SEQUENTIAL ANALYSIS OF ADULT - CHILD .
PLAY ENCOUNTERS: THE EFFECTS OF TRAIMNG .
AND PERSONAL CHARACTER‘STICS

Dissertation for the Degree of Ph. D.
MICHEGAN STATE UNIVERSITY
THOMAS M. RUTLEDGE
1974

 

 

This is to certify that the
thesis entitled A I
Seywmm fl/VALYSIS at ADULT-C ' b
(PLAY En/couwralcs: "TAE' EFFECTS 01C

TKfilNM/b KHVD Peksm/rfl. Ckn-renere/ezsncs
presented by

*‘I’Lomyrs M. (PJTLEIW‘?

has been accepted towards fulfillment
of the requirements for

W degree inf???“ L9 (051/

 

a]

 

 

Major professor
Date /

0-7639

 

 

 

ABSTRACT

SEQUENTIAL ANALYSIS OF ADULT-CHILD PLAY
ENCOUNTERS: THE EFFECTS OF TRAINING
AND PERSONAL CHARACTERISTICS

BY

Thomas M. Rutledge

The purpose of the present investigation was to
study the effects of a selection and training program
teaching undergraduates to function therapeutically in
play therapy sessions with clinic—referred children.

The selection process involved the differentiation of

two levels of potential, high and low. There were also
two training conditions: (1) those trained in specific
client-centered play therapy skills and (2) those placed
in‘a no training control group. Assessment of potential
was based upon a parent attitude inventory, a sensitivity
to children measure and a mental health questionnaire.
Control group subjects were given a list of books on play
therapy to be read if they wished. They received no
training or feedback during their play sessions. They
were observed through a one-way mirror to insure that
nothing destructive occurred. The training group met

two hours weekly for two academic years. Various play

Thomas M. Rutledge

\

therapy texts were read and discussed. Video tapes of
other play sessions were observed. Each trainee was
video taped playing with a "normal" child and such
interaction became the focus of group discussion.

The child subjects were emotionally disturbed,
nonpsychotic referrals to the Psychology Clinic of
Michigan State University, ages five to eight. They
were seen once a week for 15 weeks for a 30-minute play
session. Beginning with the first, every fifth session
was video taped. The first two of the four video taped
sessions were utilized in this study.

Categories used in coding adult and child
behaviors were essentially approach-avoidance cate-
gories. A method of sequential analysis was devised to
focus on the interactions. Each child behavior was com-
pared with the simultaneous adult behavior and with the
adult behavior one interval later. An interval was
7.5 seconds in duration.

Factor analysis of the data suggested the follow-
ing: (1) potential as assessed by the instruments was
not a factor of significance in this study, (2) trained
undergraduates exhibited significantly more positive
movement toward the child than did the untrained under-
graduates, and (3) a sequential analysis of interactions

was a feasible method. Results were discussed in terms

Thomas M. Rutledge

of implications for the future training of nonprofes-
sionals in mental health skills and in terms of future

research directions.

SEQUENTIAL ANALYSIS OF ADULT-CHILD PLAY
ENCOUNTERS: THE EFFECTS OF TRAINING

AND PERSONAL CHARACTERISTICS

BY

Thomas M: Rutledge

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Department of Psychology

1974

TO CAROLYN whose love and caring

I treasure

ii

ACKNOWLEDGMENTS

Thanks are due first of all to Gary Stollak
whose concern for children and commitment to the cause
of nurturant parenting has been contagious. I am
deeply grateful for his constant support and encourage-
ment.

Larry Messe has again helped to tease out a
number of meanings by his prodding for the meaning of
numbers. I greatly appreciate his generosity of time and
attention.

Thanks are due also to Mrs. Carmen Alioto and
Janet Strang for their conscientious clerical assistance.

Finally I wish to acknowledge my debt to my.
unborn child whose coming has pushed me into completing

my education so that I will be ready to learn how to play.

iii

Chapter

I.

II.

III.

IV.

TABLE OF CONTENTS

INTRODUCTION 0 C O O O O O C .

Nonprofessionals as Psychotherapeutic

Agents with Children. . . . .
The Selection of Nonprofessionals .
Personality Characteristics of

Therapists with Adult Clients. .
Therapists with Children . . . .
Sequential Analysis of Reciprocal

Adult-Child Interactions . . .
Some Conclusions. . . . . . .
Hypotheses. . . . . . . . .

PqE THOD O I O C C O O O O O C

Undergraduate Subjects. . . . .
Control Group. . . . . . .
Experimental Group Training Pro-
cedures . . . . . . . . .
Child Subjects . . . . . . .
Data Collected . . . . . .
Coding Adult- Child Interactions. .
Procedure for Coding Behavior Cate-
gories . . . . . . . . .
Training Assistants in the Coding
Procedure . . . . . . . .
Data Storing and Analysis. . . .

PESULTS O O O O O O O O O 0

Child Behaviors . . . . . . .
Adult Behaviors . . . . . . .

DISCUSSION. . . . . . . . . .

Overview . . . . . . . . .
Methodological Considerations . .
Potential . . . . . . . . .
Training . . . . . . . . .
Summary . . . . . . . . .
Implications for Future Research .

iv

Page

13
17
19

21

21
23

24
25
25
26

26

27
28

3O

30
32

40

40
41
48
55
59
61

Page

APPENDICES

Appendix
A. Parent Attitude Questionnaire. . . . . . 67
B. Sensitivity to Children Test (STC—T) . . . 76
C. Mental Health Scale (P-R) . . . . . . . 85

D. Categories Used in Coding the Behavior of
the Undergraduate. . . . . . . . . 87

E. Categories Used in Coding the Behavior of
the Child O O O O O O O O O O O 88

F. Analysis of Variance of Adult Behaviors Posi-
tive and Negative One Interval After a
Neutral Child Behavior . . . . . . . 89

G. Analysis of Variance of Adult Behaviors in
Categories One and Five, One Interval

After a Neutral Child Behavior . . . . 90
H. Analysis of Variance of Pdult Behaviors in
Initial Sessions . . . . . . . . . 91'
BIBLIOGRAPHY O O O O O O O O O O O I O 92

CHAPTER I

INTRODUCTION

Since reaction times were first found to vary,
the measurement of potential and aptitude has been at
least one of the Holy Grails of psychological research.
Where potential has been carefully evaluated and dif-
ferentiated there has been increased economy in the
expenditure of resources human and financial. This
present study also pursues the factor of potential as
it relates to the selection and training of undergraduates
as play therapists. If potential for the role of play
therapy exists in individuals in varying degrees which
can be differentiated with predictive utility, then the
assessment of potential deserves a full share of the
resources available. On the other hand, if training tends
to wash out any meaningful discrimination of potential,
then energies and monies can more exclusively be devoted
to the training process.

The present study attempts to answer the
questions (1) assuming that one can specify behaviors

indicative of positive mental health in children, can

we (2) select and (3) train college undergraduates who
could exhibit such behaviors to help the child? In
particular this study investigated a training program
designed to increase in adults those behaviors which have
been associated with effective adult-child interaction.
Also studied was the predictive value of the selection
process which attempted to differentiate adults with
high potential as play therapists from adults with low
potential.

Nonprofessionals as Psychother-
apeutic Agents with Children

 

 

College students have frequently been utilized
in studies of quasi—therapeutic settings. Their moti-
vation and ready availability have been taken advantage
of by behavioral researchers. For some of the same
reasons they were employed in the present study. Beyond
these pragmatic issues there is a growing body of
research to show that undergraduates have a particular
contribution to make in certain therapeutic settings.

Mitchell (l966) used student volunteers in what
he called "amicatherapy." His student volunteers were
instructed to simply make friends with disturbed chil-
dren. Clinical investigations of 74 children who par-
ticipated indicated that all the children benefited by
their contacts with these volunteers. The students

entered the child's world with some facility, Mitchell

concluded, because they, as students, were not yet
irrevocably committed to the adult world.

Peinherz (1964) utilized undergraduates in a
state hospital with children inpatients. They spent
one afternoon a week interacting with individual chil-
dren. In terms of staff reported functioning all the
children involved with the college students showed
improvements. Kreitzer (1969) confirms much the same
thing, utilizing students in interactions with hospital-
ized emotionally disturbed children. He also reported
that many behaviors regarded as inappropriate were
reduced or eliminated.

Brennan (1967) worked with a child guidance
clinic. College students were utilized as big brothers
for some of the patients. In his study, Brennan used
attrition rates as an outcome measure. Tuckman and
Lavell had reported a typical 60% attrition rate for
child guidance clinics. Brennan felt that it was a
measure of the success of the big brother program that
attrition rates had been reduced to a low of 12%.

It is clear, then, that college students are a
valuable manpower resource in a variety of roles approxi-
mating therapy. Gruver (1971) reviewed extensively the
use of college students as therapeutic agents. He also
concluded that "the use of nonprofessionals in general

and college students in particular offer promise in the

effort to provide more complete mental health services

to an ever increasing population [p. 124]." He com-
plained, however, that firm conclusions at this point

are unwarranted due to the dissimilar nature of the
studies, lack of sufficient controls, and varying amounts
of training and duration of training.

College students also seem to play a bridge role
between the professional world and the lay world. For
example, once it has become acceptable to utilize college
students in certain roles it becomes easier to gain
acceptance for others to be similarly utilized. House-
wives (Rioch, Elkes, Flint, Newman & Sibler, 1963) and
other mature women (Nichtern, Donahue, O'Shea, Marans,
Curtis & Brody, 1964; Donalue & Nichtern, 1965) have
also been trained to act as therapists for the children
of others. Parents have been trained to function as
therapeutic agents for their own children (Patterson &
Brodsky, 1966; Guerney, 1966; Stover & Guerney, 1967).

Carkhuff (1968) reviewed 20 studies regarding the
use of nonprofessionals and concluded that a minimal
amount of training produced increased ratings of empathy,
genuineness and nonpossessive warmth in hospital atten-
dants, college undergraduates, school teachers and
community volunteers. Of the 20 studies summarized,
however, only 5 demonstrated the effectiveness of the

nonprofessional in terms of outcome measures.

There is strong evidence that further research
studying the therapeutic uses of nonprofessionals in
general and college students in particular will be useful.

The study of nonprofessionals in therapeutic
settings also makes more feasible the use of control
groups. In the more typical professional setting, using
a control group, such as a no treatment group, involves
directly many sensitive ethical issues. For instance, is
it fair to the patient to be placed in a nontreatment
group? With the quasi-therapeutic arrangement in which
the nonprofessional might operate such possibilities are
somewhat expanded. Linden and Stollak (1969) used a no
training control group interacting with "normal" children
in attempting to evaluate their training procedures.

Such controls, when studying play interaction with clinic-
referred children, are more difficult to arrange.

Sometimes the recruitment of nonprofessionals
has been done on the basis of who is available. This
may be practical when the emphasis is on quantity of man-
power. A study of the quality and effectiveness of the
manpower implies paying close attention to selection pro-

cedures.

The Selection of Nonprofessionals

 

The selection of nonprofessionals has sometimes
been similar to the selection of professionals with the

emphasis on intellectual indices (Rioch, et a1., 1963).

Most of the training programs for nonprofessionals, how-
ever, have been like that of Harvey (1964) who selected
"persons who exhibit a sincere regard for others,
tolerance and ability to accept people with values
different from one's own, a healthy regard for the self,
a warmth and a sensitivity in dealing with others, and a
capacity for empathy [p. 349]." Such qualities were
detected in the Harvey study mainly through interviews
with the professional staff. Harvey concluded that the
general success of the trainees sufficiently validated
the selection process. In fact, he suggests that defects
in the selection process are more likely to lead to the
exclusion of false negatives than to the inclusion of
false positives. Lichtenberg (1956) used similar cri-
teria in the selection process and advocated that an
inverse amount of time should be allotted between
selection procedures and the training available, that
is, if the training procedures are lengthy and in depth,
selection procedures need not be as thorough. The obvious
underlying assumption seems to be that training washes
out personality differences.

Goodman (1972) in selecting undergraduates for
a companionship therapy program used a structured group
method that utilized successive sets of dyadic inter-
action as performance samples for rating. Participants

rated each other and were also rated by three staff

observers. The primary scales were understanding, open
and accepting--warm, which formed a therapeutic talent
composite. Remaining scales were labeled depressed,
quiet, rigid, and relaxed. Intercorrelation of all the
scales of the GAIT (Group Assessment of Interpersonal
Traits) suggested internal coherence. On the basis of
this method Goodman rejected 56 of 179 undergraduate
applicants.

