2000). Findings of this sort are consistent with models inwhich individual differences in both restraint/regula-tion and negative emotionality jointly contribute to thequality of individuals social functioning (e.g., Eisen-berg et al., 2000; Weinberger & Schwartz, 1990).

Emotion regulation often is viewed as a social pro-cess rather than solely an intraindividual process(Campos, Campos, & Barrett, 1989; Walden & Smith,1997). For example, the effects of different tempera-mental characteristics, including dispositional differ-ences in regulation and negative emotionality, onchildrens socioemotional functioning depend partlyon the t between childrens temperament and the so-

cial context in which they are embedded (Lerner,1984). Because of cultures role in organizing meaningin a society (Super & Harkness, 1982; Whiting, 1980),what is considered competent behavior, or optimalregulation or emotional expressivity, is partly derivedfrom cultural norms and values (Kitayama & Markus,1994). For example, in Western, industrialized cul-tures, social inhibition, which is manifested in shy,withdrawn behavior, has been viewed as reectingfearfulness and a lack of self-condence, and is oftenregarded by adults as relatively socially incompetentand immature (Rubin & Asendorpf, 1993). Thus, it isnot surprising that shyness has been associated withpeer rejection and isolation in countries such as theUnited States and Canada (e.g., Rubin, Chen, McDou-gall, Bowker, & McKinnon, 1995; Rubin, Parker, &Bukowski, 1998). In China, however, shy behavior isbelieved to indicate social maturity and understand-ing, and shyness and sensitivity are positively associ-ated with Chinese childrens peer acceptance, leader-ship, and teacher-assessed competence (e.g., Chen,Rubin, & Li, 1995; Chen, Rubin, Li, & Li, 1999).

In the present study, we examined the relations ofchildrens dispositional effortful regulation and emo-tionality (primarily negative emotionality) to theirsocial functioning and sociometric status in Indone-sia. Indonesian culture is a different context from thatof the United States with regard to values and normsfor appropriate social behavior. Hofstede (1991) ratedthe Indonesian society of Java as being on the extremeend of collectivism. Members of collectivist societiesgenerally are concerned with the consequences oftheir behavior on other members of the group andthey tend to show great willingness to engage inprosocial behavior for the good of the group (Trian-dis, 1995). Although the nature of collectivism un-doubtedly varies somewhat across cultures (Triandis,1995), maintaining personal relationships and inter-personal harmony with close others generally are keyvalues in collectivistic cultures (Markus & Kitayama,1991; Oyserman, 1993; Triandis, 1995). In general, an-ger and other negative emotions are reviewed as dis-rupting relationships and harmful. Thus, socializationin Indonesia would be expected to differ somewhatfrom socialization in individualist societies such asthe United States, in which the emphasis tends to beon individuals own needs, interests, and achieve-ments, and independence and self-initiative. Chil-dren would be expected early on to learn to conformto group norms in terms of controlling the overt ex-pression of negative emotion and emotionally drivennegative behavior, so that they behave in a mannerthat promotes group harmony and avoids interper-sonal conict. Regulation of emotion and its expres-

Eisenberg, Pidada, and Liew 1749

sion in behavior, although important to social func-tioning in Western cultures, would seem to be evenmore important on Java.

Consistent with the aforementioned expectations,traditional Indonesian society has been described asemphasizing cooperation, conformity to authority,and harmonious relationships, that is, rukun (Koent-jaraningrat, 1985). Geertz (1961) and Koentjaranin-grat (1985) asserted that Javanese children are ex-pected to be quiet, obedient, and respectful of theirparents. According to Williams (1991), Javanese par-ents socialize obedience (i.e., manut) and helping,sharing, and empathizing with others (i.e., tepaslira)as ideal human virtues.

Moreover, social scientists working in Java havedescribed Javanese society and customs in ways thatare consistent with the view that the culture empha-sizes sensitivity to others needs, self-control, andcontrol of the expression of emotion. For example,Koentjaraningrat (1985) noted that Javanese childrenwere expected to be emotionally reserved, and Geertz(1961) reported an emphasis on politeness and self-control. Mulder (1989) argued that children are social-ized to feel shame as a means of fostering conformity,self-control, and avoidance of conict and confronta-tion. The ideal of community life is to experience har-monious community, which is based on the willing-ness to adjust ones behavior to the expectations andneeds of others. Javanese believe that to become hu-man is to learn order, inwardly and outwardly so. Toknow order is to know the rules, at the very least asfar as they regulate outward behavior (pp. 2627). Acardinal ethical command is to measure at oneself(tepa slira) what ones words and actions will cause tothe feelings of others or not to do to others that whichone does not wish to have done to oneself (p. 54).This includes do commands such as do not irritatethe others and be careful not to hurt the others feel-ings (p. 34). Mulder, in a later book (1996, pp. 102103), further noted that Emotion and feeling, intu-ition, empathy and sympathy, self-consciousness andappreciation of each others dignity: these are thevalid guides in interaction, along with the suppres-sion of conict, the denial of frustration, and the mas-tery of negative emotions. Similarly, professionalscounseling in hospitals in Java have noted norms re-garding the suppression of public expression of emotionand the stoic acceptance of suffering (Van Beek, 1987).

Javanese individuals may differ from Westerners inboth the expression and experience of emotion. As al-ready noted, on Java it generally is considered inap-propriate to express negative emotions in an unregu-lated manner (Mulder, 1989). In addition, traditionallythe Javanese believe that the experience of negative

emotion (e.g., anger) makes people sick and shortenstheir lives (Geertz, 1976; Wellenkamp, 1995a). Similarbeliefs are held by the Balinese (who live on the islandnext to Java; Wikan, 1989) and people from Toraja, In-donesia (Wellenkamp, 1995b). Wikan (1989, pp. 294295) reported that the Balinese believe that shapingemotions is a collective concern, not a matter of per-sonal choice: It is rigidly enforced by stringent moralsanctions, and requires an effort to sustain. Thus, inmany parts of Indonesia, controlling the experience ofemotion is believed to be essential for communal co-operation and good health.

The limited empirical work supports the argumentthat Javanese children and adults are expected to sup-press their emotional displays. In a study of Indonesianchildrens play behavior with their mothers and oldersiblings, Farver and Wimbarti (1995) found thatmotherchild play observed in the natural context andin the experimental procedure tended to be quiet and re-served compared with siblingchild play. Thus, Indone-sian parents may impose their expectations with regardto the expression of emotion and regulation when theyare interacting with their children. Farver and Wimbarti(1995), like Williams (1991), argued that Javanese chil-dren are socialized to maintain harmonious social rela-tionships, to mask their emotions, and to be obedient,prosocial, and empathic. Consistent with their asser-tions, French et al. (2000) found that youth in the UnitedStates reported more conict in their friendships,whereas Indonesian youth reported more help-giving.

There also is limited research indicating that chil-drens social competence has some of the same corre-lates as in Western cultures. In a study of social statusand friendship, French, Setiono, and Eddy (1999)found that the relations between fth graders socio-metric status and peers, teachers, and parents ratingsof aggression were similar in the United States and In-donesia. In general, aggression (which often involvesnegative emotion) was negatively related to positivesociometric ratings and positively related to negativeratings. Findings for social withdrawal, however, dif-fered somewhat across the Indonesian and UnitedStates children. Positive sociometric ratings were neg-atively related to socially withdrawn behavior in theUnited States and to teachers (but not peers or par-ents) reports of social withdrawal in Indonesia; nega-tive sociometric ratings were positively related tosocial withdrawal in the United States and were unre-lated to negative sociometric ratings in Indonesia. Sim-ilarly, childrens anxiety was negatively related to pos-itive sociometrics and positively related to negativesociometric ratings in the United States, whereas onlypeer ratings of anxiety were associated with negativesociometrics in Java (and there were no associations for


positive sociometrics). Thus, social competence withpeers was only weakly associated with low levels ofsocial withdrawal and anxiety in Indonesiaa ndingthat is not very consistent with data on shyness and so-cial withdrawal in the United States or with the posi-tive relation between social competence and shynessfound in China (e.g., Chen et al., 1995, 1999).

In the present study, we examined the relations ofregulation and negative emotionality to multiple as-pects of childrens social functioning in an Indonesianpopulation. We know of no other study on regulationand emotionality as predictors of childrens socialfunctioning in Indonesia. A multireporter, multi-method approach was used to examine the uniqueand additive relations of regulation and negative emo-tionality to the prediction of childrens socially com-petent and externalizing problem behavior as rated byteachers, parents, and peers. In addition, prediction ofshyness from individual differences in childrens reg-ulation and negative emotionality was examined. Thegoal was to see if obtained relations were similar tothose obtained in research with Western children.

Because Javanese parents traditionally tend to so-cialize their children to be quiet, obedient, and con-trolled, regulation was expected to be a valued char-acteristic in Javanese children and, thus, related tohigh-quality social functioning (as viewed by adultsand peers). In contrast, high negative emotionality,including the expression of anger, was expected to beassociated with low social competence, problem be-havior, and low sociometric status. This pattern ofndings was expected to be at least as clear as in theUnited States because the suppression of emotionalexperience and expression is not as central to the cul-ture in the United States.

