1. family and parenting
2. children

Child Development, September/October 2003, Volume 74, Number 5, Pages 1314 – 1327
Prosocial Development in Relation to Children’s and Mothers’
Psychological Problems
Dale F. Hay and Susan Pawlby
The study tested whether children’s prosocial behavior was negatively or positively related to children’s and
mothers’ psychological problems. Participants were 149 London families when mothers were pregnant and
followed up in infancy and at ages 4 and 11. Children’s cooperation at 4 and general prosocial tendencies at 11
were negatively associated with externalizing problems but unrelated to internalizing problems. A subgroup of
children who were more prosocial than average expressed clinically significant worries about family members.
Maternal depression decreased prosocial behavior in the eyes of adults, but children of depressed mothers saw
themselves to be prosocial. Early cooperation protected children against later risk for externalizing problems,
even when their early behavioral problems were taken into account.
Are prosocial children free from psychological
problems? Different answers might be provided to
this question, depending on whether prosocial
action is seen as emotionally regulated and socially
competent behavior or, in some circumstances, as a
sign of overconcern for the well-being of others. One
aim of this study was to examine links between
children’s prosocial behavior and their emotional
and behavioral problems, from early childhood to
early adolescence. A second aim was to determine
whether mothers’ own psychological problems
inhibit or facilitate children’s prosocial behavior at
either point in development.
In addressing these questions, we defined prosocial behavior as a broad domain of ‘‘behaviors that
are positively responsive to others’ needs and
welfare. y At a descriptive level, they are behaviors
that are helpful and affiliative responses to others:
responses to signs of suffering, need, or danger in
another person or animal, such as, assisting, sharing,
being kind and considerate, comforting, cooperating,
Dale F. Hay, School of Psychology, Cardiff University; Susan
Pawlby, Section of Perinatal Psychiatry, Institute of Psychiatry.
The study began as a Ph.D. thesis by D. Sharp; follow-ups were
conducted under the leadership of R. Kumar, who is now
deceased. The research reported in this paper was supported by
the Medical Research Council through Project Grants G89202999N
and G9539876N to R. Kumar, D. Sharp, and D.Hay, and Child
Psychiatry Unit funding to D. Hay. We thank Gesine Schmücker,
Alice Mills, Helen Allen, and Anne O’Herlihy for help with data
collection, and Adrian Angold, Gordon Keeler, and Peter Wendell
at Duke University for help with the analysis of the interview
data. We are grateful to the families who have participated in the
South London Child Development Study for so many years.
Correspondence concerning this article should be addressed to
Dale Hay, School of Psychology, Cardiff University, Cardiff, Wales,
CF10 3YG, United Kingdom. Electronic mail may be sent to
haydf@cardiff.ac.uk.
protecting someone from harm, rescuing someone
from danger, and feeling empathy and sympathy’’
(Radke-Yarrow & Zahn-Waxler, 1986, p. 208). In
longitudinal analyses of an urban community
sample of British children, we used a multimethod,
multi-informant approach to study prosocial behavior at two time points.
We focused on one form of prosocial behaviorFcooperationFin early childhood. Cooperation
entails working positively with other people to meet
a mutual goal and thereby conveys benefits on
others as well as the self (Staub, 1978). Many
prosocial behaviors occur at low frequency, under
particular situational constraints, but cooperation
can be observed reliably in the context of brief
experimental tasks (Staub, 1978) and thus can be
observed systematically within a community sampling frame. Cooperation with other children and
adults, and the related concept of cheerful ‘‘committed compliance’’ to reasonable requests are
important aspects of prosocial development in the
early years (Kochanska, Aksan, & Koenig, 1995;
Rheingold, Cook, & Kowlowitz, 1987). Young children’s cooperation is positively associated with
sharing (Hay, 1979) and sympathy (Murphy, 1937),
and thus serves as a marker of more general
prosocial tendencies.
We studied a broader domain of children’s
prosocial activities at age 11, as rated by mothers,
teachers, and children. Standardized questionnaires
have been used in past research on individual
differences in prosocial behavior, showing internal
consistency across prosocial items and discriminant
validity between prosocial and problem scales, with
r 2003 by the Society for Research in Child Development, Inc.
All rights reserved. 0009-3920/2003/7405-0006
Prosocial Development and Psychological Problems
moderate consistency across informants (R. Goodman, 1994; Tremblay, Vitaro, Ganon, Piché, & Royer,
1992).
Prosocial Behavior and Children’s Psychological Problems
The evidence regarding links between children’s
prosocial behavior and psychological problems is
contradictory. Prosocial behavior is linked to emotional regulation (Eisenberg et al., 1995; Fabes,
Eisenberg, Karbon, Troyer, & Switzer, 1994), social
competence (Chen, Li, Li, & Liu, 2000; Eisenberg
et al., 1996), and moral reasoning (Carlo, Koller,
Eisenberg, DaSilva, & Frolich, 1996; Miller, Eisenberg, Fabes, & Shell, 1996). In line with these
findings, prosocial behavior, as one dimension of
psychological adjustment, should be inversely correlated with behavioral and emotional problems.
The absence of prosocial behavior in childhood
predicts disruptive behavior and criminality in later
life (Haemaelaeinen & Pulkinnen, 1996). Children
with disruptive behavior are less likely than other
children to be prosocial (Hastings, Zahn-Waxler,
Robinson, Usher, & Bridges, 2000; Hughes, White,
Sharpen, & Dunn, 2000; Vinnick & Erickson, 1992),
and children with internalizing problems show
prosocial behavior at low rates (LaFreniere, Provost,
& Dubeau, 1992). In early adolescence, emotional
distress is negatively related to prosocial behavior
(Wentzel & McNamara, 1999).
However, under some circumstances, prosocial
behavior is accompanied by behavioral and emotional problems. For example, in some samples,
children with externalizing problems showed prosocial behavior to the same extent as or more than
other children (e.g., Del’Homme, Sinclair, & Kasari,
1994; Zahn-Waxler, Cole, Welsh, & Fox, 1995). When
faced with conflict, disruptive girls showed high
levels of prosocial behavior (Zahn-Waxler et al.,
1994). Boys with attention deficit hyperactivity
disorder (ADHD) were just as likely as other boys
to show prosocial behavior (Buhrmester, Whalen,
Henker, MacDonald, & Hinshaw, 1992), although
stimulant medication reduced their levels of social
engagement. It has also been observed that fear and
sadness predict prosocial traits (Rothbart, Ahadi, &
Hershey, 1994).
Tests of associations between prosocial behavior
and psychological problems rest on the assumption
that individual differences in both domains are
robust and stable over time. It is well known that
behavioral problems in early childhood predict later
delinquency and disorder (e.g., Moffitt & Caspi,
2001). There is also recent evidence for long-term
1315
continuities in prosocial behavior (Côté, Tremblay,
Nagin, Zoccolillo, & Vitaro, 2002; Eisenberg et al.,
1999). The longitudinal design used here permits
tests for associations across the two domains at two
time points. We also examined whether prosocial
behavior in early childhood protected children
against the risk of psychopathology in early adolescence.
Maternal Mental Health Problems and Prosocial
Development
In evaluating the evidence for positive or negative
associations between children’s prosocial behavior
and psychological problems, it is important to take
into account other psychopathology in the family. In
particular, maternal depression is associated with
adverse family climates and less optimal patterns of
parent – child interaction (S. Goodman & Gotlib,
1999). Under these conditions, children’s prosocial
development may not proceed normally.
