The effects of modeling, direct instruction, and competence on children's helping responses

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content THE EFFECTS OF MODELING, DIRECT INSTRUCTION,
AND COMPETENCE ON CHILDREN'S
HELPING RESPONSES
by
Joan Helen Blacher
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(Education)
June 1981
UMI Number: DP24757
All rights reserved
INFORMATION TO ALL USERS
The quality of this reproduction is dependent upon the quality of the copy submitted.
In the unlikely event that the author did not send a complete manuscript
and there are missing pages, these will be noted. Also, if material had to be removed,
a note will indicate the deletion.
Dissertation Publ sbmg
UMI DP24757
Published by ProQuest LLC (2014). Copyright in the Dissertation held by the Author.
Microform Edition © ProQuest LLC.
All rights reserved. This work is protected against
unauthorized copying under Title 17, United States Code
ProQuest LLC.
789 East Eisenhower Parkway
P.O. Box 1346
Ann Arbor, Ml 48106- 1346
UNIVERSITY OF SOUTHERN CALIFORNIA t cA U < *
TH E G RA D UA TE SC HO O L
U N IV E R S IT Y PARK
LOS AN G ELES. C A L IF O R N IA 9 0 0 0 7 * £ ^
This dissertation, written by
. . . JOM. HELEN. BLACHER.................
under the direction of A.ER__ Dissertation Com
mittee, and approved by all its memberst has
been presented to and accepted by The Graduate
School, in partial fulfillment of requirements of
the degree of
D O C T O R O F P H I L O S O P H Y
Dean
DISSERTATION COMMITTEE
I t . '
Chairman
ACKNOWLEDGMENTS
I want to express my deepest appreciation and grati
tude to many people who helped in the completion of this
research. Dr. Myron Dembo, my chairman, provided ongoing
guidance and direction; Dr. Dennis Hocevar and Dr. Phillip
Weise, the other members of my committee, gave invaluable
assistance,
Pat Distad, Linda Damesek, Kari Damesek, and Michael
Gregson gave fine performances for the videotapes; Joan
Logan and Donna Torgeson were enthusiastic and superb
interviewers; and Joan Combes gave helpful technical
support in computerizing the data.
I am most indebted to the children, teachers, and
staff of the Conejo Valley Unified School District who
participated in the study. Without them, the research
would not have been possible,
Most of all, I want to express my thanks to my husband,
Norman Blacher, who offered continued encouragement, moral
support, and help in many ways; and to my parents Isabel
and Albert Oakholt for their faith in me.
ii
TABLE OF CONTENTS
Page
ACKNOWLEDGMENTS ........................................ ii
LIST OF TABLES.......................................... v
Chapter
I. THE PROBLEM................................... 1
Background of the Problem............... 1
Statement of the Problem Situation........... 3
Purpose of the Study ............... 7
Importance of the S t u d y ...................... 8
Hypotheses ...................................... 9
Conceptual Assumptions ........................ 9
Definitions................................... 10
Delimitations................................. 11
Organization of the Remainder of the Study . . 12
II. REVIEW OF THE LITERATURE...................... 14
Introduction ................................... 14
Conceptual Background .... 14
Methodology................................... 18
The Role of Models............................. 22
Affect and Helping............................. 36
Cognitive and Developmental Factors ......... 40
Individual Differences ........................ 44
Summary........................................ 50
III. METHODOLOGY . 54
Introduction ................................... 54
Subjects........................................ 54
Research Design ............................... 55
Materials................. 55
Social Influence Videotapes ............... 55
Competence.................... 58
Dependent Measure ............................. 59
Procedures................................. .. . 60
Initial Session ............................. 60
Delayed Measure ............................. 63
Statistical Analysis . . . 64
Chapter Page
IV. RESULTS AND DISCUSSION.................. 65
Results........................................ 65
Immediate Measure ........................... 65
Delayed Measure ............................. 68
Interpretation ................................. 72
Main Effect for Method
of Social Influence ...................... 72
Main Effect of Competence.................. 80
Other Findings............................... 84
V. SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS . . 86
Summary........................................ 86
Conclusions................................... 89
Recommendations............................... 91
REFERENCES............................................... 93
APPENDIXES
A. MEASURES OF HIGH AND LOW COMPETENCE ..... 101
B. MEASURE OF HELPING................................ 104
iv
LIST OF TABLES
Table page
1. Mean Number of Days Children Volunteered
on Immediate Measure......................... 66
2. Analysis of Variance on Immediate Measure . . 67
3. Mean Number of Days Children Volunteered
on Delayed Measure........................... 70
4. Analysis of Variance on Delayed Measure . . . 71
v
CHAPTER I
THE PROBLEM
Background of the Problem
Traditionally, educators have devoted more attention
to the teaching of academic tasks rather than to the devel
opment of social skills. While the learning of reading,
language, and arithmetic skills is, of course, an important
goal for all children, it is becoming increasingly clear
that the development of certain social skills may be
crucial to the academic instructional experience as well as
to the overall school success of the individual student,
and to the quality of children's lives wtihin the educa
tional setting (Cartledge & Milburn, 1978). Not only have
schools frequently focused primarily upon academic skills,
but additionally have viewed their role as a socializing
agent as that of suppressing antisocial activity rather
than that of developing dispositions toward prosocial
behavior. Thus, children in classrooms often have not been
taught specifically how and when to help others nor the
appropriate situation in which to offer help. Typically,
children in school know "what not to do" but not exactly
"what to do." As Bryan (1977) has pointed out, greater
1
emphasis in socialization upon the "do's" than the "don'ts"
is needed.
In order to remedy the lack of positive orientation
toward children's acquisition of social skills, teachers
will need to learn more about the socializing influences
that will facilitate helpfulness in the classroom. At
present, empirical evidence suggests that teachers have two
powerful sources of social influence at their disposal in
the classroom: modeling and verbal socialization (Rushton,
1976).
While the methods utilized by the teacher to elicit
helpfulness in his or her students are of great importance,
also to be considered are possible motivating factors that
underlie the behavior. There have been virtually no inves
tigations of the motivational aspects of children's
altruism. One possibility that has been sporadically
mentioned in the literature is the notion that feelings of
competence might play a role in influencing individuals to
act altruistically. There is some evidence (Bruner, 1972;
Kohlberg, 1969; Zigler & Yando, 1972) that a primary motive
for imitating a model is the desire to act competently.
Several investigators of prosocial behavior have shown an
interest in pursuing this line of inquiry by empirically
testing a suspected relationship between perceptions of
competence and altruism (Kazdin & Bryan, 1971; Midlarsky,
2
1968). However, at present, the research evidence is
meager and relates almost entirely to adults.
Statement of the Problem Situation
The primary question that needs to be answered for
educators is which source of social influence will produce
the most positive results in enhancing children’s prosocial
behavior: modeling or verbal socialization. Modeling, a
concept based on social learning theory, would suggest that
learning occurs through the observation and imitation of an
exemplar's behavior. There is impressive evidence that the
observation of a model can be effective in producing behav
ior change (Bandura, 1969). Verbal socialization refers
essentially to the concept of induction (Hoffman, 1970),
which means that an adult gives explanations or reasons for
requiring the child to change his or her behavior.
Examples would be pointing out the physical requirements of
the situation or the harmful consequences of the child's
behavior for himself or others. Rushton (1976) has
included in the class of verbal socialization techniques
not only induction but also labeling and preaching and has
pointed out that this aspect of social influence has not
been as well studied nor have results been as conclusive as
have those in studies of imitation.
Although there has been considerable research about
the effect of both models' deeds and their words, the
3
evidence is not conclusive. As Rushton (1975) has stated,
the relationship of verbal communication to behavior
requires much more extensive investigation since the con
ditions that maximize the effects of the model's words are
unclear. An investigation of the research directed toward
a comparison of models' performance versus their verbal
izations in producing altruistic behavior in children
offers no conclusive evidence that either method is clearly
superior.
On the one hand, a number of studies have found that a
behavioral example of helping has led to more altruistic
behavior in children than did the model's verbalizations
(Bryan 6c Walbek, 1970a, 1970b; Grusec 6c Skubiski, 1970;
Rushton 6c Owen, 1975). However, an analysis of this
research has revealed methodological problems in each case
which has cast some doubt on the results. For example, in
the Bryan and Walbek (1970a, 1970b) studies, the model's
behavioral example was verbally labeled, which meant that
there were verbalizations in each condition. Furthermore,
as Rushton (1975) has pointed out, there have been weak
nesses related to the timing and direction of the verbal
exhortations in the Bryan and Walbek (1970a, 1970b)
research and the Grusec and Skubiski (1970) study.
On the other hand, an inspection of the research which
resulted in a significant effect for the model's verbal
izations over the model's behavioral example (Elliott 6c
------ — -_________________________________ 4
Vasta, 1970; Friedrich & Stein, 1975; Midlarsky & Bryan,
1972; Rice & Grusec, 1975), uncovered an important differ
ence in design. In each of these studies, the verbal
ization of the model included an informational component
which was not present in the research wherein a significant
effect for behavioral example was found. Further research
is decidedly needed in order to clarify the discrepancies
in existing findings if training methods that will be use
ful in an educational setting are to be determined.
An additional area for fruitful inquiry into the
determinants of prosocial behavior is that of possible
motivating factors. The majority of research which has
examined altruism in children has made the assumption that
behavior and motivation are congruent, and, therefore,
little or no investigation into motivational aspects has
been conducted. In this vein, Krebs (1970) has indicated
that motivational specification is particularly important
in relation to a moral behavior such as altruism and has
viewed the lack of research in this area as a serious
omission. As Piaget (1962), Kohlberg (1964), and others
have demonstrated, it is more the intention behind an act
than its consequences that determines its moral value.
While a number of possibilities have been offered as
to why children will imitate an altruistic model, none has
been thoroughly studied. Explanations have included the
following: (1) imitating a generous model is
5
self-reinforcing (Rosenhan, 1972); (2) children learn a
social responsibility norm that one should help a dependent
(Berkowitz 6c Daniels, 1964), with the model's behavior
reminding the child of this norm; and (3) the "demand
characteristics" of the model's presence cause children to
do what they think is required of them within an experi
mental setting (Bryan, 1975). A fourth possibility which
has recently been gaining acceptance is the idea that
feelings of competence might be a contributing motivational
factor. Research into this subject, however, has focused
on adults or has only indirectly examined the effect of
competence (Berkowitz & Conner, 1966; Isen, Rosenhan 6c Horn,
1973; Kazdin 6c Bryan, 1971; Midlarsky, 1968).
Although the underlying psychological processes that
might relate competence to helping behavior aren't known,
feeling competent has been shown to enhance altruistic
responses in experiments involving adults (Kazdin 6c Bryan,
1971; Midlarsky, 1968). If feeling competent is indeed a
motivating factor for altruistic behavior, and children
learn to be competent by imitating an adult, then the
assumption could be made that children would help when they
have a model to imitate and when led to believe they are
competent to help.
One further issue that should be stated in regard to
differential methods of social influence and altruism is
that research evidence has been amassed almost exclusively
6
in experiments involving a bowling game and the donation of
money, gift certificates or candy to a nonexistent charity
(Schwartz 6c Bryan, 1971a, 1971b; Bryan 6c Walbek, 1970a,
1970b; Harris, 1971). None of these activities is likely
to occur with great regularity, if at all, in the elemen
tary school. Thus, there is a decided need to ascertain
the effects of various kinds of modeling utilizing dif
ferent procedures and other forms of altruism besides
donating as the dependent variable. It is particularly
important from an educational standpoint to examine aspects
of prosocial behavior that are more likely to take place at
school, such as helping with academic subjects or with
playground games.
This study was directed to the need for research into
the methods that will encourage children to become helpful
to one another at school and to examine the role that com
petence as a motivational factor might play in leading a
child to perform an altruistic act.
Purpose of the Study
This study was designed to generate data about the
socializing techniques that would prove valuable to the
teacher in the classroom in developing prosocial behavior
in students. The study investigated the question as to
which was the more effective procedure in eliciting help
fulness: (1) direct verbal instructions by a model, or
7
(2) a behavioral example of one individual helping another.
The effect of children's feelings of competence about their
ability to help was also considered. The subgroups of boys
and girls were examined as part of the investigation as was
the durability of children's responses.
Importance of the Study
The importance of this study was to extend the
research available on the effects of social influence and
competence on prosocial helping behavior that would be
found in a school setting. Insofar as educators are likely
to serve as powerful models to the children in their class
rooms, the nature of modeling effects are particularly
important for them to understand. Rather than relying on
common sense, opinion, tradition, or emotion to guide them,
teachers must have empirical evidence upon which to depend.
At present, published research has not conclusively
demonstrated that either a behavioral model or a model's
words are the most effective method of enhancing helpful
ness in the classroom. Evidence is greatly needed that
will lead to the development of methods that will increase
helping behavior in an educational setting. Certainly, if
social behaviors which contribute to positive personal
interaction can be demonstrated to increase with specific
antecedent events such as appropriate verbal messages about
providing help, the self-perception of competence, these
8
antecedents can be planned for in training programs which
will teach prosocial behavior to children in the classroom.
The determination of socializing methods useful in
developing children's dispositions toward helpful behavior
has broad application in schools. The present study was
designed to provide research evidence that would clarify
some of the existing discrepancies in past investigations,
and to extend the body of knowledge in regard to children's
altruistic behavior, with the hope that it would be of
benefit to those who influence the lives of children in
school.
Hypotheses
The following research hypotheses were tested:
1. A model's use of direct verbal instruction,
including specific information about the circumstances that
call for help, will result in significantly more helping
responses than will a model's physical demonstration of
help ing ano ther.
2. High competence students will help significantly
more than will low competence students.
Conceptual Assumptions
Implicit in the investigation were the following
conceptual assumptions:
1. Fourth-grade students ranging in age from 8.10 to