Based on prior work in China (Chen et al., 1995)and Indonesia (French et al., 1999), we were unclearas to what to predict with regard to shyness. Consid-ering the ndings in Chinaanother collectivistsocietyand the value of constrained and polite be-havior in Indonesia, shyness might be expected to bepositively related to regulation in Javanese children.As previously discussed, however, French et al. (1999)did not nd that social withdrawal was associatedwith high peer status in Indonesia, so shyness maynot be as valued in Indonesia as in some other socie-ties, at least in the peer context. China and Indonesianmost likely differ in the ways in which they embracecollectivistic values, and also differ in terms of theirprimary religion. Based on the discrepancies in nd-ings, we did not formulate clear hypotheses regard-ing the relation of shyness to regulation or negativeemotionality (including anger).

As in the United States, we expected prediction ofquality of social functioning to be stronger if both reg-ulation and negative emotionality were consideredthat is, we expected their effects to be additive (seeFigure 1; Eisenberg et al., 2000). Based on prior nd-ings, we also expected children who were both proneto negative emotions and unregulated to be the low-est in terms of quality of social functioning (i.e., wepredicted interaction effects).

METHOD

Participants

Participants were 127 Indonesian children (58 fe-males, 69 males; age:

M

109.41 months,

SD

4.75,

range

94132) in three classrooms from a school in

Figure 1 Joint effects of regulation and negative emotionality on childrens quality of social functioning and a list of the mea-sures of constructs. Interactions of regulation and negative emotionality also are predicted, especially for the prediction of socialskills and problem behavior.


Bandung, Indonesia. Ninety-seven percent of thechildren lived in a two-parent household, 2% lived ina one-parent household, and 1% lived with other rel-atives. Ninety-six percent were Muslim and 4% wereChristian. Many of the children were Javanese, anumber were part Sundanese, and a few came fromother islands such as Sumatera (Sumatra) andCelebes. In Bandung, and in this school, many chil-dren were of mixed heritage. The school was a privatepublic school, rather than a government public school,which means that families pay tuition that is higherthan for government-run schools. Thus, the samplewas middle class and many of the parents were pro-fessionals (e.g., bank ofcials, doctors, lawyers, andbusinessman). About 50% of the mothers of childrenat this school worked, often in white-collar jobs.

Procedure

The data were collected in Bandung, Indonesia, acity of over 1,000,000 that is about 180 kilometersfrom Jakarta on the island of Java. The city is a centerfor education and technological development (Frenchet al., 1999).

All questionnaires were translated into Indonesianby Sri Pidada and back-translated by an Indone-sian graduate student. Questionnaires were distributedto parents (124 mothers and 3 fathers), teachers, andchildren. Parents who agreed to participate were in-vited to the school on one of three Saturday sessions.The experimenter gave general instructions to par-ents, and most parents completed the measures atschool. Parents who did not have time to complete themeasures at the school took the questionnaires homeand returned them later (18 parents never completedthe questionnaires). Teachers completed measures ofchildrens social functioning, regulation, and nega-tive emotionality at home or after school. Teacherswere given instructions as a group to ensure that theyunderstood the items and the scales. For child mea-sures, an experimenter read the items to the childrenindividually. For peer nominations and sociometricmeasures, children individually nominated and rankedthe four classmates they liked most and the four theyliked least from a list of names of children in the class-room. Each child also nominated and ranked four class-mates for each of the following: those who were mostlikely to ght, get angry a lot, and be nice to others.

Three teachers rated each child. Teacher 1 was theclassroom teacher who taught the children 6 days aweek and had known them for about 8 months atthe time of data collection. Three different teachersserved as Teacher 1, and each rated children fromtheir own class (4143 children were in each). Teacher

2 was the Islam religion teacher. Teacher 2 taught thechildren once a week, but had known most childrensince they were about 32 months old. In addition toteaching religion, Teacher 2 supervised children dur-ing recess and school eld trips. Teacher 3 was thephysical education teacher and the only male teacherrater. He taught the children once a week and hadknown them for about 8 months at the time of datacollection. Teacher 2 and Teacher 3 had opportunitiesto observe children in both classroom and outdoorcontexts. To increase the reliability of the various in-dexes (Epstein, 1979) and to reduce the number ofmeasures, scores for teachers were combined for allscales, although we also looked at key associationsacross teachers (see below).

Measures

Measures pertained to one of ve constructs: chil-drens social functioning (socially appropriate behaviorand low externalizing problem behavior), shyness,emotionality, and regulation. Teachers and parentsprovided information on all constructs and childrenprovided information on their own regulation andquality of peers social functioning (ghting, proso-cial behavior) and emotionality (anger). Scores on allmeasures were computed not only for Teachers 1, 2,and 3 combined (Teacher 1/2/3), but also forTeacher 1 alone (for social functioning) and forTeachers 2 and 3 combined (Teacher 2/3; for regula-tion and negative emotionality) so that analysescould be computed across teachers. Because (as isdiscussed below) the reliabilities for Teacher 1 mea-sures of negative emotionality and regulationtended to be lower than those for Teachers 2 and 3,we opted to compute and use Teacher 2/3 measuresof these constructs and Teacher 1 measures of socialfunctioning (socially appropriate behavior and lowexternalizing problem behavior, shyness) when ex-amining relations across teachers.

Quality of Social Functioning

Teachers and parents reported on childrens socialskills (socially appropriate behavior), problem behav-ior, and shyness. Peers provided information on socio-metric status, as well as childrens problem behavior(ghting) and social skills (niceness).

Social skills.

To assess childrens social skills (i.e.,socially appropriate behavior), parents completedfour items adapted from Harters (1982) PerceivedCompetence Scale for Children (e.g., This child isusually well-behaved versus This child is not well-behaved; Eisenberg, Guthrie, et al., 1997) using Har-


ters 4-point response scale (i.e., they selected one oftwo statements and then indicated if the item wasreally true or sort of true),

.54. Although this

was low due to the small number of items, the inter-correlations averaged .23 (and this measure was com-bined with others; see below). Teachers rated chil-drens social skills using three items,

.74, .59, and.75 for Teacher 1, Teacher 2, and Teacher 3, respec-tively. An additional item was dropped for teachersbecause its inclusion in the scale lowered the

(Compared to other children this childs age, thischild has very good social skills versus Comparedto other children this childs age, this child does nothave very good social skills). Scores for Teachers 2and 3 were correlated,

r

(125)

.39,

p

.001, and werecombined by averaging the three items for Teacher 2and Teacher 3 before averaging the scale items. Sepa-rate scores were computed for Teachers 1 and Teacher2/3, so that some analyses could be computed acrossteachers. In addition, scores for Teachers 1, 2, and 3(Teacher 1s scores correlated .29 and .36 with those ofTeachers 1 and 2) were standardized and averaged foranalyses across type of reporter (i.e., with peer orteacher reports).

Sociometrics.

Children were asked to select fourpeers (one after another) whom you like the mostand four whom you like the least (i.e., dislike). So-ciometric data were calculated in two ways: as contin-uous variables and as categorical variables (i.e., pop-ular, unpopular, average, controversial, and neglectedgroups). When sociometric data were calculated ascontinuous variables, rst choices were weighted(multiplied) by 4, second choices by 3, third choices by2, and forth choices by 1, separately for same-sex peersand other-sex peers. This is similar to methods used inthe past (e.g., Eisenberg, Fabes, Karbon, et al., 1996;Hartup, Glazer, & Charlesworth, 1967). The weightednominations were summed to compute scores for totalnumber of being liked most and being liked leastby all peers (i.e., same-sex and other-sex peers).

With regard to discrete sociometric status groups,ve groups were formed using criteria similar to thatoutlined by Coie, Dodge, and Coppotelli (1982). Thetotal numbers of ratings for being liked most andbeing liked least by all peers (unweighted) werestandardized within classroom. The total score for be-ing nominated liked most was subtracted from thetotal score for being nominated liked least, thenstandardized, to arrive at a social preference score.A social impact score was calculated by summingthe total score for liked most and liked least (standard-ized within the classroom), and then standardizingthese scores. Similar to Dodge, Coie, Pettit, and Price(1990), an

SD

of .8 rather than 1.0 was used as a cutoff

in the classication. Specically, children with a stan-dardized social preference score greater than .8, aliked most score greater than 0, and a liked leastscore less than 0 were classied as popular (

n

19).Children who scored less than

.8 on their social pref-erence score, less than 0 for liked most, and greaterthan 0 for liked least, were classied as rejected (

n

18). Neglected status (

n

26) was assigned to childrenwho scored less than

.8 on their social impact score,and less than 0 on liked most and liked least. Con-troversial status (

n

10) was assigned to childrenwith social impact scores greater than 0, and scores forbeing liked most and liked least greater than .8. Chil-dren who scored between .8 above and below themean on social preference and social impact scoreswere classied as average status (

n

53).

Prosocial behavior.

Children provided ratings onpeers prosocial behavior. They were asked to selectfour classmates, one after another, who are reallynice to others. First choices were weighted (multi-plied) by 4, second choices by 3, third choices by 2,and fourth choices by 1 for same-sex and for other-sexnominations. The four weighted scores were summedand then standardized within classroom to arrive atseparate scores for same-sex and other-sex nomina-tions of being nice to others. A composite score of chil-drens prosocial behavior was computed by averagingthe standardized same-sex and other-sex nominationsof being nice to others.

