Early Childhood Research Quarterly, 4, 533-543 (19891

Helping Behaviors: An Observational Study

of Preschool Children

Dahlia F. Stockdale
Susan M. Hegland
Thomas Chiaromon te
lo wa State University

The present study examined the stability and circumstances of helping

behaviors in the preschool classroom. Frequencies of helping/nonhelping
behaviors, including the type and circumstance of helping behavior, were
observed for 51 children during free play by two judges. The frequency of
helping behaviors was not correlated with age or sex and was not stable
over four IO-min observation periods. Helping behaviors may depend on
the details of evoking situations. The majority of all helping acts are per-
formed during nonimaginary play. Each child performed at least one real
helping act: the majority of children performed at least one self-initiated
helping act. Most helping acts occur in response to a request from a teacher.
Results were interpreted to support that most preschool children are capable
of engaging in helping acts; however, in the preschool classroom, teachers
elicit most helping behaviors.

INTRODUCTION
There is a strong evidence that children at very young ages exhibit helping
behavior in a variety of settings toward teachers, parents, peers, and un-
familiar adults (e.g., Bar-Tal, Raviv, & Goldberg, 1982; Chapman, Zahn-
Waxler, Cooperman, & Iannotti, 1987; Eisenberg-Berg & Lennon, 1980;
Rheingold, 1982; Stanhope, Bell, & Parker-Cohen, 1987). Although the
presence of helping behavior in preschool children is well documented, a
number of studies have revealed that it is a behavior of relatively low fre-

This article is based on a thesis submitted in partial fulfillment for the MS. degree at Iowa
State University. The authors wish to thank Leroy Wolins, Department of Statistics, for his
valuable consultation in the analysis of the data. Journal Paper No. 440 of the Family and
Consumer Sciences Research Institute, College of Family and Consumer Sciences, Iowa State
University, Ames, Iowa.
Correspondence and requests for reprints should be sent to Dahlia F. Stockdale, Depart-
ment of Child Development, Iowa State University, Ames, IA 50011.

533
534 Stockdale, Hegland, and Chiaromonte

quency in naturalistic settings, especially preschool classrooms (Bar-Tal et

al., 1982; Yarrow & Waxler, 1976). Nonetheless, because prosocial behavior
is one of many self-evident needs of modern society, there is continued in-
terest in identifying the qualitative aspects, developmental sequences, and
instructional implications of helping behavior in children (Bar-Tal, 1982;
Mussen & Eisenberg-Berg, 1977).
The main purpose of the present study was to investigate the stability of
preschool children’s helping behaviors and the circumstances in which help-
ing behaviors occur in a preschool setting. In addition, the relationship of
competence to the frequency of displayed helping behaviors was explored.
According to Strayer (1981), research is needed regarding the stability of
individual differences in prosocial behavior. Specifically, we predicted that
more competent children would display higher frequencies of helping be-
havior as well as stability in helping acts across time.
Bar-Tal et al. (1982) theorized that social competence may be positively
associated with helping behavior. Children with higher levels of cognitive,
social perspective, moral, and motivational development will be more likely
to display higher quantities and qualities of helping behavior. Bar-Tal(1982)
defined helping as “an act which benefits others with no external rewards
promised a priori in return” (p. 102). For high-quality helping to occur
(i.e., altruism), prerequisite skills are needed (e.g., empathic skills) which
are rare in young children; therefore, young children are perceived to be
hampered in offering high-quality helping acts. Bar-Tal found that the
majority of kindergarten children helped either because an external reward
was promised or because an adult requested the help, which he described as
the first two stages, respectively, in the development of helping behavior.
The literature clearly suggests a tie between competence and helping
behavior; however, little is known regarding this association for preschool
children. Studies of adults and children have found that individuals who
perceive themselves as more competent are more likely to offer aid (Harris
& Huang, 1973; Midlarsky & Hannah, 1985). Research on children from
Grades 1, 4, 7, and 10 revealed that helping by younger children was in-
hibited by fear of inadequacy, whereas adolescents were inhibited by fear of
disapproval or the perceived potential embarassment of the potential recipi-
ent of aid (Midlarsky & Hannah, 1985). In a study of preschool children
and older children, Peterson (1983) found that, in two types of tasks, help-
ing behavior was positively affected by responsibility focus and competence.
Particularly relevant to the present study is an observational study of
preschool children’s helping behavior during free-play activity (Bar-Tal et
al., 1982). In this study, more than 65% of the children within each age
group (with the exception of the 43- to 54-month-old group) carried out at
least one helping act during the three lo-min periods; however, there were
children who did not perform a helping act. The frequency of helping acts
carried out in nonimaginary play increased with age. Bar-Tal et al. (1982)
Preschool Children’s Helping 535