Goodman attempted to be objective in selection
procedures. The literature, however, gives no evidence
concerning any control group comparisons in selection
procedures. Even where key personality traits have been
the basis for selection the effects of training have not
been compared at the different levels at which individuals
possess these traits.

Personality Characteristics of
TherapiSts with Adult Clients

 

 

The importance of an effective selection process
of professionals and nonprofessionals is emphasized by
the growing body of research on the therapeutic importance
of the therapist's personal characteristics. Epstein
(1963) concluded that the significant factors contributing
to the therapeutic ability of therapists are related to
their personality. Whitehorn and Betz (1954) initiated
a highly fruitful research vein in the explorations of

the A-B "type" distinctions which differentiated

therapists likely to be more successful with adult schizo-
phrenics from therapists likely to be more successful with
adult neurotics. Shadlow (1968) and Kirkpatrick (1969)
have continued research in this area.

Rogers (1951), in his early presentation of the
client-centered approach, proposed that the key ingre-
dients of good therapy are warmth, genuineness and
empathy. Truax and associates have assembled evidence
to suggest that these three qualities cut across divergent
theories (Truax & Wargo, 1966; Truax & Carkhuff, 1967;
Truax & Mitchell, 1968). Stoffer (1968) continued this
research direction and added dogmatism as a possible
important therapist characteristic. He was unable,
however, to find any significant relationships between
dogmatism and perceived levels of therapeutic conditions.
He found that nonpossessive warmth was the best pre-
dictor of a good therapeutic relationship.

From a psychoanalytic approach Bachrach (1968)
comes to a similar conclusion that the therapist is not
a "black box." He relates empathy to adaptive regression
and facility with primary process, both of which he views

as important therapeutic capacities.

Therapists with Children

 

Focusing on children, Moustakas (1955) drew a
parallel between the emotional growth of a child in a

play therapy relationship and the normal emotional

development of a child in a family relationship. Thus
sensitive and effective play therapists can be described
in terms appropriate to sensitive and effective parents.
Similarly the extensive literature on parental charac-
teristics provides practical information about good play
therapist's characteristics. One of the clearest examples
of this is the study of Sears, Maccoby, and Levin (1957),
who found that indulgence and too much warmth fostered
dependence in children while cold, rejecting parents
fostered insecurity in their children. Earlier Baldwin
(1949) found that a democratic home environment seemed to
foster children who were active, socially outgoing,
favored in their peer groups, and high on intellectual
curiosity, originality and constructiveness. Crandall
(1958) studied the parental influences on social com-
pliance and found that the best predictor of a child's
compliance outside the home was maternal reward for com-
pliance rather than maternal punishment for noncompliance.
In a very thorough study based heavily on
naturalistic observation, Baumrind (1967) compared three
groups of nursery school children who were assessed for
social competence. The group described as self-reliant,
assertive, self-controlled, buoyant and affiliative had
parents described as controlling, demanding, communicative
and loving. The group of children described as lacking

in self-control and self-reliance and tending to retreat

10

from novel situations had parents characterized as non-
controlling, demanding and warm. The third group of
children described as withdrawn, discontent and dis—
trustful had parents who were described as controlling
and detached.

Hoffman (1963) found that children who were
allowed to vent feelings and be accepted for them, who
are disciplined within the limits of their comprehension
and who have parents who teach and provide models for
consideration for others, appear to eXpress spontaneous,
altruistic concern for others.

Before such findings are directly applied to the
selection of potential play therapists it must be noted
that a complete review of the literature shows considerable
inconsistency. For example, Baumrind and Black (1967)
reported that punitive parental attitudes were not
associated with fearful or compliant behavior. On the
contrary, these were associated with independent and
domineering behavior, particularly with girls. They
also reported in this study that parental warmth was not
an important predictor of any child behaviors.

In a later study Baumrind (1971) found that (l)
authoritative parental behavior was associated with
independent, purposive behavior for girls but it was only
associated with such behavior for boys when the parents

were nonconforming; (2) authoritative parental control

11

was clearly associated with all indices of social responsi-
bility in boys compared to authoritarian and permissive
parental control and with high achievement in girls, but
not with friendly, cooperative behavior; (3) contrary to
expectations, parental nonconformity was not associated
with lack of social responsibility in either boys or

girls.

There are two methodological problems in parent-
child studies which may have a direct bearing on the
selection of high potential therapists. The first is
the assumed relationship between parental attitudes and
parental behavior (Bell, 1958; Mannino, Kisielewski,
Kimbro & Morgenstern, 1968; Brody, 1965; Becker, 1962).
Zunich (1962) found no high correlations between attitude
scales and actual parent-child observed interactions.
Becker, Peterson, Luria, Shoemaker and Helmer (1962)
found similar results and reported a lack of correlation
between parent attitudes and parent behavior. Becker and.
Krug (1965) did an extensive review of the research
involving the use of the Parent Attitude Research Instru-
ment (PARI). They conclude that its usefulness and
reliability varies according to the samples involved.
They report that theoretically meaningful results most
often involved homogeneous upper middle class families.
Likewise they report some of the scales influenced by

extraneous variables, for example, measures of

12

authoritarian attitudes appear influenced by an acqui-
escence-response set and by the educational level of the
respondent. Baumrind and Black (1967) found that the
pattern of correlations between parent attitudes and
observed behavior differed frequently depending on the
sex of the parent and the sex of the child. And this
pattern was not in a manner which could be predicted.
They found few outright inconsistencies between measures
of attitude and measures of behavior but the relations
were not strong.

The second methodological problem in parent-
child studies that is relevant to the selection of high
potential play therapists is that the child's behavior
is not simply a response to the parental behaviors.
Investigators have commented that the child's influence
on the parent demands close investigation (Peterson,
Becker, Hellmer, Shoemaker & Quay, 1959; Bell, 1968;
Osofsky, 1970; Medinnus, 1967; Schulman, Shoemaker &
Moelis, 1962). Harper (1971) in a review of similar
studies with mammals finds well-documented examples in
which offspring behavior is more obviously a stimulus
to a parental response rather than vice-versa. On this
basis he then makes a broad generalization and suggests
that studies confirming parental effects upon their
young might also be readily interpreted as effects of

the young upon their parents. A more moderate position

13

demands that adult-child interactions not be viewed as
unidirectional. The response of each participant in an
interaction becomes in some way a stimulus for the other.
To ignore these reciprocal influences, then, is, at best,
to ignore an important aspect of the relationship and,

at worst, to distort that relationship.

Sequential Analysis of Reciprocal
AdultiChild Interactions

 

 

One method of preserving the interactional stimu-
lation of adult-child exchanges is to make a sequential
analysis. In 1956 Moustakas, Sigel, and Schalock pub-
lished a coding system for this purpose. They used
89 adult and 82 child categories. Some of the categories
were clearly recognized as stimulus categories, e.g.,
seeking information, seeking permission, and directing
by command. Others were recognized as response cate-
gories, e.g., attentive observation, recognition, and
cooperation. The behavior of child and adult was record-
ing individually for each 5-second period. In addition
a hostility or anxiety rating accompanied all category
entries. The schedule was designed for adult—child
interactions under controlled conditions as in a play
therapy session, or in direct observation in a natural
setting as a mother and child at home.

This schedule was used in an exploratory study

by Moustakas and Schalock (1955). The investigations

III. ll" [II [I'll [ .III'II. '1 All [It I!

Il’ll'l'll' ill lllf llllllu‘l ’1 Il

14

involved the nature of the interaction of therapist and
child in a play therapy situation with particular reference
to two groups of children, one with serious emotional
problems and the other without such problems. .One
therapist was observed in two 40-minute play sessions
with each subject in each group. Reliability was estab—
lished on the recording of the behavioral category and
the anxiety-hostility ratings. Of particular importance
to the present study was the question the investigators
attempted to answer: do certain types of child behavior
consistently produce certain reactions from the therapist?
A total of 771 sequences initiated by the child were
analyzed. The categories selected for analysis were
chosen on the basis of two criteria: (1) having a
frequency of 35 or more and (2) having high stimulus
properties. Only three child categories met these cri-
teria; seeking information, directing by suggestion, and
directing by command. As regards the child seeking
information, in 73% of the instances the therapist gave
the information when requested; in 14% of the instances
he responded with a question and in 13% of the instances
he listened and observed.

There were no important differences in responses
to the two groups, i.e., the children with emotional

problems and those without emotional problems.

15

The therapist responded to directing by sug-
gestion with straightforward cooperation with no dif-
.ferences between the two groups. The therapist's pre-
dominant response to directing by command was to
cooperate; in 18% of the instances the therapist sought
information and in 11% of the instances he made an
interpretation. Again there was no particular dif-
ference in the responses to the two groups.

In a similar manner Moustakas and Schalock did
a sequential analysis in instances where the therapist
initiated the sequence. A total of 1,882 interaction
sequences initiated by the therapist were noted. Again
the criteria for selecting a category was a frequency of
35 or more and having high stimulus properties. Five
categories met these criteria; offering verbal infor-
mation, orienting the child to time, orienting the child
to his role by leaving the responsibility of decision to
the child, directing by suggestion and interpretation.
With regard to giving information, children in both
groups responded by exploring the information further,
by simply recognizing it, or by failing to recognize it.
About 80% of their responses involved some kind of
recognition or further exploration. As regards structur-
ing time, 50% of the instances resulted in some kind of

rejection. Suggestions were well received and were

16

responded to almost consistently by cooperation. With
regards to interpretation, acceptance led all child
responses.

The investigators left out several categories
from sequential analysis because they were seen primarily
as response categories. For example, attentive obser-
vation was not analyzed in this manner though it was
seen to be a major aspect of the therapist's behavior.
In fact, 5 categories were seen to account for about 85%
of the therapist's interaction with the children; atten-
tive observation, rec0gnition of stimulation, giving
information verbally, interpretation by restating ver-
balized feelings and seeking information of an impersonal
nature. These five categories accounted for more than
9,000 observations of therapist's behavior. Only 1,600
of these were included in any sequential analysis.

Caretaker-infant studies are another area where
preservation of the interactional sequence is crucial.
For the most part, however, the problem of selecting the
initiating event or stimulus is simplified by the fact
that the variety of infant behavior is more limited.

For example, Moss and Robson (1968) in a study with

infants of one to three months considered episodes of
interaction separated when one minute or more elapsed
without the infant crying or fussing. Schoggen (1963)

answered the question of who starts the interaction by

17

his definition of the "environmental unit.“ This unit is
an action of the child or parent which is directed toward
a recognizable end and is identified as such by the other.
Bell (1971) concludes that the heart of the
problem of analysis is the absence of a method of con-
ceptualizing sequences so that the contribution of both
participants is identified. The sequential analysis of
adult-child interactions has been greatly neglected.
The trend is to code and count adult and child behaviors

separately and to correlate the totals.

Some Conclusions

 

In view of the above review the following con-
clusions are drawn as specifically relevant to the
selection and training of nonprofessionals in a client-
centered approach to play therapy with clinic—referred

children:

1. A careful selection based on key personality
characteristics and attitudes could but not
necessarily will result in important differences
in the effects of training and the therapeutic
effectiveness of nonprofessionals. The key
characteristics of a client-centered therapist
have been noted to be warmth, genuineness, and
empathy. Ginott (1961) speaking more directly

of the child therapist, lists resolution of the

18

therapist's own emotional conflicts, a security
within himself which allows the expression of

all feelings of the child and which is strong
enough to prohibit the performance of some
behaviors, a stoicism and patience, and the
capacity for empathy. "A child therapist must
have the keenest ability to respond genuinely and
accurately to the child's private world, without
being infected by it, or without his own anger,

fear, or confusion being reactivated [p. 56]."

The mutual stimulus-response influences in the
adult-child interaction must be noted. With the
clinic-referred child the key interactions will
likely involve expressions of hostility and
anxiety. Moustakas (1955) compared clinic and
nonclinic referred children in play therapy and
found a major difference to be the frequency and
intensity of negative attitudes. The clinic-
referred child has a largely negative attitude
toward himself and others. He is motivated by
vague, diffuse, and undifferentiated feelings

of anger and fear. The most general expression

of these is a generalized anxiety.

19

Hypotheses

 

The present investigation attempts to provide
information about the issues discussed above. In par-
ticular the present study investigates a selection and
training program for undergraduates whereby they learn
to function therapeutically in play therapy sessions with
clinic-referred children. The selection process involves
the differentiation of two levels of potential, high and
low. There were also two training conditions: (1) those
trained in specific client-centered skills and (2) a no
training control group.