Problem behavior.

Children were rated on problembehavior by parents and teachers using the ChildProblem Behavior Checklist by Lochman and theConduct Problems Prevention Research Group (1995),which contains a subset of items for the Child Behav-ior Checklist (Achenbach, 1991). As in Eisenberg,Fabes, Guthrie, et al. (1996) and Eisenberg et al. (2000),all 24 items were included (e.g., teases other chil-dren, breaks things on purpose, deant towardadults) except for one item pertaining to setting res.Items were rated on a 4-point scale (1

never, 4

often). Alphas were .83, .93, .90, and .94 for parents,and Teachers 1, 2, and 3, respectively. Scores for thethree teachers were correlated,

r

s(127)

.41 to .48,

ps

.001, and scores for Teachers 2 and 3 were standardizedand averaged, as were scores for Teachers 1, 2, and 3. Inaddition, children were asked to nominate four peerswho ght a lot. These ratings were weighted, stan-dardized, and combined in the manner described pre-viously for the nice nominations.

Shyness.

Parents and teachers also rated children onshyness on a 7-point scale using 13 items (e.g., Actsshy around new people) adapted from the Child Be-havior Questionnaire (CBQ; Rothbart et al., 1994). Someitems were modied slightly for teachers because the


CBQ was designed for parents (Eisenberg, Fabes et al.,1997). Alphas for parents and Teachers 1, 2, and 3 were.88, .79, .91, and .79, respectively. Scores for the threeteachers were signicantly correlated,

r

s(123)

.37 to.41,

p

s

.001, and were standardized and averagedacross both Teachers 2 and 3 and Teachers 1, 2, and 3.

Emotionality

Ratings by parents and teachers were used to as-sess childrens emotionality. In addition, childrennominated peers who were likely to get angry a lot.

Adults reports of emotional intensity and anger.

Par-ents and teachers rated childrens emotional intensity(EI) with 12 items, adapted from Larsen and Diener(1987), that pertained to intensity of negative emotion(e.g., When my child [this child] gets nervous or dis-tressed, he/she gets very nervous/upset) and emotionin general (My child [This child] responds very emo-tionally to things around him/her). Items were ratedon a 7-point scale (1

never, 7

always). In addition,childrens anger was rated by parents (13 items) andteachers (11 items; e.g., Gets angry when called in fromplay before he/she is ready to quit) using the anger/frustration subscale from the CBQ (Goldsmith & Roth-bart, 1991; Rothbart et al., 1994). Items were rated on a7-point scale (1

extremely untrue, 7

extremely true).Alphas were acceptable, .68 or higher, for both

scales except for Teacher 1,

s

.43 for EI and .61 foranger. Both scales, however, pertained primarily tonegative affect and when combined, the

s for par-ents and Teachers 1, 2, and 3 were .81, .64, .88, and .91,respectively. Thus, the items for these two scales(which were all rated on a 17 scale) were averaged.Correlations among teachers ranged from .22 to .32,

p

s

.02 to .001; thus, composites for Teachers 2 and 3and for Teachers 1, 2, and 3 were computed by stan-dardizing and averaging the teachers scores. Thesescores were not used for Teacher 1 only. Althoughsome items could tap into a variety of emotions,many tapped only negative emotion; thus, this mea-sure is considered primarily an index of negativeemotionality (especially externalizing negative emo-tions) and henceforth is labeled as such.

Peer reports of anger.

Children nominated four chil-dren as most likely to get angry. Nominations weresummed in the same manner as was described for theratings of niceness or ghting.

Regulation

Measures of childrens regulation consisted of par-ents and teachers reports of childrens attentionalcontrol and inhibitory control, as well as childrens

self-reports of regulation. As for the other measures,composites of teacher scores were constructed so that itwould be possible to look at relations across teachers(Teacher 1 with Teacher 2/3) and across reporters (e.g.,Teacher 1/2/3 with parents or peers reports).

Attentional control.

Parents and teachers rated chil-drens attentional shifting and focusing using itemsfrom the CBQ (Rothbart et al., 1994). Because the CBQwas designed for parents, some items were slightlymodied for teachers. Alphas for 10-item attention-shifting scales for parents and Teachers 1, 2, and 3were .75, .70, .72, and .82, respectively. Two additionalitems that did not correlate with other items in thescale were dropped.

Parents rated seven items for attention focusing(e.g., Has difculty leaving a project he/she has be-gun)

.52; two additional reversed items weredropped due to low item-scale correlations and par-ents apparent difculty with reversed items. Al-though this

was low, when attention shifting andfocusing were combined (see below), the

was .75.Teachers rated the same seven items for attentionfocusing,

s

.61, .78, and .83 for Teachers 1, 2, and 3,respectively).

Inhibitory control.

Parents and teachers also ratedchildren on inhibitory control (e.g., Can wait beforeentering into new activities if he/she is asked to)using the inhibitory control subscale of the CBQ (13items for parents and for teachers),

s

.70 for par-ents and .84, .89, and .85 for Teachers 1, 2, and 3,respectively.

Child-reported self-regulation.

Childrens self reportof self-regulation was measured using 16 items fromRohrbeck, Azar, and Wagners (1991) Child Self-ControlScale. Children responded using a 4-point responsescale; they selected the statement that they resembledmost (i.e., Some kids nd it hard to sit still butOther kids nd it easy to sit still) and then indicatedthe degree to which they resembled the selected state-ment (i.e., Sort of like these kids or Really likethese kids). Items were averaged to arrive at a com-posite for child-reported self-regulation,

.81. Thismeasure tapped primarily inhibitory control.

Data Reduction of Peer Assessments

Principle components factor analyses with a vari-max rotation were conducted on the mean scores forthe peer ratings of prosocial behavior, ghting, beingliked, and being disliked. Prosocial behavior (.88) andbeing liked (.92) loaded on the rst factor, whereasghting (.89) and being disliked (.87) loaded on the sec-ond factor. Based on these factors, measures on therst factor were standardized and averaged to con-


struct a

positive sociometric composite.

The two indiceswere correlated,

r

(125)

.63,

p

.001,

.78 for thetwo items. The measures on the second factor werecombined in a similar manner to construct

a negativesociometric composite

,

r

(125)

.55,

p

.001,

.71 forthe two items. The high interrelations indicate thatthe children were not differentiating a lot between lik-ing and nice or between disliking and ghting.

Data Reduction of Adults Reportson Social Functioning Data

Reports on social skills and problem behavior byparents or by Teachers 1, 2, and 3, were negatively re-lated,

r

s(103, 125, 124, 125)

.35,

.63, .49, and

.72(within reporter),

p

s

.001, respectively. Thus, foreach reporter, scores were standardized (after revers-ing problem behavior) and averaged to form a com-posite score for

social skills/low problem behavior.

Thesecomposites then were standardized and averagedacross Teachers 1, 2, and 3 (social skills and problembehavior combined across teachers were correlated,

.74) to use in analyses with parent- and child- (orpeer-) reported data. Scores also were averaged acrossTeachers 2 and 3 to use in analyses with Teacher 1 data(to examine across-reporter relations). Scores for

shy-ness

were kept separate albeit standardized and aver-aged across the three reporters for most analyses.

Data Reduction for Adult-ReportedRegulation Data

The regulation data were reduced in a manner similarto that used in prior studies and the resulting compos-ites were analogous to those used in prior research inthe United States (e.g., Eisenberg, Fabes, et al., 1997;Murphy, Shepard, Eisenberg, Fabes, & Guthrie, 1999).

Attention shifting, attention focusing, and inhibi-tory control generally are all viewed as aspects of ef-fortful regulation (Ahadi & Rothbart, 1994; Rothbart,personal communication, October 2000). Moreover,these three subscales factored on a single factor infour separate principle components factor analysescomputed for parents and for each of the threeteachers. Correlations among the three scales rangedfrom .45 to .50 for parents; from .51 to .73 for Teacher1; from .64 to .70 for Teacher 2; and from .77 to .86 forTeacher 3. Thus, the three scales were standardizedand averaged within each reporter. In addition, thesecomposite scores were intercorrelated among teachers,with correlations ranging from .35 to .58. Thus, thesescores were standardized and averaged acrossTeachers 2 and 3, as well as across Teachers 1, 2, and 3.

Consistent with the conceptual distinction be-tween negative emotionality and regulation, scoresfor negative emotionality were kept separate from theregulation composite.

RESULTS

Descriptive Analyses

Outlier analyses were computed using the SPSS re-gression program (SPSS Inc., Chicago, IL). One childhad extreme scores on ratings of liking, nice, and an-ger, and, therefore, the scores on these variables weredropped for that child. In addition, a number of vari-ables were skewed; transformations were computedto improve the skew of scores for all the peer nomina-tion scores and for Teacher 1/2/3 regulation, allteacher-reported measures of social competence/problem behavior, and parents reports of shyness.

In preliminary analyses, gender and age differ-ences in the major variables were examined.

Correlations of Major Variables with Age

Age was signicantly positively related only topeers ratings of positive sociometrics,

r

(125)

.18,

p

.045. Given the number of correlations, this nd-ing may have been due to chance.