_________________________________________________________________9
9.10 years are functioning at a mature enough cognitive and
linguistic level to understand the verbal instructions
given by an adult female model.
2. Children of fourth grade would find first-grade
arithmetic problems and eighth-grade arithmetic problems of
such ease and difficulty, respectively, that they would
consider themselves to be competent to complete the former
and to lack competence for the latter.
3, Fourth-grade students are able to accurately judge
their level of academic competence in arithmetic.
Definitions
The terms defined operationally for this study were as
follows:
Competence. A theoretical construct usually describ
ing one's feelings that one is effective in dealing with
one's environment. In this study, competence is oper
ationally defined to mean the ability of fourth-grade
students to complete arithmetic problems at either the
first- or eighth-grade levels.
Direct Instruction. A videotaped adult model's verbal
direction given to a child that includes specific infor
mation about the appropriateness of helping in a particular
situation. An example of such a statement would be "John
has a flat tire on his bicycle. I would like you to help
:io
him fix it." No physical demonstration of helping accom
panies the verbal direction.
Helping. Helping refers to the rendering of aid to
another. It is operationalized, in this study, to mean
volunteering to provide tutoring help for a period of one
to five days a week to an unknown student who is having
difficulty in school with the subject of arithmetic.
Physical Demonstration. A videotaped model’s physical
demonstration of rendering assistance to an adult or child
in three episodes where help is required. The three epi
sodes depict (1) a boy with a flat tire in need of air,
(2) an adult with an abundance of packages which have
dropped on the floor, and (3) a girl who is unable to carry
out a difficult arithmetic homework assignment. In each of
the described episodes, the physical demonstration of help
is unaccompanied by verbalization.
Delimitations
The present study was subject to the following delimi
tations which served to narrow its focus and generaliz-
ability.
1, All subjects were enrolled in regular fourth-grade
classes in the Conejo Valley Unified School District.
2. At the beginning of the study, all subjects ranged
in chronological age from 8.10 years to 9.10 years.
11
3. All subjects used in the study were Caucasian,
English-speaking, and came from a community described as
middle-class and suburban.
4. The subject sample was composed of 71 girls and
75 boys.
5. Generalizations about the effect of a model's
verbalizations or physical demonstration apply only to the
procedures designed for this study.
6. Only the operationalized definition of competence
should be utilized when making generalizations about its
effect on helping.
Organization of the Remainder of the Study
The remainder of the study is organized as follows:
Chapter II presents a review of pertinent literature
related to the determinents and correlates of helping
behavior in children.
Chapter III sets forth the methodology of the study.
Included will be (1) research design and control pro
cedures, (2) subject selection and classification,
(3) materials used, (4) dependent measures, (5) statistical
analysis, (6) methodological assumptions, and (7) limi
tations of the study.
Chapter IV describes the findings of the study and
presents an interpretation of the results.
12
Chapter V offers a summary of the study, some con
clusions made about the outcomes, and recommendations based
on the results.
13
CHAPTER II
REVIEW OF THE LITERATURE
Introduction
This chapter presents the theoretical and experimental
framework upon which this investigation into the helping
behavior of children is based. To be discussed first is a
summary of the conceptual background as offered in the
literature. Next to be outlined will be an overview of the
methodology employed in investigations of this subject.
Following this section will be a discussion and critique of
studies about the determinants and correlates of prosocial
behavior that are pertinent to the present investigation.
These topics are the role of models, affect and helping,
cognitive and developmental factors, and individual dif
ferences. Lastly, a summary of the studies selected will
be given.
Conceptual Background
When examining the research, it appears that the
majority of studies which have investigated the determinants
and relationships of prosocial behavior in children have
shown theoretical and methodological similarities. Further
more , some influences on helping behavior have been well
14
investigated while others have remained virtually neglected.
For the most part, studies in helping behavior have been
based on social learning theory and have focused on the
role of observation and reinforcement upon the learning and
performance of children (Bandura, 1969). The emphasis has
been placed upon the impact of children's observing others
engaged in helping someone, the effects of such helping on
the observer, and the characteristics of the observer who
was later going to help. Children have followed the lead
of prosocial models even when the model was later absent at
the time the child had an opportunity to be charitable, and
when the model received no reinforcement for being char
itable. One explanation for these possible challenges to
social learning theory has been suggested by Kohlberg
(1969). He has postulated that the model's behavior simply
informs the child of what older, more competent people
consider the "appropriate" response in the situation.
Ostensibly, the child observer then follows the model's
example as part of his or her attempt to imitate the
behavior of competent others; in other words, emulating
competent people is intrinsically reinforcing.
Other possible explanations are that the model's
altruistic act merely reminds children of the norm of
social responsibility which they have already internalized
(Berkowitz & Daniels, 1964), or the norm of reciprocity
(Gouldner, 1960) which argues that people should help and
15
not injure those who have helped them. Finally, children
may learn that altruism is self-reinforcing when they
observe altruistic models reinforcing themselves by
expressing happiness or some other form of positive affect
when they help (Bandura, 1971; Midlarsky & Bryan, 1972).
A second major approach to the study of children’s
altruistic behavior is that of the cognitive-developmental
theorists. Their assumption is that prosocial responses
such as sharing, cooperation, giving reassurance and com
fort, and volunteering to help others should rarely occur
prior to middle childhood (ages six to twelve) because
children younger than that have not yet acquired the cog
nitive skills or abilities such as a loss of egocentrism,
empathic capabilities, and the development of role-taking
skills and more advanced levels of moral reasoning
(Kohlberg, 1969; Rushton, 1976).
Although these two theories are often viewed as con
flicting or contradictory, they do emphasize different
aspects of development. Social learning theory stresses
the role of antecedent and consequent environmental events
whereas the cognitive-developmental approach stresses the
role of cognitive structures. Rushton (1975) has used both
theoretical frameworks as a basis for a single experiment
when examining the effects of modeling and children’s moral
judgments, Shaffer (1979) suggests that both the social
learning theory and cognitive-developmental theory
16
perspectives be considered complementary rather than contra
dictory statements about the origins of altruism.
Helping behavior has been referred to in the liter
ature as altruism, aiding, donating, rescuing, bystander
intervention, sharing, and prosocial behavior, depending
upon the particular research design being employed.
A central issue in the conceptualization of helping behav
ior has been whether or not the act was unselfish and
implied self-sacrifice on the part of the benefactor.
Leeds (1963) defines altruism as a voluntary act
intended to produce a beneficial outcome for another for
which there would be no personal gain for the actor.
Macaulay and Berkowitz (1970) define altruism as that
behavior which is carried out to benefit another without
anticipation of rewards from outside sources. According to
Barrett and Yarrow (1977), prosocial behavior is an attempt
to fulfill another person's need for physical or emotional
support. These attempts would include acts of comforting
(physically or verbally expressing sympathy or reassurance);
sharing (giving materials or work space that one is using),
or giving a "turn" to another person; and helping (phys
ically assisting or offering physical assistance).
The terms altruism, prosocial, and helping behavior
have often been used interchangeably in the literature and
will be so used here, meaning in all cases a positive
social act toward another. However, it would appear that
17
the most appropriate term to use in an educational setting
would be "helping" since children often have the oppor
tunity to facilitate another's work or play but rarely have
the opportunity to "rescue" someone in distress or to
"donate" to "needy" others. An additional distinction has
been made by Severy and Davis (1971) , who define helping
behavior as having two elements: (1) task help (helping
others with tasks), and (2) psychological help (helping
with another's distressed feelings).
In summary, definitions of children's helping behavior
have, in general, depended upon the type of research being
carried out. Some meanings have implied self-sacrifice
while others have not, and the motivational factors under
lying the helpfulness have not been considered. For the
remainder of this review, the definition of helping
behavior will be considered to be a positive social act
toward another with the purpose of contributing to that
person's well being.
Methodology
The kind of helping response elicited in research
studies has generally been one of two types: either rescue
activity or donation. In the first type of experiment, the
child has been exposed to a situation wherein a peer is
experiencing some sort of distress (Staub, 1971a). In the
second type, the child has been provided with an opportunity
18
to sacrifice anonymously some prized object to a charitable
organization or to a needy child (Bryan 6c Walbek, 1970a).
It is the latter type of experiment which constitutes the
bulk of the research. Further, most research concerning
children's helping behavior has been addressed to the role
of a model's social influence in affecting altruism.
Essentially, what happens is that children have viewed
someone who does or does not help, does or does not preach
helping, or donates a small or large amount, Following
this event, measures are obtained as to how this model
influences the viewing child. The majority of research has
thus been concerned with elicitation rather than develop
ment of helping behavior (Bryan 6c London, 1970; Midlarsky,
1968) . ■
Recent efforts have begun to focus more on the
training of children to become helpful. For example,
Friedrich and Stein (1975) trained children by using both
verbal and role playing techniques, and found that training
led to an increase in helping behavior. Staub (1971b),
using role-playing situations in which children played both
helper and the one who was helped, trained children under
six years of age to show a number of helping behaviors.
This training led children to demonstrate greater altruism
in similar situations. Yarrow et al. (1973), in a natural
istic study with preschool children, found that following
training with the use of a nurturant model, children gave
19
more help, verbalized more sympathy, and were more con
sistent in their altruism.
In regard to the measurement of prosocial behavior,
the majority of studies have relied on behavioral measures,
for example, the number of pennies or candy donated
(Elliott 6c Vasta, 1970), the percentage of children who
went into another room to rescue a crying child (Staub,
1971a); the amount of time that a subject volunteered to
help at a later date (Green 6c Schneider, 1974); or the
number of children who picked up objects that an adult had
dropped (Yarrow 6c Waxier, 1976).
In addition to the lack of studies which have inves
tigated the training of helping behavior, few have been
concerned with the degree to which various measures of
helping behavior are correlated. Those that have examined
these relationships have generally found low correlations.
Rushton (197 6) points out that the interrelationships
between children’s altruistic behaviors are of the magni
tude of .3. This figure, Rushton goes on to say, could
offer support either for a general trait of altruism or a
situational specificity. While some common features of the
measures of altruism are suggested, they might be artifacts
such as response tendencies, observer bias, halo effects;
or uncontrolled variables such as IQ; internal dispositions,
i.e., empathy or cognitive structure; or.common environ
mental controlling conditions as model, cue, or reinforcing
20
stimuli. It may well be that generalizations as to the
determinants of various forms of helping behavior other than
those which have been employed in the specific experiment
are fallacious.
While most studies have taken place in a laboratory
setting, there is some evidence to suggest that laboratory-
based findings generalize to situations outside of the
laboratory. Midlarsky and Bryan (1972) did find that
children's donations in the laboratory predicted their
giving within a classroom setting.
Investigations into children helping others with
school-like tasks which would be of practical concern to
the educator are almost nonexistent. Only three laboratory
studies have been identified, The first, by Berkowitz and
Friedman (1967), included productivity on a geometric
drawing as the task. The second study (Green & Schneider,
1974) asked elementary-age students if they would be
willing to volunteer some time in order to assemble books
for poor students. In this study, also, students had an
opportunity to pick up pencils that an adult had "acciden
tally" dropped. The tests of helping in the third study
(Yarrow & Waxier, 1976) were to expose a preschool child to
an adult who purportedly spilled a box of tennis balls in
one situation, and to dropped eating utensils in another.
One naturalistic study by Severy and Davis (1971) involved
21
rating children who helped others accomplish specific
classroom tasks.
To summarize, it would appear that the majority of
studies about helping behavior have a sameness about them
in that they tap a narrow range of children's altruism in
highly contrived laboratory settings with heavy utilization
of games or mythical distress situations. Well-validated
measures of helping behavior in a variety of situations
seem to be lacking.
The Role of Models
The effect of models has been the subject of consider
able research concerned with prosocial behavior. There is
abundant evidence that children who observe helping
behavior in others are more likely to help than those who
do not, regardless of whether the model is a peer or adult
or is observed live or on videotape (Bryan, 1972). Imi
tation has been demonstrated with rescue acts (Staub,
1971a) and the donation of money or valued objects (Bryan Sc
Walbek, 1970a, 1970b; Grusec et al., 1978; Harris, 1971;
Rushton, 1975). Of considerable importance are findings
that the impact of a generous model may be relatively
enduring. A number of investigators (Grusec et al., 1978;
Israel Sc Brown, 1979; Israel Sc Raskin, 1979; Rushton, 1975;
Rushton Sc Owen, 1975) have found that children who were
exposed to a generous model in a laboratory setting were
22
not only more likely to donate within that context imme
diately after viewing the model but were also more likely
to help others at least five days later upon retesting.
When investigating the influence of models upon pro-
social behavior, a number of areas have been the subject of
study. Among them are the effects of verbal versus phys
ical demonstration when encouraging helping behavior; the
characteristics of the model such as warmth, power, or
hypocrisy; and the consequences to the model following the
model's kindness or generosity.
Studies investigating the question about which ante
cedent event, verbalization, or physical demonstration has
elicited more prosocial behavior have typically employed
the same research design. In this design, a model provided
either a physical demonstration of donating winnings from a
bowling game to a charity or preached a moral exhortation
such as "it is good to give to the poor" or "you should
give to the poor." The observing child was then given an
opportunity either in or out of the model's presence to
donate winnings, too.
An analysis of the research which has shown that a
physical demonstration has led to more altruistic behavior
in children than did the model's verbalizations (Bryan &
Walbek, 1970a, 1970b; Grusec & Skubiski, 1970; Rushton &