speculated that as children get older, parents discourage imaginary play. In

terms of circumstance, these investigators concluded that most of the sub-
jects initiated helping behavior without the promise of external reward.
In an observational study, Eisenberg-Berg and Hand (1979) investigated
children’s sharing, helping, and comforting behaviors. Results indicated
that preschool children share, help, and comfort relatively infrequently.
Yarrow and Waxler (1976) also investigated sharing, helping and comfort-
ing behavior in children ranging in age from 3 to 7’/2 years. Helping acts
occurred more frequently than sharing and comforting acts; however, it was
noted that sharing and comforting showed some consistency across both
experimental and naturalistic settings. It was concluded that although pro-
social behavior occurs in most children, it occurs at a low frequency (Yarrow
& Waxler, 1976). The authors speculated that the nursery school environ-
ment may not provide many opportunities to show generosity and sym-
pathy. Thus, in the present study, efforts were made to obtain, through
naturalistic observations, an examination of the frequency of helping acts
as well as stability within individuals.
Interactions between situational constraints and individual personal char-
acteristics are important to consider in the development of helping behaviors
(Gergen, Cergen, & Meter, 1982; Hampson, 1981). Helpful behavior is
affected by a variety of stimulus conditions and, thus, individuals tend to
show considerable variability in their behavior even when situations are fairly
similar (Bar-Tal, 1982). Eisenberg, Lundy, Shell, and Roth (1985) reported
that young children more often refused requests from peers than from adults;
however, they more often cited authority/punishment justifications to ex-
plain compliance with prosocial requests from adults. Needs-oriented and
affectional relationship justifications were used more in explaining compli-
ance with requests from peers. Eisenberg et al. found that fewer requests
from adults were prosocial than those from peers. Because adults are au-
thority figures, Eisenberg et al. speculated that compliance with adult re-
quests may be percceived by children as externally motivated and, thus, less
likely to develop the child’s internal motivation for helpful behavior.
Research is needed to understand further individual differences in helping
behavior and how situational conditions, such as the source of the request
for help, impact on a child’s behavior. Research concerning the circum-
stances surrounding the occurrence of helping behaviors is needed to fur-
ther understand how this behavior functions relative to other aspects of a
child’s prosocial development and personal characteristics. This study is an
attempt to obtain more information regarding the occurrence and stability
of helping behaviors in the preschool classroom. The three objectives of the
study were:

1. To describe characteristics of helping behaviors in early childhood

settings.
536 Stockdale, Hegland, and Chiaromonte

2. To investigate the stability of helping behaviors in young children.

3. To investigate the relationship between children’s helping behaviors
and social competence.

METHOD
Subjects
The subjects were 51 children (24 boys and 27 girls) who attended three
separate preschool laboratory programs at a large midwestern university.
The age range was 41 to 69 months with a mean of 56.4 months. These chil-
dren came from an upper-middle-class socioeconomic background.

Instruments
A social behavior observation measure was used in the present study. In-
cluded in the social behavior observation measure were categories of helping
behavior, context of helping behavior, and circumstance; these particular
behaviors were adapted from a measure developed by Bar-Tal et al. (1982).
For every 10-s interval a child was judged in each of the following cate-
gories: (a) helping or nonhelping behavior, (b) context of helping behavior,
(c) circumstance, and (d) location.
The observation measure was structured in such a way that if a helping
behavior did not occur during the 10-s interval, another type of nonhelping
behavior was scored for that time period. The following operational defini-
tions were employed for judging the children’s behavior.

Helping and nonhelping behavior

1. Donafing behavior: The child donates part or all of an object of objects
that is in his/her possession.
2. Alleviafing behavior: The child attempts to alleviate another’s emotional or
nonemotional needs through verbal or motoric behavior.
3. Other social posifive: The child acts in a positive manner through con-
versing, laughing, and smiling with peers or adults.
4. Other social negative: The child saysno, cries, pouts, or shows aggression
toward another individual.
5. Nonsocial: The child plays, eats, reads by him/herself.
Contexts of helping behavior
1. Real acfs: Acts which are performed in nonimaginary play.
2. Svmbolic acts: Acts which are performed in imaginary play.