Gender Differences in Social Functioning,Regulation, and Emotionality

Multivariate analyses were conducted to examinegender differences in measures of childrens social func-tioning, emotionality, and regulation (

M

s and

SD

s formales and females are given in Table 1). Separate multi-variate analyses were computed for the data for parents(ratings of social skills/low problem behavior, shyness,regulation, and emotionality), Teacher 1 (social skills/low problem behavior, shyness), Teacher 2/3 (socialskills/low problem behavior, shyness, regulation, emo-tionality), and peers (anger, positive sociometrics, nega-tive sociometrics). The multivariate

F

s were signicantfor the latter three:

F

(2, 124)

11.66, F(4, 120) 11.05,and F(3, 122) 8.03, ps .001, respectively, withF(4, 104) 1.83, p .13 for the parent multivariate. Ac-cording to univariate tests, females scored higher thanmales on parent-reported social skills/low problem be-havior, F(1, 107) 6.28, p .014. Teacher 1 also rated fe-males higher on social skills/low problem behavior andlower on shyness than males, Fs(1, 125) 8.47 and 4.94,ps .001 and .028, respectively. Similarly, Teacher 2/3rated females as higher on regulation and social skills/low problem behavior, and lower on negative emotion-ality than males, rs(1, 123) 16.98, 39.67, and 4.54, ps


.001, .001, and .035, respectively. Males also scoredhigher than females on negative peer sociometrics,F(1, 124) 12.30, p .001. Due to the gender differences,gender was partialed or controlled in many analyses.

Interrelations of Parent, Teacher, and PeerMeasures of Social Functioning

There generally were signicant relations betweenreporters of quality of childrens social functioning (seeTable 2). When examining correlations between parentor peer data with teacher data (across type of reporter),teacher data were combined across all three teachers.When discussing correlations between teachers (withintype of reporter), correlations were examined betweenTeacher 1 and Teacher 2/3 (correlations among teacherson the same variables were presented in the Methodsection). Patterns of correlations are emphasized wheninterpreting the results because of the possibility ofchance ndings (given the number of analyses).

Correlations among Parents and Teachersfor the Same Construct

Interrelations across raters were generally signi-cant for social functioning and for shyness. Consis-

tent with the correlations among teachers for thesemeasures (see discussion of this issue in the Methodsection), Teacher 1s and Teacher 2/3s reports ofsocial skills/low problem behavior were positivelyrelated, as were Teacher 1s and Teacher 2/3s reportsof shyness (see the bottom of Table 2). Across type ofreporter, Teacher 1/2/3s and parents reports of bothsocial skills/low problem behavior and shyness werepositively related. The correlations did not differ sub-stantially by gender.

Correlations between Similar Adultand Peer Measures

Recall that a positive sociometric composite wascomputed by averaging the standardized scores forpeer nominations for being nice and liked. A neg-ative sociometric composite was computed in a simi-lar manner for peer nominations for ghting anddisliked. Peers ratings of positive sociometricswere positively related to parents and Teacher 1/2/3s reports of social skills/low problem behavior,whereas peers negative sociometrics showed thereverse pattern of correlations (see Table 2). Thestrength of these relations argues for the validity ofthese measures. Children rated as shy by Teacher 1/2/3 scored low on peers positive and negative sociomet-rics; parent-reported shyness also was associated withlow negative sociometrics.1

Intercorrelations of Indexes of Emotionalityand Regulation

With regard to within-reporter correlations, regu-lation and negative emotionality were signicantlynegatively related for both teacher 1/2/3 and par-ents, rs(125, 107) .51 and .59, ps .001, respec-tively. Parental regulation was positively related toTeacher 1/2/3s reports of regulation and negativelyrelated to Teacher 1/2/3s reports of negative emo-tionality, rs(107) .47 and .21, ps .001 and .03, re-spectively. Parental negative emotionality was nega-tively related to Teacher 1/2/3s reports of regulation,r(107) .28, p .003, and was not signicantly re-

1 Findings for the nice versus liked (and the dislikedversus ghts) sociometric ratings were somewhat similar, al-though children rated as shy by Teacher 1/2/3 were rated as lowin liking and ghting, but not as disliked or low in niceness,rs(124, 125, 125, 124) .28, .21, 12, and .07, ps .001, .02,ns, and ns, respectively. Parent-rated social skills/low problembehavior was signicantly correlated with being liked by peersand being low in ghting, but not being nice or disliked, rs(106,107, 106, 107) .21, .19, .12, and .12, ps .03, .044, ns, and ns,respectively.

Table 1 Means of Major Variables by Gender

Variable Girls Boys

Adult report of social functioningTeacher 1 reporta .34 (.72) .29 (.94)Teacher 2/3 reporta .45 (.74) .38 (.78)Teacher 1/2/3 reporta .43 (.69) .36 (.72)Parent reporta .23 (.73) .16 (.88)

Positive peer sociometricsa .03 (.70) .08 (.65)

Negative peer sociometricsa .23 (.61) .19 (.90)

Adult report of shynessTeacher 1 report 3.39 (.57) 3.65 (.69)Teacher 2/3 reporta .04 (.93) .06 (.75)Teacher 1/2/3 reporta .11 (.78) .08 (.75)Parent report 2.85 (1.07) 3.16 (1.07)

Regulation/negative emotionalityTeachers 2/3 regulationa .31 (.82) .25 (.79)Teachers 1/2/3 regulationa .35 (.68) .29 (.77)Teacher 2/3 emotionalitya .16 (.89) .18 (.74)Teacher 1/2/3 emotionalitya .19 (.77) .19 (.67)Child-reported regulation 3.38 (.42) 3.23 (.45)Parent regulationa .10 (.77) .07 (.84)Parent negative emotionality 4.04 (.64) 4.25 (.60)Peer-reported angera .03 (.81) .10 (.72)

Note: Nontransformed means are presented for the measures.Values in parentheses are standard deviations.a This measure is a composite of standardized measures (within oracross reporters; see text).


lated to teachers 1/2/3s reports of negative emotion-ality, r(107) .14, ns. Neither parental regulation norparental negative emotionality was signicantly cor-related with peers reports of childrens anger. Peer-reported anger was positively correlated with Teacher1/2/3s reports of negative emotionality, however,r(124) .46, p .001. Childrens own reports of regu-lation also were positively related to Teacher 1/2/3sreports of regulation, r(122) .22, p .012, but werenot related to parents reports of regulation or nega-tive emotionality or peers reports of anger. Thus,there was some evidence of consistency of reportsacross respondents (i.e., parents and teachers, andteachers and peers), although relations between par-ents and peers reports of negative emotionality andregulation were nonsignicant.

Relations of Measures of Emotionality andRegulation to Social Skills/Low ProblemBehavior and Shyness

Adults reports of quality of childrens social func-tioning were generally related to their reports of chil-drens regulation and emotionality, within and acrossreporters. Both zero-order and partial correlationscontrolling for gender are presented in Table 3. We fo-cused on Teacher 1/2/3 (rather than Teacher 1 orTeacher 2/3) reports of social functioning when dis-cussing correlations of teachers reports with parentor child measures. To examine across teacher reports,Teacher 1 reports of social functioning are discussedin relation to Teacher 2/3 reports of regulation andnegative emotionality. Clearly, the number of correla-tions is far greater than chance and the pattern ofndings is emphasized.

Within Reporters

In general, high regulation and negative emotional-ity were related to social functioning and shyness inwithin-reporter correlations, with the correlations rang-ing from small to moderate size. Parents reports ofregulation correlated with their reports of high socialskills/low problem behavior and low shyness, whereastheir reports of negative emotionality were negativelyrelated to social skills/low problem behavior (but wereunrelated to shyness; see Table 3). Similarly, Teacher1/2/3s reports of regulation were positively related totheir reports of social skills/low problem behavior andnegatively related to shyness, whereas their reports ofnegative emotionality were negatively related to bothvariables. Peer-reported anger was positively and mod-erately related to negative sociometric scores.

Across Teachers

Teacher 2/3s reports of regulation were positively re-lated to Teacher 1s reports of social skills/low problembehavior and negatively related to Teacher 1s reports ofshyness. Teacher 2/3s reports of negative emotionalitywere negatively related to Teacher 1s reports of socialskills/low problem behavior and with low shyness, al-though the latter relation held only when the effects ofgender were partialed, and was weak (see Table 3).