It is hypothesized that the undergraduates of
high potential (HP) for the role of play therapist will
be more attentive and sensitive to the feelings, needs,
wishes, and desires of clinic-referred children than will
be the undergraduates of low potential (LP). Further
the trained undergraduates will be more attentive and
receptive than will be the untrained control group of
undergraduates. Each of these statements is to be inves-
tigated in terms of Specific child behaviors, that is,

a method of sequential analysis will be used. These
general statements are explicated below in four hypothe-
ses. It is also hypothesized that there will be an
interaction effect of the potential factor with the
training effects. More specifically, it is predicted

that:

20

Hypothesis 1

 

When the child reacts to the undergraduate nega-
tively the HP group will be more accepting than will
be the LP group of undergraduates. Negative child
behaviors will be those in which the child moves
away from the undergraduate or moves against the
undergraduate through threats or other hostile
behavior. The greater acceptance by the HP group
will be indicated by greater interest and movement
toward the child. The LP group in response to
similar conditions will be less interested, attentive,
and will be less likely to move toward the child.

Hypothesis 2

 

When the child exhibits positive behaviors toward the
undergraduate, again the HP group will be more
accepting and will be even physically closer to the
child than will be the LP group of undergraduates.

Hypothesis 3

 

When the child is engaged in behavior showing no
interest in the undergraduate, e.g., in general play
or object mastery situations, the HP group will con-
tinue to be more attentive, will show more interest
and will be even physically closer than will the LP
group.

Hypothesis 4

 

In similar situations involving negative, positive,
and neutral behaviors, the trained group is also
predicted to show greater movement toward the
child than the untrained group.

Hypothesis 5

 

Potential will interact in a variety of ways with
training effects. It is predicted that the HP
trained will exhibit the greatest amount of
attention, closeness, and interest and the LP
untrained group will exhibit the least amount.

No predictions are made about the other interactions.

CHAPTER II

METHOD

Undergraduate Subjects

During the 1970-1971 academic year an advertise-
ment was placed in the Michigan State University news-
paper asking for sophomore and junior level volunteers
who were interested in learning about and practicing play
therapy techniques. More than 400 students attended dis-
cussion meetings and completed three inventories: the
Parent Attitude Research Instrument (Schaefer & Bell,
1958), a Sensitivity to Children projective questionnaire
(Stollak, Scholom, Kallman, & Saturansky, 1973), and a
Personality Questionnaire designed to assess general
"mental health." (See Appendices A, B, and C.)

The Parent Attitude Research Instrument is gen-
erally scored using a variety of scales. For the pur-
poses of this research a single bipolar scale was devised
with the poles being: child-centered and adult centered.
Each item of the PARI was looked at in terms of this

scale and scored accordingly.

21

22

The Sensitivity to Children questionnaire con-
sists of 16 parent-child problem or crisis situations
to which the subject writes what he would do and say in
response to the situation. Each response is scored on
four factors: conveying understanding by reflecting an
action or feeling of the child, acceptance of the feel-
ings of the child, a statement of the writer's feelings
about the child's actions or feelings and the statement
of a constructive alternative by the writer.

From the large number of potential subjects
scores on the Sensitivity to Children test (STC) were
first differentiated in terms of the highest and lowest.
This group was then examined for their scores on the
PARI. From the group reflecting the most sensitivity
to children were selected those displaying the most child-
centered attitudes. From the group reflecting the least
sensitivity to children were selected the group displaying
the most adult-centered attitudes. The former group was
then differentiated in terms of scores on the personality
questionnaire. Those with the scores reflecting good
adjustment were then selected to be the "high potential"
group. From the latter group were selected those with
scores on the personality questionnaire reflecting poor
adjustment. This group was then designated the "low
potential" group. Care was taken to include equal

numbers of males and females in each group. From this

23

subject pool now numbering 40, 10 from the HP group and
10 from the LP group were randomly chosen to comprise
the experimental group. The remaining 10 HPs and 10 LPs
comprised the control group. Each group originally
included equal numbers of males and females. Thus,
there would be 5 subjects in each cell, i.e., 5 experi-
mental HP females, 5 experimental HP males, 5 control HP
females, etc. There was some scarcity of appropriate
child referrals plus technical problems associated with
scheduling and video-taping. The 40 subjects were
eventually reduced to 26. Only the experimental LP
males cell contained 5 subjects. The smallest cells
were the control LP male and female cells with two

each.

Control Group

 

Control group members were interviewed and
informed about the selection procedures and the neces-
sity and importance of a control condition, i.e., to
evaluate the effects of training and feedback. They were
informed that at the end of the experiment they could par-
ticipate in a course similar to that received by the
experimental group members. During the course of the
study, however, they would receive no training or feed-
back concerning their play sessions. They were told they
would be observed each session through a one-way mirror

to insure that nothing "destructive" occurred during

24

the sessions, in which case their contact with the child
would be terminated. They were also given a list of
books on play therapy to read if they wished.

Experimental Group Training
Procedures

 

 

The 20 trainees were randomly assigned to one
of the three groups. These groups were approximately
equal in the number of HPs and LPs, males and females.
Groups met two hours weekly each quarter for two
academic years. A training therapistl led each group.
Group activities initially were devoted to explaining
the nature and purpose of the role trainees would be
asked to play in the weekly play sessions they would
have with a child. Various play therapy texts were
read and extensively discussed. For three hours video-
tapes of play therapy, already collected, were presented.
The details of the session were discussed including the
basis for the therapist's actions, a critique of his
interventions, and the doubts, fears, and concerns of the
trainees about acting in such a manner. Role playing was
used extensively with the emphasis on anxiety situations
and possible problems. Each trainee was then video-taped
playing with a "normal" child and such interaction

became the focus of group discussion. Considerable

 

lStollak, Scholom, and Berliner.

25

attention was devoted to the feelings of the trainees
and their fears about assuming the role prescribed.
The major focus, however, was the possible effects of

their actions on the behavior and emotions of the child.

Child Subjects

 

Referrals were obtained through Michigan State
University Psychological Clinic. Extensive assessment
was made of the child and his environment, home, and
school. Rigorous tests and rating scales were employed
involving research interests outside the scope of this
present study. The common denominator of all child
referrals would be that they were emotionally disturbed,
but nonpsychotic, between ages of 5 to 8. After obtain-
ing parental consent for participation in the research,
the children were randomly assigned to trained and
untrained undergraduates. Children were seen weekly
for a 30-minute play session. All sessions were observed
through a one-way mirror and experimental trainees
received individual feedback and supervision immediately

after the sessions.

Data Collected

 

Besides the personality test data obtained on
all the undergraduates and the clinical assessment data
on each child, the lst, 5th, 10th, and 15th sessions

were video-taped. For this project only video tapes

26

from the lst and 5th sessions were used. In some
instances due to scheduling problems this was not

possible and the 2nd and 6th sessions were video-taped.

Coding Adult-Child Interactions

 

The behavioral categories selected for coding
were very basic. These categories describing both adult
and child behavior included: (1) strong movement toward,
(2) movement toward, (3) neutral interactional behavior,
(4) movement away, and (5) movement against. These cate-
gories are defined more precisely-in Appendices D and E.
The major reason for the selection of basic categories
was to facilitate a true test of the hypotheses. If
differentiation of the potential factor from the train-
ing factor was possible it was predicted to be more
obvious in basic or gross movements toward and away
rather than in subtle nuances. Furthermore, if train-
ing was to be truly efficacious it would need to be
apparent not only to a sophisticated judge but to the
child.

A further reason for the five categories is that
the tape quality frequently made any subtle distinctions
impossible.

Procedure for Coding Behavior
Categories

 

 

For both the undergraduate behavior and the

child behavior, the 30-minute play sessions were divided

27

into a 7.5-second interval by means of an electronic bell.
During each interval two judges observed the child
behavior while two judges simultaneously observed the
undergraduate behavior. Only one category was scored
during each interval. Obviously more than one message
might be communicated during any one interval. In these
situations the task of the judges was to determine the
main thrust or impact of the adult or the child during
that interval in terms of the five categories.

Training Assistants in the
Coding Procedure

 

 

Unaware of the hypotheses of this study four
assistants were taught the categories to be scored.
This was followed by discussions and the observation of
select examples of play therapy sessions from video-
tapes not involved in this study. Definitions were
revised and made more clear. Assistants then committed
the definitions to memory. Following this approximately
six hours were Spent viewing and scoring play therapy
tapes with the author acting as expert. Child behaviors
were easier to code than adult behaviors. This was
partly due to the fact that the one T.V. camera used was
focused more consistently on the child. The assistants
rating child behaviors were judged competent when agree-
ment scores were consistently above 90%. The extensive

verbal behavior of the undergraduates complicated the

28

task of the two assistants coding adult behaviors. It
was more demanding to select the main thrust of the adult
behavior during the brief interval. Assistants rating
the adult behaviors were judged competent when agreement
scores with author as expert were consistently above 80%.
During the actual coding of the sessions the per-
centage of agreement between judges of undergraduate
behaviors ranged from .50 to .98 with a mean score of
.83. Percentage of agreement between judges of the
child behavior ranged from .70 to 1.00 with a mean of

.92.

Data Storing and Analysis

 

Child and undergraduate behaviors were rated
simultaneously with the sequence of the behaviors care-
fully preserved, that is, each event was stored in the
same sequence in which it took place. This eliminated
some of the guess work involved in deciding that a given
behavior was a stimulus or a response. For the purposes
of this investigation the child's behavior was treated
as a stimulus. In the data analysis it was paired first
with the adult behavior which occurred during that same
interval and secondly with the adult behavior which
occurred one interval later. It was felt that this type

of analysis of the first and fifth play sessions would

29

be adequate to investigate hypotheses about the per-
sonality characteristics of the undergraduates and the
effects of training.

It was expected that the child's behavior would
fall into one of five categories as would the adult's
behavior. Thus chi-square tables would be generated

and would constitute the principal analysis.

CHAPTER III

RESULTS

The design of the present study involved two
training conditions (training vs. nontraining), and two
potential conditions (high vs. low), sampled over two
‘sessions. Child and undergraduate behaviors were coded
simultaneously, each by a pair of independent judges,
in terms of five categories. One category was chosen
by a coder to typify activity emitted within an interval

of 7.5 seconds.

Child Behaviors

 

The originally planned chi-square analysis was
predicated upon obtaining sufficient variability in
child and adult behavior to generate a five-by-five
table. The obtained child behavior, however, was not
as variable as expected. Frequency tables were generated
from the scores of the two judges observing the behaviors
of the children. These frequencies were averaged over
the two judges. Percentages were then computed and on

this basis it was found that 93% of the child's time in

30

31

the initial sessionl fell into category three (the
neutral category describing behavior exhibiting no dis-
cernible warmth or interest in the adult present). The
range was 76.6 to 99.6%. On the second taped session2
the neutral category accounted for 91.8% of the child's
time with a range of 65.1 to 100%. The overall per-
centage for the two sessions was 92.5. Of the 45 play
sessions, 18 had more than 96% of child behavior accounted
for by the neutral category. It was obvious, then, that
the originally planned chi-square analysis was not feasible.
Therefore an alternative analysis of the child
behaviors was selected: The five original categories
were collapsed to three. Categories one and two were
combined to form a positive category; category three
remained as is; categories four and five were combined
to form a negative category. Of the 26 initial under-
graduates, 19 had a second session video-taped. The
data of these 19 sessions were then examined through an
analysis of variance (2x2x2x2 with a repeated measure
on the last two factors) of the two training conditions,
two potential conditions, two scores (the proportions

of positives and negatives over the totals), and two

 

l I o u o 0
As expla1ned above, the 1n1t1al sess1ons were
either session one or two.

2 . .
As expla1ned above, the second taped sess1ons
were either session six or seven.

32

sessions. The results of this analysis are given in
Table 1. As Table 1 indicates there were no significant
F ratios for any of the main effects. The interaction
effect of training x sessions x proportions was margi-
nally significant, but subsequent tests for simple
effects (Winer, 1971, p. 302) revealed no significant
comparisons.

A separate analysis of variance was performed
on data from the 26 initial sessions. Again, the results
indicated no significant effects.

With regard to the analysis of adult behaviors

 

it was decided to focus on those behaviors which occurred
in conjunction with neutral child behaviors. In order

to increase reliability, only those neutral child
behaviors were utilized in which both judges of the

child behaviors agreed. Adult behaviors were then
tabulated with the mean frequency of the two judges of
adult behaviors used in the analyses. On this basis the
analyses account for 78.6% of the behavior in the first
session, 76.7% of the behavior in the second session,

and 77.8% over all with a range of 42.6 to 99.5%.