Owen, 1975) has revealed an imprecise methodology. In each
of these investigations, the model verbally labeled his or
.23
her behavior when donating which casts some doubt on the
finding that it was performance alone that caused the
effect. In another study (Grusec, 1972) mixed results were
found, with the model’s verbalizations being effective for
eleven-year-olds and seven-year-old girls but not for
seven-year-old boys. Grusec’s explanation for this effect
was that girls are more verbally fluent than boys at age
seven. It should be pointed out, however, that this study
also included verbal labeling of the model's performance so
results are somewhat suspect here as well.
A further methodological problem that occurred in all
of the studies cited above, as pointed out by Rushton
(1975), was that the preaching effect may have been inef-.
fective because of the weakness of the manipulation of the
preaching variable in the following areas: (1) timing,
since the modeling always occurred when the model was in
the "win" situation, and the preaching occurred either on
"nonwin" trials as in the Bryan and Walbek (1970a, 1970b)
studies, or before any trials at all as in the Grusec
(1972) and Grusec and Skubiski (1970) studies; and
(2) direction, in that verbal exhortations were not
directly addressed to the subject but were either delivered
to an empty room (Bryan 6c Walbek, 1970a, 1970b), or the
model was careful not to look at the child (Grusec, 1972).
In an attempt to rectify these methodological weaknesses,
Rushton (1975) replicated the research designs, and
____________________ 24
although he found no direct effects due to preaching on an
immediate test, he did find that preaching had a substan
tial impact on the child's donating behavior on an eight-
week retest. Based on an evaluation of this existing
research, there is hardly conclusive evidence that a
behavioral demonstration is the preferred technique.
An inspection of experimental studies which have found
a significant effect for the model's verbalizations over
the model's performance (Midlarsky 6c Bryan, 1972; Rice 6c
Grusec, 1975; Elliott 6c Vasta, 1970; Friedrich 6c Stein,
1975) has disclosed an interesting and perhaps important
difference in the research designs. In each study, the
verbalization of the model included an informational com
ponent in addition to the simple moral exhortation which
was utilized in the experiments that did not show an effect
for the model's words. To illustrate, in the Midlarsky and
Bryan (1972) study, the model's verbal message to the ten-
year-old observers contained a statement of affect indi
cating that it felt good to give to the poor. The results
of this study showed both an immediate and a ten-day
delayed effect for preaching. In their research, Rice and
Grusec (1975) had their model provide information about
what would be appropriate behavior in a situation that
allowed for donation to a charity and found that the verbal
ization treatment significantly affected nine- to ten-year-
olds' donations both on an immediate test and on a
25
four-month retest. The inclusion of an inductive verbal
direction to the model’s preaching in the Elliott and Vasta
(1970) study, e.g., "if you do something nice for someone
else it means you are a good boy,” produced more subsequent
donating in observers than the modeling condition without
the verbal elaboration. Additionally, Friederich and Stein
(1975) provided kindergarten children with labels for
behavior and feelings connected to altruistic acts and
found positive increments on measures of prosocial
responding.
Some studies have provided very specific information
about what was expected of subjects. For example, although
White (1972) referred to one of his treatment conditions as
"Guided Rehearsal," what this term really meant was that
the fourth- and fifth-grade subjects were told bluntly,
"What we would like you to do is to give one certificate to
the orphans each time you win two." There could be little
doubt in the subjects' minds as to the meaning of this
message, and, not surprisingly, the results were that those
subjects who received these strong instructions donated
more than those who did not. Similar instructions were
provided in a study of fourth- and fifth-grade girls (White
& Burnam, 1975). In this research, modeling and two types
of instruction which the investigators called (1) "per
missive" and meant that the model said, "you may give but
you don't have to" and (2) "constraining" wherein the model
26
said "what I'd like you to do is to donate the pennies you
win each time you win five." White and Burnam's findings
were consistent with White's (1972) in that the constrain
ing instructions resulted in the most donation overall with
some interaction effects in regard to the amount of
modeling that subjects observed,
A recent study by Grusec, Kuczynski, Rushton, and
Simutis (1978) involving seven- to ten-year-old children
surprisingly did not find that there was a significant
difference between the behavioral model and the model's
instructions in influencing the observers. In this inves
tigation, children were told that they were either the kind
of person who likes to help others (self-attribution) or
that they shared because the examiner expected them to
(external attribution). Instructions given to the children
were called either (1) "high certainty," which were "Now I
want you to share one of your marbles with the poor chil
dren each time you win," or (2) "low certainty" in which
the model said "if you want to you can give your marbles to
the poor but you don't have to." Despite the lack of
significant findings, a substantial amount of donation
occurred in the direct instruction (high certainty) con
dition. A further outcome in this research was that
children who were told they were kind were induced to
behave more altruistically in the modeling-only situation.
This finding was explained as showing that observing a
27
model is a more ambiguous event than being told what to do,
and so the subjects could therefore be more easily manipu
lated .
It is apparent, after a review of research investi
gating differential modeling effects on children's helping
behavior, that there is no clear evidence for either verbal
socializing techniques or modeling in producing more
helping behavior. What does seem clear, however, is that
the quality of the verbal instructions used in studies of
altruism have had a great deal to do with the outcomes.
It seems that when directions have been clear and have
contained information about what to do or what it feels
like to be charitable, these verbal messages significantly
influenced the observer to donate. As Rushton (1976) has
indicated, the verbal socializing techniques of preaching,
induction, and labeling have only recently begun to be
studied, and there appears to be little certainty about how
to conceptualize, categorize or account for them theo
retically. Since, however, most formal and informal means
of exerting influence take place, particularly in class
rooms, in a verbal medium, it would seem that much more
research attention needs to be devoted to these verbal
socialization processes.
The kinds of models that children might observe vary
in a number of ways. Among the characteristics of models
which have been the subject of study are those of the
28
model’s affective expression, warmth, power, and moral
consistency.
Since warmth has been a central concept in theories of
identification (Bandura, 1969), it would be expected that
the helping of a warm or nurturant model would be imitated
more than that of a cold or indifferent adult. This has
not always been so.
Typically, investigations studying model warmth have
involved a model-child interaction for a very brief amount
of time, with the warm model acting in a friendly, suppor
tive, interested fashion. Due to the brevity of the
interaction, however, doubt has been cast as to the rapport
established by the model. Children might well have been
suspicious or uncomfortable about the intrusiveness of the
warm model rather than responsive, and this fact might have
influenced research results.
One naturalistic study by Yarrow, Scott and Waxier
(1973) investigated rescue behavior in nursery school chil
dren for a period of six weeks and found that a warm,
nurturant model increased children's aiding behavior.
However, there was the likelihood that warmth was con
founded with the modeling of helping behavior.
Other experimental studies have had mixed results,
depending upon the kind of helping behavior being assessed.
Model warmth has been found to have no effect upon chil
dren' s willingness to imitate a model who donated to needy
29
others, and in fact, seemed to retard such imitation
(Grusec, 1971; Grusec 6c Skubiski, 1970; Rosenhan 6c White,
1 96 7 ) . One explanation for this lack of effect for model
warmth in connection with donation is that children might
find donating aversive, In these circumstances a warm
model might free the child from having to imitate that
unpleasant behavior. This hypothesis would be consistent
with studies involving model warmth and other kinds of
aversive behavior such as self-denial (Bandura, Grusec 6c
Menlove, 1968) and resistance to temptation (Parke, 1 967).
By contrast, the warmth of the model has been shown to
increase a child's willingness to rescue a peer is distress
(Staub, 1971a). In explaining this finding, Staub has
suggested that nurturance in the rescue situation may have
had a different function than in the donation and sharing
conditions. Nurturance might have indicated in the res
cuing case that the adult was not going to punish the child
for his helping behavior, i.e., leaving the room to help a
child in the next room. Staub (1970), in his earlier
research, had suggested that children might fear disapproval
for initiating an action that is clearly not permissible,
and so might not help without explicit permission.
Another hypothesis is that in a situation where a peer
is in distress, it is likely that the child will experience
distress as well through empathy (Aronfreed 6c Paskal, 1968;
Yarrow 6e Waxier,.1976). One of the means of reducing such
30
distress is by aiding the victim, thus terminating the
aversive stimulation. If rescuing means that a child must
engage in actions he believes might be associated with
negative consequences from others, a warm model may serve
to reduce the child's fears about such negative conse
quences .
Investigations into the role of the model's affect in
evoking helping behavior have generally been based on
Aronfreed's (1968) position that positive affect is a
necessary condition for an altruistic act and that helping
behavior is the result of a contiguity conditioning pro
cedure. Aronfreed suggested that if the observing child
experiences a change in affect which is increasingly pos
itive and if this change is associated in time with the
donation act, the affect will become attached to the act.
Bryan (1977) also has pointed out that if the child observes
someone who helps another and the model expresses strong
positive affect in association with the helping act, the
child will then connect his own vicariously experienced
positive affect with the donation behavior.
Bryan (1971) and Midlarsky and Bryan (1972) explored
the role of vicarious affect and found that contingent
affect expressions served to increase the child's imitative
helping behavior. Additionally, Bryan (1971) demonstrated
that minor temporal variations between the model’s expres
sion of positive affect and his donating behavior influenced
31
the child's generosity. The children were more likely to
donate if the model immediately expressed a positive
feeling after his donating response rather than when this
expression was delayed. Bryan (1977) has suggested that
the results of these studies indicate that models who are
trying it and at the same time liking it are likely to be
more influential upon the observing child than those who
are simply trying it.
While the power of the model has been demonstrated to
be an important determinant of imitation (Bandura, 1969),
not all studies have consistently found that the model's
power improved helping behavior, The conflicting findings
appear due to the manner in which the power of the model was
manipulated. When power has been directly tied to the
control and dispensations of reinforcement, the model is
likely to be imitated, otherwise not. This relationship
has been demonstrated empirically by Grusec (1971) who
found that seven- and eight-year-old children were more
likely to imitate an adult male who would be selecting a
child from the school for a special prize. It has also
been shown by Hartup and Coates (1967), who found that the
effectiveness of the child model depended upon the degree
to which the child rewarded other children within the
classroom, and the degree to which the observing child
received rewards from his classmates,
On the other hand, Bryan and Walbek (1970a) failed to
demonstrate the effect of power when the basis of the power
was the experimenter's control over the laboratory but not
over reinforcements. It is quite likely that children, in
their daily lives, interact with others who are not
entirely consistent in their verbal behavior and their
behavioral practices which are related to moral acts.
Common sense would suggest that for many individuals,
preaching morality is far easier than carrying out a moral
act, and for that reason children will undoubtedly be
exposed to many instances of moral inconsistency. Indeed,
children are probably often confronted by hypocritical
behaviors on the part of their parents, peers, and teachers.
What effect does this hypocrisy have upon the observ
ing child? Surprisingly, there is little evidence to indi
cate that the hypocrisy of a model has much effect on the
child's willingness to help. In those studies where a
child has been exposed to a hypocritical model, children
have not donated less to a needy other nor have they been
less likely to admire the hypocritical models than the non-
hypocritical model (Bryan, Redfield 6c Mader, 1971;
Midlarsky Sc Bryan, 1972; Rushton, 1975). One interesting
finding, however, in an experiment by Midlarsky, Bryan, and
Brickman (1973) with second- through fifth-grade children
was that when a model exhorted charity but failed to be
charitable, his social approval following the child's
33
donation acts served not to increase, but to lessen the
child's subsequent donations. Apparently, social approval
from one who demonstrates inconsistencies between his words
and acts and who behaves differently from the expectations
he has set for the child will become aversive to the child,
who then does not value nor want the model's approval.
Studies concerned with children's views of a helping
or nonhelping model have been rare. In those that have
been carried out, the child has been asked to rate the model
on such dimensions as niceness or goodness. Findings have
shown that the charitable model in either actions or words
has been consistently evaluated more favorably than the
selfish one (Bryan 6c Walbek, 1970a, 1970b; Rushton, 1975;
Schwartz 6c Bryan, 1971a). Additionally, these verbal
ratings have correlated with a behavioral measure of
attraction, i.e., the child chose a balloon with the
model's name on it (Schwartz 6c Bryan, 1971a).
Other aspects that have been investigated in the area
of modeling effect are the role of reinforcement and the
child's practice of a prosocial behavior in the presence of
the model. For example, if a child has observed a model
being rewarded for his generous act, the child might expect
the same reward for his own generosity. This would
undoubtedly be expected to increase the probability of
imitative helping. Results do not totally support this
possibility. Harris (1970) failed to find that fourth- and
34
fifth-grade children were more likely to imitate a generous
adult model who had received praise. Elliott and Vasta
(1970) found that only vicarious reinforcements containing
explanations of the relationship between giving and being
rewarded resulted in five-, six-, and seven-year-olds
imitating a peer. In both of these studies, donations were
anonymous. While Presbie and Coiteux (1971) found that
first-grade children were more likely to imitate a model
who either praised himself for his generous behavior or was
praised by the experimenter, the presence of the experi
menter while the children gave might have placed demands on
them to conform.
In regard to practice, two studies (Rosenhan & White,
1967; White, 1972) have demonstrated that the "rehearsal"
of sharing behavior in the presence of the model resulted
in subsequent sharing. An interpretation for this effect
might be that the child assumes that the model approves of
his behavior, and he is therefore more likely to repeat it.
Additionally, Staub (1971c) found that kindergarten children
who engaged in role playing later helped more in response
to sounds of distress from the next room than those who had
had the consequences of their behavior pointed out to them
by the model.
Not only have modeling effects been demonstrated with
live models, they have also been found when the models were