Across Type of Reporters

Parents reports of childrens regulation were posi-tively but modestly related to Teacher 1/2/3s reportsof social skills/low problem behavior and peer positivesociometrics and negatively related to peer negative so-

Table 2 Intercorrelations among Major Measures of Social Functioning across Reporters

Measures of Social Functioning 1 2 3 4 5 6

Parent (n 109)1. Social skills/low problem behavior .32*** .25** .10 .25** .19*2. Shyness .06 .35*** .03 .20*

Teacher 1/2/3 (n 127)3. Social skills/low problem behavior .03 .35*** .62***4. Shyness .21* .19*

Peer (n 127)5. Positive sociometrics .36***6. Negative sociometrics

Note: Teacher 1s reports of social skills/low problem behavior correlated .53 and .14 with Teacher2/3s reports of social skills/low problem behavior and shyness, ps .001 and .11, respectively; analo-gous correlations for Teacher 1s reports of shyness with Teacher 2/3s reports of social skills/lowproblem behavior and shyness were .08 and .44, ns and p .001, respectively.* p .05; ** p .01; *** p .001.


ciometrics (see Table 3). Similarly, Teacher 1/2/3s re-ports of regulation were positively related to parentsreports of social skills/low problem behavior andpeers reports of positive sociometrics, and negativelyrelated to peers reports of negative sociometrics; thesecorrelations sometimes were moderate in strength. Par-ents reports of negative emotionality were negativelyrelated to peer positive sociometrics. Teacher 1/2/3sreports of negative emotionality were at least margin-ally negatively related to parents reports of childrensshyness (this relation was signicant when genderwas partialed) and peer-positive sociometrics andpositively related to peer-negative sociometrics. Gener-ally, these relations across reporters for shyness weresmall in size.2

Child-reported regulation was positively relatedto Teacher 1/2/3s (but not to parents) reports of so-cial skills/low problem behavior. Child self-reportedregulation was not related to peers positive and neg-ative sociometrics or parents reports. Peers reportsof childrens anger were negatively related to par-ents and Teacher 1/2/3s reports of shyness, at p .053 or better, and negatively related to Teacher 1/2/3s reports of social skills/low problem behavior (al-though the latter relation was only marginally signif-icant in the zero-order correlation). Thus, ndings forpeer-reported anger and self-reported regulationgenerally were weak, albeit in the expected direction.Nonetheless, overall, there was some evidence,across as well as within reporters, of high regulationand/or low negative emotionality (especially theformer) being associated with social skills/low prob-lem behavior, positive peer sociometrics, and/or lownegative peer sociometrics. Shyness tended to be re-lated to low regulation and/or low negative emo-tionality, but relations were less consistent (espe-cially for parents reports of shyness).

2 Peers reports of liking were signicantly correlated withparents reports of high regulation and low negative emotional-ity, whereas analogous relations for peer nominations of nice-ness were not signicant, rs(106) .27, .19, .10, and .04, ps .005, .047, ns, and ns, respectively. In contrast, Teacher 1/2/3sreports of regulation and low negative emotionality were re-lated to peer reports of niceness, rs(124) .37 and .24, ps .001 and .007, respectively; whereas peer ratings of liking weresignicantly related to teachers reports of regulation but notnegative emotionality, rs(124) .41 and 12, ps .001 and ns,respectively. Peers reports of disliking were at least negativelycorrelated with Teacher 1/2/3- and parent-reported regulationand positively related to Teacher 1/2/3s reports of negativeemotionality, rs(125, 107, 125) .48, .49, and .17, ps . 001,.001, and .076, respectively; analogous correlations were ob-tained for peers reports of ghting, rs(125, 107, 125) .46, .49,and .27, ps .001, .001, and .005, respectively. None of thesedifferences was signicant.Ta

ble

3Z

ero-

Ord

er a

nd

Par

tial

Cor

rela

tion

s (C

ontr

olli

ng

Gen

der

) bet

wee

n M

easu

res

of S

ocia

l Fu

nct

ion

ing

and

Reg

ula

tion

or

Neg

ativ

e E

mot

ion

alit

y

Com

posi

tes

Pare

nt-R

epor

ted

Reg

ulat

ion

Pare

ntN

egat

ive

Em

otio

nalit

y

Teac

her-

Rep

orte

d

Reg

ulat

ion

Teac

her

Neg

ativ

e E

mot

iona

lity

Chi

ld-

Rep

orte

dR

egul

atio

nPe

er-R

epor

ted

Ang

er

Pare

nt (n

1

09)

Soci

al s

kills

/lo

w p

robl

em b

ehav

ior

.53*

** (.

53**

*)

.47*

** (

.45)

***

.34*

** (.

27**

)

.13

(.0

8).0

6 (.0

2).0

3 (.0

0)Sh

ynes

s

.33*

** (

.32*

**)

.14

(.12)

.0

1 (.0

4)

.17

(.2

1*)

.11

(.13)

.2

0* (

.19

)

Tea

cher

1 (n

1

27)

Soci

al s

kills

/lo

w p

robl

em b

ehav

ior

.52*

** (.

45**

*)

.31*

** (

.27*

*)Sh

ynes

s

.29*

** (

.24*

*)

.11

(.1

8*)

Tea

cher

1/

2/3

(n

127

)So

cial

ski

lls/

low

pro

blem

beh

avio

r.3

2***

(.31

***)

.1

7 (

.10)

.81*

** (.

77**

*)

.66*

** (

.64*

**)

.22*

(.16

)

.1

5 (

.24*

*)Sh

ynes

s

.12

(.1

1).1

6 (.1

4)

.25*

* (

.22*

)

.33*

** (

.38*

**)

.1

7 (

.15

)

.41*

** (

.40*

**)

Peer

(n

127)

Posi

tive

soc

iom

etri

cs.2

5**

(.24*

)

.19*

(.1

7)

.50*

** (.

50**

*)

.20*

(.1

8*)

.07

(.05)

.05

(.04)

Neg

ativ

e so

ciom

etri

cs

.26*

* (

.24*

).0

9 (.0

4)

.53*

** (

.47*

**)

.56*

** (.

53**

*)

.11

(.0

7).5

2***

(.57

***)

Not

e:V

alue

s re

pres

ent z

ero-

ord

er c

orre

lati

ons;

val

ues

in p

aren

thes

es r

epre

sent

cor

rela

tion

s pa

rtia

ling

gend

er. C

orre

lati

ons

for

Tea

cher

1 a

re e

xam

ined

onl

y in

rel

atio

n to

Tea

cher

2/

3 d

ata

(for

acr

oss-

teac

her

find

ings

). In

all

othe

r co

rrel

atio

ns in

volv

ing

teac

her

regu

lati

on o

r ne

gati

ve e

mot

iona

lity,

the

Tea

cher

1/

2/3

com

posi

te w

as u

sed

.*p

.0

5; *

*p

.01;

***

p

.001

; p

.1

0.


Sociometric Group and Differencesin Emotionality and Regulation

The relations of adults reports of childrens emo-tionality and regulation to childrens sociometricgroup status were examined with two 2 (gender) 5(sociometric status: rejected, neglected, average, con-troversial, and popular) multivariate analyses of vari-ance. Parents or Teacher 1/2/3s reports of regula-tion and emotionality were the dependent variables.Gender differences in dependent variables are not re-ported again.

For Teacher 1/2/3s reports, the multivariate Fwas signicant for sociometric status group, PillaisF(8, 232) 7.73, p .001. The univariate effect ofsociometric status was signicant for Teacher 1/2/3s reports of both regulation and negative emo-tionality, Fs(4, 116) 12.61 and 6.59, ps .001, par-tial 2 (eta squared) .30 and .19, respectively. Ac-cording to post hoc Newman-Keuls tests, p .05),the rejected group was rated lower on regulationthan all other groups; moreover, average childrenwere lower than popular children on regulation.Rejected children also scored higher than ne-glected, popular, and average children on negativeemotionality. The multivariate and univariate ef-fects of group were not signicant for the parentdata, although the univariate effect for parent-reported regulation was marginally signicant,F(4, 98) 2.07, p .09, partial 2 .026, and ac-cording to Duncans tests, rejected children wereless regulated than popular children.3

3 Adults also rated childrens sympathy with ve items (e.g.,This child often feels sorry for others who are less fortunate;Eisenberg, Fabes, et al., 1998); s for parents and Teachers 1, 2,and 3 .77, .87, .64, and .69, respectively. Scores for Teachers 2and 3 were correlated, r(114) .19, p .039. Although Teacher1s reports of sympathy were not signicantly related to Teacher2s or Teacher 3s reports, Teacher 1s reports were signicantlyrelated to the combination Teacher 2/3s reports, r(125) .22, p .014. Thus, Teachers 1, 2, and 3s reports were standardized andaveraged. Parent sympathy correlated with parents reports ofhigh regulation and low negative emotionality, rs(103) .28 and.24, ps .004 and .013, respectively; the relation with Teacher1/2/3s reports of regulation was marginal, r(103) .17, p .082. Teacher 1/2/3s reports of sympathy were positively re-lated to Teacher 1/2/3- and parent-reported regulation and neg-atively related to Teacher 1/2/3- and parent-reported negativeemotionality, rs(125, 103, 125, 107) .63, .34, .20, and .24, ps .001, .001, .025, and .012, respectively. Teacher 1s reports of sym-pathy were related to Teacher 2/3s reports of regulation, r(125) .17, p .05. Self-reported (child) regulation was positively cor-related with Teacher 1/2/3s (but not to parents) reports ofchildrens sympathy, r(122) .30, p .001. These ndings aresimilar to those found in the United States (Eisenberg, Fabes,Murphy, et al., 1996, 1998; Murphy et al., 1999).