The analysis was performed initially on adult'
behavior recorded simultaneously with the neutral child
behavior. A second analysis was performed on behavior
of the adults recorded one interval after the neutral

child behavior. It was found that these two approaches

33

TABLE 1

Analysis of Variance of Positive and Negative
Child Behaviors

 

 

Source of Variance df MS
Training (A) 1 .00459
Potential (B) 1 .00740
A X B 1 .00008
Error 15 .00400
Sessions (C) 1 .00069
A X C 1 .00005
B X C 1 .00055
A X B X C 1 .00416
Error 15 .00381
Scores (D) 1 .00256
A X D 1 .00097
B X D 1 .01782
A X B X D 1 .00399
Error 15 .00676
C X D 1 .00393
P X C X D 1 .01152
B X C X D 1 .00027
A X B X C X D 1 .00002
Error 15 .00282

Hiakaw

PWEFJH

APOA A

00
mm

A Ab.“
(“HOW

 

34

yielded approximately the same results. Tables in this
section present adult behaviors recorded simultaneously
to neutral child behaviors. Tables of adult behaviors
recorded one interval after neutral child behaviors can
be found in Appendices F and G.

The behavior of the 19 undergraduates who were
video-taped for the first two sessions was submitted to
a 2x2x2x2 analysis of variance with a repeated measure
on the last two factors. Data for the analysis were the
proportions of frequency of category one divided by the
total frequency of behaviors per session and the pro-
portions of frequency of category five divided by the
total frequency of behaviors per session. Table 2 pre-
sents the results of this analysis. Simple effects tests
were performed for the significant interactions.

Table 2 shows a significant main effect for pro-
portions. This F ratio reflects the much greater fre-
quency across all the groups of positive behaviors than
negative behaviors. There is also a significant inter-
actions effect of training x proportions. The relevant
cell means for this interaction are presented in Table 3.
A simple effects test on these data indicates that the
trained group exhibited significantly more strong posi-
tive behaviors than the untrained group. There were no

significant differences on the negative behaviors.

35

TABLE 2

Analysis of Variance of Adult Behaviors in
Categories One and Five

 

 

 

Source of Variance df MS F
Training (A) 1 .01628 2.61
Potential (B) 1 .00267 < 1
A X B 1 .00718 1.20
Error 15 .00623
Sessions (C) 1 .00221 < 1
A X C 1 .00129 < 1
B X C 1 .00058 < l
A X B X C 1 .01102 2.68
Error 15 .00411
Scores (D) 1 .35600 40.28**
A x D 1 .04687 5.30*
B X D 1 .01661 1.88
A X B X D 1 .00054 < 1
Error 15 .00884
C X D 1 .00079 < l
A X C X D 1 .00317 < 1
B X C X D 1 .01127 2.28
A X B X C X D 1 .00025 < 1
Error 15 .00495

*
p < .05

**

p < .001

36

Appendix F gives the results of a similar analysis of
adult behaviors occurring one interval after a neutral

child behavior.

TABLE 3

Mean Positive and Negative Proportions for
the Training Groups

 

 

Positive Negative
Trained .809 .014
Untrained .473 .101

 

Table 4 was generated in a similar manner as
Table 2. It shows the results of the 2x2x2x2 analysis
with repeated measures again on the last two factors.

In this case the last factor is the proportions achieved
by summing category one and two responses and dividing
by the total number of responses and by summing category
four and five responses and dividing by the total number
of responses.

Most obvious in Table 4 is the massive main
effect for proportions, which again reflects the much
greater frequency of positive behaviors rather than
negative behaviors for all the groups. There is also
a significant main effect for sessions and significant

interaction effects for training x sessions and training

37

TABLE 4

Analysis of Variance of Adult Behaviors
Positive and Negative

 

 

 

Source of Variance df MS F
Training (A) 1 .05443 4.35
Potential (B) 1 .01578 1.26
A X B 1 .01039 < 1
Error 15 .01250
Sessions (C) 1 .01006 9.67:*
A X C 1 .00751 7.22
B X C 1 .00200 1.92
A X B X C 1 .00013 < 1
Error 15 .00104
Scores (D) 1 10.59470 324.23:**
A X D 1 .15360 4.70
B X D 1 .00054 < l
A X B X D 1 .00013 < 1
Error 15 .03267
C X D 1 .01887 4.46*
A X C X D 1 .02838 6.71
B X C X D 1 .00760 1.79
A X B X C X D 1 .00046 < 1
Error 15 .00423

*

p < .05
**

p < .01
***

p < .001

38

x sessions x proportions. Table 5 presents the relevant
cell means for this latter interaction. A simple effects
test on these data indicates that the trained group
exhibited significantly more positive scores than

the untrained group. This difference is most obvious

on session one and not so dramatic on session two, as
indicated by the training by sessions interaction effect.
The test for simple effects on the threefold interaction,
however, decreases the F ratio for the effect of sessions
below the level of significance because of the magnitude
of the pooled error term. Appendix G gives the results
of a similar Analysis of variance of adult behaviors

occurring one interval after a neutral child behavior.

TABLE 5

Mean Positive and Negative Proportions for the Two
Training Groups in Two Different Sessions

 

 

 

Positive Negative
Trained
Session One 1.858 .057
Session Two 1.849 .079
Untrained
Session One 1.426 .175

Session Two 1.668 .119

 

39

Analyses of variance were run separately on the
26 initial sessions. The results were essentially the
same as those found above. The results of the analysis
of variance of the adult behavior on the initial 26
sessions are found in Appendix H.

In summary, then, the results indicate that
trained undergraduates exhibited more positive behaviors
in conjunction with neutral child behaviors than did
untrained undergraduates. There were no discernible
differences among the undergraduates in terms of nega-
tive behaviors. It was further found that potential,
as measured by three paper and pencil assessment pro-
cedures, was not a significant factor.

With regard to child behaviors, it was found
that the great majority of these reflected little or no
interest in an interaction with the undergraduate.
Furthermore, in the two sessions video-taped and coded,
there were no significant differences with respect to

positive versus negative child behaviors.

CHAPTEP IV

DISCUSSION

Overview

The role of play therapist to be learned in the
present research was formulated in general from client-
centered theories of play therapy and from selected
research studies in parent-child relations. Potential
was assessed in terms of affinity for that role and for
the skills to be developed. The assessment of potential
required more than the mere measurement of attitude.
At the same time the process needed to be efficient and
practical for a large number of individuals. For these
reasons three paper and pencil tests were utilized.

This study also made use of an interactional
methodology which attempted to account for the stimulus
and response qualities of each unit of behavior. It
was expected that a true interactional analysis would
best yield the necessary information about an adult's
mastery of a role requiring a close relationship with

a troubled child.

40

41

The results of the study do provide some infor—
mation relevant to the issues approached by the hypothe-
ses. The discussion begins with some of the methodologi-
cal problems which developed. Then the factor of potential
is discussed. Next the undergraduate behavior is viewed
in terms of the training factor. Finally results are
summarized in terms of the specific hypotheses. General
implications of the findings are discussed with the focus
on future directions in the selection and training of

para—professionals.

Methodological Considerations

 

Interactional analysis, like the human behaviors
it seeks to codify, is necessarily complex. Any single
unit of behavior has both response features and stimulus
features. Most research in human dyadic behavior has
either employed a confederate as a pre-determined
stimulus or has utilized a correlational analysis.

This latter approach clearly suggests interactional
influences but does not allow direct statements to the
effect that when one party was engaged in certain
behaviors the other party was engaged in certain stated
behaviors. But a true test of the hypotheses in this
study would seem to require the ability to make these
kinds of direct statements. For example it was expected
that when a child was being pleasant and cooperative most

adults generally would be accepting, whether trained or

42

not. It was predicted that training factors and potential
factors would be more apparent at key moments, such as
when the child was being overtly aggressive, hostile,

and anxious or very warm and affectionate. Correlational
studies are not specific enough for these questions. Nor
does a correlational approach specify which behaviors

were antecedent or consequent to a key event.

This, then, was the intent of the method employed,
to note certain child behaviors seen as crucial and to
observe the adult behaviors preceding or consequent to
such child behaviors. Child behaviors, however, did not
scatter across the five categories as predicted. As
noted earlier, over 92% of the child behavior was coded
in the neutral category describing activity with no dis-
cernible attitude toward the undergraduate other than
ignoring him. The remaining child behavior not in the
neutral category was insufficient for an interactional
analysis. Again, as noted earlier, in 18 of 45 play
sessions, less than 4% of the child behavior did not
fall into the neutral category.

Some obvious explanations for this lack of
variability need to be considered. The fault may have
been with the judges, the categories, the play setting,
or the lack of previous contacts between the undergraduates
and the children participating. The judges of the child

behaviors were experienced elementary school teachers

43

with bachelor's degrees. Each had spent several hours
in training. They could not observe the scoring pro-
cedures of each other. The agreement rate, however,
was very high. In retrospect it appears that adequate
procedures in the training of the judges were employed.

With regard to the play setting it was obviously
a strange place to the child. Furthermore, as there had
been no previous contacts, the undergraduate was cer-
tainly a stranger to the child. Neither of these two
factors were controlled for although they can be pre-
sumed to have influenced the variability of the child's
behavior.

On the other hand, with such limited distribution,
it is clear that the child categories were inadequate.
One explanation as to why this was not detected is that
video tapes from other play therapy sessions were used
in training the judges, tapes of children not all "clinic—
referred," and at various stages of involvement in play
therapy. These tapes were repeatedly stopped at certain
intervals for discussion and clarification. Examples of
each category were isolated. Throughout the training
scatter was assumed to be taking place without a check-
point to challenge this assumption.

The categories are essentially approach-avoidance.
The neutral categories were for the sake of the judges

who were in doubt as to which direction the movement

44

occurred. One theoretical foundation for the selection
of the categories was the premise of doublebind theorists
(Bateson, 1956) that two people in a room are always
communicating something about their relationship. In
reality, then, the child was not actually neutral.
Ignoring the only other person in the room or showing

no interest whatever in interacting is a generally nega-
tive stance. It also suggests considerable ambivalence
and perhaps anxiety.

Moustakas (1955) compared disturbed children with
well-adjusted children in initial play sessions. He
noted that the disturbed group had a largely negative
attitude toward themselves and others and that they were
motivated by diffuse, undifferentiated, and unfocused
feelings of anger and fear. These feelings were expressed
outwardly as a generalized anxiety. He found that as

\
therapy progressed negative attitudes became more focused
and more specific. In a later work, Moustakas (1959)
describes the internal dynamics of a disturbed child
entering therapy:

At the root of the child's difficulty is the sub-
mission and denial of his self. Somewhere along

the line of his growth and development he has given
up the essence of his being and the unique patterns
that distinguish him from every other person. The
growth of the self has been impaired because of his
rejection in important personal relationships. He
has been severely rejected by others and has come to
reject himself. He is cut off from vital self-

resources which would enable him to develop in
accordance with his own particular talent [p. 3].

45

It should have been no surprise, then, that with
so much of the child's behavior it was difficult to
determine his movement of approach or avoidance of the
adult present. This is in part a problem in determining
the intention of the child in a specific act. It is also
due in part to the child's incapacity for clear and posi-
tive direction. Rollo May (1969) sees intentionality
in a very broad sense--as the ability to mean things or
to care for--a capacity to point towards something or
someone outside oneself. His remarks seem relevant to
this discussion:
Does not intentionality give us a criterion for
defining psychological vitality? The degree of
intentionality can define the aliveness of the
person, potential degree of commitment and his
capacity, if we are speaking of a patient, for
remaining at the therapeutic task [p. 245].

Again, this is not to imply that the child is actually

neutral. Using May's terminology a disturbed child has

a low degree of intentionality. In order to differentiate

various levels of this intentionality accurately,

obviously more subtle and sophisticated categories

than those employed in this study are needed.

One remedy to the problem of distribution
encountered in this present study would be a much closer
analysis of each interval of the child's behavior by
"experts" reviewing the video tapes repeatedly until

some clear judgment about the intentionality of the

child's behavior was arrived at. It would be quite

46

time consuming, however. The great advantage would be
increased reliability without distorting the reciprocal
influences of one behavior upon another. Reliability
is, of course, always crucial. It is especially so in
making statements about two independent variables.

There is no disguising the fact that the child's
behavior was not distributed more or less equally across
the five possible categories. Such a result restricts
considerably the information available to test the
hypotheses. Nevertheless, it is strongly suggested
that the child behavior focused upon in this study
reflects a major area of therapeutic intervention with
children. Axline (1947) stated as the first of her eight
basic principles, "The therapist must develop a warm,
friendly relationship with the child, in which good
rapport is established as soon as possible [p. 73]."
This present study suggests that the impetus for such
a relationship is exclusively in the hands of the thera-
pist. Such a statement is obviously warranted by the
fact that 92% of the time in the initial sessions trained
judges could not detect any obvious interest in a
relationship with the adult present.

A wide variety of child behaviors was coded in
the neutral category. One child spent a great amount of
time working on a puzzle almost as if he was alone in

the room. Another child behaved similarly with crayons.

47

In another example, one child spent almost an entire
session expressing hostility toward the T.V. cameraman
behind the window. What these behaviors have in common
is that the undergraduate was almost totally ignored.

In these diverse situations the child is in a passive
stance as regards any interaction with the undergraduate.
This is not to say that we can accurately apportion
reciprocal stimulus—response effects. It does seem,
however, that those interactional situations have been
isolated where the child is exhibiting no high stimulus
behaviors; and the impetus for any interactional develop-
ment resides clearly in the actions of the undergraduate.

Other methodological problems developed which
are frequent in research focusing on naturalistic obser-
vations. It is difficult to assess how the presence of
the equipment may have influenced the results. It was
obvious that some of the children were acutely aware of
the camera. Their behavior in such instances, however,
did not seem significantly different from their behavior
when untaped; the camera equipment seemed to serve simply
as an alternate focus or target.

The undergraduates knew when they were being
taped. Those in the training group appeared more con-
cerned about the pressure to behave appropriately. This
pressure, however, seemed to be operative to a lesser

degree even during untaped sessions.

48

How any of these issues influenced the results
is not definitively known. The firm impression is that
the two sessions which were video-taped were represen-
tative of the ensuing play sessions.

Other mechanical factors at times seemed promi-
nent. The study was conducted using one camera. Fre-
quently the child and the undergraduate were not visible
at the same time. Sometimes the sound quality was
inadequate; at other times distortions blurred the
quality of the video tapes. These factors, however,

seemed to occur in a random fashion.

Potential

 

As assessed and differentiated by procedures in
this study, potential was in no way a factor of any sig-
nificance. Obviously such an outcome requires an
explanation. The assessment procedures appear to be
rather extensive. The sample population of over 400
seems large.

The Parent Attitude Research Inventory has been
extensively used. The major criticism regarding its
validity involves the tenuous connection between atti-
tude and the prediction of actual behavior. This has
been the albatross around the neck of all attitude scales.
For this reason a greatly simplified scoring procedure
was adopted as described earlier. In addition the

Sensitivity to Children questionnaire was utilized.

49

This questionnaire asks the individual to write down
what he would do and what he would say in problem situ-
ations involving children and adults. The third measure
of potential was a Personality Questionnaire assessing
the general "mental health" of the undergraduate. Such
self-descriptive tests give a limited perspective on
one's personality configuration. They have been shown,
however, to measure a valid, if only surface level of
emotional conflict (Leary, 1956). Relative freedom of
emotional conflict is a recurring quality mentioned by
many researchers as essential to a competent therapist.

The use of Occam's razor requires that the
simplest explanation be considered first. The simplest
explanation in this case is that potential, as related
to the specific task, does not matter. It must be noted
that the data in this present study do not directly
refute this explanation.

There are strong suggestions, however, that other
explanations account for the results more adequately.
In the first place, as reviewed in an earlier section,
the differentiation of role potential and aptitude has
been a highly successful area of research for several
generations of behavioral scientists. If potential for
the role of play therapist is irrelevant to the mastery
of that role, such a phenomenon strikes a dissonant note

in a lengthy chorus of research.

50

In the second place, Stollak, Green, Scholom,
Schreiber, and Messe (1973) focused on the same selection
and training program as that of the present study and
did find some significant effects for the potential
factor. They did not use an interactional analysis.
Undergraduate behavior was viewed in terms of empathy
defined operationally in three scales: communication
of acceptance, allowing the child self-direction, and
involvement. A rating was made in each scale during
each two minutes of the video-taped play sessions. The
ratings also included tapes from sessions 11 and 15.

As regards potential, in almost all instances differences
between HP and LP undergraduates were in the expected
direction, with the HP group evidencing greater communi-
cation of acceptance, greater allowing of self-direction,
and more involvement. As regards statistical signifi-
cance, however, most of the variance was associated with
the scale: communication of acceptance. On session 6

on this scale the HP and LP trained groups were not sig-
nificantly different. On the other hand, of the untrained
groups, the HP group showed higher levels of acceptance
than the LP group. On session 15 on the same scale the
HP group again showed higher levels of acceptance than
the LP group. Potential did not generate any significant
effects in either session 1 or 11. These results give

limited support to the importance of potential as a

51

factor. These minimal results also suggest that sample
size has reduced the yield of significant data. This
obviously applies to the present study as well as that
of Stollak et a1. (1973).

If potential cannot be appropriately dismissed
as irrelevant to the role of play therapist, further
explanations need be explored to account for the lack
of results. The small sample size has been mentioned.
Another explanation may be that potential was poorly
differentiated.

Paper and pencil tests were employed in this
study because selection procedures have to be practical.
With over 400 volunteers some kind of natural observation
procedures seem extremely unwieldy. Goodman (1972)
developed a structured group method that utilized suc-
cessive sets of dyadic interaction as performance samples
for rating (GAIT: Group Assessment of Interpersonal
Traits). He then paired fifth and sixth grade boys with
male undergraduate students in a companionship program.
Of 179 applicants, 56 or 31% were rejected. This
selection process extended, however, over nearly four
years. This was not practical in the present case.

There are strong suggestions, however, that only
natural observation will yield the information required.
Kemp (1962) conducted a study with open and close-minded

trainees in a quasi-therapy training program. He found

52

that the close minded seemed to respond to the academic
procedures and improvement was noted in paper and pencil
tests. The trainees learned to give back to the teacher
expected responses as long as they had time to consider
and reflect. In the quasi-therapy situations, however,
their responses underwent no significant changes relative
to their initial behavior. Kemp concluded that when they
did not have the opportunity to reflect and consider
their answers their customary and habitual behavior
recurred.

Applying this result to the present study, the
scores on the Parent Attitude Research Inventory and the
Sensitivity to Children test may be more a function of
the student's general knowledge and SOphistication and
not a true tap of their basic attitudes. In a similar
manner their responses to the self—description "mental
health" test may reflect similar influences, for example,
a desire to give responses perceived as suitable for
entrance to the project. This could have been remedied
by a more standard mental health adjustment test with
built-in validity scales such as the MMPI.

Another plausible explanation for the results
in this study is that perhaps all of the undergraduates
were of high potential with regard to the prescribed role.
Perhaps the assessment procedures differentiate two

levels of a high potential quality such that the outcome

53

measures could not meaningfully detect. This is a very
speculative suggestion, of course; it might similarly
be proposed that all the undergraduates were on two
levels of a very low potential quality. There is some
supporting evidence, however, for the former explanation.
Swensen (1971) reviewed some of the literature
on the personality of the successful therapist. He sum-
marized that the crucial element in the therapist's con-
tribution to therapeutic success is the therapist's
commitment to the client. He noted the repeated use of
such words as "care," "concern," and "commitment" in
the research studies. Swensen concluded that commitment
is so important that we ought to be determining which
kinds of people are able and likely to make commitments
to which other kinds of peOple.
In the present study over 400 people volunteered
to give a certain amount of time on a weekly basis for
a two-year period. This is clearly an enormous invest-
ment of time and energy. Among those selected some few
did drop out. The remaining were quite possibly a highly
motivated and committed group. Perhaps, then, in this
study a natural selection process took place in which
the kinds of people who were able to make commitments
to disturbed children did so. If commitment is a
crucial element in determining potential, then, this

study very possibly employed undergraduates of high

54

potential and no undergraduates of low potential. At
least it can be suggested that commitment, uncontrolled
for in this present study, has contaminated the effects
of potential. This may account for the nonsignificant
results in this study and for limited results obtained
by Stollak et a1. (1973).

If, because of the commitment factor all the
subjects in the present study could be classified as
high potential, which explains the poorer functioning
of the untrained groups? It is true that the untrained
group was functioning less effectively than the trained
group. This must be labeled, however, as a relative
dysfunctioning. It is not known how this untrained
group would compare with a random sample from the general
population of undergraduates.

Other studies have suggested that untrained
undergraduates in therapy or quasi-therapy situations
do not generally exhibit even minimal therapeutic skills.
Berenson, Carkhuff, and Myrus (l966) utilized under-
graduate college students in a paraprofessional role.
They found that when cast in a helping role the college
students not involved in the training program were
functioning at less than minimally facilitative inter-
personal levels. Linden and Stollak (1969) found simi-
lar results. In answer to the question, could college

students figure out for themselves a sensitive way of

55

dealing with children, they responded in the negative.
In both of these studies volunteers were used, that is,
people who committed themselves for a period of time to
dealing with others constructively in a helping role.
Linden and Stollak concluded:
Probably some people are brought up to be more
empathic than others, but the ability to communi-
cate it, which is essential to a helping relation-
ship, must be taught. Whether or not one can pro-
ject or "put himself into another's shoes" is only
part of the effective interpersonal sensitivity.
Another major variable is the communication of
whatever empathy one feels [p. 217].

The implication is that under certain circum-
stances even highly motivated and well-adjusted indi-
viduals will not exhibit some critical interpersonal
skills unless these have been taught. The situation
is not unlike that of a person with a real capacity
for learning foreign languages. Even the one with

the greatest potential cannot speak even a word until

he has been taught.

Training

It is clear in this study that training effects
yielded results in the direction predicted. The trained
undergraduates clearly exhibited more positive behaviors
than did the untrained undergraduates. This was seen
in two outcome measures. In the first of these the
two categories of positive behaviors were grouped

together. On this measure the differences between the

56

trained group and the untrained are significant on

both the first and second video-taped sessions. The
differences were more obvious and dramatic on the first
session than on the second session. In other words the
untrained group was less different by the second session.
Nonetheless, the difference remained significant.

The second outcome measure focused only on the
category of strong positive movement toward the child.
On this measure again the trained undergraduates
exhibited significantly more of these behaviors on
both of the sessions than did the untrained undergraduates.
There were no meaningful differences between the two
sessions.

There is a suggestion, then, that the untrained
group was responding somewhat more positively by session
two than they had in session one. It should be noted
that session two was actually four therapy sessions
later. This movement, however, is only a suggestion
as the test for simple effects of sessions showed no
significance due to the magnitude of the pooled error
term. Ideally this suggestion should have been followed
up in the later sessions.

With regard to negative adult behaviors, that
is, behaviors of moving away from the child or of moving
against the child, there were no discernible differences

between the trained and the untrained undergraduates.

57

Both groups exhibited far more positive type behaviors
than negative type behaviors.

It was predicted that the low potential groups
and the untrained groups would, in fact, exhibit more
negative behaviors than the high potential groups and
the trained groups. The results do not support this
hypothesis.

The neutral category of the adult carried a
different "valence" than the neutral category of the
child. By implication if not by definition the adult
went to the playroom to be "with" the child. The child
was not in the room necessarily to be "with" the adult.
He may not have known exactly why he was in the room
with the adult. If the adult behavior was such that
one could not determine any obvious interest in the
child other than simply being there, then this behavior
exhibits a greater and more obvious distancing than
the child's apparent lack of interest in the adult.

If the adult categories be viewed as a continuum
it could be said that the trained undergraduates behavior
fell more often and more characteristically toward the
category of strong, positive movement toward the child,
exhibiting noticeable interest and warmth. The untrained
undergraduate's behavior fell more often and more char-

acteristically in the direction of the neutral category.

58

Stollak et a1. (1973) utilizing the same popu-
lation likewise found significant results on the train-
ing factor. They found that there were significant
main effects for training in every session. A training
by category interaction was analyzed and simple effects
tests indicated that the effect of training was most
pronounced and always significant for the category:
communication of acceptance.

Reif and Stollak (1971), using a different popu-
lation with different categories and methodology, found
a similar importance in training effects. They helped
undergraduates to become more sensitive to "normal"
children over 20 play sessions. The experimental group
of undergraduates exhibited warmth to a greater degree
than the control group only in the latter sessions.
Warmth, insofar as it was included in the notion of
strong, positive movement in this present study was
evident from the very beginning in the trained group.

Truax and Carkhuff (1967) found that in research-
ing the conditions necessary for successful therapy,
found that the averages rather than the peaks and low
points were related to outcome. This led them to
suggest that the therapist ought not to be too cautious
about risking an occasional misunderstanding if the
effort might eventuate in greater understanding. In

this frame of reference the lack of noticeable

59

differences between the two training conditions with
regard to negative adult behaviors does not contradict
the evidence that training procedures have facilitated

the interpersonal skills of those who were trained.