on videotape or film as well (Bryan & Walbek, 1970^;
35
Rushton & Owen, 1975). In the Bryan and Walbek study,
subjects watched a videotaped model practice and preach
generosity or selfishness. Results were that the model's
acts led to more helping behavior than did the model's
words. Rushton 6c Owens’s study showed that exposing chil
dren to a TV model's generous or selfish behavior affected
those children's subsequent generosity. The effect,
however, did not endure on a two-week retest.
To summarize, it is evident that the actions of
models, either peers or adults and live or filmed, influ
ence the altruistic behavior of children, and that the
effects of the model persist. Further, the powerful model
who has some control over reinforcement is likely to affect
altruism more than one lacking such power, and the model
who exhibits a positive affect and who is rewarded for his
generous behavior will also facilitate helpfulness. Some
forms of helping, but not all, appear to be facilitated by
a warm or friendly model. At this point, however, it does
not seem clear which influences a child more, the model's
words or his or her deeds.
Affect and Helping
The relationship between affect and helping behavior
in children has been investigated by manipulating the moods
of subjects in a number of ways, One type of study has
induced a positive affect by contriving a situation in
36
which the child has a success experience of a task (Isen,
Rosenhan & Horn, 1973) . Another type of experiment has
manipulated moods by having the child recall some past
event which either produced considerable sadness or happi
ness for him (Moore, Underwood & Rosenhan, 1973). Finally,
there have been attempts to associate children's moods and
actions through conditioning procedures (Aronfreed & Paskal,
1968; Midlarsky & Bryan, 1972).
The influence of moods upon children’s helping seems
to be dependent upon how and when they have been induced
and the particular measure of helping employed, and have
thus led to inconsistent findings. For example, Moore et
al. (1973) found that seven- and eight-year-old children
were more likely to donate money after they were asked to
recall events that made them happy. The authors of this
study suggested that people who are experiencing a positive
affect tend to be kind to themselves and to others in a
variety of ways, such as rewarding themselves and others;
and even being more sensitive to words that are consistent
with their positive state.
On the other hand, negative affective states have not
been found to depress donations. In the study by Isen,
Rosenhan, and Horn (1973), third- and fourth-grade children
who experienced failure on a task donated more than non
failure subjects, but only under conditions where their
contributions were going toward a charity that was linked
37
to the experimenter. The author’s explanation for this
effect was that the children may have wanted to repair
their self-image following failure. This desire may then
have served as a motive for their altruistic behavior.
Several studies have indicated that empathic respon
siveness may be critical to prosocial behavior. Experiments
by Aronfreed and Paskal (1968) and by Midlarsky and Bryan
(1972) investigated the effects of the empathic experience
in relationship to altruistic behavior, and their findings
appear to confirm the importance of empathy as a reward
mechanism which underlies prosocial actions. While there
may have been other influences such as the demand charac
teristics of the situation or the increased nurturance of
the model, evidence seems to be that conditioning children's
empathic responses facilitates helpfulness. In effect, if
the child learns through modeling or training that a
response which will help another individual and at the same
time create a pleasant effective state within himself, he
will be more likely to aid unfortunate peers than will
children without such training,
A suspected relationship between perceptions of com
petence and altruism has been tested empirically, however,
at present the research evidence is meager and relates
almost exclusively to adults. In two of the few studies
that exist (Kazdin & Bryan, 1971; Midlarsky, 1968), the
cost to the benefactor has high, namely volunteering to
38
give blood or taking on electric shocks for another person.
In both studies, findings were that when subjects were told
they were competent on either relevant or irrelevant tasks,
their offerings of help were significantly higher than were
those subjects who were told they were incompetent. How
ever, at a later time, the subjects did not follow through
with their offers of help in terms of actual behavior.
The following explanations have been offered about a
suggested relationship between competence and helping.
Midlarsky (1968) has suggested that individuals who see
themselves as having internal control over their environ
ment may view themselves as being more competent and thus
more likely to share with others. Form and Nosow (1958)
observed that those adults who had the greatest experience
in other disasters and thus possessed specific useful tech
nical skills that applied to immediate problems, helped
more in a natural disaster, Kazdin and Bryan (1971), in
explaining the results of their research, suggested that
individuals who felt competent and then took a risk which
they were rewarded for, might be more likely to participate
in other risk-taking situations in the future.
Other research involving adults, and one with children,
has not directly examined the effect of competence.
Instead, a success factor was manipulated (Isen, Rosenhan 6c
Horn, 1973; Berkowitz 6c Conner, 1966), and when successful
subjects were more helpful, this effect was explained with
39
the suggestion that success might imply competence and that
the feeling of competence itself might have promoted char-
itability. Only one study has been discovered that con
sidered the domain of competence directly as a variable in
research with children. This study (Staub, 1971a) explored
the influence of nurturance and modeling on kindergarten
children's attempts to help and included a number of
teachers' ratings of the subjects in the measurement.
One rating scale was referred to as "Competence on Tasks,"
but unfortunately a definition of this scale was not pro
vided, so it is not clear what its meaning and purpose were.
To summarize, there is increasing evidence that chil
dren's affective states have some influence on their
helping behavior. Positive affect appears to have a
facilitating effect, while a negative state seems to cause
the child to be less helpful unless he can repair his
lowered self-image in the presence of a valued person.
Cognitive and Developmental Factors
Recent research attention to children's helping
behavior has begun to examine the cognitive and develop
mental processes involved in helping. There is evidence
that amplifying cognition within an experimental context
increases altruistic behavior (Rosenhan, 1969). In
Rosenhan's study, children who were informed about the
consequences of being orphaned contributed more to the
40
orphans' fund than did those who were not informed. There
is also reason to believe that cognitive factors bear some
relationship to altruism since there have been relatively
consistent findings that children's helping behavior is
positively correlated with the increasing age of the child
(Barnett 6c Bryan, 1974; Harris, 1971; Midlarsky 6c Bryan,
1972; Rushton, 1975), and cognitive development increases
with age.
One conflicting finding in regard to cognitive devel
opment was demonstrated in a study by Severy and Davis
(1971) wherein advanced cognitive development was nega
tively associated with helping. These researchers found
that mentally retarded children between age eight and ten
were more likely to help a peer with a task than either
older normal children or younger retardates. Moreover,
retarded children were more likely to offer psychological
comfort than were intellectually normal children. Severy
and Davis's interpretation of these somewhat surprising
findings was that older normal children were inhibited from
helping by having learned the norms of achievement and
independence which encourage competition rather than
altruism.
Since the greater number of studies have suggested a
relationship between cognition and helping, interest has
begun to focus upon the understanding of the underlying
cognitive processes, Thus research topics have been
41
children's concerns about their deservedness, their ego
centrism and role-taking skills, and the nature of their
moral judgments in affecting helping.
Children's beliefs about whether or not they deserve a
reward have been shown to have some influence on their
willingness to help others. Research evidence has demon
strated that children have given more of their experimental
pay to other experimental subjects when they felt they had
been overpaid than when they felt they had been justly paid
or when they believed they had deserved more than the
actual pay given to them (Masters, 1971; Long 6c Lerner,
1974; Miller 6c Smith, 1 9 7 7 ) . In the Miller and Smith study,
subjects in the proper-payment and underpayment conditions
gave more to the needy when the needy were portrayed as not
being responsible for their misfortune than when they were
portrayed as being responsible for it and therefore
deserving of their misfortune, Students who were overpaid
gave equally to each group suggesting that they may have
experienced some feeling of discomfort about the inequity
of payment to each group.
In studying role-taking in relationship to helping
behavior, Rubin and Schneider (1973) found that seven-year-
old children who were able to take the perspective of
others were also more likely to donate candy to poor chil
dren and to help a younger child with a task than were
children less able to assume another's role. However,
42
Elmer and Rushton (1974) failed to find a relationship
between role-taking ability and generosity among children
seven to thirteen years of age. Thus, there is no clear
evidence at this time that there is a relationship between
egocentrism and role-taking abilities and prosocial
activity in children.
There appears to be some link between the maturity of
the child’s moral judgment and helping behavior. Rubin and
Schneider (1973) found that seven-year-old children with a
higher level of moral judgment as assessed by the children's
responses to a number of moral conflict stories were more
likely to donate candies and help another child with a task
than were children with less mature judgments. Support for
this view was provided also by Elmer and Rushton (1974) who
found that the level of the child's moral judgment was
associated with the child's willingness to donate prizes to
a charity to poor children.
The research evidence regarding the relationship
between, cognitive processes and helping behavior is meager
at present and excludes aspects such as cognitive style,
and the role memory might play in observational learning.
More investigation is needed in regard to cognitive and
developmental processes underlying children's willingness
to help.
43
Individual Differences
Among the subject variables which have been investi
gated are age, sex, social class, and personality factors.
Most investigators have found, in laboratory settings, that
older children are more likely to donate and rescue than
younger ones (Barnett & Bryan, 1974; Elliott & Vasta, 1970;
Rushton, 1975). Three explanations given for this fact are
that (1) older children are less egocentric (Rubin 6c
Schneider, 1973), (2) children adhere more to norms of
social responsibility and giving with age (Bryan & London,
1970), and (3) older children may donate more than younger
ones simply because the prize to be donated is of less
value to them (Bryan, 1975).
While the majority of research suggests that as chil
dren mature they are more altruistic, there have been some
interesting exceptions. Yarrow and Waxier (1976) found no
relationship between prosocial helpfulness and age with
children ages three to seven, while Staub (1970) found a
curvilinear relationship with age and rescuing behavior
with an increase between age four and nine and a sharp drop
by age eleven. Green and Schneider (1974) sought to deter
mine whether other forms of altruism besides sharing are
positively associated with age. Boys from four age groups
--five to six, seven to eight, nine to ten, and thirteen to
fourteen--were given opportunities to be helpful in three
44
different ways. On the sharing measure, results were con
sistent with previous research in that there was a pro
gression at each age level, with older boys sharing more.
On the picking-up-pencils measure, there was a developmen
tal trend as well. However, there were no age differences
on the volunteering-to-work index with over 90 percent of
the boys in each age group willing to sacrifice some of
their play time in order to work on a project that would
benefit poor children. A suggested explanation for the
lack of age differences on the latter measure was that it
might be due to an inability on the part of younger chil
dren to anticipate or understand the costs that they would
incur (giving up free time) as a result of their helpful
ness . The Severy and Davis (1971) study earlier described
also found, in a naturalistic setting, that both age and
relatively advanced cognitive development may retard, not
facilitate, helping.
By and large, differences between girls and boys have
not been found (Grusec, 1971; Harris, 1970), or are found
to be complex and determined by variations in experimental
procedures (Grusec & Skubiski, 1970; Rosenhan & White,
1967). Rushton (1976) has reported that eleven of fifteen
recent studies found no sex differences in altruistic
behavior, and the four remaining studies either found mixed
results or a slight tendency for girls to be more altru
istic than boys on some of the measures. Thus, the belief
45
that girls are more altruistic than boys is not supported
in the literature.
The economic status of the child or the child's family
appears to bear little relation to helping behavior by
children (De Palma, 1974), However, this lack of corre
lation may be due more to the manner in which social class
was defined since Berkowitz and Friedman (1967) found in
their study of adolescents that when they divided middle
class into service-oriented and entrepreneurial categories,
children from the former group were more likely to render
aid to another than those from the latter.
Personality factors have received only passing
interest in the literature on children's helping. Inves
tigations into personality correlates of prosocial behavior
have generally focused upon one of two approaches. The
first kind of study has concerned itself with the general
question of individual consistency in helping, i.e., if the
child is helpful in one fashion, is he or she likely to be
helpful in other ways as well (Baumrind, 1971; Yarrow 6c
Waxier, 1976)? The second approach has concentrated upon
isolating particular personality characteristics which
might be associated with various forms of aiding (Long &
Lerner, 1974; Midlarsky 6e Bryan, 1972). Questions investi
gating this aspect have related to factors such as
children's need for social approval (Midlarsky, Bryan 6c
Brickman, 1973), social responsibility (Midlarsky 6c Bryan,
1972), and more recently, assertiveness (Yarrow & Waxier,
1976).
Studies which have dealt with personality factors have
differed from most research about helping behavior in that
they have been concerned with correlations rather than
experimentally induced relationships. Personality vari
ables have been typically defined by using three different
sets of criteria for measurement, which are the ratings of
others or self-ratings, scores on paper-and-pencil tests,
or behavioral measures.
Considering individual consistency first, four
naturalistic studies have asked the question, Is the child
who is helpful in one way likely to be helpful in others?
Their findings, described below, suggest that different
types of helping and consideration for others are posi
tively related. However, the kinds of measurement used
would indicate that some caution should be used in inter
preting the results.
Friedrich and Stein (1975) found that cooperative and
nurturant behaviors were correlated in their observations
of preschool boys, Dlugokinski and Firestone (1973), in
their study of fifth- to eighth-grade children, found that
children's kindness, as indexed by self-report measures,
peer ratings, and a behavioral measure involving donations
showed small but positive correlations. Baumrind (1971)
rated preschool children on sympathy and nurturance to
47
others, helping, selfishness, thoughtlessness, and
insulting behavior. All these behaviors were found to be
related in such a way as to suggest an underlying dispo
sition of consideration to others, Yarrow and Waxier (1976)
found that sharing and comforting were correlated posi
tively with each other but neither was associated with
helping as measured by the child's willingness to help an