Circumstance of helping behavior

1. Self-initiated: The child, on his/her own, performs a helping behavior.
2. Compliance with child: The child performs a helping act at the request of
another child.
3. Compliance with adult: The child performs a helping act at the request of
an adult (teacher).
Preschool Children’s Helping 537

4. I~ritorion: The child performs a helping act as a result of imitation of another

child or adult (teacher).

In addition, the judges recorded whether the child was observed inside
the classroom or outdoors in the free-play sessions. It should be noted that
originally four forms of helping behavior were to be scored. These four
helping behaviors were: sharing, giving, aiding, and comforting. In the pro-
cess of training the judges, it was discovered that giving and comforting
behaviors occurred rarely; therefore, two categories of helping behavior
were developed for purposes of this study. The category donating included
sharing and giving behaviors, and the category alleviating included aiding
and comforting behaviors. These two categories were developed from defi-
nitions offered by Bar-Tal et al. (1982). Other researchers also have reported
low frequency of comforting behaviors, which resulted in combining com-
forting with other prosocial behaviors (Eisenberg-Berg & Hand, 1979; Ian-
notti, 1985).
The California Preschool Social Competency Scale (CPSCS) was used as
a teacher rating scale of children’s competence. The CPSCS consisted of 30
items concerning children’s behaviors in home and school situations. Each
item contained four possible responses. The teacher’s response for each
item ranged from 1 to 4 (i.e., 1 =low, 4= high); therefore, scores on the
scale could range from 30 to 120. Adequate interrater agreement and internal
consistency has been reported for the CPSCS (Levine, Elzey, & Lewis, 1969).

Training of Judges and Interrater Agreement

A judge’s manual was written prior to actual data collection to provide
judges with information concerning data collection along with behavioral
definitions. Interrater agreement scores were calculated for each category to
be observed (i.e., helping behavior, nonhelping behavior, context of helping
behavior, and circumstance) both before and after the study. Interrater
agreement scores for 90 10-s intervals (i.e., nine children with 10 intervals
each) were calculated by dividing the number of agreed-upon behaviors by
the total number of intervals; each category was calculated separately.
These scores ranged from .92 to &4 before the study to .94 to 1.00 at the
conclusion of the study.

Procedure
Permission for participation in the study for each child was obtained through
a letter sent to the parents explaining the study. The helping behaviors of
each child were observed by two judges (one male and one female graduate
student in child development). Each judge observed each child twice at dif-
ferent free-play time periods. Therefore, each child was observed on four
separate occasions. A time-sampling method of 10-s intervals during a
IO-min period was used. Observation order was randomly selected prior to
538 Stockdale, Hegland, and Chiaromonte

actual data collection. Each judge observed for 10 s and then used the next
10 s for scoring. This cycle continued for 10 mm, after which time another
child was targeted for observation. This procedure allowed for the observa-
tion of 30 10-s intervals on one child during each of the four IO-min periods,
for a total of 120 10-s intervals per child. The child’s initial behavior was
recorded during each 10-s interval.
A Datamyte event recorder was used to simplify and accelerate the process
of recording the observed helping behaviors. In addition, the CPSCS was
used by the teachers to rate the children’s social competence. Each child was
rated by the teacher/teachers who knew him/her best.

RESULTS
Pearson product-moment correlations were used to calculate the stability of
each observed helping/nonhelping behavior, its type and circumstance.
Age, sex, and the frequencies of helping, other positive social behaviors,
and other negative social behaviors were correlated with social competence
as measured by the CPSCS. Because the data regarding helping behavior,
nonhelping behavior, context of helping behavior (i.e., real vs. symbolic),
and circumstance of helping behavior are not normally distributed, a square
root transformation of frequencies in each of the four time periods was used-
to stabilize the variance before summing across the four periods (Snedecor
& Cochran, 1967).
Regarding the first objective of the study, describing the characteristics of
helping behavior in early childhood settings, neither age nor sex was cor-
related with the observed frequency of helping behavior. Descriptive statis-
tics for each of the observed behaviors, including the circumstances for
helping behavior, are reported in Table 1. At least one helping behavior was
performed by each child in the study; 61% of the helping behaviors were
performed in compliance with an adult (teacher) request. Only 3% of the
helping acts were imitations of another’s helping act, and only 25% of the
children ever performed a helping act in imitation of another. A total of
65% of the children performed a helping act in compliance with another
child’s request; approximately 19% of the helping acts were in compliance
with the request of a child. A total of 65% of the children performed at least
one self-initiated helping act; approximately 17% of helping acts were self-
initiated. Each child performed at least one real helping act; 95% of all
helping behaviors were real, rather than symbolic.
The percentage of helping acts to total social acts (excluding nonsocial
acts) was computed in order to make comparisons with previous research.
Age groups were selected to match some of the age ranges reported by Bar-
Tal et al. (1982) in a study of helping behavior in preschool children. Of
specific interest are the helping behaviors shown by two groups of children
studied by Bar-Tal and his colleagues in an Israeli study. Bar-Tal et al. re-
Preschool Children’s Helping 539

Table 1. Social Contacts and Social Competence: Descriptive Statistics

M SD Median IV

Social Category
(based on 120 observations per child)
Donate 0.59 0.94 0 19 ( 35%)
Alleviate 14.21 9.29 I2 51 (100%)
Help (donate & alleviate) 14.80 9.39 12 51 (100%)
Other social positive 99.00 12.86 101 51 (100%)
Negative social 0.59 2.36 0 8 ( 16%)
Nonsocial 3.80 6.23 0 22 ( 43%)

Context of Helping Behavior

Real 13.96 8.90 11 51 (100%)
Symbolic 0.82 3.21 0 8 ( 16%)

Circumstances of Helping Behavior

Compliance with child 3.08 4.22 1 33 ( 65%)
Compliance with adult 9.94 8.23 8 49 ( %%)
Imitation 0.49 1.60 0 13 ( 25%)
Self-initiated 2.12 3.48 2 33 ( 6Wo)

Social Competence (CPSCS) %.04 16.83

Now. N = 5 1 children.
a Represents the number of children in each category who performed at least one instance
of that behavior.

ported 9.5 070of total social acts were helping acts in children 43 to 54 months
old, whereas 14.7% of total social acts were helping acts in children 55 to 66
months old. In the present study, helping acts (i.e, donating and alleviating
combined) constituted 14.5% of total social acts for the younger age group
(n = 18). whereas 11.4% of total social acts were helping acts for the older
age group (n =28).
Bar-Tal et al. also reported that 92.3% of helping acts were performed in
nonimaginary play for the younger children and 94.5% for the older chil-
dren. In the present study, 90% of all helping acts were performed during
nonimaginary play for the younger children and 95% for the older children.
The second objective of the present study related to the stability of rates
of helping behavior over time. There were no significant correlations be-
tween Judge 1 and Judge 2 over the four observational periods (i.e., two
lo-min observations per judge). This indicates that rates of helping behav-
iors (i.e., donating and alleviating) were not stable over lO-min time periods.
Three categories of nonhelping behavior were options for judges’ scoring
of children’s behavior if helping behavior did not occur in a 10-s interval.
These nonhelping behaviors consisted of the following categories: other
social positive, other social negative, and nonsocial. Of 18 correlations
computed for nonhelping behaviors, only other social positive behavior was
540 Stockdale, Hegland, and Chiaromonte

correlated between Judge 1 and Judge 2 for two different observation periods
(r= .28, p < .04). Therefore, other nonhelping behaviors also lacked stability
over the time period used in the present study.
The third objective of the study related to the relationship between ob-
served helping behaviors and social competence. Social competence, as
measured by the CPSCS, was correlated with age (r= .37, p< .Ol); older
children were rated by their teachers as more competent than younger ones.
Sex was correlated (r= - .33, p<.O2) with social competence, indicating
that, on the CPSCS, boys were rated by teachers as having fewer social skills
than girls.
Social competence was not correlated with helping behaviors and was
negatively correlated both with the frequency of nonsocial behaviors (r=
- .39, p< .Ol) and with the frequency of negative social behaviors (r= - .35,
p< .05).

DISCUSSION
Insight into the situations in which helping behaviors are likely to occur is
provided by the data on the circumstances surrounding helping behaviors.
Children most often display helping behaviors in compliance with the request
of the teacher; nearly all children perform at least one helping behavior in
response to a teacher’s request. Over 80% of all helping behaviors occur in
compliance with a request from an adult (teacher) or child. Rarely do chil-
dren imitate the helping behavior of another, adult or child. However,
although fewer than 20% of the helping behaviors are self-initiated, nearly
two thirds of the children perform at least one self-initiated helping behavior.
Thus, from the present findings, it appears that all preschool children
occasionally display helping behaviors, most frequently in response to a
request from an adult. These results are consistent with the second stage of
helping described by Bar-Tal (1982), in which children help in response to a
request from another but without a promise of reward. In addition, Eisen-
berg et al. (1985) found that children were more likely to refuse a request
from a peer than from an adult, which may explain the higher levels of help-
ing behaviors observed in response to adults than in response to peers. Ac-
cording to Eisenberg et al., the teacher’s socialization role in school settings
may also serve to increase the number of children’s helping behaviors that
comply with adult requests, as found in the present study.
The majority of children demonstrate the capability of self-initiating a
helping behavior, although this happens very rarely in the classroom setting.
Longitudinal research is needed to determine if these frequencies are stable
over time, and if the frequency of helping behaviors in response to adult
requests, peer requests, or self-initiations in younger children are predictive
of the frequency of self-initiated helping behaviors in older children.
Preschool Children’s Helping 541