Summary

This summary is given in terms of the hypothesis.
The primary focus in this study has been on the potential
factor. The hypotheses were presented with the potential
factor expected to be the main source of variance. This
did not eventuate. Potential in fact accounted for
very little of the variance in any of the analyses.
This was discussed from a variety of perspectives. It
may be that only high potential undergraduates par-
ticipated in this project. This might be explained by
the apparently high levels of commitment reflected by
volunteering for a two-year study. A possible "com—
mitment" variable was not, in fact, allowed or con-
trolled for. Another explanation was that paper and
pencil tests give subjects a chance to reflect and
consider and, thus, do not tap any roots of acceptance
or comfort with children but may merely reflect a level
of general knowledge and sophistication about what
"experts" are writing on the topic.

Such explanations, however, are quite specula-

tive. There was no statistical evidence yielding results

60

about any of the hypotheses concerned with the potential
factor. As such then, these hypotheses remain untested.

The fourth hypothesis concerned the training
condition. It was predicted that in situations involv-
ing negative, positive, and neutral child behavior the
trained group would exhibit greater interest, attention,
and would move closer to the child than would the
untrained group. It was found that there were insuf-
ficient negative and positive child behaviors for any
statistical analysis. The child's behavior fell most
often into the neutral category, thus, reflecting no
interest in interacting with the therapist. This was
discussed first as an inadequacy in the categories.
Secondly, however, it was felt that an important aspect
of the therapeutic relationship was being accurately
focused upon.

It was found that the subjects in the trained
group to a significant degree showed themselves to be
more attentive and to be ready and willing to be closer
to the child than were the subjects in the untrained
group. There was a suggestion that the untrained group
might be gaining some facility at being more attentive
over time. From the beginning, however, the trained
group very clearly demonstrated an increased capacity
for closeness to the clinic-referred children that was

not seen in the untrained group. These findings were

61

discussed in terms of other similar studies. These
more frequently used outcome measures which might be
susceptible to the mimicking of the letter and missing
the spirit of a particular approach, e.g., reflection
of verbal content, compliance clarified, reflection of
motor activity, etc. In the present study the trained
group exhibited behaviors more basically or grossly
reflecting comfort, interest, acceptance, and warmth.

Implications for Future
Research

 

This study used a combination of didactic and
experiential procedures in the training process. There
is ample evidence to suggest that both inputs are
essential in the training of a person for professional
or paraprofessional contact with "patients." This study,
however, and most of the studies employing paraprofes-
sionals in therapeutic relationships with children, look
beyond the actual placement of such trainees in clinics.
Some will be in such positions, more will be in teaching
positions with children and most will become parents.
Clearly skills learned to facilitate a child's growth
and development on the basis of one—half hour a week of
contact can be invaluable to those whose contact with
a child is much more lengthy. This entire area calls
for greater research using tighter controls. Tax-paying

citizens have shown some willingness to recognize the

62

need for institutions training therapists. Programs
for training future and present parents likewise will
require widespread moral and financial support. If
"communicated empathy" is a skill that must be taught
even to those with high potential, perhaps it should be
a "course" required for high school graduation along
with certain English courses and other communication
skills.

Ideally, students trained in this study should
be observed in interaction with their own children ten
years from now. These could be compared with the control
subjects interacting with their children. It is obvi-
ously very difficult to arrange such follow-up. It is
also obvious that much more hard evidence is required
before there can be a national commitment to such train-
ing.

On another level individual commitment as a
major aspect of potential deserves research attention.
Commitment was hypothesized to be an uncontrolled
variable in the present study. If different levels of
commitment could be differentiated then the results of
different training conditions at these different levels
could be examined. In retrospect it does not seem
possible to get 400 volunteers for a two-year project

without getting people who are quite committed. The

63

notion of an uncommitted volunteer seems contradictory
except where rewards, e.g., money, exist.

Another approach could make participation a
requirement. For example the parents of first graders
might be obligated as might a parent seeking custody
of the children in a divorce case. In a manner such
as these a sample population might be researched where
a variety of levels of commitment might exist.
Researchable questions immediately come to mind. Can
training elicit skills from those who dislike children?
Can training increase commitment in those forced to
attend sessions? How does commitment relate to other
qualities associated with therapeutic skills?

There are a variety of therapeutic approaches
and methods of intervention, each with specific require-
ments and capacity. This study attempted to select
those with the potential for mastering the role of a
nondirective play therapist with disturbed children.
This selection process reflected a lack of differen-
tiation of the potential variable. Such results are
discouraging. A more fruitful approach might be
naturalistic observation, i.e., placing each candidate
in relation to a child and evaluating his behavior
according to certain scales. This can be a difficult
procedure for a large sample population. But paper and

pencil tests do not seem to give enough information.

64

As long as the subject has a chance to reflect and con-
sider his responses it is difficult to measure actual
levels of comfort/discomfort, attention quality, warmth,
etc.

Another approach suggests the use of more than
paper and pencil tests but does not involve actual
interaction with the population to be treated. Hefele
et al. (1970) suggest the existence of a "good guy"
factor, a unitary measure that tends to identify the
therapeutically facilitative factor. Using factor
analysis he suggests that the individual most likely to
be a good therapist is a person functioning very well
in many areas, i.e., social-interpersonal, sexual-marital,
educational-vocational, and child-rearing.

Future research might profitably pursue two
lines of interest. Is it likely to lead to greater
efficiency if we select more cautiously those trained
as paraprofessionals? Will it be productive to train
all parental figures whatever their aptitude?

An interactional methodology was devised for
this research. This also has yielded results with
implications for future research. Extraneous factors,
however, restricted the testing of these procedures.
Nevertheless, it does seem viable to code interactional
events simultaneously and then to compare and correlate

simultaneous, antecedent, and consequent events. The

65

basic process was well described by Moustakas and
Schalock (1956) and tested by them (1955). But they
selected certain behaviors having "high stimulus proper-
ties." Only these few were actually involved in any
interactional or sequential analysis. The method needs
to be advanced to where any behavior in a dyad can be
discussed and compared with any other behavior. Perhaps
it would be more accurate to compare groups of behaviors.
This is not to suggest that cause and effect will finally
be adequately determined. As stated earlier each behavior
of each party has both response qualities and stimulus
qualities. It would be very useful if the sequential
flow of mutual influences could be accurately followed
and charted. Perhaps the various error terms of antece-
dent, simultaneous and consequent pairings of dyadic
behaviors would suggest which behaviors in a sequence
influenced which other behaviors. It may well be that
the stimulus qualities of a particular unit of behavior
may be felt several intervals later. Again patterns

of response may be highly individualized.

In this present study statistical comparisons
and interactional analyses were done without the use of
a computer. It is obvious from the kinds of questions
that emerge that the process needs to be computerized.

The development of a true interactional methodology has

66

implications for all dyadic research. The struggle to
understand the real nature of any therapeutic relation-

ship would be greatly facilitated.

APPENDI CES

APPENDIX A

PARENT ATTITUDE QUESTIONNAIRE

APPENDIX A

NAME

 

AGE

 

SEX

 

DATE

 

QUESTIONNAIRE

On the following pages are statements about parents
and children. Indicate your opinion by drawing a circle
around the "A" if you strongly agree, around the "a" if
you mildly agree, around the "d" if you mildly disagree,
and around the "D" if you strongly disagree with the
statement. It is best to work rapidly. Give your first
reaction. If you read and reread the statements, it tends
to be confusing and time-consuming.

There are no right or wrong answers, so answer
according to your own opinion.

It is very important that all questions be
answered, even if they don't seem relevant to your imme-
diate life. Many of the statements will seem alike,
but all are necessary to show slight differences of
opinion.

67

68

QUESTIONNAIRE: page 2 Strongly Mildly
Agree Agree

1. Children should be
allowed to disagree
with their parents if
they feel their own
ideas are better. A a

2. It's best for the child if
he never gets started won-
dering whether his
mother's views are right. A a

3. Parents should adjust to
the children some rather
than always expecting
the children to adjust
to the parents. A a

4. Parents must earn the
respect of their chil-
dren by the way they
act. A a

5. Children would be
happier and better
behaved if parents would
show an interest in
their affairs. A a

6. Some children are just
so bad they must be
taught to fear adults
for their own good. A a

7. Children will get on any
woman’s nerves if she has
to be with them all day. A a

8. One of the worst things
about taking care of a
home is a woman feels
she can't get out. A a

9. If you let children talk
about their troubles they
probably will end up com-
plaining even more. A a

Mildly
Disagree

Strongly
Disagree

69

QUEST¥0NNAIRE Strongly Mildly
(continued) page 3 Agree Agree

10. There is nothing worse
for a young mother than
being alone while
going through her
first experience with
a baby. A a

11. Most children should be
toilet trained by 15
months of age. A a

12. The sooner a child
learns to walk the
better he's trained. A a

13. A child will be grate-
ful later on for strict
training. A a

14. A mother should make it
her business to know
everything her children
are thinking. A a

15. A good mother should
shelter her child from
life's little diffi-
culties. A a

16. There are so many things
a child has to learn in
life there is no excuse
for him sitting around
with time on his hands. A a

17. Children should be
encouraged to tell
their parents about
it whenever they feel
family rules are
unreasonable. A a

18. A parent should never
be made to look wrong
in a child's eyes. A a

19. Children are too often
asked to do all the
compromising and
adjustment and that
is not fair. A a

Mildly
Disagree

Strongly
Disagree

70

QUESTIONNAIRE Strongly Mildly
(continued) page 4 Agree Agree

20. As much as is
reasonable, a parent
should try to treat
a child as an equal. A a

21. Parents who are inter-
ested in hearing about
their children's par-
ties, dates, and fun
help them grow up
right. A a

22. It frequently is neces-
sary to drive the mis-
chief out of a child
before he will behave. A a

23. Mothers very often feel
that they can't stand
their children a
moment longer. A a

24. Having to be with chil-
dren all the time will
probably give a woman
a feeling her wings
have been clipped. A a

25. Parents who start a child
talking about his
worries don't realize
that sometimes it's
better to just leave
well enough alone. A a

26. It isn't fair that a
woman has to bear just
about all the burden of
raising children by
herself. A a

27. The earlier a child is
weaned from its
emotional ties to its
parents, the better it
will handle its own
problems. A a

28. A child should be weaned
away from the bottle or
breast as soon as possible. A a

Mildly
Disagree

Strongly
Disagree

71

QUESTIONNAIRE Strongly Mildly

(continued) page 5 Agree

29. Most young mothers are
bothered more by the
feelings of being shut
up in the home than by
anything else. A

30. A child should never keep
a secret from his
parents. A

31. A child should be pro-
tected from jobs which
might be too tiring or
hard for him. A

32. Children who don't try
hard for success will
feel that they have
missed out on things
later on. A

33. A child has a right to
his own point of View
and ought to be allowed
to express it. A

34. Children should never
learn things outside the
home which make them
doubt their parents'
ideas. A

35. There is no reason
parents should have
their own way all the
time any more than that
children should have
their own way all the
time. A

36. Children seldom express
anything worthwhile;
their ideas are usually
unimportant. A

37. If parents would have
fun with their children,
the children would be
more apt to take their
advice. A

Agree

Mildly
Disagree

Strongly
Disagree

72

QUESTIONNAIRE Strongly Mildly
(continued) page 6 Agree Agree

38. A wise parent will
teach a child early
just who is boss. A a

39. It's a rare mother who
can be sweet and even-
tempered with her chil-
dren all day. A a

40. Children pester you with
all their little upsets
if you aren't careful
from the first. A a

41. A wise woman will do
anything to avoid being
by herself before and
after a new baby. A a

42. Children's grades in
school are a reflection
of the intelligence of
their parents. A a

43. It is more effective to
punish a child for not
doing well than to reward
him for succeeding. A a

44. Children who are held to
firm rules grow up to
be the best adults. A. a

45. An alert parent should
try to learn all her
child's thoughts. A a

46. Children should be kept
away from all hard jobs
which might be dis-
couraging. A a

47. Parents should teach
their children that the
way to get ahead is to
keep busy and not waste
time. A a

Mildly
Disagree

Strongly
Disagree

73

QUESTIONNAIRE Strongly Mildly
(continued) page 7 Agree Agree

48. A child's ideas should
be seriously considered
in making family
decisions. A a

49. The child should not
question the thinking
of the parent. A a

50. No child should ever set
his will against that of
his parents. A a

51. Children should fear their
parents to some degree. A a

52. When you do things
together, children feel
close to you and can
talk easier. A a

53. Children need some of the
natural meanness taken out
of them. A a

54. Raising children is a
nerve-wracking job. A. a

55. One of the bad things
about raising children
is that you aren't free
enough of the time to do
just as you like. A a

56. The trouble with giving
attention to children's
problems is they usually
just make up a lot of
stories to keep you
interested. A a

57. Most women need more time
than they are given to
rest up in the home after
going through childbirth. A a

58. A child never sets high
enough standards for
himself. A. a

Mildly
Disagree

Strongly
Disagree

74

QUESTIONNAIRE Strongly Mildly

(continued) page 8 Agree

59. When a child does
something well we can
start setting his
sights higher. A

60. It is a mother's duty
to make sure she knows
her child's innermost
thoughts. A

61. I liked my child best
when I could do every-
thing for him. A

62. The sooner a child learns
that a wasted minute is
lost forever, the better
off he will be. A

63. When a child is in
trouble, he ought to know
he won't be punished for
talking about it with his
parents. A

64. Parents should be careful
lest their children
choose the wrong friends. A

65. A child should always
accept the decision of
his parents. A

66. Children should do
nothing without the
consent of their
parents. A

67. Children should have a
say in the making of
family plans. A

68. It is sometimes neces-
sary for the parent to
break the child's will. A

69. It is natural for a mother
to "blow her top" when
children are selfish
and demanding. A

Agree

Mildly
Disagree

Strongly
Disagree

75

QUESTIONNAIRE Strongly Mildly Mildly Strongly
(continued) page 9 Agree Agree Disagree Disagree

70. A young mother pro-
bably feels "held
down" because there
are lots of things
she wants to do
while she is young. A a d D

71. Children should not
annoy their parents
with their unimportant
problems. A a d D

72. Taking care of a small
baby is something that
no woman should be
expected to do all by
herself. A a d D

73. Some children don't
realize how lucky they
are to have parents
setting high goals
for them. A a d D

74. If a child is pushed into
an activity before he is
ready, he will learn that
much earlier. A a d D

75. Unless one judges a
child according to
strict standards, he will
not be industrious. A a d D

76. It is a parent's business
to know what a child is
up to all the time. A a d D

77. Children are better off
if their parents are
around to tell them what
to do all the time. A a d D

78. A child should be rewarded
for trying even if he does
not succeed. A a d D

GS:rb

APPENDIX B

SENSITIVITY TO CHILDREN TEST (STC-T)

APPENDIX B

STC-T

NAME AGE SEX (M or F) DATE

 

 

Grade Teaching (If Applicable)

 

Instructions

 

A series of situations will be found on the following
pages. You are to pretend or imagine you are the teacher
of the child or children described. All the children in
the following situations are to be considered between
£133 and eight years old.

Your task is to write down exactly how you would
respond to the child or children in each of the situations,

in a word, sentence or short paragraph. Write down your

 

exact words and/or actions, but please do not explain why

 

you said or did what you described. Again, write down your
exact words or actions as if you were writing a script

for a play or movie (e.g., do not write "I would reassure
or comfort him," instead, for example, write "I would

smile at him and in a quiet voice say, 'Don't worry,

Billy, you'll do better next time.'")

12/12/69 IFS 807:GES Rev.

76

77

Just before school is to begin you are having a
friendly talk with a fellow school teacher in your
room. Carl rushes in and begins to interrupt your
conversation with a story about a friend.

After hearing a great deal of giggling coming from a
rear corner of the gymnasium during recess, you go
there and find Mary, Clark, and Sharon with their
underwear off. It appears that they were touching
each other's sexual parts before you arrived.

78

Barbara, who is normally bright-eyed and gay, has just
arrived in school silent, sad-faced, and dragging her
feet. You can tell by her manner that something
unpleasant has happened to her.

There have been complaints of students missing some
money. Upon returning to your classroom after recess
you find Lillian putting down another student's
wallet with a quarter in her hand.

79

5. You are helping Bruce with an arithmetic problem
and he seems to be having difficulty. He suddenly
exclaims: "I am so stupid. I never know the answers

to any of the questions you ask. I don't want to come
to school anymore.

6. As the students file out after class, Fred comes over

to you and asks in a quiet, concerned voice, "Do you
love me?"

80

Al is full of anger. The class had been scheduled to
go to the zoo for weeks and he was very eager to go.
However, it rained today and the trip had to be
rescheduled. He angrily exclaims in front of a number
of students, "I hate school. Just because it rained
we couldn't go."

Upon arriving in school, Joe excitedly tells you about
how his friend, Gary, who is in another class, was
pushed into a rain-filled puddle by some older boys.
Joe says that they were just walking to school when
all of a sudden three sixth graders ran up from

behind and shoved Gary into the puddle and ran away
laughing.

10.

81

Nancy has returned a signed form from her parents
requesting that flouride treatments be administered
by the school dentist during school hours. All
morning she has been unable to work and it seems
that she is becoming increasingly upset as her turn
to visit the dentist is approaching. It is now her
time to go and she refuses and begins to cry, "I
don't want to go! I don't want to go!"

Danny, who has been told that he must miss recess
because of some misbehavior, immediately begins to
scream loudly and starts to kick you in a display
of temper.

ll.

12.

82

You have entered your classroom and greet Barbara.
As you are talking she says "Let me see how you look
without your glasses on." Wanting to be friendly
you remove them. She says, laughing loudly, "You
look just like I thought you would . . . goofy."

All the boxes of paints have been used many times in
class and none bears a quality of newness any longer.
Although other children accept this fact and busily

engage themselves with painting, Dick refuses to accept

a messy box of paints. He clasps his arm and says,
"I want a new box."

l3.

14.

83

Karen refuses to clasp hands with Susie so that the
class can begin a circle game. When you ask why she

says, "My mother says I'm not to play with her because
she's black."

The children have been asked to draw pictures of
anyone they like. As they complete them, they place
the pictures on the chalkboard ledge. John has
finished a picture of you but has inscribed it
boldly with a taboo word.

15.

16.

84

Robert, one of the brighter students, is busy checking
papers for other children. As they finish he goes to
their desks quietly so that they do not need to wait
for the teacher who is with another group. While he
is gone from his desk, you notice Eddie scribbling
with crayon all over one of Robert's workbooks. You
tell Eddie to stop and when Robert, who takes great
pride in his work, sees what Eddie has done, he begins
to cry. Eddie laughs and calls Robert a baby.

The children have decided to play kickball at recess
and captains are chosen. The captains begin to choose
their teams. When one captain, Billy, names Frank,

a poorly coordinated boy, to his team, the other
members of his team insist loudly that he not choose
Frank because he is a "lousy" player.

APPENDIX C

MENTAL HEALTH SCALE (P-R)

Name

ing

APPENDIX C

P-R SCALE

Date

 

 

For each of the characteristics found below, circle

the number under it which you feel represents the amount
that you feel you possess of that characteristic, accord-
' to the following code:

I possess little or none of this characteristic, much
less of it than most people I know.

I possess some or at least a noticeable amount of this
characteristic, but less of it than most people I know.

I possess an average amount of this characteristic,
about the same amount of it as most people I know.

I possess more than an average amount of this char-
acteristic, more of it than most people I know.

I possess a great deal of this characteristic, much
more of it than most people I know.

Anxiety

1 2 3 4 5
Assertiveness
l 2 3 4 5
Autonomy

l 2 3 4 5
Considerateness
l 2 3 4 5
Curiosity

l 2 3 4 5
Dominance

l 2 3 4 5
Enthusiasm

l 2 3 4 5

8.

10.

ll.

12.

l3.

l4.

Empathy

l 2 3 4 5
Genuineness
l 2 3 4 5
Hostility

l 2 3 4 5
Imaginativeness
l 2 3 4 5
Independence
l 2 3 4 5
Intelligence
1 2 3 4 5
Neuroticism

1 2 3 4 5

85

15.

l6.

17.

18.

19.

20.

21.

Obedience

l 2 3 4 5
Open-mindedness
l 2 3 4 5
Possessiveness
l 2 3 4 5
Self-confidence
l 2 3 4 5
Sensitivity

l 2 3 4 5
Shyness

l 2 3 4 5
Spontaneity

l 2 3 4 5

22.

23.

24.

25.

Suspiciousness
l 2 3 4 5
Tolerance

l 2 3 4 5
Trust

1 2 3 4 5
Warmth

l 2 3 4 5

86

APPENDIX D

CATEGORIES USED IN CODING THE BEHAVIOR OF

THE UNDERGRADUATE

 

Il,ll‘llll|i||lllllllill

APPENDIX D

Categories used in coding the behavior of the undergraduate:

(l)

(2)

(3)

(4)

(5)

Strong positive movement toward. The undergraduate
expresses considerable warmth, positive interest or
gives undivided attention to the child. Ex: A moves
physically toward C with a big smile; A gives praise
and encouragement to C; A engages in the child's fan-
tasies following the lead of C; A reflects the feel-
ings of C in an empathic manner.

 

Positive movement toward. Undergraduate exhibits some
clear Interest in the child. Ex: A attentively
observes C at play; A is engaged in play with C; A
talks with C in a manner suggesting some interest.

 

Neutral behavior. The undergraduate exhibits no

clear movement toward the child nor away from the
child. Ex: A engaged in parallel play while C

plays; A silently observes C while looking bored,
checking watch frequently; A ignores C and gazes about
the room.

 

Movement away. The undergraduate attempts to put some
diStance between himself and the child, emotionally

or physically. Ex: A moves backwards away from C;

A interposes tables or chair between himself and C;

A reminds C of limits in a nonfeeling, mechanical
manner.

 

Movement against. The undergraduate engages in
behavior expressing aggression against the child.
Ex: A sharply criticizes C; A shoots darts at C;
A attacks C in fantasy with puppet.

 

87

 

l I .llll’ l I "llll lull In! till] ('1 dllllll 'I llll in! I II.

APPENDIX E

CATEGORIES USED IN CODING THE BEHAVIOR

OF THE CHILD

 

 

.‘l‘ll’l‘l'll‘il‘l 11‘"! .‘zl’

APP ENDI X E

Categories used in coding the behavior of the child:

(1)

(2)

(3)

(4)

(5)

Strong positive movement toward. The C (child) moves
toward the A (adult) physically or emotionally or both
in a manner suggesting considerable interest and
warmth. Ex: giving therapist undivided attention with
a big smile; reaching out to touch or hold hand of A;

C says, "I like you."

 

Positive movement toward. The C shows some interest,
attention or warmth in the undergraduate. Ex: physi-
cally moves closer to A; including the A in play;
initiating a personal conversation with the A.

 

Neutral behavior. The C gives no indication of move-
ment toward or obviously away from the A. The C seems
to ignore the A, acting almost as if he were not

there. Ex: C is totally involved in a game or puzzle
by himself; C seems not even to hear the comments of A;
C is preoccupied with something in the playroom to the
complete exclusion of A.

 

Movement away. Physically, verbally or emotionally
the child appears to be avoiding the A. Ex: C turns
his back on A and walks away; C rejects play activity
initiated by A; C ignores questions from A.

 

Movement against. The Child shows obvious anger or
aggression toward the undergraduate. Ex: C shoots
darts at A; C threatens A with gun; C calls A names.

 

88

APPENDIX F

ANALYSIS OF VARIANCE OF ADULT BEHAVIORS POSITIVE AND

NEGATIVE ONE INTERVAL AFTER A NEUTRAL

CHILD BEHAVIOR

APPENDIX F

F-l
Analysis of Variance of Adult Behaviors Positive and

Negative One Interval After a Neutral
Child Behavior

 

Source of Variance df MS F

 

 

Training (A) 1 .04896 4.26
Potential (B) 1 .01815 1.59
A x B 1 .01578 1.37
Error 15 .01148
Sessions (C) 1 .00551 5.45:
A x c 1 .00743 7.36
B x c 1 .00096 < 1
A x B x c 1 .00062 < 1
Error 15 .00101
Scores (D) 1 10.77890 305.26**
A x 0 1 .15264 4.33
B x D 1 .00031 < 1
A x B x 0 1 .00243 < 1
Error 15 .03525
c x D 1 .00994 1.80*
A x c x 0 1 .02532 4.60
B x c x 0 1 .00189 < 1
A x B x c x D 1 .00311 < 1
Error 15 .00551

*

p < .05

*‘k
p < .001

89

APPENDIX G

ANALYSIS OF VARIANCE OF ADULT BEHAVIORS IN CATEGORIES ONE

AND FIVE, ONE INTERVAL AFTER A NEUTRAL CHILD BEHAVIOR

APPENDIX G
G-l

Analysis of Variance of Adult Behaviors in Categories One
and Five, One Interval After a Neutral Child Behavior

 

 

 

Source of Variance df MS F
Training (A) 1 .02004 3.28
Potential (B) 1 .00179 < l
A X B 1 .00576 < 1
Error 15 .00611
Sessions (C) 1 .00121 < l
A X C 1 .00054 < l
B X C 1 .00104 < l
A X B X C 1 .00826 2.23
Error 15 .00371
Scores (0) 1 .38192 41.69:*
A X D 1 .05147 5.62
B X D 1 .01098 1.19
A X B X D 1 .00063 < 1
Error 15 .00916
C X D 1 .00050 < l
A X C X D 1 .00213 < l
B X C X D 1 .00768 2.21
A X B X C X D 1 .00234 < 1
Error 15 .00348

*

p < .05
**

p < .001

90

APPENDIX H

ANALYSIS OF VARIANCE OF ADULT BEHAVIORS IN

INITIAL SESSIONS

li'll’k'l

 

APPENDIX H

H-l
Analysis of Variance of Adult Behaviors in Initial Sessions,
Category One

 

 

 

Source of Variance df MS F
Training (A) 1 .04781 4.19
Potential (B) 1 .01545 1.35
A X B 1 .02678 2.35
Error 22 .01144

H-2

Analysis of Variance of Adult Behaviors in Initial Sessions,
Category Five

 

 

 

Source of Variance df MS F
Training (A) 1 .01025 3.11
Potential (B) 1 .00920 2.79
A X B 1 .00880 2.67
Error 22 .00329

H-3

Analysis of Variance of Adult Behaviors in Initial Sessions,
Categories One and Two

 

 

 

Source of Variance df MS F
Training (A) 1 .07568 2.78
Potential (B) 1 .01724 < l
A X B 1 .00001 < 1
Error 22 .02729

H-4

Analysis of Variance of Adult Behaviors in Initial Sessions,
Categories Four and Five

 

 

Source of Variance df MS F
Training (A) 1 .00781 2.12
Potential (B) 1 .00095 < l
A X B 1 .00173 < 1
Error 22 .00369

 

91

BIBLIOGRAPHY

BIBLIOGRAPHY

Axline, V. M. Play therapy. Boston: Houghton Mifflin,
1947.

 

Bachrach, Henry. Adaptive regression, empathy and psy-
chotherapy. Psychotherapy: Theory, Research and
Practice, 1968, 5, 203-209.

 

Baldwin, A. L. The effect of home environment on nursery
school behavior. Child Development, 1949, 29,
49-61.

 

Bateson, G. Toward a theory of schizophrenia. Behavioral

 

Science, 1956, 1, 251-264.

Baumrind, D. Child care practices anteceding three
patterns of pre-school behavior. Genetic Psy-
chology Monographs, 1967, 15, 43-88.

 

 

Baumrind, D., & Black, A. Socialization practices associ-
ated with dimensions of competence in preschool
boys and girls. Child Development, 1967, 38,
291-328. ‘—

 

Baumrind, D. Current patterns of parental authority.
Developmental Psychology, 1971, 4, 1—103.

 

Becker, W. Consequences of different kinds of parental
discipline. In M. L. Hoffman and L. W. Hoffman,
Review of child development research. New York:
Russell Sage, 1964.

 

Becker, W., & Krug, R. The parent attitude research
instrument: A research review. Child Develop—
ment, 1965, 36, 329-365.

 

Becker, W. C., Peterson, D. R., Luria, Z., Shoemaker,
D. J., & Helmer, L. A. Relationship of factors
derived from parent interview ratings to behavior
problems of five year olds. Child Development,
1962, 33, 509-535. -

 

92

93

Bell, R. O. A reinterpretation of the direction of
effects in studies of socialization. Psycho-
lggical Review, 1968, 15, 81-95.

 

Bell, R. Q. Stimulus control of parent or caretaker
behavior by offspring. Developmental Psychology,
1971, 4, 73-88.

 

Bell, R. W. Retrospective attitude studies of parent-
child relations. Child Development, 1958, 29,
323-338.

 

Berensen, B. G., Carkhuff, R. F., & Myrus, P. The
interpersonal functioning and training of
college students. Journal of Counseling Psy-
chology, 1966, 13, 441-446.

 

Brennan, E. C. College students and mental health pro-
grams for children. American Journal of Public
Health, 1967, 51, 1767-1771.

 

Brody, Grace. Relationship between maternal attitudes
and behavior. Journal of Personality and Social
Psychology, 1965, 2, 317-323.

 

 

Carkhuff, R. R. Lay mental health counseling: Prospects
and Problems. Journal of Individual Psychology,
1968, 24, 88-93.

 

Crandall, V. J., Orleans, 8., Preston, A., & Rabson, A.
The development of social compliance in young
children. Child Development, 1958, 29, 429-443.

 

Donahue, G. T., & Nichtern, S. Teaching the troubled
child. New York: Free Press, 1965.

 

Epstein, N. B. Recent advances in the study of behavior
change. In Proceedings of the Academic Assembly
on Clinical Psychology. Montreal: McGill Uni;
versity, 1963, 161-162.

 

 

Ginott, H. Group psychotherapy with children. New York:
McGraw-Hill, 1961.

 

Goodman, G. Companionship therapy. San Francisco:
Jossey—Bass, 1972.

 

Gruver, G. G. College students as therapeutic agents.
Psychological Bulletin, 1971, 16, 111-127..

 

94

Guerney, B. G. Filial therapy. Description and rationale.
Journal of Consulting Psychology, 1964, 33, 304-310.

Guerney, B. G. (Ed.). Psychotherapeutic agents: New roles
for nonprofessionals, parents and teachers. New
York: Holt, Rhinehart and Winston, 1969.

Harvey, L. V. The use of nonprofessional auxiliary
counselors in staffing a counseling service.
Journal of Counseling Psychology, 1964, 33, 348-
351.

Harper, L. V. The young as a source of stimuli controlling
caretaker behavior. Developmental Psychology,
1971, 3, 68-72.

 

Hefele, T. J., Collingwood, T., & Drasgow, J. Therapeutic
facilitativeness as a dimension of effective
living: A factor analytic study. Journal of
Clinical Psychology, 1970, 3E, 121-123.

 

 

Hoffman, M. L. Parent discipline and the child's consid-
eration for others. Child Development, 1963, 33,
573-588.

 

Kemp, C. G. Influence of dogmatism on the training of
counselors. Journal of Counseling Psychology,
1962, 3, 155-157.

 

Kirkpatrick, P. J. A further exploration of the A-B
"type" distinction. Dissertation Abstracts,
1969, 33, 849.

 

Leary, T. Multilevel measurement of interpersonal
behavior. Berkeley: PsyChological ConsfiI-
tation Service, 1956.

 

Kreitzer, S. F. College students in a behavior therapy
program with hospitalized emotionally disturbed
children. In B. G. Guerney, Jr. (Ed.), Psycho-
therapeutic agents: New role for nonprofes-
sionals, parents, and teachers. New York: Holt,
Rinehart and Winston, 1969.

 

 

Lichtenberg, Phillip. Time perspective and the initiation
of cooperation. Journal of Social Psychology,
1956, i3, 247-260.

 

95

Linden, J. L., & Stollak, G. E. The training of under-
graduates in play techniques. Journal of
Clinical Psychology, l969,'33, 38—45.

 

 

Mannino, F., Kisielewski, J., Kimbro, E., & Morganstern,
B. Relationships between parental attitudes and
behavior. The Family Coordinator, 1968, 237-240.

 

May, Rollo. Love and will. New York: Horton, 1969.

 

Medinnus, Gene. Readings in the psychology of parent-
child relations. New York: John Wiley and Sons,
1967.

 

 

Mitchell, W. E. Amicatherapy: Theoretical perspectives
and an example of practice. Community Mental
Health Journal, 1966, 3, 307-314.

 

 

Moustakas, C. E. The frequency and intensity of negative
attitudes expressed in play therapy: A comparison
of well adjusted and disturbed children. Journal
of Genetic Psychology, 1955, 33, 309-325.

 

Moustakas, C. E. Psychotherapy with children. New York:
Harper, 1959.

 

Moustakas, C. E., & Schalock, H. D. An analysis of
therapist-child interaction in play therapy.
Child Development, 1955, 33, 143-157.

 

Moustakas, C. E., Sigel, I., & Schalock, H. An objective
method for the measurement and analysis of child-
adult interaction. Child Development, 1956, 33,
109-134.

 

Moss, H. A., & Robson, K. S. The role of protest
behavior in the development of the mother-infant
attachment. In J. L. Gerwirtz (Chm.), Attach—
ment behaviors in humans and animals. Symposium
presented at the meeting of the American Psycho-
logical Association, San Francisco, September,
1968.

Nichtern, S., Donahue, G. T., O'Shea, J., Marans, M.,
Curtis, M., & Brody, C. A community educational
program for the emotionally disturbed child.
American Journal of Orthopsychiatry, 1964, 33,
705-713.

 

Osofsky, J. The shaping of mother's behavior by children.
Journal of Marriage and the Family, 1970, 33,

 

96

Patterson, G. R., & Brodsky, G. A behavior modification
program for a child with multiple problem behaviors.
Journal of Child Psychology, l966,3, 277-295.

 

Peterson, D. R., Becker, W. C., Hellmer, L. A., Shoemaker,
D. J., & Quay, H. C. Factors in parental behavior
and personality as related to problem behavior in
children. Journal of Consulting Psychology, 1959,
33, 107-118.

 

Reif, T. F., & Stollak, G. E. Sensitivity to young chil-
dren: Training and its effects. East Lansing:
Michigan State University Press, 1972.

 

 

Reinherz, H. The therapeutic use of student volunteers.
Children, 1964, 3, 137-142.

Rioch, M., Elkes, E., Flint, A., Udansky, B., Newman, B.,
& Sibler, E. NIMH pilot study in training mental
health counselors. American Journal of Ortho-
spychiatry, 1963, 33, 678-689.

 

 

Rogers, C. R. The necessary and sufficient conditions of
therapeutic personality change. Journal of Con-
sulting Psychology, 1951, 33, 95-103.

 

 

Shoggen, P. Environmental forces in the everyday lives
of children. In R. G. Barker (Ed.), The stream
of behavior: Explorations of its structure and
content. New York: Appleton-Century-Crofts, 1963.

 

 

Schulman, R., Shoemaker, D., & Moelis, I. Laboratory
measurement of parental behavior. Journal of
Consulting Psycholggy, 1962, 33, 109-114.

 

 

Sears, R. F., Maccoby, E. E., & Levin, H. Patterns of
child rearing. Evanston, 111.: Row, Peterson, 1957.

 

 

Shadlow, George, W. The Whitehorn-Betz A-B scale as related
to differential motivational effectiveness and
interpersonal perception. Dissertation Abstracts,
1968, 33, 1181.

 

Stoffer, D. L. Investigation of therapeutic success as
a function of genuineness, nonpossessive warmth,
empathic understanding and dogmatism in the
helping person. Proceedings of the 76th Annual
Convention of the APA, I968, 3, 619—620.

 

Stover, L., & Guerney, B. G. The efficacy of training
procedures for mothers in filial therapy. Psy-
chotherapy, Theory, Research and Practice, I967,
3, 111-116.

 

97

Stollak, G. E. The experimental effects of training
college students as play therapists. Ps cho-
therapy: Theory, Research and Practice, I968,
3, 77-86.

 

Stollak, G. E., Green, L., Scholom, A., Schreiber, J.,
& Messe, L. A. The process and outcome of play
encounters between undergraduates and clinic-
referred children: Preliminary findings. Paper
presented at 1973 meeting of Society for Psy-
chotherapy Research, Philadelphia.

Swensen, C. H. Commitment and the personality of the
successful therapist. Psychotherapy: Theory,
Research and Practice, 1971, 3, 29-36T

 

 

Truax, C. B., & Carkhuff, R. F. Towards effective coun-
seling and psychotherapy: Training and practice.
ChiCago: AIdine PubIIShing Company,71967.

 

 

Truax, C. B., & Wargo, D. G. Psychotherapeutic encounters
that change behavior: For better or worse.
American Journal of Psychotherapy, 1966, 33,
499-520.

 

Truax, C. B., & Mitchell, K. M. Human encounters that
change behavior. In Marvin Feldman (Ed.),
Buffalo studies, New York: State University of
New York at Buffalo, 1968, 55-81.

 

Tuckman, J., & Lavell, M. Attrition in psychiatric
clinics for children. Public Health Report,
1969, 33, 309-315.

 

Whitehorn, J. C., & Betz, B. J. A study of psycho-
therapeutic relationships between physicians
and schizophrenic patients. American Journal
of Psychiatry, 1954, 333, 321-331.

 

 

Zunich, M. The relation between parental attitude toward
child rearing and child behavior. Journal of
Consulting Psychology, 1962, 33, 197.

 

 

109 ms UNIV. men/miss
0476443