adult who had spilled some materials,
It would seem that children who are helpful in one way
are likely to be helpful in others, and that children can
be reliably differentiated on the general dimension of
consideration for others. However, the amount of research
which has been carried out in this area is small and the
correlations are modest which suggests that caution should
be used in making general assumptions.
The evidence concerning individual personality charac
teristics is also based upon only a few studies where there
has been some doubt about the validity of the measures used.
For example, the Social Responsibility Scale (Harris, 1957),
which was designed as a measure of altruism, has been found
to be highly correlated with measures of social desirabil
ity. Therefore, the finding that people who give socially
desirable answers receive high altruism scores casts some
doubt on the validity of the Social Responsibility Scale.
Midlarsky and Bryan (1972), using Harris's scale of
Social Responsibility, found small correlations between
48
donations and social responsibility in both an immediate
testing situation involving the donation of money and a
later testing session involving candies. However, as
indicated above, the Social Responsibility Scale may not be
a valid measure, In the same study, social desirability
and trust were not found to correlate significantly with
children’s donations.
Long and Lerner (1974) found that children with high
scores on a delay-of-gratification test donated more
undeserved rewards than children who had lower scores on
delay of gratification. The relationship of delay of gra
tification to donations was also found to depend upon
whether the obtained rewards were deserved and whether or
not someone else would know of the donation.
While studies have been few, it has not been found
that children high in the need to gain approval donate more
or tell others that they have donated than children with a
lower need (Staub & Sherk, 1970).
Yarrow and Waxier (1976) found that young children who
were assertive were more likely to render aid than those
who showed high aggression or hostility or those who showed
little aggression. Barrett and Yarrow (1977) also found
that for both boys and girls ages five to eight, assertive
ness was positively and significantly related to prosocial
behavior.
49
A survey of the literature has indicated that there is
a paucity of evidence in regard to the relationship between
helping behavior and personality. Much more needs to be
learned about the influence of personality on helping
behavior.
Summary
The picture which emerges from a review of the liter
ature concerning prosocial behavior is that helping
behavior can be elicited and developed, and that a number
of antecedent variables have been identified. However, the
concept of helping behavior has been defined somewhat
narrowly, with the definition generally operationalized in
order to fit the research design. Typically, helping
behavior has meant rescuing a child or an adult in distress,
or either sharing with or donating prized objects to needy
others.
Early researchers often tended to be more interested
in the self-sacrificing aspect of helping behavior, while
later research did not always include the element of
sacrifice.
Among the variables that have been associated with
positive social behavior are the effect of models, affect,
cognitive-developmental factors, and individual differences.
Helping behavior has been demonstrated to be heavily
influenced by models, both adult and child, either live or
________________________________________________________________ 50
on television. There can be little doubt that children
will engage in helping behavior if they see a model who
helps in some way. It appears, too, that both a warm and
an indifferent model will increase helping behavior,
depending upon the kind of help that is expected from the
child. A warm model has not been of influence when the
expected help might be aversive, but has had an impact if
the helping behavior has not meant great cost to the
helper. Further, a powerful model who is able to dispense
reinforcement will be able to influence a child to help, as
will a model who expresses enjoyment after having helped.
While the impact of a model's verbalizations versus
physical demonstrations of help have been investigated, the
results are not clear cut. At times a model's verbal
messages produce more prosocial behavior than the model's
deeds, and at other times the opposite is true. Expla
nations offered for these differences have related to the
kinds of directions given and to sex differences, with
girls more likely than boys to help if verbal directions
are used.
Hypocritical models do not increase or decrease a
child's helping behavior; however, the child's desire for
approval from that model diminishes.
A child's positive mood, stemming from a successful
experience, will lead to increased helping behavior. On
the other hand, when a child wants to improve his
_________ 51
self-image after having failed in some way, his negative
mood will also result in increased altruism, if an indi
vidual important to him is there observing when he has the
opportunity to help.
Older children and girls are more likely than boys or
younger children to help. As yet, social class does not
appear to be related to helping behavior; however, this
lack of effect may be due simply to the methods employed
in the investigations.
Children who believe they have been given something
they do not deserve are more likely to share with others,
as are children whose moral judgments are more mature. The
relationship between egocentrism and helping behavior has
not been clearly established.
In regard to personality and helping behavior, there
is little evidence about state or trait factors, although
there are suggested trends that children who are kind and
considerate in one way may be so in others. Individual
personality characteristics such as feelings of social
responsibility and trust for others have not been shown to
relate to helping; but the ability to delay gratification
and to be assertive appear to correlate with prosocial
behavior.
More research in prosocial behavior has focused upon
the elicitation of helpfulness rather than its development.
52
However, there is some evidence to suggest that children
can be trained to be more helpful.
There is extensive research into the area of prosocial
behavior, based for the most part upon experimental studies
which have examined the role of the model. Most of the
research has been carried out by developmental psychol
ogists, while in an educational context studies are
virtually nonexistent. Further, despite the fact that many
experiemnts have been conducted with school-age children,
the kinds of behaviors elicited from them have been those
which children in school are unlikely to have many oppor
tunities to carry out. For example, rescuing a child in
distress or donating money to orphans has been heavily
emphasized, while studies of helping behavior related to
school-like tasks have been rare.
53
CHAPTER III
METHODOLOGY
Introduction
This chapter details the methodology employed in the
study and outlines the research design used. Discussion is
presented in the following sequence: (a) subjects,
(b) research design, (c) materials, (d) dependent measure,
(e) procedures, (f) statistical analysis, (g) method
ological assumptions, and (f) limitations of the study.
Subj ects
Subjects participating in the study were 163 white,
English-speaking, fourth-grade students (81 boys and 82
girls) for whom parental permission was obtained. All
children were randomly selected from four middle-class,
suburban elementary schools in the Conejo Valley Unified
School District in Ventura County. A statified sampling
procedure was employed in order to acquire both boys and
girls in each group. Anticipating some attrition, more
subjects were selected for the experimental groups than the
control groups so as to ensure an adequate sample for
analysis. Ages of the children ranged from 8.10 years to
9.10 in October 1980 when the study began,
34
Research Design
A 3 x 2 x 2 factorial design was employed, with the
factors being method of social influence (verbalization,
physical demonstration, no model-control), competence (high
versus low), and sex. In order to assess the durability of
the children's helping responses, each subject participated
in two experimental sessions which were separated by
approximately ten days.
The dependent variable was the amount of time subjects
were willing to volunteer in order to tutor an unknown
student in arithmetic skills. The outcome measure was
presented after the independent variables were given,
following the posttest-only control group design described
by Campbell and Stanley (1966, p, 25). This procedure was
considered suitable in order to control for the reactive
effect of testing in which subjects' sensitivity to the
experimental variable might have been increased if there
were a pretest,
Materials
Social Influence Videotapes
Videotaped models were utilized in order to assure
standardized presentations of the models. A description of
the videotapes follows. Two experimental videotapes
(verbalization and physical demonstration) and one control
videotape of short duration (all were under four minutes in
________________________________________________________________55
length) were designed and produced for the present study.
Equipment used to film the sequences included a Sony black-
and-white camera Model AVC 3200, Concord videotape recorder
Model 820, and Concord monitor MR 900. Two elementary
school teachers viewed the completed films for suitability
of content.
Actors in the videotapes were two adult females, one
fourth-grade girl, and one fourth-grade boy, all of whom
portrayed a similar role in each sequence of the three
videotapes. All of the actors live in a community distant
to that of the subjects and were unknown to them.
Each of the two experimental tapes depicted three
incidents in which an individual, either adult or child,
was in need of help from another person. The three scenes
portrayed (1) a boy standing next to his bicycle and trying
without success to pump up a flat tire, (2) an adult female
walking along the street with an oversupply of packages
which she kept dropping, and (3) a girl seated at a desk
trying and failing to complete an arithmetic homework
assignment. In each sequence, the actor in need of assis
tance did not indicate, when help was offered, any affec
tive response nor thank the benefactor in order to avoid
the confounding effects of the variables of affect and
reinforcement. For the same reason, the actor who por
trayed the benefactor avoided any display of feeling.
The three videotapes were varied in the following
manner:
_____________ 56
Direct verbal instruction tape. In this tape, each of
the scenes described above was filmed as follows. The
adult model verbalized twice upon seeing the individual in
need of help (to equalize the amount of time shown in this
tape and in the physical demonstration tape) to the boy or
girl model who accompanied her as she walked along the
street. The statements were:
Scene 1: "John has a flat tire. I would like you to
help him fix it."
Scene 2: "Mrs. Smith has too many packages to carry.
I would like you to help her."
Scene 3: "Nancy is having trouble with her homework.
I would like you to help her with it."
Each scene ended after the above directions were given so
that no physical demonstration of help was seen.
Physical demonstration tape. In this tape, the adult
female indicated to the child model walking with her that
(1) John had a flat tire, (2) Mrs. Smith had too many
packages to carry, and (3) Nancy couldn't do her homework
assignment. The adult model made no further comment;
however, after hearing these comments and seeing the indi
viduals in need of help, the accompanying child then
provided the following assistance:
Scene 1: The girl helped pump up the bicycle tire for
John.
Scene 2: The boy helped Mrs, Smith pick up her
packages.
Scene 3: The boy helped Nancy with her arithmetic
problems.
No model-control tape. In this tape, the same actors
as above played in three scenes. The scenes included a boy
with a bicycle, a woman carrying packages, and a girl doing
her arithmetic assignment. In none of the scenes, however,
was a need for help depicted, with the actors merely making
small talk together.
In all three videotapes, an introductory message was
given by an adult female actor. In this introduction, the
watching children were asked to pay careful attention to
the film because after seeing it they were going to be
asked some questions about it at a later time. At the end
of each tape, the adult female again reminded the children
to remember what they had seen,
Competence
The competence manipulation was determined by an easy
(high competence) or difficult (low competence) page of
arithmetic problems. This page was shown to subjects prior
to their choosing whether or not to volunteer to tutor
another student in arithmetic. Subjects in the "low"
competence condition were shown a page of problems similar
to those that would be found in an eighth-grade arithmetic
book (Appendix A). In the "high" .competence condition,
subjects were shown a page of problems that would be
expected in a first-grade arithmetic book (Appendix A).
Those two grade levels were selected after having examined
the norm tables for the Wide Range Achievement Test (Jastak
& Jastak, 1976) to determine the level of arithmetic
problems that would be considered easy or difficult for
fourth-grade students, The norms indicated that only one
percent of students in the age range of subjects in the
present study would be able to complete problems at the
eighth-grade level, while ninety-seven percent of the same-
age students could be expected to complete problems at the
first-grade level. These two grade levels, therefore, were
chosen to be of a sufficient ease and difficulty level to
be used for the competence manipulation. A pilot study
later confirmed this choice as being an appropriate one.
Dependent Measure
The outcome measure of helping used in the present
study was the amount of time subjects were willing to
volunteer to tutor, at some time in the future, an unknown
child. This anonymous child was to be given help in
arithmetic. The measure is a modification of one of those
used in the Green and Schneider (1974) study. In that
experiment, boys were asked if they would be willing to
volunteer their time to work on a project to help poor
children. If the subjects volunteered, they were told that
they could choose to work on one, two, three, four, or five
days. In the present investigation, in order to eliminate
_________________________________________________________________________ 59
"demand characteristics," the experimenter did not ask
subjects whether or not they wanted to volunteer during an
individual interview. Rather, subjects made their choice
without an onlooker. They privately circled the number of
days they were willing to volunteer on a piece of paper
labeled "Number of Days I Will Tutor" (see Appendix). This
paper showed the days of the week in both written and
graphic forms. A graph was utilized in addition to names
of the days of the week because it was expected that chil
dren of eight and nine would require a concrete represen
tation in order to help them understand an abstract time
concept. Since graphs are utilized in arithmetic programs
by at least third grade, this graphic presentation seemed
to be a suitable format.
Procedures
Initial Session
Subjects, who had been randomly assigned to one of the
two experimental social influence conditions or to the
control group, were sent by their teachers in groups of
approximately eight to another classroom in the school
where they viewed the videotapes. The children were met at
the door by a female experimenter, who, after making some
welcoming remarks, asked them to take seats. The experi
menter then showed one of the videotapes described above.
Following the viewing, students were asked not to talk about
60
the contents of the film with their classmates and then
asked to return to their classrooms. Subjects were also
told that someone at a later time during the day would be
taking them from their classroom in order to ask them
questions about the film.
Subsequent to the viewing, subjects were sent indi
vidually from their classrooms to another room where they
were greeted by one of two female experimenters. In order
to control for experimenter bias, neither of these two
experimenters had knowledge of the nature of the videotapes,
nor the method of social influence conditions to which the
subjects had been assigned. When the subjects came to the
room, the experimenter greeted them and then asked each one
for a color-coded card with the student's name on it. This
card indicated the competence condition to which the sub
ject had been randomly assigned. Following these initial
procedures, the experimenter conducted a standardized
interview with each subject, The interview was designed
in part to check the subjects' attention level and their