Neither age nor sex is correlated with the frequency of helping behaviors.
Although Bar-Tal (1982) predicted that, with age, the frequency of helping
behaviors should increase, the age range of preschool children observed in
the present study may have been too restricted to show that expected increase.
In the present study, observed frequencies of helping behaviors in younger
and older preschoolers were similar to those reported by Bar-Tal et al. (1982)
in a study of children in Israeli preschools. The high percentage of helping
behaviors observed in real, rather than imaginary, play settings also agrees
with the percentages of real helping behaviors observed by Bar-Tal et al.
These similarities suggest that reliability exists in the rates of helping behav-
iors in young children across cultures.
The results of the present study indicate that helping behaviors among
preschool children occur with low frequency. Other investigators (Bar-Tal
et al., 1982; Eisenberg-Berg & Hand, 1979; Yarrow & Waxler, 1976) have
also reported that helping behaviors occur infrequently. In addition, it can
be speculated that because the frequency of helping behavior is low, the
stability of those behaviors might also be low. Lack of significant correlations
between judges’ observations of helping behaviors indicates that instability.
Results of the present study indicate that helping behavior is not a stable
characteristic of children. To further examine the stability of helping behav-
ior, research might be conducted outside the structured laboratory preschool
programs. It might be theorized that the structure imposed during free play
via types of materials and activities provided will influence both the types
and frequencies of behaviors at any one time. In other words, the variety of
activities offered within the framework of free play is undoubtedly a potent
variable. In support of this hypothesis, Mischel (1968) argued that the fre-
quency of helping behaviors in young children is determined by situational
factors, not stable personality traits. According to Mischel, behaviors such
as helping, sharing, or comforting are highly situation-specific and depend
on the details of the evoking situations and the methods chosen to evaluate
the response.
The operational definition of helping behavior needs continued, careful
scrutiny, particularly in the classroom. In the present study, the child’s
response to a question was considered helpful behavior; the delineation of
the intent of the question may be useful information in future research.
Eisenberg et al. (1985) suggested that, in the classroom, requests from the
teacher are seen by the child more as a demand for obedience than as a need
for help.
Teacher ratings of social competence are significantly related to age and
sex of the child. As expected, older children are rated by their teachers as
more socially competent. Hartup (1983) suggested that older preschool chil-
dren have more opportunities for peer interaction and are more socially
mature. In the present study, teachers rated boys as having fewer social
542 Stockdale, Hegland, and Chiaromonte

skills. These results regarding age and sex are congruent with results reported
by Froeschner (1983). Female teachers may rate girls more favorably than
boys because they find girls’ school-related behaviors more to their liking
(Minuchin & Shapiro, 1983).
Contrary to predictions, there are no correlations between teacher ratings
of social competence and the observed frequencies of helping behaviors;
however, teacher ratings of social competence are negatively correlated with
observed frequencies of nonsocial and negative social behaviors. Teacher
ratings of social competence may be less influenced by helping behaviors,
which are rarely self-initiated and usually occur in response to a teacher re-
quest, than by the more salient nonsocial and negative social behaviors.
In summary, helping behaviors in the preschool classroom are infrequent
and unstable. All children display at least one helping behavior; however,
helping behaviors occur most often in response to a request from a teacher.
Children may not be cognizant of when help is required. Teachers of young
children may need to incorporate curricular activities designed to promote
the recognition of the need for help. Researchers (Finkelstein, 1982; Has-
kins, 1985) reported decreased frequencies of aggressive behaviors follow-
ing the implementation of prosocial curricula, such as the My Friends and
Me materials developed by Davis (1977).
The results of the present study suggest that future research should ad-
dress how teachers can help children respond to needs of others without a
teacher request. For example, teachers may demonstrate how children ask
for help directly from their peers, or teachers may describe the needs of
another without specifically asking for the child’s help. Research is also
needed to delineate how specific classroom variables promote or inhibit
children’s helping behaviors. Although it is clear from the present study
that young children exhibit the potential for helping behavior, it also sug-
gests that teachers may play an important role in facilitating that behavior.

REFERENCES
Bar-Tal, D. (1982). Sequential development of helping behavior: A cognitive-learning ap-
proach. Developtnenial Review, 2, 101-124.
Bar-Tal, D., Raviv, A., & Goldberg, M. (1982). Helping behaviors among preschool children:
An observational study. Child Development, 53, 396-402.
Chapman, M., Zahn-Waxier, C., Cooperman, G., & lannotti, R. (1987). Empathy and respon-
sibility in the motivation of children’s helping. Developmental Psychology. 23, 140-145.
Davis, D.E. (1977). My friends and me. Circle Pines, MN: American Guidance Service.
Eisenberg, N., Lundy, T., Shell, R., & Roth, K. (1985). Children’s justifications for their adult
and peer-directed compliant (prosocial and nonprosocial) behaviors. Developtnenral
Psychology, 21, 325-33 I.
Eisenberg-Berg, N., & Hand, M. (1979). The relationships of preschoolers’ reasoning about
prosocial moral conflicls to prosocial behavior. Child Development, 50, 356-363.
Eisenberg-Berg, N., & Lennon, R. (1980). Altruism and the assessmenl of empathy in the
preschool years. Child Development, 51, 552-557.
Preschool Children’s Helping 543

Finkelstein, N.W. (1982). Aggression: Is it stimulated by day care? Yolrng Children, 37(6), 3-9.
Froeschner, E.L. (1983). Social esperience. social compelence. peer group acceptance and
stability o/-friendship in preschool children. Unpublished master’s thesis, Iowa State
University, Ames.
Gergen, K.J., Gergen, J.M., & Meter, K. (1982). Individual orientation to prosocial behavior.
Journal of Social Issues, 28, IOS- 130.
Hampson, R. (1981). Helping behavior in children: Addressing the interaction of a person-
situation model. Developmental Review, I, 93-l 12.
Harris, M., & Huang. L. (1973). Competence and helping. Journal of Social Psychology, 89,
203-210.
Hartup, W.W. (1983). Peer relations. In P.H. Mussen (Ed.), Handbook of childpsychology
(Vol. 4, pp. 104-196). New York: Wiley.
Haskins. R. (1985). Public school aggression among children with varying day-care experience.
Child Development, 56, 689-703.
lannotti, R. (1985). Naturalistic and structured assessments of prosocial behavior in preschool
children: The influence of empathy and perspective taking. Developmental Psychology,
21. 46-55.
Levine, S., Elzey, F.F., & Lewis, M. (1969). Calijornia Preschool Social Competency Scale.
Palo Alto, CA: Consulting Psychologists Press.
Midlarsky, E., & Hannah, M. (1985). Competence, reticence, and helping by children and
adolescents. Developmental Psychology, 21, 534-541.
Minuchin, P.P., & Shapiro, E.K. (1983). The school as a context for social development. In
P.H. Mussen (Ed.), Handbook of childpsychology (Vol. 4, pp. 197-274). New York:
Wiley.
Mischel, W. (1968). Personality and assessment. New York: Wiley.
Mussen, P.H., & Eisenberg-Berg, N. (1977). Roots of caring, sharing, and helping. San Fran-
cisco: W.H. Freeman.
Peterson, L. (1983). Influence of age, task competence, and responsibility focus on children’s
altruism. Developmental Psychology, 19, 141-148.
Rheingold, H.L. (1982). Little children’s participation in the work of adults, a nascent pro-
social behavior. Child Development, 53, 114-125.
Snedecor, G.W., & Cochran, W.G. (1967). Statistical methods. Ames, IA: Iowa State Univer-
sity Press.
Stanhope, L., Bell, R., & Parker-Cohen, N. (1987). Temperament and helping behavior in pre-
school children. Developmental Psychology. 23. 347-353.
Strayer, F.F. (1981). The nature and organization of altruistic behavior among preschool chil-
dren. In J.P. Rushton and R.R. Sorrentino (Eds.), Altruism and helping behavior:
Social, personality and developmental perspectives (pp. 33 1-351). Hillsdale, NJ:
Erlbaum.
Yarrow, M., & Waxler, C. (1976). Dimensions and correlates of prosocial behavior in young
children. Child Development, 47, 118-125.