Examination of Unique and MultiplicativePrediction by Regulation and Emotionality

The unique and multiplicative (i.e., interacting)effects of regulation and emotionality in predictingchild outcomes were examined with regression anal-yses (see Table 4). In a series of separate hierarchicalregression analyses, adults reports of social skills/low problem behavior, shyness, or peer positive ornegative sociometrics, were predicted by teachersor parents reports of regulation and emotionality, aswell as with the multiplicative term for the interac-tion of regulation and emotionality. Teacher 1/2/3sreports of regulation and negative emotionality wereused to predict parent or peer measures of social func-tioning (or vice versa for parent data) whereasTeacher 2/3s reports of regulation and negativeemotionality were used to predict Teacher 1s mea-sures of social functioning. Age and gender were en-tered in the rst step; teachers (or parents) reports ofregulation and negative emotionality were entered inthe second step, and the multiplicative term for theinteraction of regulation and negative emotionalitywas entered in the last step. The latter step was signif-icant in only one analysis (about the number expectedby chance) and thus this nding was not examined.The lack of signicant interactions is not surprising;power for teachers and parent predictorsgiven anestimated effect size of .03 (as in Eisenberg, Fabes,Guthrie, et al., 1996), .05was .49 and .43 (or .61and .54 for an estimated effect size of .04).

Teachers Regulation and Negative Emotionality

As can be seen in Table 4, Teacher 1/2/3s reportsof regulation and negative emotionality had uniqueeffects when used to predict Teacher 1/2/3s reportsof social skills/low problem behavior and peers neg-ative nominations. Teacher-reported social skills/lowproblem behavior generally was predicted by highregulation and low emotionality, whereas the reversepattern was found for negative sociometrics. Teacher1s reports of social skills/low problem behavior wereuniquely predicted only by Teacher 2/3s reports ofregulation. Only Teacher 1/2/3s reports of regula-tion predicted parent-reported social skills/low prob-lem behavior and peer positive sociometrics when theunique effects of regulation and negative emotional-ity were examined.

Teacher 2/3s reports of low regulation and lownegative emotionality both predicted unique vari-ance in Teacher 1s reports of shyness; similar nd-ings were obtained when Teacher 1/2/3s reports ofregulation and negative emotionality were used to


predict their own reports of shyness. Parent-reportedshyness was uniquely predicted only by Teacher 1/2/3s reports of low negative emotionality. As can be seenin Table 4, the s were modest to moderate in size.

Parents Reports of Regulationand Negative Emotionality

Parents regulation and negative emotionality alsowere used to predict parents and Teacher 1/2/3s re-ports of social/skills/low problem behavior and shy-ness, as well as peer sociometrics. As can be seen inTable 4, there were unique additive effects of the twovariables when predicting parents own reports of so-cial skills/low problem behavior, with high regula-tion and low negative emotionality predicting highsocial skills. In contrast, only parents reports of regu-lation predicted Teacher 1/2/3s reports of high socialskills/low problem behavior, as well as high positiveand low negative peer sociometrics and parents re-ports of childrens shyness. There were few unique ef-fects of parents reports of negative emotionality.

DISCUSSION

For the most part, the ndings in the present studywere consistent with those obtained in the UnitedStates and other Western countries (e.g., Caspi et al.,1995; Eisenberg et al., 2000; Pulkkinen & Hamalainen,1995), which suggests that the processes involved inemotion regulation and its relation to social function-

ing are similar across diverse cultures. Teachers andparents reports of high regulation and/or low nega-tive emotionality (i.e., intense emotions and especiallyanger) generally were associated with parents,teachers, and peers reports of social skills/low prob-lem behaviors. This pattern of ndings was foundacross reporters and sometimes across settings (homeversus school) as well. In addition, in some cases,teacher-reported negative emotionality and regula-tion, albeit correlated constructs, accounted for uniquevariance in quality of social functioning, especially asreported by teachers and peers. Thus, in Java as in theUnited States, children who were well regulated or lowin negative emotionality (e.g., intense negative emo-tions and peer-reported anger) were viewed as sociallyskilled and/or adjusted by adults and peers. Althoughparental reports of negative emotionality sometimescorrelated with measures of childrens social function-ing, the number of correlations were fewer than forteacher-report data and there were few unique effectsof parent-reported emotionality (only for their ownreports of socially appropriate behavior/low problembehavior) when regulation also was a predictor. None-theless, in general, the ndings support the view thattemperamental differences in childrens effortful regu-lation and emotionality contribute to individual differ-ences in social functioning in Indonesia as well as inWestern cultures. The pattern of ndings, however,was not so strong that one could conclude that these in-dividual differences are stronger predictors of socialfunctioning in Java than in Western cultures.

Table 4 Prediction of Social Functioning from Parents and Teachers Reports of Regulation and Negative Emotionality

R2 Change Regulation Negative Emotionality F for Step 2

Parent Teacher Parent () Teacher () Parent () Teacher () Parent Teacher

Social skills/lowproblem behavior

Parent .30*** .08* .40*** .32** .22* .07 F(2, 104) 24.51*** F(2, 104) 4.57*Teacher 1 .19*** .41*** .11 F(2, 122) 17.01***Teacher 1/2/3 .08** .51*** .33** .57*** .12 .32*** F(2, 104) 6.30** F(2, 122) 138.10***

ShynessParent .11*** .05 .38*** .09 .11 .26* F(2, 104) 6.50*** F(2, 104) 2.86

Teacher 1 .13*** .37*** .29** F(2, 122) 9.95***Teacher 1/2/3 .03 .32*** .05 .53*** .15 .61*** F(2, 104) 1.62 F(2, 122) 29.83***

PeerPositive sociometrics .07* .23*** .23* .54*** .06 .03 F(2, 103) 4.15* F(2, 121) 19.33***Negative sociometrics .07* .32*** .32*** .30* .14 .40*** F(2, 104) 4.13* F(2, 122) 32.55***

Note: Parent reports of negative emotionality/regulation were examined only in relation to Teacher 1/2/3, not Teacher 1. Teacher 2/3s re-ports of regulation and negative emotionality were used in analyses in which the criterion variables were Teacher 1s reports (so the analy-ses are across reporter); Teacher 1/2/3s reports of regulation and negative emotionality were the predictors in all other analyses involvingteacher regulation and negative emotionality. Age and gender were controlled on the rst step; data presented are for the main effects ofregulation and negative emotionality on the second step.* p .05; ** p .01; *** p .001; p .10.


Unlike research performed in the United States, in-teraction effects generally were not obtained whenpredicting indexes of quality of social functioningfrom both regulation and emotionality. The samplesize in this study was relatively small, and, thus,power to detect a relatively small effect size wasweak. In the United States, the effect size for suchndings in an elementary school sample was about.02 to .04, and power in the present study for a .03 ef-fect size was less than .50. Moreover, measures ofemotionality and regulation were moderately related,which would reduce the likelihood of obtaining inter-action effects. It is possible that interactions effectswould be apparent in a larger sample. It also is possiblethat regulation is so important in Javanese culture that,unlike in the United States (Eisenberg, Fabes, et al.,1997; Eisenberg, Guthrie, et al., 1997), it predicts out-comes equally well for unemotional and emotionalchildren. Nonetheless, the ndings for teachers re-ports of negative emotionality and regulation partlysupport Eisenberg and Fabes (Eisenberg et al., 2000)heuristic model in which individual differences innegative emotionality and regulation are importantfor the prediction of quality of social functioning.

The ndings for sociometric status groups, likethose for other measures of social functioning, gener-ally were consistent with those in Western countries(Coie, Dodge, & Kupersmidt, 1990; Maszk et al., 1999;Newcomb et al., 1993). Children who were classied asrejected were rated lower by teachers on regulationthan were all other sociometric groups. Further, re-jected children were higher on negative emotionalitythan were popular, neglected, and average children. Itappears that children who display more extreme nega-tive and intense emotions (e.g., teacher-reported anger)and low regulation are rejected by their peers. Childrenwho are more regulated and less negative are rela-tively likely to behave in a prosocial manner (Eisen-berg, Fabes, Murphy, et al., 1996) and simply may bemore pleasant in social interactions. Moreover, angeris likely to be linked to reactive aggression and otherexternalizing problems that can alienate peers (Coieet al., 1990; Newcomb et al., 1993).

Peers reports of anger did not relate consistentlyto adult-reported social skills/low problem behavior,although they were positively related to negative peersociometrics. The latter nding is consistent with datain different cultures indicating that aggression tendsto be associated with negative and not positive peerevaluations (Chen et al., 1995; Coie et al., 1990; Frenchet al., 1999). Peers reports of anger were moderatelycorrelated with teachers reports of negative emotion-ality, suggesting that peers perceptions were valid.

Adult-reported shyness was not consistently related

to adult-reported social skills/low problem behavior.Similar to French et al.s ndings (1999), however,teachers reports of childrens shyness were nega-tively related to peers reports of liking and prosocialbehavior combined (but only liking, when examinedseparately); peers also viewed teacher- and parent-rated shy children as low in negative sociometrics(disliked/ghts). Thus, shy children appeared to beneglected by peers in their nominations (in fact,teacher-rated shy children were signicantly moreoften neglected than controversial children in amultivariate analysis of variance). Shy children alsowere rated by adults as somewhat unregulated andwere rated by teachers as low in intense emotions orexternalizing negative emotions. Thus, in Indonesia,there seems to be some relation between social with-drawal in the classroom and low peer liking of achild, although shy children also were not disliked.Although shy children are viewed as low in effortfulattentional and inhibitory control, they are not viewedas expressing much negative emotion. These ndingsclearly differ from those obtained in China (e.g., Chenet al., 1995) where shyness has been associated withpositive peer status and teacher ratings. It is possiblethat the ndings differ because Chens measure of so-cial withdrawal or shyness included peer nomina-tions for sensitivity (someone whose feelings gethurt very easily) and sadness (e.g., someone whois usually sad) as well as shyness (someone who isvery shy). Thus, the content of their scale differedsomewhat from ours and, in the Chen et al. study(1995, 1999), peers, rather than teachers, reported onshyness. In future work, it will be useful to determineif it was the measure, the reporter, or cultural factorsthat were responsible for the different patterns ofndings in this sample and in China. In any case, itdoes not appear from our data that shy children in In-donesia are well liked and regulated; if anything, asin the United States, the reverse seems to be true.