ability to understand the nature of the film. From the
answers given by students, it was apparent that they had
watched the film closely and remembered what they had seen.
The interview script was as follows:
"I'm going to ask you some questions about the film
you saw today. Did you like the film? Can you tell me
61
why? Can you tell me what happened in the film? Do you
think other boys and girls would enjoy it?"
Following the attention check, the experimenter said:
"Now I want to ask you about something else. Some
times boys and girls at school help other students who are
having trouble with their school work. I would like to get
an idea of how many boys and girls at this school would
like to help other boys and girls by tutoring them in
arithmetic."
At this point, the experimenter presented either the
"high" or "low" competence arithmetic paper described above
and said,
"Here is the kind of work students might need help
in."
The experimenter then showed the "Days I Will Tutor"
paper described above, explained how to complete it, and
made sure the subject understood what to do. When the
subject indicated that he or she knew what to do, the
experimenter pointed to a box on a table across the room
and said,
"Don't put your name on the paper. I don't need to
know the names of boys and girls who might want to help,
just how many at this school who are interested in helping.
There are some pencils there on the table for you to use in
circling the number of days you will help. After you do
that, fold the paper and put it in the box."
62
These procedures guaranteed that the subjects would
make their choice to help anonymously. While, as Kazdin
and Bryan (1971) have pointed out, volunteering is not
doing, and that equating volunteering with aiding should be
made with caution, subjects were not asked to tutor another
student directly in this study but rather to volunteer to
do so at some time in the future. This procedure was con
sidered important in order to avoid the confounding vari
ables regarding the recipient of help such as attractive
ness of appearance or manner, affect, and sex.
Delayed Measure
Approximately ten days after the initial measure was
taken, subjects were again sent from their classrooms one
at a time to another room in the school. Here, one of the
two experimenters greeted them and asked each subject to
fill out another dependent measure. While subjects were
not spontaneously given the reason for having to do this
again, if they asked, they were told that the experimenter
wanted them to and that it was important. Those few who
asked about the reason for the second measure seemed satis
fied with the answer given them.
Before the subjects made their choice, the experi
menter reviewed the paper with them, explaining again how
to circle their choice. As before, the subject made the
decision anonymously. After this second session, all
63
subjects were debriefed and thanked for their part in the
experiment as were their teachers.
Statistical Analysis
In order to test the research hypotheses, a 3 x 2 x 2
fixed-effect analysis of variance was conducted on the
immediate and delayed measures. Because of the unequal
cell frequencies, complete linear model analyses were
carried out using the Statistical Package for the Social
Sciences ANOVA (Option 9) computer program (Nie, Hull,
Jenkins, Steinbrenner 6c Bent, 1975). In this program, the
unequal effects of each factor are evaluated, controlling
for all other factors. Independent variables were method
of social influence (verbalization, physical demonstration,
no model-control), high versus low competence, and sex.
The dependent variable was the number of days (0 to 5) that
the subject volunteered to tutor. The significance level
was set at .05.
64
CHAPTER IV
RESULTS AND DISCUSSION
Analysis of the data is based upon only the 146 sub
jects who completed the investigation, since absences,
family mobility, or failure to complete the dependent
measure properly resulted in a loss of 18 students from the
original sample of 163 who began the study. Data will be
presented first for each of the outcome measures, immediate
and delayed. Following this presentation, interpretative
comments will be offered.
Results
Immediate Measure
The dependent variable consisted of the mean amount of
time that children volunteered to help, with zero being the
minimum, and five days the maximum. The mean number of
days, standard deviations, and number of subjects in each
cell are presented in Table 1.
An analysis of variance of the mean number of days was
carried out on the effects that method of social influence
and competence had on children's willingness to volunteer
to tutor another student. Results of this analysis are
given in Table 2,
65
TABLE 1
MEAN NUMBER OF DAYS CHILDREN VOLUNTEERED
ON IMMEDIATE MEASURE
Method of Social Influence
Group Verbalization Physical Demonstration No Model-Control Total
Boys Girls Boys Girls Boys Girls Boys Girls Total
High
Competence
n 15 10 17 17 7 6 39 33 72
M 2.47 3.20 2.82 3.82 3.85 4.00 2.87 3.66 3.23
SD 1.92 1.47 2.03 1.59 1.34 1.67 1.86 1.59 1.78
Low
Competence
n 18 15 11 16 7 7 36 38 74
M 3.00 2.53 2.45 3.25 2.85 2.57 2.80 2.84 2.82
SD 1.87 1.72 2.11 1.84 1.95 0.97 1.95 1.63 1.78
Total
n 33 25 28 33 14 13 75 71 146
M 2.75 2.80 2.67 3.54 3.35 3.23 2.84 3.22 3.02
SD 1.89 1.62 2.06 1.71 1.64 1.29 1.90 1.60 1.79
CT\
TABLE 2
ANALYSIS OF VARIANCE ON IMMEDIATE MEASURE
Source of Variation
Sum of
Squares df
Mean
Square F
P
Method 5.39 2 2.70 0.84 0.43
Competence 10.63 1 10.63 3.32 0.07
Sex 3.19 1 3.19 0.99 0.32
Method x competence 5.97 2 2.98 0.93 0.40
Method x sex 6.13 2 3.06 0.96 0.39
Competence x sex 2.91 1 2.91 0.91 0.34
Method x competence x sex 1.86 2 0.93 0.29 0.75
Explained 36.73 11 3.34 1.04 0.41
Residual 429.16 134 3.20
Total 465.88 145 3.21
C F '
It is evident upon examining Table 2 that there was no
main effect for method of social influence, F(2,..134) =
0.842, £ > .05. Therefore, contrary to prediction in
Hypothesis 1, a model’s verbalizations did not result in
significantly more volunteering to help than did a model's
physical demonstration of help. There was no significant
main effect for competence, although the difference between
the high and low competence groups approached significance,
F(l, 134) = 3.318, £ < .08, providing tentative support for
Hypothesis 2. In line with the prediction, there was a
tendency for students with high competence to help more
than students with low competence. There was no inter- .
action between method of social influence and competence,
F(2, 134) = 0.981, £ > .05.
Although the hypotheses examined in this investigation
did not include any which dealt with effect of sex, F
ratios were generated for this factor as well as a by
product of data analysis. As can be seen in Table 2, there
was no significant main effect of sex on amount of volun
teering, F(l, 134) = 0.996, £ > .05, nor were there inter
action effects between sex and method of influence,
F(2, 134) = 0.957, £ > .05, nor between sex and competence,
F(1, 134) = 0.908, £ > .05,
Delayed Measure
In order to determine the durability of children's
choices, a Pearson's r correlation coefficient was computed
68
between the scores on the initial measure and the scores on
the second measure which was administered approximately ten
days later. A strong positive relationship (r = 0.73,
£ < .001) was found, suggesting that children's choices
persisted from one session to the next.
Again, the dependent measure was the mean number of
days of volunteering to help. Mean scores for the number
of days, standard deviations, and number of subjects in
each cell are given for the delayed measure in Table 3.
An analysis of variance was carried out on the delayed
measure to determine the effects of method of social influ
ence and competence on children's willingness to volunteer
to tutor another child. The results of this analysis are
presented in Table 4. It is apparent that, as on the ini
tial measure, there was no main effect for method of social
influence on the delayed measure, F(2, 134) = 0.460,
£ > .05. Neither was there a significant effect for com
petence, F(l, 134) = 1.746, £ > .05. As on the initial
measure there were no interaction effects between method of
influence and competence, F(2, 134) = 0.205, £ > .05.
Again, there was no main effect for sex, F(l, 134) = 0.006,
£ > .05, nor were there interaction effects between sex and
method of social influence, F(2, 134) = 0.861, £ > ,05, nor
sex and competence, F(l, 134) = 1.350, £ > .05.
69
TABLE 3
MEAN NUMBER OF DAYS CHILDREN VOLUNTEERED
ON DELAYED MEASURE
Method of Social Influence
Group Verbalization Physical Demonstration No Model-Control Total
Boys Girls Boys Girls Boys Girls Boys Girls Total
High
Competence
n
M
SD
15
3.40
1.84
10
3.80
1.62
17
3.11
1.93
17
3.88
1.76
7
3.42
1.61
6
3.50
1.76
39
3.28
1.83
33
3.79
1.71
72
3.51
1.75
Low
Competence
n
M
SD
18
3.72
1.74
15
3.07
2.09
11
2.63
2.20
16
3.06
1.98
7
3.43
2.07
7
2.57
0.97
36
3.33
1.94
38
2.97
1.83
74
3.14
1.89
Total
n
M
SD
33
3.57
1.78
25
3.36
1.90
28
2.92
2.03
33
3.47
1.86
14
3.42
1.84
13
3.00
1.33
75
3.31
1.88
71
3.35
1.77
146
3.33
1.83
"■vl
o
TABLE 4
ANALYSIS OF VARIANCE ON DELAYED MEASURE
Source of Variation
Sum of
Squares df
Mean
Square F
P
Method 3,19 2 1.59 0.46 0.63
Competence 6.04 1 6.04 1.75 0.19
Sex 0.02 1 0.02 0.01 0.94
Method x competence 1.42 2 0.71 0.20 0.81
Method x sex 4.67 1 4.67 1.35 0.25
Method ,x„competence x sex 1.00 2 0.50 0.14 0.87
Explained 22.82 11 2.07 0.60 0.83
Residual 463.39 134 3.46
Total 486.21 145 3.35
Interpretation
Main Effect for Method
of Social Influence
It was hypothesized in this study that a model's
verbalization would be more effective in eliciting helpful
ness from children than would a model's physical demon
stration of help. This prediction was not confirmed by the
data. Although both methods of social influence resulted
in subsequent offers of help from the subjects, neither
method proved to be more effective than the other. This
finding was not totally surprising since prior research had
provided no clear evidence that either method was superior.
In a number of studies (Elliott 6c Vasta, 1970; Friedrich 6c
Stein, 1975; Midlarsky 6c Bryan, 1972; Rice 6c Grusec, 1975)
verbalizations were more effective in producing subsequent
helpfulness; while, on the other hand, an almost equal
number of studies (Bryan 6c Walbek, 1970a, 1970b; Grusec 6e
Skubiski, 1970; Rushton 6c Owen, 1975) found that a behav
ioral exemplar was more successful. The results of the
present study suggest that either method would be capable
of eliciting helping responses from children. A cautionary
note must be offered, however, before this interpretation
can be considered. This note of caution is due to the
somewhat surprising outcome that the subjects in the
experimental groups did not differ significantly from the
control group. Although this finding would suggest that
72
one could interpret the results to mean that neither verbal
models nor behavioral exemplars do affect children's
helping responses, there is an abundance of evidence, as
earlier cited, to the contrary. Indeed, models have had
considerable effect on the subsequent altruistic behavior
of children. The interpretation, therefore, that models
are not effective does not then seem well-founded. What
would seem more plausible is that there were other vari
ables which diluted the modeling effect in some manner.
Possible explanations for this suggested reduction in
strength follow,
Two aspects that may be linked to a possible dilution
of the modeling effect are (1) the nature of the modeling
procedures employed in the study, and (2) the issue of
measurement. Each of these will be discussed in turn, with
a number of considerations offered for each.
First, although the modeling procedures used in the
present investigation were based on those utilized in other
investigations, in retrospect, there were a number of
limitations that should be eliminated in future research.
In the verbalization condition, the adult gave a very
clear direct message to the child, stating in no uncertain
terms that she wanted the child to help the individual in
need of help. While the message might have seemed a
request to some-children, others might have perceived the
words as a demand. If so, this direct message could well
73
have activated an oppositional tendency on the part of the
observers in that they then volunteered to tutor for as few
days as possible without actually refusing to volunteer.
Staub (1971c), drawing on Brehm's (1966) notion of reac
tance, offered this explanation when, in his study of
induction's influence on helping, it was found that induc
tion resulted in a decrease in help. Staub postulated that
subjects may have perceived the verbal message as pressure
to behave according to the experimenter’s suggestion. This
supposed pressure placed upon the subjects to help may have
had exactly the opposite influence in that subjects decided
to provide less help. In the present study, too, subjects
may have felt pressure through the vicarious experience of
watching the videotaped model give a direct message to the
child, and then have resisted themselves when they were
expected to volunteer. If directions have been less direct,
i.e., "You may want to help John," this suspected opposi
tional tendency could well have been minimized.
A related plausible explanation to the one offered
above concerns a procedural limitation in the verbalization
videotape. In order not to confound verbalization with a
behavioral example, the child model portrayed in the verbal
ization tape took no action after the adult model said,
"I would like you to help." Although the scene ended
shortly after the adult's statement, the child was seen for
a few seconds simply standing still rather than complying
74
with the adult’s request. Because of this inaction in the
videotape, subjects may have interpreted the child's
response as resistance and may then have modeled their own
behavior after it. If the camera had been shifted in some
way so that the child was not seen after the adult made her
request, this interaction would not have been viewed.
A third possibility deals with the physical demon
stration of help. In the physical demonstration videotape,
the model’s behavioral example may not have provided enough
information about expected behavior. Some researchers
(White, 1972) have suggested that a behavioral example
provides merely an ambiguous model to use as a guide and is
not an efficient way of informing subjects about how they
are expected to behave under similar circumstances.
Extending this possibility to the present study, it might
well be that subjects who viewed the behavioral example may
have found that the physical demonstration of help failed
to provide them with cues about what to do. Additionally,
subjects received no direction from the experimenter at the
time they anonymously made a decision about volunteering to
help. For this reason, subjects may well have been at a
loss as to what to do, and without clear standards of con
duct, some concern about social approval may have arisen.
This concern may have centered on the need for their
teacher's approval for leaving the classroom, since this
potential absence was implied in the activity of tutoring
75
another student. Subjects in this study did not know
whether or not their teacher would sanction this activity.
As Bryan (1975) has pointed out, it is reasonable to sup
pose that children who are concerned with social approval
are less likely to be helpful in those situations where
there exists any ambiguity regarding the acceptability of
behavior. Children in the present study who viewed an
ambiguous model and who had no clear expectations about
what to do may have been unwilling to risk social approval
in order to be helpful. This unwillingness may have
limited the amount of time they were willing to volunteer.
A fourth plausible explanation for the suggested
dilution of the modeling effect relates to the notion of
"demand characteristics," that is children will do what
they think is expected of them in an experimental situation
(Bryan, 1975). In the present study, an attempt was made
to control for this variable in that subjects made their
choice to help anonymously. This control was deemed impor
tant since a number of researchers had suspected that
modeling effects in experimental studies may have been due
only to these "demand characteristics" (Krebs, 1970).
Recent research (Israel & Brown, 1979) has provided some
support for this notion that the presence of an onlooker
had influenced subjects to be generous. In Israel and
Brown's investigation, when subjects were without a
model's surveillance, the modeling effect was reduced.
76
In the present study, modeling effects may well have been
reduced because there was no pressure placed on the sub
jects by a live model.
A fifth explanation for the lack of effect for a video
taped model may relate to the fact that the model was on
film. While prior research, as earlier cited, has demon
strated that videotaped models have led to helping
responses in children, a recent replication study by