In general, positive home environments foster
prosocial tendencies. Prosocial behavior is more
frequent in children whose parents provide inductive reasoning (Hart, DeWolf, & Burts, 1992),
acceptance of the child (Chen & Rubin, 1994), and
support for the child’s relationships with peers
(Ladd & Hart, 1992). Children’s prosocial behavior
is associated with authoritative parenting (Dekovic
& Janssens, 1992; Hastings et al., 2000) and with the
modeling and reinforcement of sympathetic behavior by mothers (Eisenberg et al., 1992). Maternal
depression might interfere with the provision of
sensitive, authoritative parenting and therefore disrupt prosocial development. If so, the children of
depressed mothers might show more psychological
problems and less prosocial behavior. Support for
this view was provided by analyses of a large British
community sample, which demonstrated a significant negative correlation between mothers’ depressive symptoms and children’s prosocial tendencies,
both measures rated by mothers (Dunn, DeaterDeckard, Pickering, O’Connor, Golding & the
ALSPAC Study Team, 1998).
Nevertheless, prosocial behavior can also emerge
under stressful circumstances. Prosocial behavior
between siblings occurs when parents quarrel
(Cummings & Smith, 1994), and siblings become
important sources of support for each other when
their parents separate and divorce (e.g., Kier &
Lewis, 1998). More generally, prosocial behavior
may flower under highly adverse social conditions,
such as war: Elevated rates of prosocial behavior are
reported by children whose lives were disrupted by
1316
Hay and Pawlby
war in Lebanon (Macksoud & Aber, 1996) and
Croatia (Raboteg-Saric, Zuzul, & Kerestes, 1994).
In line with this body of evidence, it is possible
that the needs of a depressed mother may foster
prosocial tendencies, at least for a while. Under
some circumstances, maternal depression is linked
to children’s empathic reactions (T. Dix, June 28,
2002, personal communication; Radke-Yarrow,
Zahn-Waxler, Richardson, Susman, & Martinez,
1993; Zahn-Waxler, Cummings, McKnew, & RadkeYarrow, 1984). Indeed, children of depressed parents
may become preoccupied with the parents’ problems, thus shouldering excessive prosocial responsibilities (Zahn-Waxler, Cole, & Barrett, 1991) and in
some cases exhibiting internalizing problems themselves (Klimes-Dougan & Bolger, 1998). To the extent
that the children of depressed mothers are generally
at risk for depression and other disorders (S. Goodman & Gotlib, 1999; Rice, Harold, & Thapar, 2002), it
seems possible that these children’s prosocial behavior might be associated with their own psychopathology.
In addressing these issues, it is important to
examine qualitative as well as quantitative features
of children’s responses to other people. Children of
depressed mothers who may have to take care of
siblings and the mother herself may do so at some
cost to themselves. In particular, they may become
anxious about the possible harm that might befall
their family members. We examined whether children were more likely to worry about family
members when their mothers were depressed,
whether their worries were positively or negatively
associated with other forms of prosocial behavior,
and whether worrying about others was more
characteristic of girls than of boys. We recorded
the incidence of clinically significant worries about
other people, which is a clinical symptom of
separation anxiety disorder but might also be
viewed as an excessive form of prosocial concern
for others. Children who show such extreme worry
about their loved ones might show other forms of
prosocial behavior but might also have other
internalizing problems (Klimes-Dougan & Bolger,
1998).
Maternal depression may exaggerate gender
differences in prosocial behavior (Zahn-Waxler et
al., 1991), and girls’ worries about other people may
sometimes verge on the unmitigated communion
shown by some women who focus on others at cost
to their own health (Helgeson & Fritz, 1998). Thus,
we examined whether girls and boys were equally
likely to worry about the well-being of family
members.
The Importance of the Child’s Perspective
The fact that informants about children’s prosocial behavior show only moderate consistency
(e.g., R. Goodman, 1994; Tremblay et al., 1992)
suggests that the perception of prosocial behavior
in oneself or another is itself a complex social
judgment. When children, parents, and teachers
were asked to classify behaviors as prosocial or
antisocial, less agreement was found in the classification of prosocial as opposed to antisocial behavior,
both within and between each category of informant
(Warden, Christie, Kerr, & Low, 1996).
It is particularly important to canvass the child’s
perspective. Children who grow up under adverse
circumstances may see themselves as cooperative
and sympathetic, even if their behavior is not always
judged to be so by other observers. For example, in
response to a story completion task, maltreated
children saw themselves as likely to show positive
responses to people in distress; this was despite the
fact that maltreated children are generally not seen
to be sympathetic in response to peers’ distress
(Macfie et al., 1999). A similar discrepancy between
the child’s view of the self and views by others may
characterize the children of depressed mothers.
Thus, in the present study, we tested for differences
among informants as well as combining information
from different sources to reveal the links between
children’s prosocial behavior and their psychological
problems.
Method
Design
Analyses were undertaken within a representative urban sample of British children. Children’s
prosocial behavior and psychological problems were
assessed at two time points, when the children were
nearly 4 and 11 years of age. The mother’s mental
health was assessed at various points throughout the
study, allowing for a diagnosis of maternal depression in the child’s lifetime and a rating of her
psychological functioning at each time point.
The Families
The families participated in an ongoing, prospective longitudinal study of child development in two
communities in South London (Sharp, 1993; Sharp
et al., 1995). The families were recruited at routine
prenatal check-ups, with 179 chosen at random for
full psychiatric assessments before and after childbirth. In the families, 171 women gave birth to live
Prosocial Development and Psychological Problems
infants and 149 (87%) who were assessed at 3
months postpartum were followed up at later time
points. When the child was 11 years old, 134 (90%)
were available for follow-up, and 89% of those had
also participated when the child was 4 years old. In 1
family the child had died, in another the child had
been taken into care by the Social Services and access
was denied, 2 had withdrawn from the study before
the first birthday, 7 had moved abroad or could not
be traced, and 4 were not willing to participate.
The mother’s mean age at the start of the study
was 25.8 (SD 5 5.1, range 5 16 – 43). Characteristics of
the sample are presented in Table 1. In many ways,
the sample is representative of the general population in Great Britain, where approximately 90% of
the population lives in urban areas (Office for
National Statistics, 2003). Family structure was
almost identical to national norms at birth (Office
for National Statistics, 2000) and, at 4 years of age, to
findings in a very large community sample in the
English county of Avon (Dunn et al., 1998). The rate
of children’s psychological problems at age 4 was
also in line with norms from that study. In other
ways, the sample is representative of the sociocultural mix of a large metropolitan area. National
Table 1
Characteristics of the Sample
Variable
Child’s sex
Child’s birth order
Sibling status at age 11
Marital status at birth
Marital status at age 11
Parent in household at 11
Social class
Maternal education
Ethnicity
Parental employment
Description
53% female
47% firstborn
90% with siblings (median 5 2,
range 5 1 – 7)
63% married, 29% cohabiting, 8%
single
58% married (89% to biological
father)
14% cohabiting (39% to biological
father)
28% single-parent household
56% two biological parents
40% biological mother
2% biological father
2% other guardians
89% working class (Goldthorpe &
Hope, 1974)
72% basic qualifications, 14% further
education
72% White British, 6% White, nonBritish, 22% other (Caribbean,
African, South Asian, East Asian)
75% at least one parent employed
58% maternal employment
1317
surveys show that, in the United Kingdom as a
whole, the majority of families (62%) view themselves as working class, as opposed to middle class
(K. Prandy, June 26, 2002, personal communication). The proportion of working class families is
higher in this sample, which also includes several
high-status, affluent families. The sample is also
more culturally diverse than the United Kingdom as
a whole, where approximately 7% of the population
belongs to ethnic minorities (Office for National
Statistics, 2000).
Procedure
A multimethod, multi-informant strategy was
adopted. Experimental, observational, interview,
and questionnaire methods were used, and informants included parents; teachers; testers; and, at age
11, the children themselves.
Pregnancy and the first postnatal year. The mothers
were interviewed at 14 and 36 weeks of pregnancy,
and at 3 and 12 months postpartum (for details see
Sharp, 1993). The Clinical Interview Schedule (CIS;
Goldberg, Cooper, Eastwood, Kedward, & Shepherd,
1970) was used to assess the mother’s psychopathology.
Fourth birthday. The families were visited at home
when the children were approaching their fourth
birthday. The mother’s current and past psychopathology was assessed, using the Schedule of
Affective Disorders and Schizophrenia (SADS – L;
Spitzer, Endicott, & Robbins, 1978). Mothers also
rated the child’s psychological problems, using the
Child Behavior Checklist (CBCL; Achenbach, 1988).
The assessment of the child at age 4 consisted of a
standardized cognitive test, the McCarthy Scales of
Children’s Abilities (McCarthy, 1972), and an experimental cooperation task. After the cognitive test, the
tester, who had not interviewed the mother and was
unaware of her psychiatric history, rated the child’s
cooperativeness. In the experimental task, pairs of
mothers and children were asked to cooperate to
copy a picture of a house, using an Etch-A-Sketch
toy (Stevenson-Hinde & Shouldice, 1995). Mother
and child were each assigned a dial to use, and told
that they might not use the other person’s dial so
that each person could only draw vertical or
horizontal lines. By using their dials at the same
time, they could produce diagonal lines. This
procedure creates a situation in which two people
must work together to meet interdependent goals
and therefore meets the criteria of the classic
definition of cooperation (Deutsch, 1949), which
provides the conceptual basis for most experimental
1318
Hay and Pawlby
cooperative tasks (e.g., Brownell & Corriger, 1990;
Rose-Krasnor, Rubin, Booth, & Coplan, 1996).
11th birthday. When the child was age 11, one
researcher assessed the mother’s mental health
currently and retrospectively to the last visit, and
her level of functioning in the week preceding the
interview. The mothers also reported on the child’s
psychological problems and prosocial behavior,
using the Strengths and Difficulties Questionnaire
(SDQ; Goodman, 1999). The mother was also
interviewed in much more detail about the child’s
clinically significant psychological problems, using
the Child and Adolescent Psychiatric Assessment
(CAPA; Angold et al., 1995).
A second researcher, unaware of the mother’s
psychiatric history, interviewed the child. The SDQ
was read to the child as part of an interview that also
included the CAPA. All parents gave permission for
the class teachers to complete the SDQ. Teachers’
reports were available for all but one child, who was
taught at home.
Measures
Cooperativeness in early childhood. The primary
measures of prosocial behavior at age 4 were the
tester’s rating of cooperativeness during the cognitive test and an observational measure of cooperation with the mother during the Etch-A-Sketch task.
The Tester’s Rating of Children’s Behavior (TRCB;
Wolke, 1990) had been adapted from a rating scale of
infant behavior (Wolke, Skuse, & Mathisen, 1990).
Cooperativeness with the tester was rated on a
9-point scale ranging from very resisting and uncooperative (1) to very cooperative, readily and enthusiastically entering into the tasks (9).
Behaviors shown by mothers and children during
the Etch-A-Sketch task were rated using an observational coding system specifically designed for the
study (Sharp et al., 1995). Two 9-point scales,
measuring the child’s engagement in and persistence at the cooperative task, were summed to
produce a composite measure of the child’s contribution to the cooperative task. Behaviors indicating
the child’s cooperation with the mother included
looking at and listening to her, as signs of mutual
engagement; incorporating her suggestions, directions, and demands; and turn taking and reciprocity.
In other words, these ratings reflect the extent to
which the child engaged with the mother in the
mutual task and persisted in attempting to attain the
mutual goal. Observers of the video records were
unaware of the mother’s diagnostic status. Observer
agreement, measured by intraclass correlation coef-
ficients, was .87 and .72 for engagement and
persistence, respectively.
General prosocial tendencies at age 11. When the
child was age 11, a more general assessment of
prosocial behavior was made, using the SDQ. The
SDQ is a 25-item questionnaire on prosocial behaviors and symptoms of childhood disorder (R. Goodman, 1999; R. Goodman, Meltzer, & Bailey, 1998).
The prosocial scale consists of 5 items: (1) considerate of other people’s feelings; (2) shares readily with
other children (treats, toys, pencils, etc.); (3) helpful
if someone is hurt, upset or feeling ill; (4) kind to
younger children; and (5) often volunteers to help
others (parents, teachers, other children). Past
research has demonstrated internal consistency of
the prosocial scale and its discriminant validity in
terms of negative associations with the problem
scales of the SDQ (e.g., Dunn et al., 1998). In the
present study, the prosocial scale showed acceptable levels of internal consistency (a 5 .68 for
children, a 5 .85 for teachers, and a 5 .72 for
mothers).
Children’s ratings of prosocial behavior were
correlated with teachers’ ratings, r(129) 5 .26,
po.002, and with mothers’ ratings, r(126) 5 .19,
po.03. Teachers’ ratings were significantly correlated with mothers’ ratings, r(125) 5 .44, po.001.
Measures of the child’s psychological problems. When
the child was age 4, mothers rated the child’s
behavioral and emotional problems on the CBCL.
Higher scores indicate more problems. Because the
children in this study were not quite 4 years of age at
the time of assessment, the 2- to 3-year-old version of
the CBCL was used. The internalizing and externalizing subscales were highly correlated in this
sample, r(114) 5 .78, po.001.
Measures of the child’s psychological problems at
age 11 were obtained from the SDQ and CAPA. The
SDQ scales provide continuous measures of children’s behaviors across the normal and abnormal
range. For the purpose of the present study, the
conduct problems and hyperactivity scales were
summed to form a general measure of externalizing
problems. The composite measure showed good
internal consistency, a 5 .72 for children, a 5 .90 for
teachers, and a 5 .82 for mothers. In all cases the
consistency of the composite externalizing measure
was better than that achieved by the original scales.
The emotional problems scale was used as a
measure of internalizing problems (a 5 .60 for
children, a 5 .72 for teachers, and a 5 .66 for
mothers).
The CAPA is a semistructured diagnostic interview that takes on average 2 to 3 hr to complete. The
Prosocial Development and Psychological Problems
interviewer asks a series of questions about family
structure and relationships, peer relations, and
problems in the domains of several clinical disorders, including ADHD, conduct disorder, oppositional-defiant disorder, anxiety disorders, and
depression. The parent and child, who were interviewed separately, were asked to recount details
about particular behaviors and events during the
last 3 months. Interviewers used operational criteria
to determine if and when a symptom of clinical
significance occurred, when it had begun, and how
long it persisted. Thus, whereas the SDQ provides
continuous measures about the child’s behavior
throughout the normal range, the CAPA is designed
to detect symptoms of childhood psychiatric disorder, according to Diagnostic and Statistical Manual
of Mental Disorders (DSM) criteria.
The analysis of the CAPA measures was conducted on the subsample of 122 families in which
both parent and child had successfully completed
the interview. Measures of externalizing and internalizing symptoms reported by parent and child
were obtained, using definitions retained from the
Diagnostic and Statistical Manual of Mental Disorders
(3rd ed., rev. [DSM – III – R]; American Psychiatric
Association, 1987), in which the distinction between
internalizing and externalizing problems had been
made. Because most children did not manifest
severe psychiatric symptoms, these distributions
were skewed; therefore, logarithmic transformations
were used in subsequent analyses.
We also identified children with particularly
severe problems, meeting Diagnostic and Statistical
Manual of Mental Disorders (4th ed. [DSM – IV];
American Psychiatric Association, 1994) criteria for
emotional disorders and disruptive behavior disorders (DBD). Mothers’ and children’s joint reports
were used to identify children meeting caseness
criteria. Computer algorithms were used to determine which children met the operational definitions
of DSM – IV disorders.
‘‘Worries about family members’’ at age 11. When the
child was age 11, a self-report measure of worrying
about family members was obtained from the child’s
CAPA interview. In the separation anxiety section of
the CAPA, a series of questions tap the child’s
excessive worries about possible harm that might
befall attachment figures. Symptomatic behaviors
include ‘‘unrealistic and persistent worry or fear
about possible harm befalling major attachment
figures’’ or ‘‘rising at night to check that attachment
figures are still present and/or free from harm.’’ The
interviewer probes for particular instances of such
behavior and uses operational criteria to determine
1319
whether the child was showing this symptom of
separation anxiety. Diagnostic algorithms used information about this symptom in combination with
other symptoms to assess the child’s caseness for
separation anxiety disorder. The occurrence of this
symptom alone does not provide a diagnosis of
separation anxiety disorder. Here, we examined the
presence of this clinically significant symptom as an
index of the child’s self-reported tendency to feel
great anxiety about the welfare of family members.
Measures of maternal depression. The interviews
used in pregnancy and immediately after the birth of
the child using the CIS permitted contemporaneous
diagnoses of depression using International Classification of Diseases (ICD-9) criteria. The interviews
given at 4 and 11 years of age, using the SADS – L
provided contemporaneous and lifetime diagnoses
of depression in line with research diagnostic criteria
(RDC) criteria. Both sets of criteria identify clinically
significant cases of depression.
The CIS was developed for use in community
settings (Goldberg et al., 1970). In this study, the
women were asked about 10 specific psychiatric
symptoms occurring in the 2-week period before the
interview; the interviewer also rated 12 manifest
abnormalities of behavior and affect shown during
the interview. A rating was made on a 5-point scale
for each of the 10 symptoms, after establishing the
frequency, severity, and duration of each. The ratings
produced a total weighted score and a case – noncase
distinction. An ICD-9 diagnosis of depression was
then assigned to cases (World Health Organization,
1978).
The SADS – L (Spitzer et al., 1978) was used at the
4- and 11-year-old assessments to provide detailed
information about the respondent’s mental health,
both current and retrospective. Diagnoses of depression in line with RDC were made after discussion
with an adult psychiatrist (R. Kumar). In addition, a
continuous measure of the mother’s psychological
functioning in the weeks preceding the 4- and 11year assessments was provided by the Global
Assessment Scale (GAS; Endicott, Spitzer, Fleiss, &
Cohen, 1976). This rating scale evaluates the overall
functioning of a participant during a specified
period on a continuum (1 to 100) from psychological
or psychiatric sickness to health. Higher scores
reflect better levels of functioning.
The contemporaneous and retrospective assessments of maternal depression at various points in
the study produced two summary variables: (a)
current functioning (the mother’s GAS scores at the
4- and 11-year assessments) and (b) depression
during the child’s lifetime (caseness for depression,
1320
Hay and Pawlby
using either ICD-9 or RDC criteria, at any point since
the child’s birth).
Data Analysis
Construction of composite variables. When examining correlations over time and across the domains of
prosocial behavior and psychological problems,
reliability is increased if composite measures are
used. The assessments of cooperation at age 4 were
analyzed separately, as they were only modestly
correlated and were conducted in very different
contexts. However, because the reports of prosocial
behavior at age 11 all derive from the SDQ, a
principal components analysis was conducted,
yielding a single prosocial factor, which accounted
for 53% of the variance in the informants’ reports.
The factor score was used in subsequent tests for
association between prosocial behavior and psychological problems.
Similarly, to achieve more reliable measures of the
children’s psychological problems, we combined
information across informants and methods. The
three informants’ reports of externalizing and internalizing problems on the SDQ and the logarithmically transformed measures of externalizing and
internalizing symptoms on the CAPA were entered
into a principal components analysis. Because of the
theoretical distinction between internalizing and
externalizing problems, a two-factor solution was
sought. The analysis indeed yielded two orthogonal
factors, accounting for 50.2% of the variance in the
informants’ reports on the two instruments, which
were then used as composite measures of internalizing and externalizing problems.
Tests for links between prosocial behavior and psychological problems. To address the first aim of the study,
the analysis for links between children’s prosocial
behavior and their psychological problems, the
composite measure of prosocial behavior was
correlated with the composite measures of internalizing and externalizing symptoms. The children’s
severe problems of clinical concern were dichotomous variables (diagnoses of disorder and presence
of the key symptom of worrying about family
members). Therefore, the relationship between these
dichotomous measures of the child’s mental health
and prosocial behavior was assessed through tests of
differences between ill and well children, using
t tests and analysis of variance (ANOVA) in cases in
which the child’s sex was included in the model.
Tests for the influence of maternal depression. To
meet the second aim of the studyFanalysis for the
effects of the mother’s illness on prosocial behavior
FANOVAs were used to test for the effects of
maternal depression on each measure of cooperation
at age 4. When the child was age 11, because of the
importance of retaining the child’s perspective, the
effects of the mother’s illness were assessed with
respect to each informant’s report. The main effect of
informant and the interaction of informant with
maternal depression were tested in a repeatedmeasures ANOVA.
Long-term prediction of children’s psychological problems. A final set of analyses addressed both aims of
the study longitudinally, testing whether cooperative
behavior in early childhood predicted the composite
measures of externalizing and internalizing symptoms in early adolescence, taking into account the
mother’s history of depression. Multiple regression
analyses controlled for the effects of the child’s
gender, the mother’s depression, and the child’s
earlier problems.
Results
Links Between Children’s Prosocial Behavior and Their
Psychological Problems
Cooperativeness at age 4. Children who were more
engaged in the cooperative task with their mothers
had significantly fewer problems on the CBCL,
r(112) 5 – .18, po.05. In particular, they had fewer
externalizing problems (Table 2), although that
association only approached statistical significance.
The tester’s rating of cooperativeness was not
significantly linked to the child’s problems.
General prosocial tendencies at age 11. Prosocial
children were no more or less likely than other
children to have emotional problems. There was no
significant association between the composite prosocial score and the composite measure of internalizing problems (Table 2). Furthermore, children
diagnosed with clinically significant emotional disorder were no more or less likely than other children
to show prosocial behavior. Twenty children (16.4%
of those with complete CAPA data) met diagnostic
criteria for an emotional disorder on the basis of the
mothers’ and children’s joint reports on the CAPA
interview. A 2 2 ANOVA was used to test for the
effects of diagnostic status and the child’s sex on the
composite measure of prosocial tendencies. There
was no effect of the child’s diagnosis of an emotional
disorder, nor of the child’s sex, nor a significant
interaction.
In contrast, prosocial behavior was inversely
related to externalizing problems (Table 2). Furthermore, the negative relation between externalizing
Prosocial Development and Psychological Problems
1321
Table 2
Pairwise Correlations Between Measures of Prosocial Behavior and Problems at Ages 4 and 11 (Pearson r, N)
Measure
1.Cooperation with
mother at 4
2. Cooperation with
tester at 4
3. Prosocial score at 11
4. CBCL externalizing
score at 4
5. CBCL internalizing
score at 4
6. Externalizing factor
score at 11
7. Internalizing Factor
score at 11
Variable M
Variable SD
1
2
0.19 (115)
3
4
5
6
7
0.09 (109)
0.18+ (112)
0.13
(112)
0.30 (105)
0.07 (105)
.00 (107)
0.12 (114)
0.16+
(114)
0.15
(107)
0.07 (107)
0.24 (109)
0.21 (109)
0.78 (114)
0.56 (118)
0.42 (106)
0.07 (118)
0.10 (106)
0.30 (106)
0.19+ (106)
0.00 (118)
12.0
4.8
5.3
2.0
0.0
1.0
49.6
10.2
49.9
9.8
0.0
1.0
0.0
1.0
po.10. po.05. po.01. po.001.
+
problems and prosocial behavior was also found
with respect to clinically significant disorder. Twenty
children (16.4% of those with complete CAPA data)
were diagnosed with DBD according to DSM – IV
criteria. Of these, 9 (45%) were comorbid for an
emotional disorder, kappa 5 .34, po.001. A 2 2
ANOVA tested for the effects of diagnosis with DBD
and the child’s sex on children’s general prosocial
tendencies. Children with DBD were significantly
less prosocial than other children, M 5 – 0.66
(SD 5 1.17) versus M 5 0.14 (SD 5 0.88), F(1,
114) 5 10.75, po.001. There was no effect of the
child’s sex and no interaction between sex and
diagnosis.
Worries about family members. Extreme anxiety
about others, as measured by self-reported worries
about harm that might befall attachment figures,
was shown by 14 children in the sample (10.9%): 7
boys (11.7%) and 7 girls (10.3%). Links between
extreme worries about family members and more
general prosocial tendencies, as measured by the
prosocial factor score, were examined (Table 3).
Children who worried about the well-being of
family members were significantly more prosocial
than other children, t(119) 5 – 2.34, po.02.
The children who worried about family members
had significantly fewer externalizing problems than
other children, t(20) 5 2.30, po.03, (degree of freedom is adjusted because equality of variance was not
assumed), but experienced a greater number of
internalizing problems, t(116) 5 – 3.74, po.001. Their
problems were severe. Most of the children who
worried about family members (69%) met DSM – IV
criteria for an emotional disorder, as opposed to only
10% of other children, w(1) 5 29.64, po.001, odds
ratio 5 20.04 (confidence interval 5 5.29 – 75.98). (The
symptom of worrying about family members was
not by itself sufficient to diagnose separation anxiety
disorder.) This association was significant for both
sexes, with 67% of boys and 71% of girls who
worried about family members meeting criteria for
an emotional disorder, w(1) 5 12.75, po.001 for boys
and w(1) 5 16.91, po.001 for girls.
Maternal Depression and Children’s Prosocial Behavior
Are the associations between children’s prosocial
behavior and their own psychological problems
mirrored by links between children’s prosocial
behavior and their mothers’ own problems?
Cooperativeness at age 4. There was no association
between maternal depression and cooperation at age
4. Fifty-seven women (51.8% of those with complete
data) met criteria for depression between the child’s
birth and fourth birthday, as measured by the CIS at
3 and 12 months, or by the SADS – L, using RDC
criteria to diagnose depression between 12 months
and 4 years. A 2 2 multivariate analysis of variance
(MANOVA) testing for the main effects of gender
and the mother’s diagnosis on the two measures of
cooperativeness showed that the children of depressed mothers were no more or less cooperative
than other children. The two sexes did not differ, nor
did the mother’s history of depression interact with
the child’s sex to influence cooperation at age 4.
1322
Hay and Pawlby
Table 3
Prosocial and Problem Factor Scores for Children Who Do and Do Not
Express Clinically Significant Worries About Family Members
Children with
extreme
worries
Other children
(N 5 14)
(N 5 114)
Dependent variable
Prosocial composite
Externalizing problems
Internalizing problems
M
SD
M
SD
M
SD
0.56
0.79
0.36
0.66
0.83
1.03
0.05
0.94
0.04
1.03
0.10
0.95
Note. The values represent factor scores on each dependent
variable for the two groups of children.
po.05. po.001.
Furthermore, the mother’s current mental state, as
measured by the GAS rating, was not associated
with the child’s cooperation with the mother at age
4, r(114) 5 .04, nor with the tester, r(116) 5 – .06.
Children’s general prosocial tendencies at age
11. There was no effect of maternal depression on
the prosocial factor score. Rather, as predicted, the
direction of the relation between maternal depression and the children’s general prosocial tendencies
at age 11 depended on the informant being questioned (Figure 1).
By the time the child was 11 years old, 75 mothers
in the sample (57.7%) had met criteria for depression
on at least one occasion. A 2 2 3 repeatedmeasures MANOVA, with the mother’s caseness
and the child’s sex as between-subjects factors and
informant as a within-subjects factor, revealed a
significant interaction between the mother’s caseness and the identity of the informant, multivariate
F(2, 118) 5 3.11, po.05. Planned contrasts indicated a
significant difference between mothers’ and children’s own ratings of prosocial behavior, F(1,
119) 5 6.08, po.02. Mothers with a history of
depression saw their children as less prosocial, but
the children of depressed mothers saw themselves as
more prosocial than other children. There was no
three-way interaction with the child’s sex.
A similar pattern was found with respect to
mothers’ current psychological functioning. Mothers
who were functioning better rated their children as
more prosocial, r(125) 5 .20, po.02; teachers also
rated children of well-functioning mothers as more
prosocial, r(126) 5 .20, po.02. However, children’s
own assessments of their prosocial tendencies were
unrelated to their mothers’ current psychological
functioning, r(127) 5 .01.
Figure 1. The three informants’ reports of prosocial behavior at age
11, as a function of the mother’s depression during the child’s
lifetime.
Worrying about family members. Children’s worries
about family members were not significantly associated with the mother’s current mental state, nor
with the mother’s caseness for depression. Extreme
anxiety about family members was not shown by the
4 children whose mothers had attempted suicide in
the last 7 years.
Does Maternal Depression Moderate the Links Between
Children’s Prosocial Behavior and Psychological
Problems?
We tested the possibility that children of depressed mothers might be especially likely to show a
profile of behavior that includes both prosocial
tendencies and problematic behavior. In other
words, for the children of depressed women,
prosocial behavior might be positively related to
psychological problems. However, no evidence was
found in support of this proposal. Prosocial tendencies and externalizing problems were negatively
related, r 5 – .45, po.001 for the children of well
mothers, and r 5 – .63, po.001 for the children of
depressed mothers. Against expectations, a positive
association between prosocial behavior and emotional problems was found for the children of
mothers who had not been depressed, r 5.44,
po.001. Subsequent analyses showed that this was
especially true for the sons of the well mothers,
r(24) 5 .59, po.002.
Early Cooperation and Risk for Later Psychological
Problems
Finally, we tested for long-term relations between
cooperation in early childhood and psychological
problems in early adolescence. Children who were
more cooperative with their mothers at age 4 had
Prosocial Development and Psychological Problems
fewer externalizing problems at age 11, r(108) 5 – 30,
po.002. There was no association between cooperation at age 4 and internalizing problems at age 11.
The tester’s rating of cooperativeness did not predict
externalizing or internalizing problems.
The relation between early cooperation and later
externalizing problems was explored in more detail.
Linear regression was used to evaluate the contribution of early cooperation to the prediction of later
externalizing problems, controlling for the child’s
sex, the mother’s history of depression, the interaction between those two variables, and the child’s
early behavioral problems and current prosocial
tendencies. The dependent variable for this analysis
was the composite measure of externalizing problems at age 11. The child’s sex, along with the
mother’s diagnosis of depression and the interaction
between sex and diagnosis, centered with respect to
the means of each variable, were entered on the first
step. Both measures of early cooperation were
entered on the second step, and the total CBCL
score was entered on the third step (the high
correlation between the externalizing and internalizing scales introduced collinearity problems if both
were included in the model). Finally, as a measure of
current prosocial behavior, the prosocial factor score
was included at the final step. Results from the final
model are presented in Table 4.
Introduction of predictors at each step led to
significant R2 change; thus, the inclusion of the
measure of observed cooperation with the mother
was a significant predictor of the child’s later
externalizing problems, even when the child’s sex
and the mother’s diagnosis had been taken into
account, t 5 – 3.22, po.002, R2 change 5 .09, po.005.
Furthermore, when the measure of early behavioral
Table 4
The Prediction of the Child’s Externalizing Problems at Age 11
Predictor variable
B
Child’s sex
Mother’s caseness
Caseness Sex
Cooperation with mother
Cooperation with tester
CBCL problems at 4
Prosocial score at 11
.24
.20
.14
.04
.02
.02
.47
(95% Confidence
Interval)
(
(
(
(
(
.56
.12
.75
.07
.10
(.00
( .63
to
to
to
to
to
to
to
.08)
.52)
.46)
.01)
.06)
.04)
.31)
b
.12
.10
.04
.19
.04
.19
.48
Note. The table summarizes the contribution of each predictor
variable at the final step, N 5 104, R2 5 .49, adjusted R2 5 .44,
standard error of estimate 5 0.75. B 5 unstandardized weight in
regression analysis; CBCL 5 Child Behavior Checklist.
po.05. po.01. po.001.
1323
problems and the age 11 composite measure of
prosocial tendencies were incorporated in the
regression model at the final step, the level of
cooperation with the mother at age 4 still remained
a significant predictor of externalizing problems at
age 11 (Table 4). In the final model, the child’s sex
was no longer a significant predictor of externalizing
problems at age 11. Rather, externalizing problems
were predicted by the level of cooperation with the
mother at age 4, the mother’s rating of the child’s
problems at age 4, and the child’s current prosocial
behavior.
An analogous linear regression analysis revealed
that internalizing problems at age 11 were not
significantly influenced by the predictor variables.
The only significant univariate predictor was the
mother’s caseness for depression, t 5 2.25, po.03,
which was no longer significant when the child’s
CBCL problems score at age 4 was included in the
model. Thus, early cooperation predicted externalizing but not internalizing problems at age 11.
Discussion
The preceding longitudinal analyses clearly demonstrate that prosocial children show low rates of
externalizing problems. This association began to
emerge in early childhood and was clearly observed
in early adolescence, across instruments and informants. In this sample, early cooperation was a
protective factor against later disruptive behavior,
and its unique effects were not mediated by
continuities in either disruptive or prosocial behavior over time.
Active cooperation in the preschool years reflects
emotionally regulated, socially competent engagement with the social world and may set the child on
a course of successful interactions that promote
general psychological adjustment. These findings
add to the growing body of information about longterm continuities in prosocial behavior over the
childhood and adolescent years, which have focused
on helpfulness (Côté et al., 2002) and sharing
(Eisenberg et al., 1999). However, conclusions to be
drawn are necessarily constrained by the fact that, in
the present study, only two limited measures of one
form of prosocial behavior were taken at the age 4
assessment. It is not clear whether cooperative
behavior with siblings or peers, as opposed to
adults, would have similar predictive power, nor
whether a similarly long-term protective function
could be discerned for other forms of prosocial
behavior. These data suggest, however, that longitudinal studies of prosocial development might
1324
Hay and Pawlby
profit from a focus on cooperation as well as
other, more extensively studied forms of prosocial
behavior.
Although prosocial children were not very disruptive, it was not the case that prosocial behavior
was invariably a marker of psychological adjustment. Some prosocial children were well adjusted
and others had emotional problems. Early cooperativeness did not protect the children from developing emotional problems in early adolescence. It is
important to note that prosocial boysFparticularly
those who were not exposed to maternal depressionFmay be at special risk for emotional problems.
Longitudinal analyses of a large sample of children
in Québec showed that only a minority of boys
showed prosocial behavior at consistently high rates
across childhood (Côté et al., 2002). To the extent that
prosocial boys’ behavior runs counter to gender
stereotypes and actual peer group norms, they may
experience peer rejection. Alternatively, boys with
emotional problems may be more likely than other
children to use prosocial strategies to gain peer
acceptance. In future work it will be important to
study gender differences not just in the overall level
of prosocial behavior but in their relation to other
traits and problems.
This study demonstrated that children who
reported being extremely anxious about loved ones
often manifested serious emotional disorder but also
were more likely to show prosocial behavior than
other children. The worries expressed by these
children were very serious, indeed. Qualitative
analysis of these children’s comments during the
interview indicated that they were worried about
their family members being hit, mugged, raped, or
killed. The children’s reports also revealed aspects of
self-sacrifice, with the children interrupting their
own sleep or giving up pleasurable activities to
check on the well-being of their family members. For
example, one child expressed concerns that his
mother ‘‘could slip on the kitchen floor and be
unconscious and nobody would know.’’ He commented that ‘‘I try to get home as soon as I can,’’
even to the extent of not playing soccer with his
friends. These analyses have identified an important
subgroup of emotionally troubled boys and girls in
this community sample who were significantly more
prosocial and less disruptive than average.
This study provided mixed evidence regarding a
link between maternal depression and children’s
prosocial development. The direction of the relationship depended on the period of development being
examined, the nature of the prosocial measure, and
the informant being questioned. In early childhood,
no significant links between maternal mental health
and children’s cooperation could be discerned. This
stands in contrast to the findings of a major
community study of British 4-year-olds, where
maternal depression was negatively associated with
mothers’ ratings of children’s prosocial behavior
(Dunn et al., 1998), but in that study both measures
were based on maternal questionnaires. The finding
also stands in contrast with suggestions in other
studies that there might sometimes be a positive
association between maternal depression and young
children’s responses to distress (e.g., T. Dix, June 28,
2002, personal communi Radke-Yarrow et al., 1993).
Finally, in view of the possibility that the children of
depressed mothers might develop excessive forms of
prosocial behavior, it is important to note that the
children of depressed mothers were not more likely
than other children to express clinically significant
worries about family members.
It proved important to canvass the opinions of the
children themselves as well as the opinions of their
mothers and teachers. In early adolescence, the adult
informants reported that the children of depressed
mothers were less prosocial than other children. In
contrast, the children of depressed mothers saw
themselves as more prosocial than other children.
This is similar to the finding that maltreated children
see themselves as prosocial agents in response to
other people’s distress, even though their observed
behaviors are not particularly prosocial (Macfie
et al., 1999).
Mothers and, especially, teachers no doubt compare children’s prosocial tendencies against their
own expectations and social norms. A long history
of interaction with a depressed parent (and the
related family upheaval that often entails) may
indeed promote disruptive behavior and reduce
prosocial tendencies, perhaps especially in response
to the challenges of the classroom. Yet, children who
grow up with depressed parents may also be called
on to assist and comfort their parents and siblings in
many ways. They may see themselves quite rightly
as helpful, sympathetic people.
It is tempting to see the children’s reports as less
reliable than those of adults, but the degree of
consistency across informants argues against this
view. Rather, it seems that children’s reports of
prosocial behavior are drawing on different sources
of evidence than those used by the adult raters.
Different informants use different implicit definitions
of prosocial behavior (Warden et al., 1996). Children’s
perceptions of themselves as prosocial are important
in their own right, and the self-perceptions of
children who have experienced family adversity
Prosocial Development and Psychological Problems
are of particular interest. They may lead to distorted
working models of one’s responsibility for the care
of others (Macfie et al., 1999), but positive selfperceptions might also provide a source of resilience
in the face of difficult family circumstances.
In future work, it will be important to examine the
effects of maternal health problems using a broader
array of measures of prosocial behavior, including
direct observation, experimental tasks, and peer
nominations. Although standardized measures of
prosocial tendencies, such as the SDQ prosocial
scale, show good levels of reliability (R. Goodman,
1994; Ladd & Profilet, 1996; Tremblay et al., 1992), it
is clear that different measures are differentially
sensitive to particular causal influences. In particular, factors that influence self-reports differ from
those that influence more objective measures (e.g.,
Eisenberg & Lennon, 1983). Situational reactions,
especially as assessed by psychophysiological measures, do not always correlate with dispositional
prosocial behavior (Holmgren, Eisenberg, & Fabes,
1998). A broad approach to the assessment of
prosocial behavior in childhood and adolescence is
clearly needed.
In summary, our prospective longitudinal study
of an urban British sample demonstrated complex
links between prosocial tendencies and psychological problems as children moved from childhood to
adolescence. Low rates of prosocial behavior were
characteristic of children with DBDs, but higher
than average rates of prosocial behavior, plus
extreme levels of concern for family members, were
shown by a subgroup of children with severe
emotional problems. Some troubled children are
prosocial, and prosocial children, though unlikely to
be disruptive, are not universally free of mental
health problems.
References
Achenbach, T. M. (1988). Child Behavior Checklist for ages
4 – 16. Burlington: University of Vermont.
American Psychiatric Association. (1987). Diagnostic and
statistical manual of mental disorders (3rd ed., rev.),
Washington, DC: Author.
American Psychiatric Association. (1994). Diagnostic and
statistical manual of mental disorders (4th ed.), Washington,
DC: Author.
Angold, A., Prendergast, M., Cox, A., Harrington, R.,
Simonoff, E., & Rutter, M. (1995). The Child and
Adolescent Psychiatric Assessment (CAPA). Psychological Medicine, 25, 739 – 753.
Dunn, J., Deater-Deckard, K., Pickering, K., O’Connor,
T. G., Golding, J., & the ALSPAC Study Team. (1998).
Children’s adjustment and prosocial behaviour in step-,
1325
single-parent and non-stepfamily settings: Findings
from a community sample. Journal of Child Psychology
and Psychiatry, 38, 965 – 980.
Brownell, C. A., & Corriger, M. S. (1990). Changes in
cooperation and self-other differentiation during the
second year. Child Development, 6, 1164 – 1174.
Buhrmester, D., Whalen, C., Henker, B., MacDonald, V., &
Hinshaw, S. P. (1992). Prosocial behavior in hyperactive
boys: Effects of stimulant medication and comparison
with normal boys. Journal of Abnormal Child Psychology,
20, 103 – 121.
Carlo, G., Koller, S. R., Eisenberg, N., DaSilva, M. S., &
Frolich, C. B. (1996). A cross-national study on the
relations among prosocial moral reasoning, gender role
orientations, and prosocial behaviors. Developmental
Psychology, 32, 231 – 240.
Chen, X., Li, D., Li, Z-Y, & Liu, M. (2000). Sociable and
prosocial dimensions of social competence in Chinese
children: Common and unique contributions to social,
academic, and psychological adjustment. Developmental
Psychology, 36, 302 – 314.
Chen, X., & Rubin, K. H. (1994). Family conditions,
parental acceptance, and social competence and
aggression in Chinese children. Social Development, 3,
269 – 290.
Côté, S., Tremblay, R. E., Nagin, D., Zoccolillo, M., &
Vitaro, F. (2002). The development of impulsivity,
fearfulness, and helpfulness during childhood: Patterns
of consistency and change in the trajectories of boys and
girls. Journal of Child Psychology and Psychiatry, 43,
609 – 618.
Cummings, E. M., & Smith, D. (1994). The impact of anger
between adults on siblings’ emotions and social
behavior. Journal of Child Psychology and Psychiatry, 34,
1425 – 1433.
Dekovic, M., & Janssens, J. M. (1992). Parents’ childrearing style and child’s sociometric status. Developmental Psychology, 28, 925 – 932.
Del’Homme, M. A., Sinclair, E., & Kasari, C. (1994).
Preschool children with behavioral problems: Observation in instructional and free play contexts. Behavioral
Disorders, 19, 221 – 232.
Deutsch, M. (1949). A theory of cooperation and competition. Human Relations, 21, 129 – 152.
Eisenberg, N., Fabes, R. A., Carlo, G., Troyer, D., Speer,
A. L., Karbon., M., et al. (1992). The relations of maternal
practices and characteristics to children’s vicarious emotional responsiveness. Child Development, 63,
583 – 602.
Eisenberg, N., Fabes, R. A., Karbon, M., Murphy, B. C.,
Wosinski, M., Polazzi, L., et al. (1996). The relations of
children’s dispositional prosocial behavior to emotionality, regulation, and social functioning. Child Development, 67, 974 – 992.
Eisenberg, N., Fabes, R. A., Murphy, B., Maszk, P., Smith,
M., & Karbon, M. (1995). The role of emotionality and
regulation in children’s social functioning: A longitudinal study. Child Development, 66, 1360 – 1384.
1326
Hay and Pawlby
Eisenberg, N., Guthrie, I. K., Murphy, B., Shepard, S.,
Cumberland, A., & Carlo, G. (1999). Consistency and
development of prosocial dispositions: A longitudinal
study. Child Development, 70, 1360 – 1372.
Eisenberg, N., & Lennon, R. (1983). Gender differences in
empathy and related capacities. Psychological Bulletin,
94, 100 – 131.
Endicott, J., Spitzer, R. L., Fleiss, J. L., & Cohen, J. (1976).
The Global Assessment Scale. A procedure for measuring overall severity of psychiatric disorders. Archives of
General Psychiatry, 33, 766 – 771.
Fabes, R. A., Eisenberg, N., Karbon, M., Troyer, D., &
Switzer, G. (1994). The relations of children’s emotional
regulation to their vicarious emotional responses and
comforting behaviors. Child Development, 65, 1678 – 1693.
Goldberg, D., Cooper, B., Eastwood, M. T., Kedward, H. B.,
& Shepherd, M. (1970). A standardised psychiatric
interview for use in community surveys. British Journal
of Preventive Social Medicine, 24, 18 – 23.
Goldthorpe, J. H., & Hope, K. (1974). The social grading of
occupations. Oxford, England: Oxford University Press.
Goodman, R. (1994). A modified version of the Rutter
Parent Questionnaire. Including extra items on children’s strengths: A research note. Journal of Child
Psychology and Psychiatry, 35, 1483 – 1494.
Goodman, R. (1999). The extended version of the Strengths
and Difficulties Questionnaire as a guide to child
psychiatric caseness and consequent burden. Journal of
Child Psychology and Psychiatry, 40, 791 – 799.
Goodman, R., Meltzer, H., & Bailey, V. (1998). The
strengths and difficulties questionnaire: A pilot study
on the validity of the self-report version. European Child
and Adolescent Psychiatry, 7, 125 – 130.
Goodman, S. H., & Gotlib, I. H. (1999). Risk for
psychopathology in the children of depressed mothers:
A developmental model for understanding mechanisms
of transmission. Psychological Review, 106, 458 – 490.
Haemaelaeinen, M., & Pulkinnen, L. (1996). Problem
behavior as a precursor of male criminality. Development
and Psychopathology, 8, 443 – 455.
Hart, C. H., DeWolf, M., & Burts, D. C. (1992). Linkages
among preschoolers’ playground behavior, outcome
expectations, and parental disciplinary strategies. Early
Education and Development, 3, 265 – 283.
Hastings, P. D., Zahn-Waxler, C., Robinson, J., Usher, B., &
Bridges, D. (2000). The development of concern for
others in children with behavioral problems. Developmental Psychology, 36, 531 – 546.
Hay, D. F. (1979). Cooperative interactions and sharing
between very young children and their parents. Developmental Psychology, 15, 647 – 655.
Helgeson, V. S., & Fritz, H. L. (1998). A theory of
unmitigated communion. Personality and Social Psychology Review, 2, 173 – 183.
Holmgren, R.A, Eisenberg, N., & Fabes, R. A. (1998). The
relations of children’s situational empathy-related emotion to dispositional prosocial behaviour. International
Journal of Behavioural Development, 22, 169 – 193.
Hughes, C., White, A., Sharpen, J., & Dunn, J. (2000).
Antisocial, angry, and unsympathetic: ‘‘Hard-to-manage’’ preschoolers’ peer problems and possible cognitive
influences. Journal of Child Psychology and Psychiatry, 41,
169 – 179.
Kier, C., & Lewis, C. (1998). Preschool sibling interaction in
separated and married families: Are same-sex pairs or
older sisters more sociable? Journal of Child Psychology
and Psychiatry, 39, 191 – 201.
Klimes-Dougan, B., & Bolger, A. K. (1998). Coping with
maternal depressed affect and depression: Adolescent
children of depressed and well mothers. Journal of Youth
and Adolescence, 27, 1 – 15.
Kochanska, G., Aksan, N., & Koenig, A. L. (1995). A
longitudinal study of the roots of preschoolers’ conscience: Committed compliance and emerging internalization. Child Development, 66, 1752 – 1769.
Ladd, G. W., & Hart, C. (1992). Creating informal play
opportunities: Are parents’ and preschoolers’ initiations
related to children’s competence with peers? Developmental Psychology, 28, 1179 – 1187.
Ladd, G., & Profilet, S. M. (1996). The Child Behavior Scale:
A teacher-report measure of young children’s aggressive, withdrawn, and prosocial behaviors. Developmental
Psychology, 32, 1008 – 1024.
LaFreniere, P. J., Provost, M. A., & Dubeau, D. (1992). From
an insecure base: Parent-child relations and internalizing behaviour in the preschool. Early Development and
Parenting, 1, 137 – 148.
Macfie, J., Toth, S. L., Rogosch, F. A., Robinson, J., Emde,
R. N., & Cicchetti, D. (1999). Effect of maltreatment on
preschoolers’ narrative representations of responses to
relieve distress and of role reversal. Developmental
Psychology, 35, 460 – 465.
Macksoud, M. S., & Aber, J. L. (1996). The war experiences
and psychosocial development of children in Lebanon.
Child Development, 67, 70 – 88.
McCarthy, D. (1972). Manual for the McCarthy scales of
children’s abilities. New York: Psychological Corporation.
Miller, P. A., Eisenberg, N., Fabes, R. A., & Shell, R. (1996).
Relations of moral reasoning and vicarious emotion to
young children’s prosocial behavior toward peers and
adults. Developmental Psychology, 32, 210 – 219.
Moffitt, T. E., & Caspi, A. (2001). Childhood predictors
differentiate life-course persistent and adolescent-limited antisocial pathways among males and females.
Development and Psychopathology, 13, 355 – 375.
Murphy, L. (1937). Social behavior and child personality. New
York: Columbia University Press.
Office for National Statistics. (2000). Social trends pocketbook.
Wales: Her Majesty’s Stationers Office.
Office for National Statistics. (2003). From http://
www.statistics.gov.uk.
Raboteg-Saric, Z., Zuzul, M., & Kerestes, G. (1994). War
and children’s aggressive and prosocial behaviour.
European Journal of Personality, 8, 201 – 212.
Radke-Yarrow, M., & Zahn-Waxler, C. (1986). The role
of familial factors in the development of prosocial
Prosocial Development and Psychological Problems
behavior: Research findings and questions. In
D. Olweus, J. Block, & M. Radke-Yarrow (Eds.), Development of antisocial and prosocial behavior: Research, theories,
and issues (pp. 207 – 233). New York: Academic Press.
Radke-Yarrow, M., Zahn-Waxler, C., Richardson, D. T.,
Susman, E., & Martinez, P. (1993). Caring behavior in
children of clinically depressed and well mothers. Child
Development, 65, 1405 – 1414.
Rheingold, H. L., Cook, K., & Kolowitz, V. (1987).
Commands activate the behavior and pleasure of
two-year-old children. Developmental Psychology, 23,
146 – 151.
Rice, R., Harold, G., & Thapar, A. (2002). The genetic
aetiology of childhood depression: A review. Journal of
Child Psychology and Psychiatry, 43, 65 – 79.
Rose-Krasnor, L., Rubin, K. H., Booth, C. L., & Coplan, R.
(1996). The relation of maternal directiveness and child
attachment security to social competence in preschoolers. International Journal of Behavioral Development, 19,
309 – 325.
Rothbart, M. K., Ahadi, S. A., & Hershey, K. L. (1994).
Temperament and social behavior in childhood. MerrillPalmer Quarterly, 40, 21 – 39.
Sharp, D. (1993). Childbirth-related emotional disorders in
primary care: A longitudinal prospective study. Unpublished doctoral dissertation, University of London.
Sharp, D., Hay, D. F., Pawlby, S., Schmücker, G., Allen, H.,
& Kumar, R. (1995). The impact of postnatal depression
on boys’ intellectual development. Journal of Child
Psychology and Psychiatry, 36, 1315 – 1336.
Spitzer, R. L., Endicott, J., & Robbins, E. (1978). The Schedule
for Affective Disorders and Schizophrenia – Lifetime version
(SADS – L) (3rd ed.), New York: Biometric Research.
Staub, E. (1978). Positive social behavior and morality: Social
and personal influences. New York: Academic Press.
Stevenson-Hinde, J., & Shouldice, A. (1995). Maternal
interactions and self-reports related to attachment
classification at 4.5 years. Child Development, 66,
583 – 596.
Tremblay, R. E., Vitaro, F., Gagnon, C., Piché, C., & Royer,
N. (1992). A prosocial scale for the Preschool Behaviour
Questionnaire: Concurrent and predictive correlates.
1327
International Journal of Behavioural Development, 15,
227 – 245.
Vinnick, L. A., & Erickson, M. T. (1992). Relationships
among accumulated lifetime life events, prosocial skills,
and behavior problems in elementary school children.
Journal of Child and Family Studies, 1, 141 – 154.
Warden, D., Christie, D., Kerr, C., & Low, J. (1996).
Children’s prosocial and antisocial behaviour, as perceived by children, parents and teachers. Educational
Psychology, 16, 365 – 378.
Wentzel, K. R., & McNamara, C. C. (1999). Interpersonal
relationships, emotional distress, and prosocial behavior in middle school. Journal of Early Adolescence, 19,
114 – 125.
Wolke, D. (1990). Tester’s rating of child behavior: Scoring
manual. Unpublished manuscript.
Wolke, D., Skuse, D., & Mathison, B. (1990). Behavioural
style in failure to thrive infants: A preliminary communication. Journal of Pediatric Psychology, 15, 237 – 253.
World Health Organization. (1978). Mental disorders:
Glossary and guide to their classification in accordance with
the ninth revision of international classification of diseases.
Geneva, Switzerland: Author.
Zahn-Waxler, C., Cole, P., & Barrett, K. C. (1991). Guilt and
empathy: Sex differences and implications for the
development of depression. In K. Dodge & J. Garber
(Eds.), Emotion regulation and dysregulation (pp. 243 –
272). Cambridge, England: Cambridge University Press.
Zahn-Waxler, C., Cole, P. M., Richardson, D. T., Friedman,
R. J., Michel, M. K., & Belouad, F. (1994). Social problemsolving in disruptive preschool children: Reactions to
hypothetical situations of conflict and distress. MerrillPalmer Quarterly, 40, 98 – 119.
Zahn-Waxler, C., Cole, P. M., Welsh, J. D., & Fox, N. A.
(1995). Psychophysiological correlates of empathy
and prosocial behaviors in preschool children with
behavior problems. Development and Psychopathology, 7,
27 – 48.
Zahn-Waxler, C., Cummings, E. M., McKnew, D., & RadkeYarrow, M. (1984). Altruism, aggression, and social
interactions in young children with a manic-depressive
parent. Child Development, 55, 112 – 122.