It is worth noting that there was considerableacross-reporter consistency among parents, teachers,and peers with regard to their reports of childrens so-cial functioning. Moreover, adults reports of chil-drens shyness generally were signicantly related,especially among teachers. There also was some agree-ment among parents, teachers, and children (self-reports) with regard to childrens level of regulation.Some relations between variables were higher be-tween teachers and peers than between parents andpeers, quite possibly because peers and teachers ob-served children in the same setting.

The gender differences in childrens regulation andsocial skills/low problem behavior were very similarto those often obtained in the United States (e.g.,


Eisenberg, Fabes, Murphy, et al., 1996; Eisenberg,Shepard, et al., 1998). In general, males were ratedhigher than females on intense and negative emotion-ality and negative sociometrics and were rated lowerin regulation and social skills. The cross-cultural con-sistency in these ndings could be due to either bio-logically based gender differences (e.g., in tempera-ment) or consistencies across cultures in socialization(or both). Regardless, given that these gender differ-ences likely contribute to how children are viewed byothers, they may contribute to similar behaviors oroutcomes in cultures as different as Java and theUnited States.

In conclusion, our results with regard to thepatterns among regulation, negative emotionality,and social functioning in a sample of Indonesianchildren generally are consistent with data fromchildren in the United States. Although Indonesiaand the United States present very different socio-cultural contexts, initial ndings support the viewthat it is useful to consider both individual differ-ences in negative emotionality and regulation whenpredicting social outcomes for children. Of course,because the data were correlational, we cannotdraw conclusions about causal relations. In addi-tion, some of the correlations were modest in size.Moreover, the ndings would likely differ if themeasure of emotionality tapped emotions such asfear, guilt, or shame rather than primarily external-izing negative emotions. Finally, because there aremany diverse ethnic groups in Indonesia, our nd-ings may not apply to other ethnic and languagegroups in that country.

ACKNOWLEDGMENT

This research was supported by grants (1 R01HH55052 and 1 R01 MH 60838) and a Research Sci-entist Award from the National Institute of MentalHealth (K05 M801321) to the rst author. This workalso was facilitated by a Fulbright grant to Sri Pi-dada. The authors wish to thank the parents,teachers, and children involved in this study, andAmanda Shefeld Morris for her comments on themanuscript.

ADDRESSES AND AFFILIATIONS

Corresponding author: Nancy Eisenberg, Depart-ment of Psychology, Arizona State University, P.O.Box 871104, Tempe, AZ 85287-1104; e-mail: [email protected]. Sri Pidada is at Padjadjaran Uni-versity, Bandung, Indonesia; Jeffrey Liew is also atArizona State University.

REFERENCES

Achenbach, T. M. (1991). Manual for the Child Behavior Check-list/4-18 and 1991 prole. Burlington, VT: University ofVermont Department of Psychiatry.

Ahadi, S. A., & Rothbart, M. K. (1994). Temperament, devel-opment, and the big ve. In C. F. Halverson, Jr., G. A.Kohnstamm, & R. P. Martin (Eds.), The developing struc-ture of temperament and personality from infancy to adult-hood (pp. 189207). Hillsdale, NJ: Erlbaum.

Asendorpf, J. B. (1987). Videotape reconstruction of emo-tions and cognitions related to shyness. Journal of Person-ality and Social Psychology, 53, 542549.

Block, J. H., & Block, J. (1980). The role of ego-control andego-resiliency in the organization of behavior. In W. An-drew Collins (Ed.), Development of cognition, affect, and so-cial relations. The Minnesota Symposia on Child Psychology(Vol. 13, pp. 39101). Hillsdale, NJ: Erlbaum.

Campos, J. J., Campos, R. G., & Barrett, K. C. (1989). Emer-gent themes in the study of emotional development andemotion regulation. Developmental Psychology, 25, 394402.

Caspi, A. (1998). Personality development across the lifecourse. In N. Eisenberg (Ed.), W. Damon (Series Ed.),Handbook of child psychology: Vol. 3 Social, emotional, andpersonality development (5th ed., pp. 311388). New York:Wiley.

Caspi, A. (2000). The child is father of the man: Personalitycontinuities from childhood to adulthood. Journal of Per-sonality and Social Psychology, 78, 158172.

Caspi, A., Henry, B., McGee, R. O., Moftt, T. E., & Silva, P. A.(1995). Temperamental origins of child and adolescentbehavior problems: From age three to age fteen. ChildDevelopment, 66, 5568.

Chen, X., Rubin, K. H., Li, B.,& Li, D. (1999). Adolescent out-comes of social functioning in Chinese children. Interna-tional Journal of Behavioral Development, 23, 199223.

Chen, X., Rubin, K. H., & Li, Z. (1995). Social functioningand adjustment in Chinese children: A longitudinalstudy. Developmental Psychology, 31, 531539.

Coie, J. D., & Dodge, K. A. (1988). Multiple sources of dataon social behavior and social status in the school: Across-age comparison. Child Development, 59, 815829.

Coie, J. B., Dodge, K. A., & Coppotelli, H. (1982). Dimen-sions and types of social status: A cross-age perspective.Developmental Psychology, 18, 557570.

Coie, J. D., Dodge, K., & Kupersmidt, J. B. (1990). Peergroup behavior and social status. In S. R. Asher & J. D.Coie (Eds.), Peer rejection in childhood. Cambridge, U.K.:Cambridge University Press.

Derryberry, D., & Rothbart, M. K. (1988). Arousal, affect,and attention as components of temperament. Journal ofPersonality and Social Psychology, 55, 958966.

Dodge, K. A., Coie, J. D., Pettit, G. S., & Price, J. M. (1990).Peer status and aggression in boys groups: Develop-mental and contextual analyses. Child Development, 61,12891309.

Eisenberg, N., Cumberland, A., Spinrad, T. L., Fabes, R. A.,Shepard, S. A., Reiser, M., Murphy, B. C., Losoya, S. H., &


Guthrie, I. K. (2001). The relations of regulation andemotionality to childrens externalizing and internaliz-ing problem behavior. Child Development, 72, 11121134.

Eisenberg, N., Fabes, R. A., Guthrie, I. K., Murphy, B. C.,Maszk, P., Holmgren, R., & Suh, K. (1996). The relationsof regulation and emotionality to problem behavior inelementary school children. Development and Psychopa-thology, 8, 141162.

Eisenberg, N., Fabes, R. A., Guthrie, I. K., & Reiser, M.(2000). Dispositional emotionality and regulation: Theirrole in predicting quality of social functioning. Journal ofPersonality and Social Psychology, 78, 136157.

Eisenberg, N., Fabes, R. A., Karbon, M., Murphy, B. C.,Wosinski, M., Polazzi, L., Carlo, G., & Juhnke, C. (1996).The relations of childrens dispositional prosocial behav-ior to emotionality, regulation, and social functioning.Child Development, 67, 974992.

Eisenberg, N., Fabes, R. A., & Murphy, B. (1995). Relationsof shyness and low sociability to regulation and emo-tionality. Journal of Personality and Social Psychology, 68,505517.

Eisenberg, N., Fabes, R. A., Murphy, B. C., Karbon, M.,Smith, M., & Maszk, P. (1996). The relations of childrensdispositional empathy-related responding to their emo-tionality, regulation, and social functioning. Developmen-tal Psychology, 32, 195209.

Eisenberg, N., Fabes, R. A., Shepard, S. A., Murphy, B. C.,Guthrie, I. K., Jones, S., Friedman, J., Poulin, R., &Maszk, P. (1997). Contemporaneous and longitudinalprediction of childrens social functioning from regula-tion and emotionality. Child Development, 68, 642664.

Eisenberg, N., Fabes, R. A., Shepard, S. A., Murphy, B. C.,Jones, J., & Guthrie, I. K. (1998). Contemporaneous andlongitudinal prediction of childrens sympathy from dis-positional regulation and emotionality. DevelopmentalPsychology, 34, 910924.

Eisenberg, N., Guthrie, I. K., Fabes, R., Reiser, M., Murphy,B. C., Holmgren, R., Maszk, P., & Losoya, S. (1997). Therelations of regulation and emotionality to resiliency andcompetent social functioning in elementary school chil-dren. Child Development, 68, 295311.

Eisenberg, N., Shepard, S. A., Fabes, R. A., Reiser, M., Mur-phy, B. C., & Guthrie, I. K. (1998). Shyness and childrensemotionality, regulation, and coping: Contemporane-ous, longitudinal, and across context relations. Child De-velopment, 69, 767790.