Rushton and Owen (1975) showed a less durable and weaker
overall effect for a TV model than did the prior study by
Rushton (1975) which used a live model.
A final explanation related to the modeling procedures
employed concerns the kinds of verbal messages utilized in
the verbalization condition. This possibility was alluded
to earlier in a different context. It may well be that
simply asking children to carry out an act, without pro
viding reasons for engaging in that act, or without
including some form of reinforcement in the message, is not
appropriate to influence children to be helpful. However,
as Rushton (1976) has pointed out, there is little research
about verbal socializing techniques and, therefore, little
understanding about how to account for their influence.
Thus, lack of adequate information precludes further inter
pretation about the kind of verbal messages which would
have the greatest influence on helping, Research evidence
is greatly needed in this area.
77
In sum, a number of explanations have been offered as
to the impact the modeling procedures used in this investi
gation might have had on children’s willingness to volun
teer. Some or all of the procedural limitations could well
have interfered with the power of the modeling effect.
Future research which controls for these possible con
founding variables is needed.
A second interpretation of the results is related to
measurement problems encountered in operationalizing
altruism into behavior that would be likely to occur at
school. This was no easy task. As indicated by Rushton
(1976), there are no well-validated measures of altruism.
For this reason, most studies have relied on behavioral
outcome measures, which measures, as Rushton has further
pointed out, have not been found to correlate well with one
another (r = ,3). While the outcome measure developed for
this study was essentially a behavioral measure similar to
others that were used in previous research, it considered a
new, unexplored aspect of helping. While, in a pilot study,
the measure was found to discriminate, it was not a measure
of donations or sharing as were essentially all of those
used in prior research studying the effects of models on
helping. Instead, the measure used in the present study
was the amount of time children would be willing to tutor
another child in arithmetic. Because the kind of helping
assessed here was different from that in past research, the
78
assumption could be made that the modeling effect does not
generalize to other forms of altruism. On the other hand,
the assumption could also be made that the measure of
helping used in the present study was not sensitive enough
to assess the "true” effects of the independent variable.
In other words, the lack of effect may have been an arti
fact of measurement,
A related possibility in regard to measurement deals
with the amount of resources which subjects had available
to them. Sims (1978) demonstrated that the availability of
resources had an impact on at least the sharing aspect of
helping, in that children in her study who possessed few
candies donated proportionately less than those who had
more. In the present investigation, time rather than candy
was the available resource, and children's lack of avail
able time may have influenced their decisions on the out
come measure. Children's comments to the experimenter
during the interviews lent some support to this possibility.
A number of children expressed concern to the experimenter
about having to leave their classroom in order to tutor
another student. The expressed reason for their concern
was that they were involved in a number of extracurricular
school activities, i.e., chorus, band, special enrichment
classes, etc., which would limit the amount of time they
could devote to tutoring. The children who brought up this
subject seemed worried about a potential conflict and
______________ 79
expressed reluctance at volunteering for very many days.
This reluctance due to lack of time may well have caused a
number of subjects to reduce the amount of time they volun
teered to help. The cost of tutoring may thus have been
too high. This problem could probably have been remedied
if a pretest about time availability had been administered.
With both a pretest and a posttest, proportional amounts
could have been calculated, resulting in a more precise
measure of children's helpfulness.
While interpretations about the effect of measurement
have been offered, and suggestions made for refinement of
measures, the problem of measurement remains a serious one
in regard to children's altruistic behavior. It is a
problem that has not yet been solved as attested to by
previous research which has used the same measure almost
exclusively. It is mandatory that future research studies
address this problem.
Main Effect of Competence
It was hypothesized that high competence children
would help more than low competence children; however, this
prediction was not supported. While there was not a sig
nificant main effect for competence on helping, there was a
difference between the high and low competence groups which
approached significance. Those in the high competence
group had a tendency to help more than did those in the low
80
competence group. This finding would suggest that the
hypothesis in regard to competence needs further investi
gation. However, the near significant effect did not
endure as evidenced by the lack of effect on the delayed
measure. Interpreting this weak effect which did not
persist is somewhat difficult, particularly since there is
no prior research in this area about children with which to
compare the present findings.
One plausible explanation for the lack of clear
findings in regard to children’s feelings of competence may
well stem from the concept of competence, which was oper
ationalized in this study to mean academic competence.
High competence was a first-grade arithmetic paper, and low
competence was an eighth-grade paper, assumed to be,
respectively, easy or difficult for fourth-grade students
to understand, In using this format, it was assumed that
children would look at papers presented to them and make a
self-evaluation as to their level of academic competence to
teach another child those particular kinds of arithmetic
problems. It was further assumed that those subjects who
viewed the first-grade papers would judge themselves to be
competent, and that those who saw the eighth-grade papers
would make the self-appraisal that they were not competent.
This evaluative judgment was to be made before the subjects
made the decision to tutor, Thus, children's ability to
accurately appraise their arithmetic skills was an
81
important mediating factor between viewing the paper and
deciding to tutor. How students made this self-appraisal
may have been a key factor in their decision making. The
self-appraisal may not have been accurate enough to cause
perceptions of competence to be strong enough to matter.
In other words, children who were shown eighth-grade papers
may have had an inflated appraisal of their ability and
thus have chosen to tutor as much time as did those chil
dren who viewed the first-grade paper. What cognitive
processes were taking place are not clearly understood.
Little is known about how young children assess their
own competence; however, there is evidence that children do
use social comparison (Veroff, 1969) to accomplish this
appraisal. It is also known that by the age of seven or
eight, children are comparing themselves to their peers
(Pepitone, 1972; Ruble, Feldman 6c Boggiano, 1976). In
addition, Feldman and Ruble (1977) have found that age-
related differences in children's method of appraisal
depended on the particular situation in which social com
parison is needed. An important situational factor in
this process is the presence or absence of an objective
criterion. Having an objective criterion was an important
feature of Festinger's (1954) original formulation of
social comparison theory. According to Festinger, when
objective, nonsocial means are not available, people
82
evaluate their abilities by comparison with the abilities
of others,
In the present investigation, subjects had neither
objective nor social means with which to evaluate their
ability level in regard to the arithmetic papers presented
to them, This lack of criteria for evaluating their com
petence level, particularly in regard to the eighth-grade
paper, may have resulted in subjects disregarding this
paper when preparing to make a decision, This likelihood
would have been especially true if having to make a judg
ment meant some loss in self-esteem, which certainly was a
possibility for those subjects in the low competence group.
The importance of social comparison in the decision
making process was demonstrated by Jones and Regan (1974),
who found that people want information about their ability
level prior to making decisions because that critical
knowledge concerns not only the level of their ability, but
also what they can and cannot accomplish with that ability.
Thus, if individual A knows how B's ability level compares
with his own and also knows the successes and failures B
has experienced which are related to that ability, then A
is in a good position to estimate the consequences of his
own decision and action alternatives.
Since investigations of children's self-appraisal have
shown that children use social comparison by third grade,
it could be assumed that subjects in the present study
83
would use this method to help them judge their academic
competence. If this be true, they, then, would have
benefitted by knowing what their own referent group could
accomplish in arithmetic. If the experimenter, for example,
had said, in presenting the arithmetic paper to them, "This
is an eighth-grade arithmetic paper which is usually hard
for most fourth-graders," and likewise, "This is a first-
grade paper which is usually easy for fourth-graders,"
subjects would have gained both objective information and a
social comparison group. Having this information might
have led to a clearer appraisal of competence, which in
turn may have led to a stronger effect for this variable.
Other Findings
There were no significant differences between boys and
girls on either the immediate or the delayed measure.
These findings are clear and need little interpretation
since they are consistent with prior research which also
has found no differences (Rushton, 1976). Because this
aspect of prosocial behavior has been well demonstrated in
numerous investigations, there seems little purpose in
planning future studies about helping where sex is a
factor.
Persistence of children's choices was an important
adjunct finding. Although it has been suggested that situ
ational factors have often accounted for the results in
84
studies of modeling effects on children's helping, the
present findings confirm more recent investigations
(Rushton, 1975; Israel & Brown, 1979) which have demon
strated that children's prosocial behavior persists. The
importance of the apparent durability of children's altru
istic behavior is that training methods can be developed
which will persist beyond the laboratory setting.
85
CHAPTER V
SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS
Summary
The development of social skills is becoming increas
ingly important for students, both in relation to academic
success and in regard to the social climate within the
classroom. Because educators have traditionally focused
more on suppressing antisocial behaviors, teachers are not
as knowledgable about methods that will foster positive
social behavior. Two techniques that teachers have at
their disposal are verbal socialization and behavioral
example. Although there has been extensive research about
models' words and deeds, the evidence as to which is the
superior method is not clear. A further important aspect
in regard to developing children who are helpful to others
is gaining an understanding of children's motivation for
behaving in helpful ways. One possible motivating factor
that has been suggested but little investigated in adults,
and not at all in children, is the notion that feeling
competent to help another will result in altruistic
behavior. In other words, one might be helpful if one has
the skills to help. While there has been considerable
research about children's helping behavior, investigations
86
have narrowly focused on two kinds almost exclusively--
donating to the poor and rescuing a child in distress.
There has been virtually no research about helping behavior
that would be likely to be found at school.
The present investigation was conducted in order to
examine determinants of helping behavior in elementary
school children. The purpose of the study was (1) to
clarify discrepancies in past research regarding the dif
ferential effect of a model's verbal instruction and a
model's behavioral example on children's subsequent altru
istic behavior, (2) to investigate the influence that
competence might play on children's willingness to help
others, and (3) to utilize a behavioral outcome such as
peer tutoring that would be likely to be found in elemenT
tary schools.
It was hypothesized that a model's use of direct
verbal instruction, including specific information about
the circumstances that call for help, would result in
significantly more helping than would a model's physical
demonstration of helping, It was further hypothesized that
high competence students would help significantly more than
low competence students.
The sample in the present study consisted of 146 (71
girls and 75 boys) fourth-grade, white, English-speaking
students randomly selected from four elementary schools in
a suburban, middle-class school district.
87
All subjects viewed one of three videotapes (physical
demonstration model, verbal instruction model, and no-model
control), following which they were interviewed by an
experimenter about the film. During the interview, subjects
were asked if they would be willing to volunteer to tutor
other students in arithmetic. Subjects were then shown the
"kind of work the students need help in." In the high
competence condition, subjects saw an easy first-grade
arithmetic paper while in the low competence condition,
they were shown a difficult eighth-grade arithmetic papbr. .
Following an appraisal of the arithmetic papers, subjects
anonymously filled out a paper indicating the number of
days they would be willing to volunteer. They could choose
zero days or a maximum of five days. Approximately ten
days later, subjects again were asked to indicate the num
ber of days they would be willing to volunteer to tutor
another student.
In order to test the two hypotheses, a 3 x 2 x 2
analysis of variance for both the initial and delayed
measures was carried out, with modeling, competence, and
sex as the factors, The dependent variable was the mean
number of days that students volunteered to help. In
addition, a correlation coefficient was computed in order
to discover the relationship between the immediate and the
delayed measure.
88
Findings, as presented within the framework of the
hypotheses associated with them, are summarized as follows:
1. A nonsignificant main effect for modeling was
obtained. There were no significant differences between
the experimental groups and the control group on both the
immediate and delayed measures as reflected by a small dif
ference between mean volunteering scores for the groups.
2. A nonsignificant main effect for competence was
obtained on the immediate measure with mean scores of the
high group showing a marginal elevation over the low com
petence group. A nonsignificant main effect was also found
on the delayed measure.
3. Although not directly relevant to either of the
two hypotheses posed, a nonsignificant effect of sex was
obtained, with no differences between the mean scores for
boys or girls.
4. A highly significant, strong positive correlation
was obtained between the immediate measure of helping and
the delayed measure, suggesting that children's choices
persisted for at least ten days.
Conclusions
An evaluation of findings in light of the assumptions,
limitations, and delimitations led to the following con
clusions and implications associated with them.
________________________________________________________________ 89
1. Although no significant findings were obtained for
the effect bf modeling, this lack of effect might be due to
other factors which tended to dilute the effects of the
model. Suggested factors which may have caused this
dilution were the nature of the videotaped models employed
in this study, the activity.or lack of it as portrayed by
models in the videotapes, the type of verbal messages given
by models, a possible psychological reactance on the part
of subjects who observed a model verbalize, or an artifact
of measurement. The observing children may have been
affected by one or all of these factors, and it is not
possible to ascertain that fact from the research design
employed.
2. Despite the fact that there were no significant
findings for the effect of competence, the near significant
effect on the immediate measure suggests that this factor