Epstein, S. (1979). The stability of behavior: 1. On predictingmost of the people most of the time. Journal of Personalityand Social Psychology, 37, 10971126.

Farver, J. A. M., & Wimbarti, S. (1995). Indonesian childrensplay with their mothers and older siblings. Child Devel-opment, 66, 14931503.

French, D. C. (1988). Heterogeneity of peer-rejected boys:Aggressive and nonaggressive subtypes. Child Develop-ment, 59, 976985.

French, D. C., Jansen, E., Fosco, A., Rianasari, M., Pidada, S.,& Nelwan, P. (2000). Friendship qualities of United Statesand Indonesian children and adolescents. Manuscript sub-mitted for publication.

French, D. C., Setiono, K., & Eddy, J. M. (1999). Bootstrap-ping through the cultural comparison mineeld: Child-hood social status and friendship in the United Statesand Indonesia. In W. A. Collins & B. Laursen (Eds.), Re-lationships as developmental contexts: The Minnesota Sym-posia on Child Psychology (Vol. 30, pp. 109131). Mahwah,NJ: Erlbaum.

Geertz, C. (1976). The religion of Java. Chicago: University ofChicago Press.

Geertz, H. (1961). The Javanese family: A study of kinship andsocialization. Prospect Heights, IL: Waveland.

Goldsmith, H. H., & Rothbart, M. K. (1991). Contemporaryinstruments for assessing early temperament by ques-tionnaire and in the laboratory. In A. Angleitner & J. Stre-lau (Eds.), Explorations in temperament (pp. 249272).New York: Plenum.

Gottman, J. M., Katz, L. F., & Hooven, C. (1996). Parentalmeta-emotion philosophy and the emotional life of fam-ilies: Theoretical models and preliminary data. Journal ofFamily Psychology, 10, 243268.

Harter, S. (1982). The perceived competence scale for chil-dren. Child Development, 53, 8797.

Hartup, W. W., Glazer, J. A., & Charlesworth, R. (1967). Peerreinforcement and sociometric status. Child Development,38, 10171024.

Hofstede, G. (1991). Cultures and organizations: Software of themind. London: McGraw-Hill.

Izard, C. E., Libero, D. Z., Putnam, P., & Haynes, O. M.(1993). Stability of emotion experience and their rela-tions to traits of personality. Journal of Personality and So-cial Psychology, 64, 847860.

Kitayama, S., & Markus, H. R. (1994). Introduction to cul-tural psychology and emotion research. In S. Kitayama& H. R. Markus (Eds.), Emotion and culture: Empiricalstudies of mutual inuence (pp. 119). Washington, DC:American Psychological Association.

Kochanska, G., Coy, K. C., Tjebkes, T. L., & Husarek, S. J.(1998). Individual differences in emotionality in infancy.Child Development, 64, 375390.

Kochanska, G., Murray, K., & Coy, K. C. (1997). Inhibitorycontrol as a contributor to conscience in childhood: Fromtoddler to early school age. Child Development, 68, 263277.

Kochanska, G., Murray, K., Jacques, T. Y., Koenig, A. L., &Vandegeest, K. A. (1996). Inhibitory control in youngchildren and its role in emerging internalization. ChildDevelopment, 67, 490507.

Koentjaraningrat. (1985). Javanese culture. New York: Ox-ford University Press.

Larsen, R. J., & Diener, E. (1987). Affect intensity as an indi-vidual difference characteristic: A review. Journal of Re-search in Personality, 21, 139.

Leary, M. R. (1986). Affective and behavioral components ofshyness: Implications for theory, measurement, and re-search. In W. H. Jones, J. M. Cheek, & S. R. Briggs (Eds.),Shyness: Perspectives on research and treatment (pp. 2738).New York: Plenum.

Lerner, J. V. (1984). The import of temperament for psycho-social functioning: Tests of a goodness of t model.Merrill-Palmer Quarterly, 30, 177188.


Lerner, J. V., Hertzog, C., Hooker, K. A., Hassibi, M., &Thomas, A. (1988). A longitudinal study of negativeemotional states and adjustment from early childhoodthrough adolescence. Child Development, 59, 356366.

Lochman, J. E., & the Conduct Problems Prevention Re-search Group. (1995). Screening of child behavior prob-lems for prevention programs at school entry. Journal ofConsulting and Clinical Psychology, 63, 549559.

Markus, H. R., & Kitayama, S. (1991). Culture and the self:Implications of cognition, emotion, and motivation. Psy-chological Review, 20, 568579.

Maszk, P., Eisenberg, N., & Guthrie, I. K. (1999). Relations ofchildrens social status to their emotionality and regula-tion: A short-term longitudinal study. Merrill-PalmerQuarterly, 45, 468492.

Mulder, N. (1989). Individual and society in Java: A culturalanalysis. Yogualarta, Indonesia: Gadjah Mada UniversityPress.

Mulder, N. (1996). Inside Indonesian society: Cultural change inIndonesia. Amsterdam, The Netherlands: The Pepin Press.

Murphy, B. C., Shepard, S. A., Eisenberg, N., Fabes, R. A., &Guthrie, I. K. (1999). Contemporaneous and longitudinalrelations of young adolescents dispositional sympathyto their emotionality, regulation, and social functioning.Journal of Early Adolescence, 19, 6697.

Newcomb, A. F., Bukowski, W., & Pattee, L. (1993). Chil-drens peer relations: A meta-analytic review of popular,rejected, neglected, controversial, and average sociomet-ric status. Psychological Bulletin, 113, 99128.

Oyserman, D. (1993). The lens of personhood: Viewing theself, others, and conict in a multicultural society. Journalof Personality and Social Psychology, 65, 9931009.

Pulkkinen, L., & Hamalainen, M. (1995). Low self-control asa precursor to crime and accidents in a Finnish longitu-dinal study. Criminal Behaviour and Mental Health, 5, 424438.

Rohrbeck, C. A., Azar, S. T., & Wagner, P. E. (1991). Childself-control rating scale: Validation of a child self-reportmeasure. Journal of Clinical Child Psychology, 20, 179183.

Rothbart, M. K., Ahadi, S. A., & Evans, D. E. (2000). Temper-ament and personality: Origins and outcomes. Journal ofPersonality and Social Psychology, 78, 122135.

Rothbart, M. K., Ahadi, S. A., & Hershey, K. L. (1994). Tem-perament and social behavior in childhood. Merrill-Palmer Quarterly, 40, 2139.

Rothbart, M. K., & Bates, J. E. (1998). Temperament. InN. Eisenberg (Ed.), W. Damon (Series Ed.), Handbook ofchild psychology: Vol. 3. Social, emotional, and personality de-velopment (5th ed., pp. 105176). New York: Wiley.

Rothbart, M. K., Posner, M. I., & Hershey, K. L. (1995). Tem-perament, attention and developmental psychopathol-ogy. In D. Cicchetti & D. Cohen (Eds.), Manual of develop-mental psychopathology (Vol. 1, pp. 315340). New York:Wiley.

Rubin, K. H., & Asendorpf, J. B. (1993). Social withdrawal,inhibition, and shyness in childhood: Conceptual and

denitional issues. In K. H. Rubin & J. B. Asendorpf(Eds.), Social withdrawal, inhibition, and shyness in child-hood (pp. 317). Hillsdale, NJ: Erlbaum.

Rubin, K. H., Bukowski, W., & Parker, J. G. (1998). Peer in-teractions, relationships, and groups. In N. Eisenberg(Ed.), W. Damon (Series Ed.), Handbook of child psychol-ogy: Vol. 3. Social, emotional, and personality development(5th ed., pp. 619700). New York: Wiley.

Rubin, K. H., Chen, X., McDougall, P., Bowker, A., &McKinnon, J. (1995). The Waterloo Longitudinal Project:Predicting internalizing and externalizing problems inadolescence. Development and Psychopathology, 7, 751764.

Saarni, C. (1990). Emotional competence: How emotionsand relationships become integrated. In R. A. Thompson(Ed.), Socioemotional development (pp. 115182). Lincoln:University of Nebraska Press.

Stocker, C., & Dunn, J. (1990). Sibling relationships in child-hood: Links with friendships and peer relationships.British Journal of Developmental Psychology, 8, 227244.

Super, C. M., & Harkness, S. (1982). The development of af-fect in infancy and early childhood. In D. A. Wagner &H. W. Stevenson (Eds.), Cultural perspectives on child de-velopment (pp. 119). San Francisco: W. H. Freeman.

Triandis, H. C. (1995). Individualism and collectivism. Boul-der, CO: Westview.

Van Beek, A. M. (1987). Pastoral counseling challenges inthe Javanese hospital: A cross-cultural comparison. Pas-toral Psychology, 36, 112122.

Walden, T. A., & Smith, M. C. (1997). Emotion regulation.Motivation and Emotion, 21, 725.

Weinberger, D. A., & Schwartz, G. E. (1990). Distress and re-straint as superordinate dimensions of self-reported ad-justment: A typological perspective.

Eisenberg, Pidada, Liew - Child Development 2001

Documents

Transcript of Eisenberg, Pidada, Liew - Child Development 2001