may well be associated with altruism in some way. At
present, not enough is known about the concept of compe
tence to postulate many possible explanations, although one
that seems plausible concerns the way in which children
appraise their own competence. The puzzling outcome that
the competence effect did not persist is also difficult to
explain and will need to be clarified in future research.
3. The lack of significance for sex suggests that
both males and females share an equal propensity to help,
and that the same underlying processes in regard to
90
prosocial behavior operate in an equivalent manner for both
sexes.
4. Children's helpfulness seems to persist and is,
therefore, not due totally to situational factors. Methods
utilized to enhance helpfulness could then be expected to
endure.
Recommendations
Based on the results and conclusions of this investi
gation, the following recommendations for future research
are made:
1. Inquiry should be pursued into the nature of
verbal socialization. More research is needed in order to
understand how a model's verbal messages affect children's
subsequent helping. Possible areas for fruitful study
relate to the kinds of verbal messages given, the role
of verbal reinforcement, and the part that attributional
messages might play. In addition, the utilization of
both verbal messages and a physical demonstration of help
together rather than either one or the other might
strengthen the power of the model.
2. Refinement of modeling procedures is important.
Live models should be used unless videotaped models can be
improved.
3. The notion of competence should be pursued on many
fronts. The present study is merely a beginning, and at
________________________________________________________________ 91
this point little is known about the relationship between
competence and helping in children. What are some possible
reasons that competence might enhance helping? A few
possibilities suggest themselves. First, being competent
may reduce the cost of helping for the benefactor in that a
higher level of skill may reduce stress. Second, the .
exercise of one's competencies may in itself be rewarding
and associated with feeling good. Third, feeling competent
may be related, in some way, with one's self-attributions.
Whatever the reasons, more needs to be learned about these
aspects of competence and helping. Other important areas
for research are a consideration of developmental factors
that might link competence and helping, the cognitive
processes underlying children's self-appraisal of com
petence, and racial and socioeconomic factors. Finally,
a variety of competence measures should be developed in
both academic and social areas, so that more than one
measure is available to use in studies involving competence.
4. Further inquiry into valid and reliable measures
of prosocial behavior is critical. This pursuit will not
be easy, but the development of prosocial behavior in chil
dren is important enough to be given high priority.
92
REFERENCES
Aronfreed, J. Conduct and conscience. New York: Academic
Press, 1968.
Aronfreed, J., 6c Paskal, V. Altruism, empathy, and the
conditioning of positive affect. In J. Aronfreed,
Conduct and conscience. New York: Academic Press,
19681
Bandura, A. Principles of behavior modification. New York:
Holt, Rinehart 6c Winston, 1969.
Bandura, A. Social learning theory. Morristown, New
Jersey: General Learning Press, 1971.
Bandura, A., Grusec, J. E., 6c Menlove, F, L. Some social
determinants on self-monitoring reinforcement systems.
Journal of Personality and Social Psychology, 1968, 8,
99-108.
Barnett, M. A., 6e Bryan, J. H, Effects Of competition
outcome feedback upon children’s helping behaviors.
Developmental Psychology, 1974, ID, 838-842.
Barrett, D. E., 6e Yarrow, M, R. Prosocial behavior, social
behavior, social inferential ability, and assertive
ness in children. Child Development, 1977, 48,
475-481.
Baumrind, D. Current patterns of parental authority.
Developmental Psychology Monograph, 1971, 4, 1-103.
Berkowitz, L., 6e Connor, W, H. Success, failure, and
social responsibility. Journal of Personality and
Social Psychology, 1966, 4, 664-669.
Berkowitz, L . , 6c Daniels, L. Affecting the salience of the
social responsibility norm: Effects of past help on
the response to dependency relationships. Journal of
Abnormal and Social Psychology, 1964, 68^, 275-281.
Berkowitz, L., 6c Friedman, P. Some social class differ
ences in helping behavior. Journal of Personality and
Social Psychology, 1967, .5, 217-225,
93
Brehm, J. W, A theory of psychological reactance.
New York: Academic Press, 1966.
Bruner, J. S. Nature and uses of immaturity. American
Psychologist, 1972, 2l_y 687-708.
Bryan, J. H. Model affect and children's imitative
behavior. Child Development, 1971, 4v2, 2061-2065.
Bryan, J. H. Why children help: A review. Journal of
Social Issues, 1972, 28^, 87-104.
Bryan, J. H. Children's cooperation and helping behaviors.
In E. M. Hetherington (Ed.), Review of child develop
ment Research (Vol. 5). Chicago: University of
Chicago Press, 1975.
Bryan, J. H. Prosocial behavior. In H. L. Horn, Jr. , 6c
P. A. Robinson (Eds.), Psychological processes in
early education. New York: Academic Press, 1977.
Bryan, J. H., 6c London, P. Altruistic behavior by
children. Psychological Bulletin, 1970, 7_3, 200-211.
Bryan, J. H., Redfield, J., 6c Mader, S. Words and deeds
about altruism and the subsequent reinforcement power
of the model. Child Development, 1971, 42., 1501-1508.
Bryan, J. H., 6c Walbek, N. H. The impact of words and
deeds concerning altruism upon children. Child
Development, 1970, 41, 747-757, (a)
Bryan, J. H., 6c Walbek, N. H. Preaching and practicing
generosity: Children's actions and reactions. Child
Development, 1970, 4_1, 329-353. (b)
Campbell, D. T., 6e Stanley, J. C. Experimental and quasi-
experimental designs for research. Chicago: Rand
McNally, 1963.
Cartledge, G., 6c Milburn, J, F. The case for teaching
social skills in the classroom: A review. Review of
Educational Research, 1978, 48, 133-156.
De Palma, D. J, Effects of social class, moral orien
tation, and severity of punishment on boys' moral
response to transgression and generosity. Develop
mental Psychology, 1974, 10, 890-900.
Dlugokinski, E., 6e Firestone, I. J. Congruence among four
methods of measuring other-centeredness. Child
Development, 1973, 44, 304-308.
94
Elliott, R., & Vasta, R. The modeling of sharing: Effects
associated with vicarious reinforcement, symbol
ization, age and generalization. Journal of Experi
mental Child Psychology, 1970, 10, 8-15.
Elmer, N. P., & Rushton, J. P. Cognitive-developmental
factors in children's generosity. British Journal of
Social and Clinical Psychology, 1974, 13, 277-281.
Feldman, N. S., & Ruble, D. N. Awareness of social com
parison interest and motivation: A developmental study.
Journal of Educational Psychology, 1977, 69_, 579-585.
Festinger, L. A theory of social comparison. Human
Relations, 1954, 7, 117-140.
Form, W. H., & Nosow, S. Community in disaster. New York:
Harper, 1978.
Friedrich, L. K., & Stein, A. H. Prosocial television and
young children: The effects of verbal labeling and
role playing on learning and behavior. Child Develop
ment, 1975, 4j5, 27-38.
Gouldner, A. W, The norm of reciprocity: A preliminary
statement. American Sociological Review, 1960, 25,
161-178.
Green, F. P., 6e Schneider, F« W. Age differences in the
behavior of boys on three measures of altruism.
Child Development, 1974, 45^, 248-251.
Grusec, J. E, Power and the internalization of self-
denial. Child Development, 1971, 42, 93-105.
Grusec, J. E, Demand characteristics of the modeling
experiment: Altruism as a function of age and
aggression, Journal of Personality and Social
Psychology, 1972, 22, 139-148.
Grusec, J. E., Kuczynski, L., Rushton, J. P., 6c Simutis,
Z. M, Modeling, direct instruction and attributions:
Effects on altruism. Developmental Psychology, 1978,
14, 51-57.
Grusec, J, E . , 6c Skubiski, L. Model nurturance, demand
characteristics of the modeling experiment and
altruism. Journal of Personality and Social
Psychology, 1970, 14, 352-359,
95
Harris, D, B. A scale for measuring attitudes of social
responsibility in children. Journal of Abnormal and
Social Psychology, 1957, !55, 322-326.
Harris, M. B. Reciprocity and generosity: Some deter
minants of sharing behavior. Child Development, 1970,
41, 313-328.
Harris, M. B. Models, norms and sharing. Psychological
Reports, 1971, 29, 147-153.
Hartup, W. W., & Coates, B. Imitation of a peer as a
function of reinforcement from the peer group and the
rewardingness of the model. Child Development, 1967,
38, 1003-1016.
Hoffman, M. L. Moral development. In P. H. Mussen (Ed.),
Carmichael's manual of child psychology (3rd ed.)
(Vol. 2). New York: John Wiley & Sons, 1970.
Isen, A. M., Rosenhan, D. L., 6c Horn, N. Effects of
success and failure on children's generosity. Journal
of Personality and Social Psychology, 1973, 27,
239-247.
Israel, A. C., 6c Brown, M. S. Effects of directiveness of
instructions and surveillance on the production and
persistence of children's donations. Journal of
Experimental Child Psychology, 1979, 27_, 250-261.
Israel, A. C., 6c Raskin, D. A, Directiveness of instruc
tions and modeling: Effects of production and
persistence of children's donations. Journal of
Genetic Psychology, 1979, 135, 269-277.
Jastak, J. F., 6c Jastak, S. R. Wide range achievement test:
Manual of instructions (Rev” ed.). Wilmington,
Delaware: Guidance Associates of Delaware, 1976.
Jones, S. C., 6c Regan, D. T. Ability evaluation through
social comparison. Journal of Experimental Social
Psychology, 1974, 1(), 133-146.
Kazdin, A. E., 6c Bryan, J. H, Competence and volunteering.
Journal of Experimental Social Psychology, 1971, 7,
87-97.
Kohlberg, L. Development of moral character and moral
ideology. In M. L. Hoffman 6c L, W. Hoffman (Eds,),
Review of child development research (Vol. 1),
New York: Russell Sage Foundation, T964,
96
Kohlberg, L. Stage and sequence: The cognitive-
developmental approach to socialization. In D. A.
Goslin (Ed.), Handbook of socialization theory and
research. Chicago: Rand McNally, 1969.
Krebs, D. Altruism: An examination of the concept and a
review of the literature. Psychological Bulletin,
1970, 73, 258-302.
Leeds, R. Altruism and the norm of giving. Merrill-Palmer
Quarterly, 1963, 9_, 229-240.
Long, G, T., & Lerner, M. J, Deserving, the "personal
contract" and altruistic behavior by children.
Journal of Personality and Social Psychology, 1974,
29, 551-556.
Macaulay, J., & Berkowitz, L. (Eds.). Altruism and helping
behavior. New York: Academic Press, 1970.
Masters, J. C. Effects of social comparison upon children's
self-reinforcement and altruism toward competitors and
friends. Developmental Psychology, 1971, .5, 64-72.
Midlarsky, E. Aiding responses: An analysis and review.
Merrill-Palmer Quarterly, 1968, 14, 229-260.
Midlarsky, E., & Bryan, J. H, Affect expressions and
children's imitative altruism. Journal of Experi
mental Research in Personality, 1972, 6>, 195-203.
Midlarsky, E., Bryan, J. H., & Brickman, P. Aversive
approval: Interactive effects of modeling and rein
forcement on altruistic behavior. Child Development,
1973, 44, 321-328.
Miller, D. T., & Smith, J. The effect of own deservingness
and deservingness of others on children's helping
behavior, Child Development, 1977, 48, 617-620.
Moore, B. S., Underwood, B., & Rosenhan, D. L. Affect and
altruism. Developmental Psychology, 1973, 8, 99-104.
Nie, N. H., Hull, C. He, Jenkins, J. G., Steinbrenner, K.,
6c Bent, D, H. Statistical package for the social
sciences (2nd ed.). New York: McGraw-Hill, 1975.
Parke, R. D. Nurturance, nurturance withdrawal and
resistance to deviation. Child Development, 1967,
35, 1101-1110.
97
Pepitone, E. A. Comparison behavior in elementary age
children. American Educational Research Journal,
1972, 9, 45-63.
Piaget, J. The moral judgment of the child. New York:
Collier, 1962.
Presbie, R. J., 6c Coiteux, P. F, Learning to be generous
or stingy: Imitation of sharing behavior as a function
of model generosity and vicarious reinforcement.
Child Development, 1971, 42, 1033-1038.
Rice, M. E., 6c Grusec, J. E. Saying and doing: Effects on
observer performance. Journal of Personality and
Social Psychology, 1975"! 32, 584-593.
Rosenhan, D. L. The kindness of children. Young Children,
1969, 25, 30-44,
Rosenhan, D, L. Learning theory and prosocial behavior.
Journal of Social Issues, 1972, 28, 151-163.
Rosenhan, D. L., 6c White, G, M. Observation and rehearsal
as determinants of prosocial behavior. Journal of
Personality and Social Psychology, 1967, .5, 424-431.
Rubin, K. H . , 6c Schneider, F. W, The relationship between
moral judgments, egocentrism, and altruistic behavior.
Child Development, 1973, 44, 661-665.
Ruble, D. N., Feldman, N. S., 6c Boggiano, A. G. Social
comparison between young children in achievement
situations. Developmental Psychology, 1976, 12,
192-197.
Rushton, J. P. Generosity in children: Immediate and long
term effects of modeling, preaching and moral judgment
Journal of Personality and Social Psychology, 1975,
31, 459-466.
Rushton, J. P. Socialization and the altruistic behavior
of children, Psychological Bulletin, 1976, 83,
898-913.
Rushton, J. P., 6c Owen, D, Immediate and delayed effects
of TV modelling and preaching on children’s generosity.
British Journal of Social and Clinical Psychology,
1975, 14, 309-310.
Schwartz, T., 6c Bryan, J. H, Imitation and judgments of
children with language deficits. Exceptional
Children, 1971, 38, 157-158. (a)
98
Schwartz, T., & Bryan, J. H. Imitative altruism by deaf
children. Journal of Speech and Hearing Research,
1971, 14, 453-461. (b)
Severy, L. J., 6c Davis, K. E, Helping behavior among
normal and retarded children. Child Development,
1971j 42, 1017-1031.
Shaffer, David R. Social and personality development.
Monterey, Calif.: Brooks/Cole Publishing, 1979.
Sims, S. A, Sharing by children: Effects of behavioral
example, induction and resources. Journal of
Psychology, 1978, 100, 57-65,
Staub, E. A, A child in distress: The effects of focusing
responsibility on children in their attempts to help.
Developmental Psychology, 1970, 2^ 152-154.
Staub, E. A. A child in distress: The influence of nur
turance and modeling on children's attempts to help.
Developmental Psychology, 1971, 5^, 124-132. (a)
Staub, E. A. Helping a person in distress: The influence
of implicit and explicit "rules" of conduct on chil
dren and adults. Journal of Personality and Social
Psychology, 1971, 17, 137-144. (bj"^
Staub, E. A. The use of role playing and induction in
children's learning of helping and sharing behavior.
Child Development, 1971, 42_, 805-816. (c)
Staub, E. A., 6c Sherk, L. Need for approval, children's
sharing behavior, and reciprocity and sharing.
Child Development, 1970, 41, 243-252.
Veroff, J, Social comparison and the development of
achievement motivation. In C. P. Smith (Ed.),
Achievement-related motives in children. New York:
Russell Sage Foundation, 1969.
White, G. M. Immediate and deferred effects of model
observation and guided and unguided rehearsal on
donating and stealing. Journal of Personality and
Social Psychology, 1972, 2_1, 139-148.
White, G. M., 6c Burnam, M. A. Socially cued altruism:
Effects of modeling, instructions, and age on public
and private donations. Child Development, 1975, 46,
559-563.
99
Yarrow, M. R. , Scott, P. M, , 6c Waxier, C, Z. Learning
concern for others. Developmental Psychology, 1973,
8, 240-261,
Yarrow, M. R, , 6c Waxier, C. Z. Dimensions and correlates
of prosocial behavior in young children. Child
Development, 1976, 47, 118-125,
Zigler, E., 6c Yando, R. Outerdirectedness and imitative
behavior of institutionalized and noninstitutionalized
younger and older children. Child Development, 1972,
43, 413-425.
100
APPENDIX A
MEASURES OF HIGH AND
LOW COMPETENCE
101
Find each missing number.
x is 30% of 1 80. 8% of 25 is a. 66 o% of 120 is m.
5% of 740 is g. v is 74% of 320. 10% of 60 is d.
1 2
1 1 * X
12
4 2 ) 3 2 8 9 9 >268 5 4 ) 4 8 2 9 3
2 5 >15 50 3 7 >1 8 3 9 2
8(6) + 7 14 + 8(7) - 10
8(6 + 7) (14 + 8)7 - 10
102
2 + 2 =
3 - 2 =
4 - 2 + 3 +
5 - 3 = + 4 — 5 -
4 + 2 - - 2 =
+ 3 + 6 + 3 + 7
+ 4 + 7 “I - 0 + 4 + 2 + 4
+ 2 + 6 + 4 + 3
+ 4
+ 2 + 5 + 8
103
APPENDIX B
MEASURE OF HELPING
104
Days I will tutor
None Monday Monday
Tuesday
Monday
Tuesday
Wednesday
Monday
Tuesday
Wednesday
Thursday
o
Ln
Monday
Tuesday
Wednesday
Thursday
Friday

Asset Metadata

Creator Blacher, Joan Helen (author)

Core Title The effects of modeling, direct instruction, and competence on children's helping responses

Contributor Digitized by ProQuest (provenance)

Degree Doctor of Philosophy

Degree Program Education

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag education, educational psychology,OAI-PMH Harvest

Language English

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c26-486905

Unique identifier UC11245245

Identifier usctheses-c26-486905 (legacy record id)

Legacy Identifier DP24757.pdf

Dmrecord 486905

Document Type Dissertation

Rights Blacher, Joan Helen

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA