TEWDTWAV """"" K '''''''''''

3107036

 

 

LJBRARJES remove this checkout from

A MSU * RETURNING MATERIALS:
Place in book drop to
4-m— your record. FINES will

 

be charged if book is
returned after the date
stamped be10w.

 

‘? .

-._ _, 2 ,, ,.
mzrwé
r}. 190 FA‘lL 112,3 , .

i135: ' 9 "‘

.5 .A;

a ' ‘V

‘aewa

 

 

 

A COMPARATIVE INVESTIGATION OF THE
EFFECTIVENESS OF TWO FOSTER PARENT

TRAINING PROGRAMS

BY

Jean-Claude Dutés

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Counseling,
Educational Psychology and
Special Education

1985

ABSTRACT
A COMPARATIVE INVESTIGATION OF THE

EFFECTIVENESS OF TWO FOSTER PARENT
TRAINING PROGRAMS

BY

Jean-Claude Dutés

Over the last 50 years, foster family care has become
the most widespread from of substitute care for children
with special needs and those requiring parental surrogates.
With this reliance has come the realization that foster
parents require specialized training if they are to respond
adequately to the needs of children under their care. While
a great many training programs have been deve10ped, their
evaluations have not been conclusive.

It was the purpose of this investigation to assess the
differential effects of two 16-hour behaviorally-oriented
foster parent training programs on perceptions of family
climate, parenting efficacy, placement stability and
dependence on agency staff. One curriculum emphasized child
behavior management and self-management, while the other
concerned only behavior management. The self-management
component was designed to train parents in techniques of

self-control that would allow them to reduce emotional

Jean-Claude Dutés

arousal and facilitate the use of behavioral principles in
crisis situations.

A group by time research design was used to evaluate
the two curricula. Pretest, posttest and follow-up
measurements were made, the latter 60 days after completion
of training.

The subjects were licensed foster parents affiliated
with a child placement agency, who were required to attend
training to maintain their licenses. The majority of the
subjects wre urban black, female, single heads of
households. It was hypothesized that subjects trained in
both parenting and self-management skills would outperform
those trained in parenting skills only.

Using univariate analysis of repeated measures analysis
on the data, it was found that the two training groups did
not significantly differ on the outcome measures. It was
reasoned that the results could not be explained solely in
terms of the manipulated variables, but that selection bias
due to attrition, poor attendance, insufficient training
time, subjects' inadequate reading and writing skills and
trainer's inexperience with persons of low educational
levels, offered competing explanations of the results. It
is suggested future efforts at foster parent training
curriculum development take; into consideration the learning
styles and needs of foster parents with low educational

levels.

Dedication

To my parents, especially my mother, Julie Mariellie,
whose belief in my abilities helped me to believe in myself.
To my wife, Margareth, for her patience, understanding and
devotion and to our son, Jean-Paul, whose presence has

enriched my understanding of what it means to be a parent.

ii

Acknowledgements

I would like to thank the members of my doctoral
committee, Dr. Gloria Smith, Dr. John Schweitzer, Dr. Lee
June, Dr. Esther Fergus and Dr. Richard Thomas for their
guidance and support. I am especially grateful to my
chairperson, Dr. Smith, for her contributions to my
professional growth and Dr. Schweitzer, my research adviser,
for his incessant patience, encouragement and friendship.

A special thanks goes to Dr. Richard Coelho and Ira
Washington, III for their help with the computer analysis of
the data.

Finally, I want to thank my wife for countless hours of
typing and her sustaining devotion, my brother Eric for
reaching out to me away from home and all those who

contributed to the realization of this work.

iii

TABLE OF CONTENTS

Chapter Page
I. THE PROBLEMOOOOOOOOOOOOOOOOOOOOOOOOOOOO0.0.0.0... 1

Introduction.....................................
Need for the Study...............................
Replacements and Foster Children............
Replacements and Caseworkers................
Interventions...............................
Purpose.......................................... 16
Statement of the Problem and
Research Questions............................. 17
Hypotheses....................................... 18
Definitions of Terms............................. 19
Theoretical Foundations.......................... 20
Social Learning Theory...................... 25
Cognitive Behavioral Approach............... 26
Overview......................................... 28

\OGU‘wH

II. REVIEW OF THE LITERATUREOOOOOOOOOOOOOOOOOOOOOOOOO 29

Historical Overview...............................29
Group Case Studies.......................... 43
Correlational Evaluations................... 45
Quasi-Experimental Studies.................. 47
Experimental Studies........................ 49
Summary..................................... 52

Related Literature............................... 54
Behavioral Parent Training.................. 54
Family Perceptions.......................... 63
Self Efficacy............................... 67

Summary.......................................... 69

III. HETHODOLWYOOOOO00.......0...OOOOOOOIOOOOOOOOOOOO 72

Introduction..................................... 72
Design........................................... 72
Sample Selection................................. 73
Assignment....................................... 75
Training Curricula............................... 79

iv

Setting............................... ..... ...... 84
Trainers......................................... 85
Training Implementation.......................... 87
Implementation Difficulties...................... 88
Training Related Concerns........................ 91
Instrumentation.................................. 93
Data Collection..................................100

Pretesting..................................100

Posttesting.................................lOl

Follow-Up...................................102
Data Analysis....................................103

Iv. ANALYSIS OF DATAOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO.104

Hypothesis Testing...............................113
Data Analysis for Subjects with Completed Scaled
Measures at Posttest............................131
Supplementary Analysis...........................l40

V. SUMMARY, DISCUSSION AND IMPLEMENTATION...........175

Summary of Results...............................178
Discussion.......................................183
Supplementary Analysis...........................188
Implications for Future Research.................189

Appendices.OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO000.00192

A. Foster Parent Profile Questionnaire..........192
B. Moos Family Environment Scale................196
C. Foster Parent Parenting Efficacy Scale.......199
D. Utilization and Usefulness Scale.............200
E. Poster Placement Information Sheet...........201
F. Tests........................................202
G. Trainer's Manual.............................207
H. Informed Consent.............................262

References........................... ....... ..........263

Table

bub“ U
C O I O O
H

o o 0
tom \3 0‘ U" :5 U NHN

h-fi fink :5 :5 uh ab :5
O

0
HH
HO

:5
O

H
N

4.13

4.14
4.15

4.16
4.17

4.18
4.19

4.20
4.21

4.22
4.23

4.24

LIST OF TABLES
Page

Number of Subjects Initially Assigned and

Those Responding...................................77
Subjects' Attendance and Completion of Scales......78
Summary of E-Tests Results on Pretest Data....106
Results of Fisher's Exact Test for Presenting
Problems at Time 1................................107
Results of Chi Square and Fisher's Exact Test

for Presenting Problems at Time 2.................108
Chi Square Results for Presenting Problems at

Time 3............................................109
Repeated Measures Analysis of Variance of Foster
Children..........................................110
Mean Number of Poster Children for Training
Conditions........................................lll
Repeated Measures Analysis of Variance of Scores

on the Cohesion Subscale..........................1l4
Means of Scores on Cohesion Subscale..............114
Repeated Measures Analysis of Variance of Scores

on the Conflict Subscale of the FES...............116
Means of Scores for Conflict Subscale.............116
Repeated Measures Analysis of Variance of Scores

on the Control Subscale of the FES................118
Means of Scores for Control Subscale..............118
Results for Repeated Measures Analysis of Variance

of Scores on the FPPES............................120
Means of Efficacy Scores..........................120
Repeated Measures Analysis of Variance on Number

of Requests for Removal...........................121
Mean Number of Requests for Removal...............122
Repeated Measures Analysis of Variance for

Number of Unrequested Removals....................123
Mean Number of Unrequested Removals...............124
Results of Repeated Measures Analysis of Variance

for Runaway Children..............................125
Mean Number of Runaway Children...................125
Repeated Measures of Variance for Number of Home
Contacts..........................................127
Means of Home Contacts............................127
Results of Repeated Measures of Variance for

Number of Office Contacts.........................129
Means of Office Contacts..........................129

vi

4.25

4.26
4.27
4.28
4.29
4.30
4.31

4.32
4.33

4.34
4.35

4.36
4.37

4.38
4.39

4.40
4.41

4.42
4.43

4.44
4.45

4.46
4.47

4.48
4.49

4.50
4.51

4.52
4.53
4.54
4.55
4.56

4.57

Results of Repeated Measures of Analysis of

Variance for Telephone Contacts...................130
Means of Telephone Contacts.......................13l
Summary of t-Test Results on Pretest Data.........133
Chi Square for Presenting Problems at Time 1......134
Chi Square for Presenting Problems at Time 2......135
Chi Square for Presenting Problems at Time 3......136
Repeated Measures Analysis of Variance of Foster
Children..........................................138
Means of Foster Children..........................138
Repeated Measures of Analysis of Variance of

Scores on the cohesion Subscale of the FES........141
Mean of Cohension Subscale Scores.................l41
Repeated Measures of Analysis of Variance of

Scores on the Conflict Subscale...................l42
Means of Conflict Subscale Scores.................143
Results of Repeated Measures of Analysis of

Variance of Scores on the Control Subscale........144
Means of Control Subscale Scores..................144
Repeated Measures of Analysis of Variance of

Scores on the FPPES...............................145
Means of Efficacy Subscale Scores.................l46
Repeated Measures of Analysis of Variance on

Number of Requests for Removal....................147
Mean for Removal Requests.........................147
Repeated Measures of Analysis of Variance for
Unrequested Removals..............................148
Means of Unrequested Removals.....................l49
Repeated Measures of Analysis of Variance

for Runaway Children..............................150
Means for Runaway Children........................151
Repeated Measures of Analysis of Variance of

Home Contacts.....................................152
Means of Home Contacts............................152
Repeated Measures of Analysis of Variance

of Office Contacts................................153
Means of Office Contacts..........................154
Repeated Measures of Analysis of Variance of
Telephone Contacts................................155
Means of Telephone Contacts.......................155
Summary of t-Tests Results of Pretest Variables...158
Chi Square Results for Presenting Problems at

Time l............................................1S9
Chi Square Results for Presenting Problems at

Time 2............................................160
Chi Square Results for Presenting Problems at

Time 3............................................161
Repeated Measures of Analysis of Variance of

Number of Foster Children.........................162

vii

4.58
4.59

4.60
4.61

4.62
4.63

4.64
4.65

4.66
4.67

4.68
4.69

4.70

Means for Number of Foster Children...............l63
Repeated Measures Analysis of Variance on Number

of Requests for Removal...........................165
Mean Number of Requests for Removal...............165
Repeated Measures Analysis of Variance for

Number of Unrequested Removals....................l66
Mean Number of Unrequested Removals...............l67
Repeated Measures Analysis of Variance for
Runaways..........................................168
Means of Runaways.................................168
Repeated Measures of Analysis of Variance for

Number of Home Contacts...........................l70
Means of Home Contacts............................17O
Repeated Measures of Analysis of Variance for

Number of Office Contacts.........................171
Means of Office Contacts..........................172
Repeated Measures of Analysis of Variance for
Telephone Contacts................................173
Mean Number of Telephone Contacts.................173

viii

CHAPTER ONE

THE PROBLEM

Introduction

 

Over the past fifty years, foster family care has
become the preferred form of substitute care for neglected,
abandoned and abused children. Fanshel (1970) indicated
that in 1933, foster parents cared for 42 percent of
children in need of surrogate parenting. By the late 19708,
75 percent of such children were served by foster family
care (Borejsi, 1979). Of 1.8 million children under the age
of 18 who were receiving social services in 1977,
more than half a million were involved with foster family
care (Shyne & Schroeder, 1978) and in 1974 it was estimated
that about 160,000 foster families per year served as
surrogate families (Child Welfare League of America, 1974).

This increased reliance on foster family care has
presented foster care professionals and foster parents with
new challenges. Today's foster children often come from
home environments that have contributed to developing
behavioral and emotional difficulties (Horejsi, 1979;

Kaplan & Seitz, 1980). These include disorganized home

environments (Maas & Engler, 1959; Thomas, 1961), where
family members possess relatively few personal and social
resources (Horejsi, 1979). Foster children, thus, are
likely to have histories of parent-child conflict,
aggressive behavior, self destructive tendencies, disruptive
community behavior, school difficulties and emotional
disturbance (Gillham, 1983; Larson et al., 1974). In
addition, a large proportion of foster children are reported
to be physically, medically and mentally disabled (Bryant,
1981; Gillham, 1983).

The difficulties inherent in working with problem
children have turned foster parenting into a more demanding
job than it formerly seemed to be. Poster chidren appear to
require much more than a stable home environment and
parental supervision. Compared to other children, a foster
child is likely to require more of the foster parents' time,
effort and commitment according to Dinnage and Pringle
(1969). Gillham's observations (1983) appear to corroborate
those of Dinnage and Pringle. She indicates that ”providing
care to children with special medical or emotional needs
requires not only extra time, but also greater patience,
skill and endurance in dealing with the child's demands and

their effects on the foster parents personal life " (p. 9).

Need for the Study '

The foster care field is lagging in its ability to
provide foster parents with the skills needed to
successfully respond to the needs of today's foster children
and to reduce the effects of those needs on their personal
lives. The difficulties presented by foster children and
the inadequate preparation of foster parents have combined
to create situations that are likely to result in more harm
than good for both children and foster parents. For
example, Hampson and Tavormina (1980) question the ability
of foster parents to manage behaviorally and emotionally
disordered children therapeutically. Because of these
discrepancies between foster parents' skills and children's
needs, the stability and continuity of care that are so
crucial to the behavioral and emotional rehabilitation of
children, is in jeopardy. This is especially evident when
foster parents become disillusioned and withdraw from family
foster care (Boyd & Remy, 1978). The realization that
foster home care entails much more then they had anticipated
appears to underlie the decisions of many to drop out. Such
parents appear overwhelmed and completely bewildered by
their foster parenting experiences (Cautley, 1980). Boyd &
Remy (1978) reviewed the findings of a 1973 study which
revealed a one-year dropout rate of 25 to 50 percent. Other
parents, who do not drop out of foster family care when

faced with an emotionally or behaviorally disordered child,

may request the removal of the child but remain potential
foster parents. These parents, however, feeling deceived by
the placement agency, are likely to become cautious about
accepting new referrals and relucant to accept new
placements (Cautley, 1980; Horejsi, 1979).

When a foster parent drops out or requests the removal
of a foster child, the placement is a failure and
necessitates the replacement of the child. Garrett (1977)
found that foster children are likely to experience 2.7
moves or new placements before leaving the foster family
care system. Stone and Stone (1983), after an examination
of 64 case records, found that 48.5 percent of the children
had experienced unplanned transfers. A majority of the
unplanned transfers occurred within the first four weeks of
the placements. Cautley (1980), in a study of new foster
parents, found that one-third of the parents requested the
removal of their foster children. Of 39 parents, 15 percent
requested removal of the foster child within three months of
the placement. This situation has been found to be related
to adverse emotional states in foster children and foster
parents (Bauer & Heinke, 1976; Dinnage & Pringle, 1969: Fein
& Maluccio, 1984; Horejsi, 1979; Kline & Overstreet, 1972;
Stone & Stone, 1983), which severely impede the delivery of

quality care to foster children (Stone & Stone, 1983).

Replacements and Foster Children

It has been argued that multiple placements contribute
to feelings of rejection, insecurity and a sense of
rootlessness in foster children (Horejsi, 1979). Being
moved from one home to another fosters a distrust of adults
and strengths feelings of being unacceptable and unwanted.
Children who experience such replacements are at risk of
developing identity problems, suffering fears of separation
and experieincing difficulties in forming close
relationships (Horejsi, 1979). Moreover, replacements have
been found to be related to adjustment difficulties that
tend to increase the risk of failure in future placements
(Boyd & Remy, 1978; Maas & Engler, 1959). In addition,
failed placements may delay a foster child's return to his
natural family or render permanent placement more difficult
when return to the family of origin is not possible or
preferable. Fein & Maluccio (1984) reported that children
with more foster placements did not do as well as those with
fewer placements on measures of family adjustment, behavior
and school functioning (p.428). While the investigators did
not indicate what constituted "more” and ”fewer” placements,
it appears that failed placements have the potential for

transforming temporary placements into a way of life.

 

Replacements and Caseworkers

Unsuccessful placements also affect child care workers
adversely, seeming to exacerbate the strains that exist in
the professional lives of caseworkers (Barrett & McKelvy,
1980; Bryant, 1981). These professionals usually carry large
caseloads and are under constant pressure to locate new
homes (Horner, 1981). A failed placement may deprive the
worker of a home, while creating an immediate need for
replacement of a child, sometimes on the spur of the moment.
Such situations keep caseworkers constantly "putting out
brushfires' and prevent even distribution of their time
among the foster parents on their caseloads. Because of
this, the child care worker appears to respond to foster
parents only in times of crises, that is, when a change in
placement is needed or in reaction to a complaint needing
investigation, with little time left to provide support and
reassurance to foster parents who seem to be doing well. By
inadvertently neglecting foster parents who are not in
crisis, the child care worker runs the risk of being
perceived as uncaring and unsupportive of these parents'
efforts to tackle a difficult, frustrating and seldom
rewarding task. Aldridge et al., (1974) report that parents
become demoralized, worry about not doing a good job or feel
that they are being used. In apparent agreement, Stone &
Stone (1983) found that the degree of contact, rapport

building and energy expended by placement staff were

strongly associated with placement success. The opposite is
often true, however, when parents question the agency's
support. It has frequently been noted that a foster
parent's request for the removal of a child is motivated
less by that child's behavior than by a perceived lack of
support from placement agency personnel (Kline & Overstreet,
1972).

Another area likely to feel reverberations from the
unsuccessful placement of a foster child is the foster
family. The introduction of a new child into a family is
bound to be associated with some disruption of that family's
equilibrium, which brings about changes in the family's
usual patterns of relating. As Wilkes (1974) points out, the
addition of one foster child to a family of three increases
the possible subgroups within that family from 4 to 11.
With increased possible relationships, foster family members
have less time to be with each other. Wilkes explains,
“each family member now has a number of groups within the
home in which he may be included or from which he may be
excluded, and he must work out his position with each
group ” (p. 374). Thus changes in family composition,
resulting from frequent placements and premature
terminations, may create a sense of family disorganization
and confusion among family members. Moreover, the
relatively large amount of time that foster parents may

Spend working with problematic children may engender

resentful feelings in spouses and natural children,
contributing to conflictual family environments. This
situation is threatening to the biological members of the
family and puts the foster community at risk.

The effects of failed placements on foster parents,
while perhaps less severe than on foster children, are far
from being innocuous. Mash & Johnston (1983) found that
parents of hyperactive children, when compared with those of
normal children, perceive themselves as less competent and
derive less pleasure from the parenting role. Mothers of
hyperactive children were found to be significantly more
likely to experience depression, social isolation, self
blame, role constriction and detachment from their children
than mothers of normal children. Moreover, significantly
higher levels of stress were associatd with parent-child
interactions for mothers of hyperactive children than for
those of normal children. Observations of the functioning
of foster parents, although lacking empirical
substantiation, would appear to agree with Mash and
Johnston's findings. Jocobs (1980) notes a pervasive sense
of powerlessness and isolation among foster parents and
Stone & Stone (1983) suggest that failed placements
contribute to burnout in foster parents. Moreover, the
caution and reluctance to accept new children that
characterize foster parents who have experienced a failed

placement (Horejsi, 1979) suggest that such parents may

experience some degree of role constriction. McCoy and
Donahue (1961) noted that foster mothers tended to blame

themselves for placement failures.

Interventions

In light of the previous discussion, various aspects of
the foster care system could be targeted for intervention.
At the point of placement, child care agencies can make
efforts to reduce the discrepancies between parents'
strengths and foster children's problems. A great number of
investigations have been undertaken to identify correlates
of successful foster parents, families, and homes in
attempts to provide empirical information to be used in
developing successful parent-child matches (Cautley, 1980;
Cautley 8 Aldridge, 1975; Fanshel, 1966; Levant & Geer,
1981). While reported results are informative, they seem to
have had very little impact on the foster care field. The
ability of child care workers to match foster parents and
foster children, based on research-derived information or
in any sensible manner, is seriously undermined by the
limited availability of foster homes and the increasingly
large numbers of children in need of foster home placements
(Horner, 1981; Taylor a Star, 1967). Horner (1981) observes
that ”many foster care programs are hard pressed simply
to find sufficient numbers of placements for the' children

thought to need such help, and are rarely able to be as

10

selective as the literature would suggest ' (p. 42).
Potential parents that child care agencies may turn down
when there is a sufficient supply of homes may be hurriedly
recruited in times of need (Fanshel, 1970; Horner, 1981).
In short, child-parent matching is not very practical unless
or until the demand for foster family homes decreases.

An alternative approach would be for child care workers
to develop and maintain regular contact with the parents
affiliated with their agency, regardless of how the
placement is unfolding. This would provide foster parents
with needed support and alter perceptions of the agency
staff as preoccupied mainly with foster parents'
transgressions (Cautley, 1980). This solution would require
that agencies employ more child care workers so as to
decrease the ratio of foster families to workers and has two
major weaknesses. First, there are few placement agencies
capable of offering the degree of staff support needed on a
consistent basis (Horner, 1981). Second, this approach
could lead to increased dependence on agency staff and thus
prevent the professionalization of foster parents. The
greater the parents' dependence on agency staff, the more
likely it is that parents will be treated as clients instead
of coworkers.

The third intervention approach, which appears to be
the most promising at this time, is to render foster parents

more effective in handling foster children and in coping

11

with the effects of the job. Increasing parents'
effectiveness is likely to decrease their dependence on
agency workers, thereby freeing workers to distribute their
time more evenly and consistently among families in their
caseloads. Furthermore, more effective parenting is likely
to facilitate the adjustment of the foster child in ways
that do not overly tax the resources of other family
members. It is also likely to reduce conflictual
relationships in the home environment. In short, training
appears to be the best strategy to bring about placement
stability, increase parents' sense of parenting competency
and reduce dependence on agency workers. Moreover, it has
been argued that it is the least expensive way to prevent
parent dropout and failed placements (Boyd & Remy, 1978).
Given these potential advantages, it is not surprising that
an increasing interest in the development and evaluation of
foster parent training programs has been noted.

The need for foster parent training that has long been
recognized by parents and foster family care professionals
(Boyd & Remy, 1978; Goldstein, 1967) has generated a number
of training programs, emanating from different theoretical
perspectives. Some of these have been empirically evaluated
(Brown, 1980; Cobb et al., 1982; Guerney, 1976; Levant et
al., 1981).

Cobb and his colleagues (1982) trained parents in

communication, conflict resolution and behavior management

12

skills. Levant and his colleagues emphasized human
relations skills, while Guerney's training concentrated on
effective commnication and parenting skills geared toward
children under 12 years of age. A fourth training program
helped foster parents develop greater awareness of the
difficulties of foster children and increased understanding
of foster family care (Ryan, 1979).

Evaluative studies of the above training programs have
produced results that are encouraging but far . from
conclusive. The results suggest that foster parents are
able to acquire the skills taught in training (Cobb et al.,
1982) and that training is related to positive attitudes
toward children (Brown, 1980; Guerney, 1976). Training does
not, however, appear to be associated with changes in
children's behavior or to affect placement stability (Levant
et al., 1981). Only one study reported training to have a
positive effect on stability of placement and children's
behaviors (Boyd & Remy, 1978), but even those findings must
be interpreted with caution as the experimental and control
groups were constructed two years after training had
occurred. Training consisted of a behaviorally-based
parenting skills program.

It is of interest that none of the evaluative studies
assesses the effectiveness of training in reducing foster
parents' dependence on placement agency staff. While the

failure of agency staff to respond promptly to foster

13

parents' needs apparently influences parents' decisions to
retain the children placed in their care, little attention
has been paid to this variable. Clearly, there is a need for
prospective evaluations of foster parent training programs
which include number of contacts between foster parent and
agency staff as an outcome variable, as this would provide
an indication of the effects of training on foster parents'
abilities to function with relatively high degree of
independence following training.

Recent developments in the field of parent training
(Wolfe et al., 1981) and stress innoculation theory
(Meichenbaum, 1977) suggest that the inclusion of a self-
management component in foster parent training programs is
likely to enhance foster parents' effectiveness. Meichembaum
and his colleagues (1982) have reviewed results of
laboratory studies which suggest that many individuals fail
to respond appropriately in stressful situations, not
because of a lack of appropriate skills, but because of a
failure to use those skills due to the interference of non
task-oriented cognitions. Thus, it appears that providing
foster parents with parenting skills is only part of the
solution. Since foster parents working with emotionally and
behaviorally disturbed children operate under stressful
conditions (Gillham, 1983), it appears that training which
combines parenting and self-management holds greater promise

than that which stresses only parenting skills. It was

14

thought that such training would allow
foster parents to become aware of the cognitions that
mediate their responses to problem behaviors and provide
them with an opportunity to alter these in favor of more
task-oriented responses. In addition, training would equip
foster parents with coping skills that could be used
to reduce emotional arousal. Parents who receive such
training, as compared to parents who do not, can be expected
to make more effective use of learned skills in crisis
situations and to show greater tolerance for behaviorally
and emotionally disturbed children.

In summary, it appears that foster parents are not
adequately prepared to respond to the needs of behaviorally
and emotionally disturbed children. This situation results
in placement failures, which further strain the resources of
foster family care. The effects of replacements or failed
,placements prevade all aspects of foster family care.
Foster children, foster parents and placement workers are
adversely affected. Failed placements drain the field of
experienced parents, a much needed but limited resource, and
compound the problems of finding and training an adequate
supply of foster parents to provide children with stable
care. Foster parents become dependent on agency staff for
guidance, support and reassurance while the latter are

constantly responding to perpetual crises which hinder their

3':

Us

15

ability to respond to the on-going needs of foster parents
-not experiencing crises.

It appears that foster parent training constitutes the
best form of intervention to address the problem of
placement stability and its consequences. While a number of
training programs have been developed, there is a need for
continuing program development, particularly since existing
training programs have not been found effective in reducing
placement breakdown in carefully designed evaluative studies
(Slattery, 1980; Slobodian, 1980). There is also a need
to evaluate training programs with respect to more tangible
outcome measures. More important, however, is the
possibility that current training curricula may be
inadequate in preparing foster parents to work with problem
behavior. Recent developments in the treatment of abusive
parents through training and in the field of stress
innoculation indicate that self-management techniques may
hold some promise of efficiency for future foster parent
training programs, as foster parents are known to be exposed
to psychologically and physically threatening situations
that may stimulate cognitions that are likely to interfere
with their ability to use acquired skills in times of

stress.

16

Purpose

The major purpose of this study has been to determine
the differential effectiveness of two behaviorally-based
foster parent training programs on measures of (1)
perceptions of family environment, (2) foster parents' sense
of parenting efficacy, (3) placement stability, and (4)
dependence on agency staff. Several investigators have
indicated that foster parents have expressed a need for
training in managing foster children's behaviors (Cautley,
1980; Hampson 8 Tavormina, 1980). Both parenting training
and parenting plus self-management training programs have
been found useful in helping parents handle behavior
problems (Karoly 8 Rosenthal, 1977; Wolfe et a1, 1981). Of
the two programs, one provides both parenting and self-
management skills, while the other provides parenting skills
only.

Another reason for this particular investigation was
that while both of these types of training have been found
effective, there have not been any comparative
investigations of their differential effectiveness. The
need to provide the field of foster family care with
empirically tested and efficacious training programs
provided another reason fo this comparative study.

The fourth and last purpose of this project was to

provide foster parents with training in behavior-based

17

parenting techniques that have been found helpful with other
parent groups (Karoly 8Rosentha1, 1977; Whaler et al., 1973,
Wolfe et al., 1981). Moreover, although the need for
universal training of foster parents has long been
recognized (Goldsteinn, 1967; Vassaly, 1976), many foster
parents continue to receive little or no training at all or

receive training of dubious quality.

Statement of the Problem
and Research_guestions

This study was conducted in an effort to evaluate the
differential effectiveness of two 16-hour foster parent
training programs. The models compared in the study
consisted of:

(1) Parenting and self-management skills training

(2) Parenting training only.

Four research questions based on the previous
discussion were formulated to guide the inquiry.

1. Does training in parenting and self-management
skills, as compared to training in parenting
skills, lead_ to improved family environment as
perceived by foster parents?

2. Does training in parenting and self-management
skills, as compared to training in parenting
skills only, lead to an increased sense of
parenting efficacy in foster parents?

3. Does training in parenting and self-management

skills, as compared to training in parenting
skills only, lead to fewer placement breakdowns?

18

4. Does training in parenting and self-management
skills, as compared to training in parenting
skills only, lead to fewer contacts with agency
staff?

Hypotheses

Based on the research questions, ten experimental
hypotheses were formulated.

The following hypotheses were related to Question #1.

H1:l Foster parents trained in parenting and self-
management skills will score significantly higher
on the Cohesion scale of the Family Environment
Scale than those trained in parenting skills only.

Hl:2 Foster parents trained in parenting and self-
management skills will score significantly lower
on the conflict scale of the Family Environment
Scale than those trained in parenting skills only.

Hl.3 Foster parents trained in parenting and self-
management skills will score significantly higher
on the control scale of the Family Environment
Scale than those trained in parenting skills only.

The second hypothesis is related to Question #2.

H2:1 Participants trained in parenting and self-
management skills will score significantly higher
on the Foster Parent Parenting Efficacy Scale than
parents given parenting training only.

Hypotheses related to Question #3 are:

H3:l Participants trained in parenting and self-
management skills will express significantly fewer
requests for removal than those trained in
parenting only.

H3:2 Participants trained in parenting and self-
management skills will experience significantly
fewer unrequested removals than those trained in
parenting only.

19

H3:3 Participants trained in parenting and self-
management skills will have significantly fewer
children running away from their homes than those
trained in parenting only.

Hypotheses related to Question #4 are:

H4:1 Participants trained in parenting and self-
management skills will need significantly fewer
home visits than those trained in parenting only.

H4:2 Participants trained in parenting and self-
management skills will have significantly fewer
office contacts than those trained in parenting
only.

H4:3 Participants trained in parenting and self-
management skills will have significantly fewer
problem-related telephone contacts than those
trained in parenting only.

Definitions of Terms

 

The following terms are defined in the context in which
they were used in this study.

Fosteryparent.--a person who is licensed by the state
to provide foster care services.

Request for removal.--a verbal or written communication

 

from foster parent asking for the removal of a foster child.

Home contact.--any documented visit to foster parent

 

home.

Office contact.--any documented office visit by a
foster parent.

Phone contact.--any documented telephone contact

with foster parents, for purposes of resolving a problem

20
situation, monitoring treatment progress or establishing
rapport with foster parent.

Unrequested removal.--the removal of a child from a
foster home prior to the contracted termination date, due to
violations of foster care rules or to foster parents'
inability to function as parents.

Incidence of child running away.--the number of times a
foster child has failed to return to his foster home within
a 24-hour period.

Cohesion.--the degree of commitment, help, and support
family members provide for one another (Moos 8 Moos, 1981).

Conflict.--the amount of openly expressed anger,
aggression, and conflict among family members (Moos 8 Moos,
1981).

Control.--the extent to which set rules and procedures

are used to order family life (Moos 8 Moos, 1981).
Theoretical Foundations

This research study derived its conceptual framework
from family systems, social learning and cognitive behavior
theories. The assumptions underlying each theoretical
perspective will be briefly reviewed in light of their
implications for this investigation.

In the view of systems theory (Miller, 1965; Skynner,
1976), the family is made up of interdependent subsystems
that constitute individual family members in the many

combinations that they can form. Some subsystems derive

21

their identity from biological characteristics like gender,
mother or children, while others appear to be subsystems as
a result of their functions and the role they play in the
family system. Thus, the adults in one family, that is the
husband and wife, can be thought of as belonging to the
spousal, parent and provider subsystems.

Since the subsystems are interdependent, changes in one
affect all of the others, and thus the family as a whole.
Each subsystem has its needs, which compete with and
complement those of the other subsystems. This competition
is resolved by the development of patterns of behaviors
within each family, which function to regulate interactions
between subsystems so as to protect the viability of the
family. Thus, each family has its rules for meeting the
needs of its members. Everyone knows what is permitted, how
far to go and what he or she is likely to get (Minuchin,
1981). These rules evolve as a family expands through the
inclusion of new members and/or the adoption of new
functions by existing family members. For example, a child
whose family expands through the birth of another child
becomes part of the older brother or sister subsystem, while
also a member of the sibling subsystem. The child, by
virtue of his family's enlargement, acquires new
responsibilities and new privileges, which contribute to the
need for adjustment in existing patterns. By responding to

such events successfully, a family continues to provide its

22

members with the sense of stability that is needed for
security, yet remain flexible enough to allow for individual
growth (Minuchin, 1981).

This suggests that the family as a system is not
static, that it is able to respond with some degree of
flexibility to the growth of its members or other change-
inducing events. The family, however, can afford to be
flexible only within some preset range, apparently
determined by the nature of its structure, or the
established patterns of the relationships among its
subsystems. Each family has a limit but that limit differs
for individual families. This threshold appears to
function as a feedback mechanism that activates the
hemostatic responses of the family system. In other words,
the family has the ability to tolerate and grow from changes
up to a point. If change threatens the system beyond its
level of endurance and endangers its existence, however, the
system is likely to reject change or develop other
mechanisms to reestablish some sense of equilibrium among
its relationships and thus preserve its structure.
Subsystems that are not integrated with the family system
are scapegoated or rejected (Skynner, 1976).

The responsibility for protecting the system, while it
is diffused throughout the subsystems, rests primarily with
the decider-subsystem (Skynner, 1976). This function tends

to rest with the parental subsystem. According to systems

 

23

theory (Miller, 1965), this subsystem gathers information
from other systems and its own subsystems, evaluates the
information in light of the overall systems' interests, and
if the need exists, it initiates corrective action to bring
deviant subsystems' behavior in accordance with the
interests of the whole system. Minuchin (1981), in
describing the responsibilities and the rights of parents,
states:

The adult in the parental system have

responsibility to care for, protect, and

socialize the children, but they also have

rights. The parents have the right to make

decisions that are related to the survival of the

total system, in such matters as relocation,

selection of schools and the determination of
rules that protect all family members. (p.18)

In light of the above discussion, the introduction of
a child into a foster home necessitates changes in the
existing patterns of relationships among the subsystems of
the foster family. It would appear that the family, as a
system, has the potential to adjust to the resulting changes
and integrate the foster child within its existing
subsystems to the extent that it is not pushed beyond its
capacities for endurance or tolerance. Foster family care
professionals have attempted to prevent such overloading of
a family's structure by careful matching of children and
families, but this has not worked effectively because of the

relative scarcity of foster families with respect to demand.

24

Thus, many foster families receive children that severely
tax their resources. Resulting placement breakdowns are
perceived as a failure by the family and/or foster child to
develop successful integration strategies that meet the
needs of the incoming child and those of the family members.
Moreover, it is contended that such breakdowns reflect the
family system's attempt to restore its equilibrium by
getting rid of a divisive element.

The role of the parental subsystem makes it the logical
choice for intervention efforts aimed at the amelioration of
conflictual family climate. This is all the more important
in view of the contention that parents' perception of the
family climate has motivational potential (VenDerVeen,
1965). Increasing recognition of the influence of cognitive
processes on behaviors underscores this point (Bandura,
1981). Moreover, findings from several studies indicate
that parents' perception of problem behaviors do not agree
with those of trained observers (Karoly 8 Rosenthal, 1978;
Arnold et al., 1975). This suggests that parents may be
responding to much more than observable behaviors, that the
meaning attributed to a problem behavior is affected by
factors other than the actual behavior.

The parental subsystem Optimally reflects the interests
of the family as a whole in its decisions. Its purpose is
to serve, as well as to guide the other subsystems. Thus, a

foster parent may request the removal of a foster child with

25

whom he or she is able to work, because other family members
are so distraught over the child's behavior that the parent
feels that removing the child will be best for the family.
Parental perceptions function as a mediating variable that
has the potential to influence foster parent behavior.
Findings suggesting that foster parents' perceptions of
family climate are positively related to placement stability
appear to lend credence to this contention (Levant 8 Geer,
1981).

Since the foster parenting experience is likely to
produce some disturbance in family equilibrium (Wilkes,
1974) it is likely that foster families experience changes
in perceived cohesion, conflict and control within the
family. Researchers have found that distressed families
experience less cohesion and more conflict (Scoresby and

Christensen, 1976).

Social Learning Theory

The view of the family as an interactive system is also
shared by proponents of social learning theory who postulate
a reciprocal influence in parent-child interaction. Because
parents' behaviors toward their children can be rewarded,
reinforced, punished or extinguished, it is thought that
children teach their parents just as parents teach their
children. Patterson et al.(1975), found that interaction

patterns in families with aggressive children to be aversive

26

for both parents and children and to increase in
aversiveness over time.

Since behaviors usually occur within a social context,
their reinforcement depends to a large measure on the
behaviors of others in that context. This implies that
helping foster parents become aware of how they unwittingly
contribute to the acquisition and maintenance of socially
maladaptive behaviors will lead to the adOption of more
adaptive behaviors by both parents and children.
Behaviorally-oriented parent training has been found
successful in the elimination of undesirable behaviors and
in producing positive changes in family functioning (Arnold

et al., 1975; Karoly 8 Rosenthal, 1977).

Cognitive Behavioral Approach

While training foster parents in behavioral principles
is likely to be influential in improving family environment,
reducing placement breakdowns and other placement related
events, there is reason to believe that parenting training
alone may not be sufficient. Recent developments in the
cognitive behavioral approach have highlighted the impact of
cognitive variables on behavior. Bandura (1982), in
discussing the significance of cognitive processes, notes
that "knowledge, transformational operations and component
skills are necessary but insufficient for accomplished

performances” and that "people often do not behave

27

optimally, even though they know what to do " (13122).
This, he explains is because self-referent thoughts mediate
the relationship between knowledge and action. It is
assumed that psychological interventions influence behaviors
through their effect on self perceptions of efficacy
(Bandura, 1977). Other investigators have reported findings
suggesting that self perceptions of personal efficacy are
more accurate predictors of performance than past behaviors
(Lee, 1982) and are more reliable predictors of performance
across treatment modes (Bandura et al., 1977; Brown 8
Inouye, 1978; Schunck, 1981).

Viewed from this perspective, this investigation
attempted to influence self perceptions of efficacy and thus
the behavior of foster parents through two modes of
treatment. One mode utilized parenting training alone, the
other combined parenting training with training in self-
management skills.

Meinchembaum (1982) indicates that effective responding
requires ”attention to the needed skills and to the factors
that contribute to the facilitation or inhibition of these
skills ' (p.293). It has been hypothesized that a person's
response to a given situation is influenced more by his
perception of that situation then by the intrinsic
characteristics of the situation (Bandura, 1982;
Meichenbaum, 1977, 1982). This perception is thought to

have the potential to trigger emotional and physiological

28

reactions, which can adversely affect performance,
regardless of ability. Thus, it was reasoned that a skills
training curriculum which combined parenting and self-
management, would facilitate the application of learned

child management principles in crisis situations.

Overview

Chapter Two will present a review of the historical
development of foster care and a discussion of the
literature related to this investigation. The research
design and procedures will be described in Chapter Three.
Chapter Four will present the results of the statistical
analysis of the data. Lastly, a discussion of the results
and implications for future research will be presented in

Chapter Five.

CHAPTER TWO

REVIEW OF THE LITERATURE

This chapter provides an overview of the historical
development of foster family care and a discussion of
evaluative studies of foster parent training curricula. It
includes a review of literature pertinent to this research

investigation.

Historical Overview

Foster family care has existed in various forms
throughout recorded history (Kadushin, 1974). Each
generation appears to have dealt with the task of providing
for parentally-absent children in accordance with its
needs and understanding of what constitutes the best long
term interest of society. It is not suprising that
practices that are deemed salutary by one generation become
questionable and may even be considered injurious by later
generations.

For most of modern times until the latter part of the
nineteenth century, the survival of children with special
needs depended to a great extent on their ability to provide
labor in exchange for food and shelter. Indenture, as a
form of substitute care sanctioned by Elizabethan laws, was

29

30

widely practiced in the nineteenth century and began to
decline only in the latter part of the century as
manufacturing shifted from homes to industrial sites
(Kadushin, 1974).

Declining interest in indenture coincided with an
increase in the number of children in need of parental
supervision in the industrialized cities of the eastern
United States. This situation was dealt with by referring
such children to adult institutions. Bryant (1981)
indicated that "homeless and neglected children commonly
were assigned the mixed almshouse and workhouse for the
poor, or to adult prisons if their behavior was considered
delinquent " (p.10).

Dissatisfaction with indenture and mixing children with
adults in institutions provided an impetus for the
forerunner to foster family care. The New York Children's
Aid Society, under the leadership of Charles Loring Brace,
in 1853 undertook to place homeless and poor children from
New York with families interested in providing loving care
in return for child labor. Children were placed with
families living in New York and in the midwestern states
(Langsam, 1967). The ”placing-out" program, as it was
called, represented only a slight improvement on the
practice of indenture. Poor parents who could not provide
for their children had to give up legal custody permanently

so their children could be placed (Kadushin, 1974). There

31

was little information on the capacity of individuals
wantinga child to provide surrogate parenting, other than
their willngness to accept a child and no adequate
supervision followed the placement of a child. Bryant
(1981) aptly summarizes the general operating procedures of
the "placing-out" and other family care programs of that

period.

Children were selected for placement almost
exclusively on the basis of the poverty of their
backgrounds, even when their natural families
were intact. Prospective homes were not
investigated, or, for that matter even identified
for the individal child until the day and hour of
placement. Once in the home, the child was left
to the good graces otherwise of his new family.
(p.10)

The shortcomings of this form of foster care, which
neglected the individual needs of children, provided the
impetus for the professionalization of foster family care.
The Boston Children Aid's Society, under the leadership of
Charles Birtwell, introduced a more comprehensive, child-
oriented approach that was to become the foundation for the
modern foster family care system (Wolins, 1963). In
contrast the Brace's approach, Birtwell's emphasized the
individual needs of each child. ”The society advocated for
children by emphasizing the need for investigation and
careful selection of prospective foster homes, continuing

placement supervision, thorough recordkeeping and periodic

32

review of cases " (Bryant, 1981, p. 11). Foster care was
seen as only a temporary solution and children placed out of
their homes were to be returned to their parents once the
parents became able to provide for them. Birtwell's
approach preserved the family ties after placement that
Brace's approach terminated.

With the increased professionalization of foster family
care, came a change in the role of the foster parents. In
1866, Massachusetts began ”paying board money to foster
families for the maintenance of children who might otherwise
have been placed in institutions and who were too young to
be profitably indentured " (Kadushin, 1974, p. 399k By
paying foster parents to care for- children, placement
agencies acquired a wider selection of homes, greater
stability of homes and experienced fewer problems in
supervising foster homes (Kadushin, 1974). More important,
it eliminated the need for foster children to exchange their
labor for food and shelter. Foster parents became temporary
custodial caretakeres instead of employers.

From the early years of this century until the late
19508, foster family care was considered the preferred form
of substitute care for parentally-deprived children. Prior
to the Second World War, such care was generally reserved
for children who were free of physical disabilities,
emotional disturbances or behavioral problems. Since foster

children tended to be relatively free of problems, foster

 

 

 

33

parents were not provided with any special training. What
was thought to be important was the accurate matching of
foster parent and foster child. Training, when it did
occur, took the form of advice and suggestions on how to
handle specific placement-related problems. By and large,
foster parents functioned on their own and were relatively
independent of placement agencies because the limited
resources of the latter did not allow placement personnel to
make frequent visits to foster homes. McCoy and Donahue
(1961) aptly describe the position of the placing agencies

of that period.

We are convinced that our foster homes are the
most valuable resource for the care of our
children. But, like most placing agencies,
particularly public agencies with a large number
of children in placement, we have long been aware
that we have foster parents of vastly different
capacities, who vary widely in knowledge and
understanding of the problems children have and
in methods of handling them. Also, like many
other agencies across the country, we are
sometimes faced with a lack of staff which makes
it impossible to reach all of the foster homes as
often as we would like. (p.29)

Foster parents responded to the task in light of their own
perceptions of children's needs and their own experience
with children.

While this strategy may have been adequate for
relatively well-adjusted children, it clearly .became

inadequate when the traditional clientele of foster family

care began to change. Due to a combination of factors, one

 

 

 

34

of which was greater understanding and appreciation of the
importance of family relationships for personality
development (Bryant, 1981), child care professionals, who
were predominantly trained in the psycholanalytic approach,
began to place maladjusted children in foster family care at
about the time of World War Two (Bryant, 1981), without
expanding the capacity of the system or improving its
services. Child welfare professionals were asking foster
family parents to undertake a task for which they were ill
prepared.

Results of empirical investigation of the effects of
foster care clearly reflected the system's inadequacy in
responding to the needs of foster children and families.
Maas and Engler (1959) conducted an investigation of 4,281
children's case records in nine communities in an attempt to
identify conditions which kept children in foster care and
precluded adoptions. Their findings suggested that children
were being maintained in permanent impermanence upon
entering the system. Children were found to experience two
to three placements and to spend two to five years in care.
The findings of a study comparing 140 foster children
referred for psychiatric evaluation to 440 foster children
who had not been referred for such evaluation suggested that
length of placement was related to referral for psychiatric
evaluation and multiple placements (Eisenberg, 1962). The

longer the time spent in placement, the greater was the

 

35

likelihood of replacement. Replacement was also found to be
related to behavioral problems. Children with behavior
problems tended to be moved more often than those who did
not display such difficulties. The results of these two
studies, which were replicated by other investigators,
indicate that foster parents were experiencing difficulties
in discharging their roles. Thomas (1961) noted that
"foster parents have not expected or been prepared to deal
with children showing such a magnitude of problems.‘ (p.
218)

Empirical support of foster parents' inadequate
preparation to handle behaviorally disordered children, was
provided by a study undertaken by Ambinder and his
colleagues (1965). Fifty 'foster parents caring for
delinquent boys ranging in age from eight to twelve years of
age were asked to describe incidents they experienced and
how they responded to them. Seventy-three percent of the
techniques used by the parents were those considered
unlikely to enhance personal control or to encourage
development. Furthermore, it was felt that some might even
be harmful. Sixteen percent were judged likely to
facilitate control and eleven percent were thought to be
neither farmful nor helpful.

Professionals in the field responded to this situation
with three remedies. One approach attempted to provide

empirical evidence which could be used in the selection of

 

~__

36

prospective foster parents. Investigators tried to identify
demographic, attitudinal and personality correlates of
successful foster parents and families. It has been
reported that families that are willing to participate in
new experiences, have the ability to find adapting to new
and different situations enjoyable, and allow family members
to be independent are positively associated with stable and
enduring placements (Levant and Geer, 1981). These
researchers also found that prior experience with natural
children of comparable age to the foster child, prior foster
care placements and fewer contacts with social workers, were
related to placement stability (Levant 8 Geer, 1981).
Cautley and Aldrige (1975) reported that high religiousness
and the presence of one or more school-aged children in the
home were negatively associated with successful placements.
Positive placement outcomes were found to be associated with
familiarity with child care and ability to cope with
problematic behaviors (Cautley and Aldrige, 1975). Babcock
(1963) found that foster parents who enjoyed caring for
preschool aged children tended to come from large families,
had limited education, married early and led stable, non-
eventful lives. He described them as stable and healthy but
somewhat rigid in their approach. Fanshel (1961) indicated
that foster parents who appear to enjoy caring for
behaviorally disordered children, obtained high ratings from

caseworkers for emotional maturity, democratic family

37

relationships and identification with the role of the foster
parent. Those foster parents who cared for non-aggressive
physically disabled, mentally retarded and emotionally
disturbed children tended to prefer younger children. These
parents were found to obtain lower ratings for maturity and
democratic family relationships than those who cared for
aggressive youngsters. Both groups of parents identified
with the foster parent role and had had a good deal of
experience with their own children.

The second approach consisted of providing more
specialized services to foster children with physical
disabilities, behavioral problems and emotional
disturbances, which led to the development of a specialty
within foster family care known as specialized foster care
(Gillham, 1983). Although it was first introduced in the
19508, the concept of specialized foster care received its
main thrust during the 19708. Gillham (1983) indicated that
depressed economic situations and questionable treatment
outcomes provided the impetus for community-based treatment
alternatives. The relatively high cost of maintaining
children in institutions and the undesirable effects of
institutionalization in furthering maladaptive behaviors
fueled the deinstitutionalization movement of the 19608
(Bryant, 1981; Gillham, 1983). Other factors which
generated that movement included a reappraisal of the

importance of family life and greater appreciation and

38

understanding of how behaviors are generalized from one
situation to another. If the goal of the habilitation and
rehabilitation of delinquent and disturbed children was
their eventual return to the community with the ability to
function adaptively, it was felt they needed to undergo
treatment in a minimally restrictive environment where they
could acquire the skills that facilitate adaptive
functioning (Bryant, 1981; Gillham, 1983). To keep such
children in institutions was considered tantamount to
condemning them to perpetual institutionalization, as
institutions have their own demands, behavior codes and
dynamics which are of questionable value to independent and
adaptive adjustment in the real world. More important,
however, is the issue of the transfer of adaptive behaviors
from the artificial and restricted setting of the
institution to the larger community. Keeping children in
institutions deprives them of the opportunity to acquire
skills needed for community living. Gillham (1983) points
out ”the recognition that most children will eventually
return to a home situation and that the home provides
training for future parenting skills provided further
justification for the development of community based foster
care programs ' (p. l). The significance of modeling for
behavioral acquisition lends credence to the importance of

treating disturbed children in the community (Bandura,

1977).

39

The introduction of formerly institutionalized children
into foster family care meant that added specialized
services were needed to enable foster parents and agency
personnel to address the treatment needs of the children and
the parenting needs of the parents. Efforts have been made
to reduce the ratio of children to workers, thereby allowing
workers to spend more time per child. Many such foster
children also receive therapy on an ongoing basis as an
integral part of their placement (Bryant, 1981).

The severity of the disturbances presented by
previously institutionalized children has contributed to a
reconceptualization of the foster parent role (Bauer 8
Heinke, 1976). Foster parents are no longer simply
custodial caregivers. An increasing number of child care
professionals are calling for foster parents' integration
into the treatment team as co-professionals with agency
workers (Bauer 8 Heinke, 1976; Larson et al., 1978; Murray,
1982). Larson et a1 (1978) describe a specialized foster
care program for severely disturbed children aged 8 to 16,
in which parents function as parent-counselors. Alberta
parent counselors have combined aspects of institutional
treatment and regular foster care in a goal-directed
treatment program that is thought to have the advantages of
a residential treatment center while children reside in
regular foster homes. Foster parents play a crucial role in

the development and implementation of the treatment program.

40

Prior to receiving any children, the parent-counselors
receive 35 to 40 hours of training in such areas as
communication skills, behavior management, work with natural
families, theories of personality development, sexuality and
legal departmental policy. After training and the placement
of foster children, these parent-counselors meet as a group
for support and supervision. As in regular foster care,
parent-counselors receive a daily allowance for each child,
plus payment for other expenses such as clothing, school
supplies and training costs.

While other specialized foster care programs do not
appear to give foster parents the same degree of
responsibility as the Alberta Parent-Counselors Program,
they all stress the need for training and recognition of
foster parents as a crucial element of the treatment team.
Freeman (1978) indicated that foster parents involved in a
special foster care program for mentally retarded children,
in Pittsburg, were paid salary and considered regular
staff. Foster parents in the program received in-service
training and enjoyed the frequent contacts with agency
staff. Another program that addressed the needs of
emotionally disturbed, behaviorally disordered and mildly
retarded chidlren aged 3 to 18, provided foster parents with
ten hours of formal didactic training that emphasized
orientation to the agency's program, nature of the foster

parent role, and child behavior management. In addition to

41

the pre-service training, foster parents receive on-the-job
training through frequent contacts with service program
managers. Trained foster parents receive payment in
addition to the regular room and board payment in
recognition of their level of skill and competence (Bryant,
1981).

While specialized foster care has emphasized increased
supportive services to both foster children and foster
parents, a third approach has emphasized increased
supportive services to foster parents only. In recognition
of the need to integrate foster parents into the treatment
team, Thomas (1961) acknowledged that changes in the types
of children entering foster care implied changes in the role
of foster parents. She recognized that foster parents
needed to do more than provide food, shelter and love and
argued they needed to be "part of a child-care team of
experts, helping children who are emotionally upset and who
face more than the usual problems in reaching adulthood '
(Thomas, 1961, p.218). In order to provide foster parents
with the needed skills, group training was introduced to
supplement (but not replace) the consultative contacts that
occurred between foster parents and agency workers (McCoy 8
Donahue, 1961; Soffen, 1962; Thoma8,196l). These training
groups provided exposure to a variety of topics related to
the foster care field and foster care parenting. Taylor and

Starr (1967), in summarizing the benefits of group training,

42

indicated that such training was thought to contribute to a
decreased sense of social isolation, increased ability to
discuss feelings and problems with the caseworker, improved
role classification between foster parents and social
workers with respect to the child's life, increased
recognition of foster parent as colleagues instead of
clients, greater identification with agency goals, greater
opportunity to gather additional diagnostic information,
regarding foster children, and improved ability to determine
professional and policy problems (p. 371).

Soffen (1962) evaluated the effectiveness of a 32-week
group training program in which the curriculum depended on
the contributions of the participants but covered such areas
as the problems children present and how to handle them,
relationships with natural parents, agency policies and the
role of the caseworker. The training sought to give parents
problem-solving skills, through emphasizing the purposive
nature of behavior and testing approaches to respond to
problem behaviors. The 20 foster parents who completed the
program were compared to S4 foster parents who had not had
the program. Experimental and control groups were compared
with respect to (1) relationship with casework and agency,
(2) understanding growth needs of children and skills in
dealing with these needs, (3) understanding meaning of
behavior and skill of dealing with these, (4) motivation for

becoming foster parents, (5) psychologaical and emotional

43

climate in the family, (6) understanding the importance of
natural parents to foster children and ability to respond
appropriately to natural parents. Using caseworker ratings
as a measure, Soffen found that the experimental group
scored significantly higher than the control gorup on five
of the six dependent measures. No significant differences
were obtained in ratings of family climate. Flaws in the
design of this study, including selection of the subjects,
suggest caution in interpreting its results. Foster parents
in the experimental group may not have been comparable to
those in the control group.

While foster parent group training continues to be
offered by agencies, there has been a trend toward better
focused, single topic, skills-oriented training. Many of
these training programs were developed in universities and
combined didactic instructions with group learning (Stove 8
Hunzeker, 1978).

Evaluative studies of several types of training
programs will be reviewed in the succeeding pages. Studies
are categorized by their designs, so that group case studies
will be discussed first, followed by those with

correlational, quazi-experimental and experimental designs.

Group Case Studies

Penn (1978) evaluated the effectiveness of an 11-week,

behaviorally-oriented foster parent training program.

44

Subjects consisted of seven foster parents who were selected
for their learning potential. Experience as foster parents
varied from none to six years. Completed Child Behavior
Inventories on 15 children of these parents indicated that
131 specific behaviors improved and 27 deteriorated during
the duration of the training. Improvements and
deterioriation were noted for both natural and foster
children. The eight natural children who were observed
improved on an average 4.8 behaviors and deteriorated on
1.25 behaviors. An average 13 instances of positive and 2.5
cases of negative change were reported for the seven foster
children in the study. As there was no comparison group,
the results of this study are, to a large extent,
uninterpretable.

Similarly, Lloyd, using a one-group design, attempted
to evaluate the effects on foster parents of parenting
delinquents. A group of six families, that were judged to
have a ”conductive home environment," were selected from 18
volunteers. Selected parents attended one week of intensive
training on working with delinquents, met weekly for group
counseling and participated in weekly family therapy
sessions. Therapeutic activities employed an interpersonal
approach. Comparison of pre- and posttest scores, six
months after training revealed that male foster parents
experienced growth in the areas of self worth and reliance

as compared to foster mothers. As in the previous case, the

45

lack of a control group suggests caution with respect to
these results. Moreover, the extensive amount of supportive
services provided to the foster families render this
particular program so atypical that it is unlikely it could

be carried on a large scale for any extended period of time.

Correlational Evaluations

Relationships between foster parent training and (l)
placement disposition, (2) length of placement, (3) foster
parent rating of placement success, and (4) licence
retention, were examined by Boyd 8 Remy (1978), who examined
267 placements in four groups,- two years after an intensive
l6—week behaviorally-oriented foster parent training
program. The groups consisted of (1) parents who received
training prior to their first placements and who kept their
children for two years or less, (2) those who did not
receive training and kept their children for less than two
years, (3) those who received training after they had their
placements and retained their foster children for more than
two years, and (4) those who did not receive training and
retained their foster children for two years or more.

V Recognizing the weaknesses inherent in the design of
their study, Boyd 8 Remy (1978) assessed the comparability
of their groups with respect to foster parents' experience,
environmental stress, and children's characteristics. The

two groups which had received training prior to placement

46

did not differ on environmental stress, number of prior
placements and behaviors of foster children. They differ,
however, with respect to experience, foster children's sex
and age at placement. These two groups consisted of 55
trained and 113 untrained parents and were considered the
primary sample. Multiple regression analyses were used
in order to determine the effects of training, experience,
environmental stress and foster children's characteristics
on the dependent measures. The results revealed that
training was a significant predictor of placement
disposition, placement success rating and license retention.

Analyses of the data pertaining to foster parents who
received training after obtaining their placements indicated
that trained parents worked with more difficult children.
Training was a significant predictor of licensing retention
and placement disposition but the relationship was not as
strong as those obtained with the primary sample.

While the authors of this study have done a fairly good
job in attempting to assess the equivalence of the two
groups of trained and untrained parents, selection bias
precludes clear interpretation of the results. It can be
argued, as the researchers themselves noted, that the
results may have been due to personal differences between
the two groups or to a combination of training and personal
attributes. This is especially probable since the foster

parents volunteer for the training. Moreover, a large

47

percentage of the trained parents participated in other
training or educational activities after the behavioral
training ended.

Since many foster parents participated in more than one
type of training over the two-year span, the results cannot
be attributed solely to the original behaviorally-oriented
program. The results, rather, point to the efficiency of
training in general and are perhaps more generalizable to
foster parents who have attended more than one training
activity. Boyd and Remy (1978) report that most of the
trained, as compared to a few of the untrained, parents
participated in three or more activities offered through the
Foster Parent Association. In spite of its limitations,
this study represents an effort to evaluate training with
respect to outcome measures that are relevant and important

to the field of foster family care.

Quasi-Experimental Studies

 

Cobb et al., (1982) compared 30 foster parents who
received training in communication and conflict resolution
skills to 18 matched "no contact“ controls. They found that
trained parents differed significantly on measures assessing
communication and problem solving skills from control group
parents. Assessment measures consisted of audio tape

recorded responses to stimulus situations.

48

Another group of investigators (Levant 8 Geer, 1981;
Levant, Slattery 8 Slobodian, 1981) conceived of the foster
parent's role as that of a paraprofessional mental health
counselor. In accordance with this premise, they selected a
group of foster parents who displayed relatively high levels
of empathy, genuiness and capacity to be nonjudgemental. Of
the chosen parents, 9 were assigned to the training
condition and 6 to a 'no contact” group based on subjects'
time availability. The experimental group received training
in helping and communication-oriented parenting skills. No
significant differences were found between the two groups
with respect to foster child's self esteem, behavior
adjustment and placement stability. The only significant
result was related to emotionality and tension. The
investigators suggest that this result may be spurious. No
significant improvement in helping and parenting skills were
obtained and it is thought (Levant, Slattery 8 Slobodian,
1981) that the findings were due to difficulties encountered
by the subjects in their daily lives, which interferred with
the instructional plan.

A few years earlier, Guerney (1976) reported that
parents trained in the Foster Parent Skill Training Program,
showed more favorable attitude of parental acceptance than
matched controls on the Porter Parental Acceptance Scale.
Similar results were obtained on Sensitivity to Children

Questionnaire. The investigator relied on a sample of 132

49

subjects, 57 of whom constituted a ”no contact" control

group.

Experimental Studies

 

Brown (1980) compared two training programs with
respect to their effectiveness in affecting parental
attitudes sensitivity to children and problematic behaviors.
One program, the Foster Parent Skill Training Program
(FPSTP), developed by Guerney (1976) emphasized
communication and methods of handling problematic behaviors.
The other, Issues in Fostering (IF), was developed by Ryan
(1977) and was concentrated on issues related to foster
parenting. Using an initial sample of 76 volunteers, Brown
randomly assigned subjects to one of five treatment groups
and found there were no differences between the training
groups and a "no contact" control group on parental
attitudes measured by the Foster Parental Acceptance Scale,
Foster Parent Attitude Survey and the Sensitivity to
Children Scale. A difference, favoring the FPSTP, was
observed with respect to sensitivity to children when
compared to the IF program. These results, however, were
obtained from a sample of only 59 subjects, decreased due
to attrition, mainly of subjects assigned to the control
condition.

In evaluating the study, Brown noted several flaws

that may have affected the results. It was observed that

50

pretest scores of the control subjects were higher than
those of any of the other groups.' He reasons that, since
control subjects did not receive any form of training or
experimenter-derived reinforcement, the control subjects who
returned the assessment instruments might have been those
who were most highly motivated and verbally skilled.
Moreover, it may be argued that those foster parents might
have represented those who felt more competent in their
roles as foster parents. As Brown suggested, the relatively
high attrition of 50 percent of the control group could have
adversely affected the initial randomization.

Another area of possible bias identified by Brown
concerned the measuring instruments, which were time-
consuming and required verbal ability and skill, which may
have affected some subjects' ability to respond. Brown also
explained that the ”written responses may not translate into
behavioral changes, and conversely these instruments may be
insensitive to some behavioral changes that did occur as a
result of training ' (p. 106).

Corroboration of the results by Penn (1978), reported
under ”Group Case Studies," was provided by the findings of
another comparative study of the differential effectiveness
of two foster parent training modes done by Hampson and
Tavormina (1980a). Behaviorally-based parent training was
found to be more effective in improving interactional

patterns between foster parents and children and in reducing

51

problem behaviors, as measured by observation and foster
parent reports than reflective group counseling and a "no
contact" control group. Additionally, behaviorally trained
parents were found to view themselves as causal agents in
changing foster children's behaviors and felt more satisfied
with their role. Reflective training was found superior in
improving parental attitudes.

These results carry greater significance than those
reported by Penn (1978) and Boyd 8 Remy (1978). Of the 42
subjects who volunteered for training, 17 were randomly
assigned to the behavioral training group and 16 to the
reflective training condition. Nine foster parents who
were unable to attend were placed in a waiting list control
group. Subjects were members of lower socioeconomic
classes.

In contrast to the investigations conducted by Penn
(1978) and Boyd 8 Remy (1978), the results of‘ this
investigation were not confounded by the selection process.
Foster parents' participation did not depend on their
learning potential as in Penn's study. Thus, they were more
likely to be representative of parents who volunteer for
training. By comparing volunteers with volunteers, the
validity of the results of this study was strengthened by
eliminating selection as a plausible explanation. More
important, however, this study constitutes an evaluation of

behavioral versus other types of training for foster

52

parents, while Boyd 8 Remy's does not. Since the parents
trained in their sample tended to participate in at least
three educational activities, the investigation can be
construed as an evaluation of the effectiveness of training
versus no training.

Unlike Boyd 8 Remy (1978) who followed their subjects
over a two year period, Hampson and Tavormina (1980a) did
not include follow up measures in their design. Subjects
were tested prior to and after training ended. Training

lasted eight weeks.

Summary

Review of the evaluative studies of skills-focused
parent training programs reveals several issues that need to
be noted and addressed. First, it appears that evaluative
investigations are becoming more rigorous in their designs.
This is a positive trend in a field in which there is a
burgeoning interest in training program development.

The second issue of concern is an apparent reluctance
to evaluate training programs' effectiveness on placement-
related outcomes. Most of the evaluative studies have
focused on assessing changes in parental attitudes and
foster children's behavior. Changes in parental attitudes
did not appear to lead to changes in parenting nor to
changes in foster children's behavior according to the

results from Hampson and Tavormina's (1980) study. The

53

training that led to significantly improved parental
attitudes did not generate significant changes in parents'
or children's behaviors.

While the interest in assessing behavior change as a
result of training is understandable in light of its
relationship to placement related problems (Pardeck, 1982),
it does not preclude the need to evaluate training with
respect to placement breakdown and other placement related
difficulties. Since the family functions as an interactive
set of subsystems, improvement in foster children's behavior
may not be sufficient in improving placement stability. In
discussing the results of his study, Penn
reported that one parent requested the removal of her foster
child in spite of improvement in the child's behavior,
because of deterioration in the behavior of her natural
children. It is thus suggested that evaluative programs
follow the example of Boyd 8 Remy (1978), who evaluated
training on outcome variables that are crucial to helping
foster families attain the objecives of foster family care.

The third point of interest concerns the apparent
superiority of behaviorally-oriented training programs.
These programs appear to lead to behavior change in both
children and parents and thus to changes in family
functioning (Hampson and Tavormina, 1980; Penn, 1978).
Moreover, they appear to be effective among various

subgroups of the foster parent population. Penn (1978) has

54

reported their effectiveness with a selected group of
parents, while Hampson 8 Tavormina's (1980) report indicates
their applicability to lower class, rural foster mothers.
Similar results have been obtained with other parent
populations and will be reviewed briefly in the next

section.

Related Literature

 

The literature reviewed in this section is that which
pertained to the parent pOpulation in general. The areas
that are covered, abbeit briefly, include (1) behavioral
parent training, (2) perceptions of family functioning after

treatment, (3) self-efficacy studies.

Behavioral Parent Training

 

Karoly 8 Rosenthal (1977) compared a group of nine
families that had undergone a didactic 10-week behaviorally-
oriented training program to a wait-control group consisting
of eight families and found that trained parents differed
significantly from the controls on measures of parental
perceptions of family environment, child problem behaviors,
and a number of deviant behaviors that were recorded by
trained observers using the Patterson Behavioral Coding

System.

55

Results of the experiment showed that parents who had
received training perceived their family environments as
more cohesive than those who had not. That is, the members
were seen to be more committed to family interests and goals
and more likely to be helpful to and supportive of other
family members. A decrease in parental perceptions of
maladaptive behaviors was also noted after training. Home
observations corroborated these parents' perceptions of the
decrease in their children's maladaptive behaviors relative
to children of control parents. Observers noted a reduction
of 59 percent in deviant behaviors, while parents perceived
a decline of 34 percent. Based on this evidence, the
researchers suggest that a training program which combines
cognitive restructuring and systematic parent training, may
constitute a powerful intervention tool.

The study subjects, who were self-referred because of
difficulties with chidlren who were exhibiting non-
compliance, temper tantrums and/or aggressive behaviors,
were randomly assigned to experimental and control
conditions. Experimental subjects received training in
applying behavioral methods such as positive reinforcement,
consistency, charting, pinpointing, extinction, timeout,
response const and contracting, as well as training in
communication techniques. The training format consisted of
lectures, role playing and large and small group

discussions.

56

A few years earlier, Wiltz and Patterson (1974)
compared six parents who received behavioral training to six
parents on a waiting list with respect to their children's
rates of deviant behavior and found that children of parents
with training experienced significant decreases in observed
aggressive behaviors. Similarly, children of parents who
had had training, compared to those assigned to a control
group, were observed to display significantly less conduct
disordered behavior (Walter 8 Gillmore, 1973).

Of greater interest, however, are studies whose results
indicate that training parents in behavioral techniques has
an enhancing effect on family functioning. This positive
effect appears to be the result of behavioral change on ‘the
part of the parents. As parents change their behavior
through the application of learning principles, these
changes tend to alter previous maladaptive interactional
patterns and produce positive changes in their children
(Hampon 8 Tavormina, 1980; Wahler, 1965). Further support
for this hypothesis is found in the results of an
investigation into the effects of parent behavioral training
on the behaviors of siblings of identified target children
(Arnold et al., 1975). Identified and non-identified
children did not differ significantly with respect to
observed deviant behaviors at baseline. Both groups,
however, showed significant reduction in deviant behaviors

over time. Siblings who were not the target of treatment

57

showed significant reduction in deviant behaviors at
posttest, which were maintained six months later.

The investigators suggested several explanations for
these findings. It was felt that the training provided the
parents with skills which they used in responding to all
children in the family. Training might have also provided
more than instructions in equipping parents with sufficient
understanding and a sense of mastery that enabled them to
apply the principles to children other than those involved
in the clinic referrals.

Another explanation takes into consideratioon the
interactive characteristics of the family. When one child
in a family is experiencing behavior problems, the situation
is likely to trigger negative behavior among siblings. The
chain of events that evolves may be characterized as
reciprocal maladaptive behaviors by the sibling subsystem.
These may be maintained, to some extent, through reinforcing
contingencies that are partly retained by parents. If,
however, the parents break that chain, behavioral
improvement is likely to be the result for all involved.

In an effort to maximize the latter response to
children's problem behavior, parent training programs have
begun to emphasize self-control as an integral component of
parents' behavioral instruction. This follows from an
understanding that children's aversive behaviors trigger

angry responses in parents, which then contribute to

58

coercive interactional patterns between parents and
children. Wolfe and Sandler (1981), for example, observed
three patterns of parent-child interaction among abusive
parents. One pattern is described as highly deviant child
behavior reciprocated with relatively less aversive parental
behavior; the second is characterized by high rates of
reciprocal aversive behaviors and the third involves
relatively more aggressive behavior on the parents part
(Wolfe and Sandler, 1981). Similar observations were
reported by Patterson and Reid (1970).

The interactional nature of parental response and
discipline derives from the basis that children are able to
train parents and vice versa. This helps to explain the
seemingly contradictory premise that, in attempting to
extinguish children's problem behavior, parents can be
trained to respond in ways that foster maladaptive
interaction. Another explanation, on which most
behaviorally-oriented parent trainers agree, is that
parents' responses reflect lack of information about child
development which contributes to unrealistic expectations of
children's behavior. A contributing factor is parental
reliance on aversive disciplinary techniques. Aversive
techniques, by virtue of their ability to suppress
unacceptable behaviors in children, may be readily employed
by some parents, but tend to lose some of their

effectiveness over time due to habituation on the part of

59

children. Parents employing aversive discipline then tend
to increase the intensity of punishment to achieve the same
level of compliance. This situation has been described by
Patterson and Reid (1970) as a coercive process in which the
aversive behaviors of both parents and children escalate
over time. Parental dependence on such aversive
disciplinary methods appears to be related to a lack of
knowledge of effective positive child management techniques
(Sandler, VanDercar and Milhoan, 1978).

Poor impulse control has also been hypothesized to be a
factor in aversive parental response patterns (Steele 8
Pollack, 1968; Wasserman, 1967). Children's behaviors
arouse negative emotions in parents who then respond
aversively, both to eliminate the deviant behaviors and to
reduce their own unpleasant emotions. Thus, the parents'
behavior is intended, not only to teach the child to
suppress undesirable behaviors, but to relieve their own
annoyance, irritation, frustration and anger (Welfe 8
Sandler, 1981). Without other methods to help them vent or
overcome these emotions, punishment serves a dual role of
emotion ventilation for the parent and of reducing the
child's deviant behavior through coercion.

At this point is should be noted that while most of the
research in this area has centered on abusive parents, it
has been found to apply equally well to non-abusive parents

who have aggressive children (Patterson 8 Reid, 1970).

60

While abusive parents with problem children tend to use
harsher methods of discipline, both parent groups operate
within the framework of reciprocally coercive family
interactions.

Relying on the premise that abusive parents need
training in both behavior management and self-control
skills, several behavioral science researchers have
experimented with training individual parents to alter their
behaviors and those of their children (Crozier 8 Katz,
1979; Denicola 8 Sandler, 1980; VanDercar 8 Milhoan, 1978;
Wolfe 8 Sandler, 1981). While these researchers relied on
single subject designs in evaluating their interventions,
their results have been supported by the results of group
design evaluative studies.

Wolfe, Sandler and Kaufman (1981) conducted a study in
which 16 court-referred abusive parents were randomly
assigned to control and treatment groups. Dependent
measures consisted of parental perceptions of child
behaviors, caseworkers' ratings of children's problem
behavior and home observations. The treatment subjects
received group training in parenting augmented by individual
home-based training. Control parents received caseworker
supervision only. Group training included topics in human
development and child management, problem solving, self-
control, and modeling of appropriate child behavior. The

self-control training that is particularly of interest here

61

consisted of teaching parents deep muscle relaxation
techniques and impulse control procedures specific to their
individual problem situations. Teaching parents deep muscle
relaxation techniques provided them with coping behavior
that could be used to ameliorate negative emotional arousal.
When they can respond to stressful encounters in a
relatively relaxed and tension-free manner, parents are more
apt to retain control of the situation and less prone to
respond impulsively. Bandura (1980) postulates that
physiologic reactions are one source of beliefs of self-
efficacy. The experience of negative emotional arousal and
increased autonomic activity are likely to decrease self
perceptions of efficacy, while the ability to approach
stressful situations in a relatively calm manner tend to
bolster confidence in one's capacity to handle such
situations.

Training in impulse control focuses on restructuring
parents' appraisal of stress-inducing behaviors. This
training is designed to help parents generate task-oriented
self instructions when facing difficult encounters. It
teaches parents to replace negative self-statements with
positive ones that are likely to facilitate rather than
hinder the task. Michenbaum (1977) observed that when
individuals are faced with tasks that they appraise as
beyond their abilities or for which they lack appropriate

coping responses, they tend to generate task-irrelevant

62

self-statements which reinforce avoidance behavior by
convincing them they will fail instead of succeed. In the
case of abusive parents, such training is aimed at helping
them to develop greater confidence in their ability to
control their responses in anger-producing interactions with
their children.

As indicated earlier, the results supported the
findings of the single subject studies. It was found that
parents .who received behavioral and self-control training,
as compared to those who received caseworker supervision
alone, experienced signficantly fewer child rearing
problems. Moreover, the child rearing difficulties they did
experience were significantly less intense than those
experienced by control group parents. Follow-up measures,
at 10 weeks posttreatment, revealed that treatment effects
were maintained.

The results of these studies suggest that behaviorally
based training methods are effective with different parent
populations. They indicate that parental fuctioning is
enhanced by training and that it seems to have positive
effects on family interactions. Parents who have had the
training are able to contribute to the development of
interactional patterns that tend to reduce deviant child
behaviors. With abusive parents, training not only reduced
child rearing problems, but also the intensity of children's

problems. Abusive parents, who have been found to be rigid

63

and predisposed to the use of discipline to control their
children (Friedman et a1, 1981) were found to be able to
expand their parenting repertoires after training. This
pertains also to other parents in the studies reviewed. In
general, parents who received training acquired parenting
skills that were not present prior to training. Training,
it might be said, helps parents to broaden their parenting
repertoires, thus allowing them to be more flexible and at
the same time more systematic in their parenting.

At this time, it is difficult if not impossible to
evaluate the contribution of self-control to the
effectiveness of training. While Wolfe and his colleagues
compared the results of their interventions with agency
supervision alone, those comparisons are too inclusive to
allow any conclusions to be drawn about the relative
effectiveness of any specific component of the intervention.
A comparison of behaviorally-based parent training with an
intervention that combines behavioral and self-control
skills training would permit an assessment of the

contributions of the self-control component.

Family Perceptions

Parental perceptions of family environment are thought
to be important for long term maintenance of adaptive system
functioning (Karoly 8 Rosenthal, 1977). These researchers

argue that the discrepancies noted between parents'

64

perceptions of deviant child behaviors and those made by
trained outside observers, suggest that parents' perception
may be affected by their judgements, attitudes or personal
biases. Arnold et a1 (1975) noted that children identified
by parents as problem children do not differ from their
siblings with respect to rates of deviant behaviors, a
finding that supports Karoly and Rosenthal's contention.
Further support is provided by findings that indicate
discripancies between parents' and trained home observers'
assessments of changes in problem behaviors. Karoly 8
Rosenthal (1977) report a 15 percent discrepancy favoring
trained observers' assessments of change following
intervention. The hypothesis that deviant behaviors are
viewed differently by parents and trained observers is
consistent with theoretical expectations. Parents, as the
subsystem responsible for the welfare of the family, can be
expected to view behavior in relation to its effects on
other subsystems and members of the family. As a result, it
is likely that parents view their children's behavior in the
context of its meaning for the family as a whole. Unlike
trained observers, they do not see the behavior of the child
as a discrete objective entity, apart from its effects on
and meaning for the rest of the family.

Karoly 8 Rosenthal (1977) hypothesized that training in
behaviorally-based parenting would generate perceptions of

an increased sense of cohesion, decreased level of conflict

65

and greater sense of control. Behaviorally-trained parents,
as compared to parents without training, perceived their
families as significantly more cohesive. No significant
differences were found with respect to conflict and control.

The findings of two other studies reveal that training
of parents is related to improved perceptions of family
environment by family members who were not involved in
training. These findings further corroborate the contention
that families function as interactive systems in which
changes in one member contribute to changes in other
members. Haffey 8 Levant (1984) compared the perceptions of
children whose parents received training in behavioral
parenting to those whose parents received communication-
oriented skills training. Children of behaviorally trained
parents perceived their parents as significantly more
congruent than children of parents trained in communication
skills. The authors indicated that this finding was
anticipated and reasoned that behavioral training's
emphasis on following through with consequences, both
positive and negative, may contribute to greater consistency
in parents' behavior. Reiter 8 Killman's (1978) study did
not involve behavioral training, but reported that husbands,
whose wives received counseling to improve the quality of
communication within the family, perceived their families as
significantly more integrated after counseling than those

whose wives did not receive training. From these studies it

66

appears that perceptions of family environments are
sensitive to intervention-induced changes.

The results of an investigation conducted by VanDerVeen
(1965) suggest that parental perceptions have motivational
effects on parents' willingness to seek help. VanDerveen
(1965) explains that family concept, that is, the way a
person perceives his family which includes feelings,
attitudes, and expectations, influences behavior. Moreover,
he states that the family's self concept can be referred to,
measured and changed as a result of experience (p. 197).

Parents of children judged low in personal adjustment
who were involved with a child guidance clinic, perceived
their families significantly less positively than parents of
children also judged low in personal adjustment but not
involved with a child guidance clinic. Fathers of poorly
adjusted children not involved with a clinic were noted to
be more satisifed with their families than mothers. This
situation was not observed in parents of clinic involved
children who were both dissatisfied with their families.

A comparative study of family perceptions between
families of poorly adjusted children who attended clinics,
poorly adjusted children who did not and well adjusted
children revealed that parents of clinic patients were more
likely to perceive conflict in their families and to
experience difficulties than the other groups. Parents of

poorly adjusted non-patients were found to display a

67

tendency to deny the problems and to present a false picture
of harmony within their families. Families of well adjusted
children, on the other hand, perceived their families as
warm and positive and free of problems (Berkowitz, 1963).
The results of this study and that of Vanderveen (1965)
appear to suggest that parents' perceptions of their
families' functioning may determine to a great extent their
willingness to seek help.

In summary, it would appear that perceptions of family
environment are crucial in understanding how parents come to
define behavior as deviant. Perceptions of family
environments seemed sensitive to intervention-induced
changes and appeeared to motivate help-seeking behavior in

parents.

Self-Efficacy

Social learning theory postulates that psychological
interventions lead to behavior change by strengthening self
perceptions of efficacy (Bandura, 1977). This construct
refers to a person's judgement of his capacity to
successfully undertake a given task. Efficacy expectations
can be altered through performance achievement, vicarious
experiences, verbal persuasion and arousal control (Bandura,
1977). Support for the relationship assumed to exist

between perceptions of self efficacy and behavior change has

68

been provided by the results of several studies (Bandura,
1982).

In an attempt to test the hypothesized role of the self
efficacy construct, adults suffering from severe snake
phobias were assigned to three conditions (Bandura, Adams 8
Beyer, 1977). Two groups received treatments based on
performance mastery and vicarious experiences, the other
received no treatment. Subjects' efficacy expectations were
measured at pre and posttest. Subjects who received
participant modeling treatment as compared to those in the
modeling only conditions obtained significantly higher self
efficacy scores. No significant change over time was noted
for subjects who received no treatment. Modeling was found
to lead to moderate increases in self perceptions of
efficacy. In addition, it was found that regardless of the
treatment mode, the stronger the self perception of
efficacy, the greater the probability that the subjects
would successfully undertake a given activity.

Similar results were obtained by Schunk (1981), who
reported that self perceptions of efficacy predicted
arithmetic performance in children across levels of
difficulty and treatment conditions. The children's
judgements of their capacities reflected their performance.
Schunk reported that the more problems his subject thought
they could solve, the more they subsequently solved.

Moreover, he observed that the greater the subjects self

69

perceptions of arithmetic efficacy, the greater their
persistence on arithmetic tasks and achievement in such

tasks. Weinberg, Gould 8 Jackson (1979) reported results

that corroborate those reported by the Bandura and Schunk

studies.

There appears to be, however, a dearth of information
pertaining to perceptions of self efficacy in the parent
training literature. In light of the motivational effects
of this variable, it is surprising that it has not been
included as a dependent measure in studies evaluating the
effectiveness of parent training.

In summary, it would appear self perceptions of
efficacy play a controlling role in producing behavior
change. While such perceptions are capable of influencing
behavior, they are succeptible to being influenced by
psychological interventions. Perceptions of self efficacy
appear to determine the degree of effort exerted on a task
and the duration of such effort. Clearly, future evaluation
of the effectiveness of training would greatly enhance
understanding of how parents bring about behavior change, if

this variable is included as a dependent measure.

Summary

The purpose of this investigation was to differentially
evaluate the effectiveness of two behaviorally-based foster

parent training curricula, with respect to parental

70

perceptions of family environment, foster parenting
efficacy, placement stability, dependence on agency
personnel. The review of the literature reveals that the
role of foster parents is changing and that present and
future foster parents need more specialized skills. A
number of training programs have been developed to respond
to that need. Results of their evaluation appear to suggest
that behaviorally-based training is somewhat superior to no
training and communication oriented training programs.
Behaviorally-oriented training programs appear to -be
effective, not only in bringing about behavior change in
foster children, but seem to enhance the functioning of the
family system.

Developments in the field of behavioral training for
parents have begun to include a self-control or stress
reduction component in programs designed for abusive
parents. The self-control component is aimed at
facilitating the employment of behavioral principles by
reducing impulsive or anger triggered behaviors in response
to aversive behavior by children. Since foster parents are
continually faced with behaviors that are likely to trigger
angry reactions, it is felt that training parents in self
management would increase their ability to apply parenting
techniques learned in training. It has been found that the

greater and the more varied a person's coping responses are

71

the less stress he is likely to experience (Pearlin 8
Schooler, 1978).

In accordance with the literature reviewed, this study
evaluated the training programs with respect to variables
that are critical to the objectives of foster family care,
but which appears to have received little attention. While
a fair amount of attention has been paid to the effects of
training on foster parents' attitudes toward children and
perceptions of changes in children's behavior, little
interest has been shown in the effects of training on
perceptions of family climate, foster parents sense of
parenting efficacy and placement related outcomes. This
investigation attempted to address this situation by
evaluating the effectiveness of training with respect to

these outcome measures.

CHAPTER THREE

METHODOLOGY

Introduction

 

This chapter presents the methodology employed in this
investigation. It is divided into several subsections
describing the design of the study, the procedures used in
selecting and assigning subjects, the training curricula
compared and the research setting. In addition, information
pertaining to trainers, assessment measures and procedures,

and data analysis is presented.

Design

A group X time research design was employed in this
study. Two independent variables were manipulated: type of
training and time period. Type of training had two levels:
(1) parenting and self management skills training and (2)
parenting skills training without self management. Time
period had three levels: (1) pretest, (2) posttest, and (3)
follow-up time periods.

Outcome measures consisted of (1) score on the
Cohesion, Conflict and Control subscales of the Family

72

73

Environment Scale (FES) (Moos 8 Moos, 1981); (2) score on
the Foster-Parent Parenting Efficacy Scale (PFFES); (3)
number of unsuccessfull terminated placements or placement
breakdowns, and (4) number of contacts with agency
personnel.

Outcome was measured at three points: (1) prior to the
onset of training, (2) immediately upon the completion of
training, and (3) two months after training ended. The FES
and the FPPES were administered on the first and last day of
training, and within a two-week period sixty days after
termination. Outcome measures based on archival data were
obtained after the completion of training, which covered the
two months prior to the onset of training, the two months of

training, and the two months after training.

Sample Selection

The subjects who took part in this study were licensed
foster parents, who were affiliated with a private, non-
profit child placement agency. The latter, as part of its
contractual responsibilities with the State of Michigan, is
required to provide its foster parents with training
designed to enhance their ability to parent children with
special needs. Foster parents are required to attend
training as part of the requirements for maintaining their

license.

74

In addition to meeting its contractual
responsibilities, the agency felt that its foster parents
needed specialized training in child management and
discipline techniques. Agency personnel indicated that a
significant percentage of their foster parents had expressed
a desire for such training.

Foster parents were informed in person and by mail of
the availability of training. Six months prior to the onset
of the parenting classes, the need for more rigorous,
skills-focused training was noted by agency personnel in a
general foster parent meeting. At that time, the foster
parents were promised that such training would be
forthcoming in the near future. Two months prior to the
beginning of training, agency personnel who had daily
contacts with the foster parents informed them of the
upcoming classes and encouraged them to participate. During
these contacts, it was indicated that parents would have the
opportunity to receive 'training in two behaviorally-
oriented, parenting skills training programs that had been
found helpful to other parents. Three weeks prior to the
beginning of this project, the agency notified its foster
parents, by letter, of the date and time of training. The
parents were asked to contact the agency if the assigned
time and day were not convenient. Training was mandatory.
Parents were told that they would receive a certificate upon

successful completion of training. This certificate was

75

contingent on attendance at least six sessions, completing
all homework assignments and obtaining an average score of
3.0 out of a possible score of 4.0 on all quizzes. Five
parents met these requirements. Of those five, three were

control subjects and two were experimental subjects.

Assignment

Ninety-eight licensed foster parents were randomly
assigned to the two training curricula. Foster parents
affiliated with the agency were listed and each one was
given a number. Using a table of random numbers, potential
subjects were assigned to one of the two training
conditions. A coin was tossed to decide which of the two
groups would receive the first subject.

This procedure resulted in two groups of 49 subjects
each. The subjects in each training condition were than
randomly assigned to either a morning or an afternoon class.
In this way, four classes were created, with each training
group having a morning and an afternoon class on different
days. Subjects were allowed to attend either a morning or
an afternoon class on their day to facilitate greater
consistency in attendance and accommodate parents who could
not attend one session or the other.

Of the 98 foster parents who were initially assigned,
52 attended at least one training session. Thirty-seven

foster parents did not respond, that is they did not report

76

for training nor contact the agency. Seven foster parents,
three from the experimental and four from the control
condition, contacted the agency to indicate that they could
not attend at the assigned time. Two parents, one from the
experimental and one from the control group declined to
participate in evaluation of the training curricula, but
attended training. Participation in the curricula
evaluation was voluntary.

Of the 52 parents who attended at least once, 27
attended four or more sessions, while 25 participated in
three or fewer training classes. The 27 parents who attended
four or more sessions consisted of 15 experimental and 12
control group subjects. These 27 parents were considered
successful subjects and constituted the experimental sample.
This procedure was used because it was felt that subjects
needed to participate in at least four sessions in order for
training to have an effect. Similar procedures were
employed by Brown (1980), Slattery (1980), and Guerney
(1976). Table 3.1 presents the number of subjects lost due
to the various forms of attrition that occurred.

Another form of attrition that occurred in this study
concerned the completion of scaled assessment measures.
Many subjects declined to complete the scaled measures,
although they agreed to participate in the evaluation of
training. These subjects tended to request permission to

complete the scales at home and then did not return. Table

77

Table 3.1

 

Number of Subjects Assigned and Those Responding

 

 

 

Number Number Re- No Time Re-
Group Assigned sponding Response Conflict fused
Exp. 49 28 17 3 l
Cont. 49 24 20 4 1
Total 98 52 37 7 2

 

3.2 shows attendance and the number of subjects who
completed scaled measures.

Of the 27 subjects who constituted the experimental
sample, 23 had completed data on the main measures at post-
test, 13 experimental, 10 controls. At follow-up, thirteen
subjects agreed to complete assessment measures, seven from
the control group and six from the experimental group.

Of the four qualified subjects who did not complete the
PBS and the FPPES at posttesting, one declined to do so
because she did not like tests, another consistently forgot
to complete the scales and two wanted to delay the
examination until the situations in their homes improved.
Attempts to persuade these subjects to complete the scales

expeditiously were not successful.

78

Table 3.2

 

Subjects' Attendance and Completion of Scales

 

 

 

 

 

Attendance Completed FES, PEPES
Less 4 or Post- Follow-
Group than 4 more test up
Experi-
mental 13 15 13 6
Control 12 12 10 7
Total 25 27 23 13

 

Every subject who had completed the scaled measures at
posttest was contacted and given the PBS, the FPPES and the
Utilization and Usefulness scale at follow-up. Of the
twenty-three who were contacted and given the scales, three
were angry at the agency and refused to complete the scales,
and five consistently forgot to complete them or did not
have them completed when contacted by the examiners. The
other six indicated the tests were too long, asked why they
had to do them again after having done them twice before and
complained they had no time.

Using Campbell and Stanely (1963) symbols, the design
can be represented as shown in Figure 3.1.

Although it was initially planned that foster parents
would be asked to volunteer, the agency made participation

in training mandatory. Agency personnel explained that its

79

Legend:

X = Parenting + Self Management Skills Training

1
X = Parenting Skills Training Only
2
01-02-03 = Pretest, posttest and follow-up for Parenting

and Self Management curriculum

-0 = Pretest, posttest and follow-up for Parenting

01-0
4 5 6 Training only

R = Random assignment of subjects

Figure 3.1: Study Design Using Campbell and Stanley Symbols

foster parents had a poor history of voluntary compliance
with training, and argued that, since all of its foster
parents needed training, voluntary participation would give
the message that training was optional. By making training

mandatory, the agency hoped to enhance attendance.

Training Curricula

Trainers' manuals for each curriculum are included in
the appendix. A brief description and an outline of the

content of each program will be presented.

80

Both training curricula are based on operant and social
learning principles. The contents of the training programs
were adapted from the works of Burchard and Leintenberg
(1978), Abidin (1976) and Patterson (1976). Each curriculum
is designed for eight weekly sessions of two hours per week
of very structured and skill-oriented programs. Both types
of programs have been found successful with parents (Hampson
& Tavormina, 1980a: Karoly & Rosenthal, 1977: Wolfe et al.,
1981) and the first four sessions of each curriculum were
identical. The Parenting and Self Management curriculum
covered the following areas:

Sessions 1: Acquisition and maintenance of behavior:

modeling and reinforcement: parenting

foster children.

Session 2: Selecting target behaviors: pinpointing,
observing, recording, graphing.

Session 3: Methods for decreasing behaviors:
extinction, response cost, overcorrection,
positive practice, time out.

Session 4: Contingency contracting

Session 5: Conflict resolution and self management
skills: managing feelings.

Session 6: Conflict resolution and self management

 

skills: deep ’Breathing, relaxation
techniques.
Session 7: Developingibehavioral program and self-

 

 

management skills: relaxation techniques.

 

Session 8: Problem-solving: combining child
management and self-management skills.

 

The Parenting Skills training curriculum constituting

the comparison group included the following areas:

81

Session 1: Acquisition and maintenance of behavior:
modeling and reinforcement; parenting
foster children.

 

Session 2: Selecting target behaviors: pinpointing,
observing, recording and graphing.

 

Session 3: Methods for decreasing behaviors:
extinction, response cost, overcorrection,
positive practice, time out.

 

Session 4: Contingency contracting
Session 5: Conflict resolution

Session 6: Conflict resolution

 

Session 7: Developing behavioral programs

Session 8: Problem solving

 

The goal of the curriculum that combined parenting and
self-management skills training was to render foster parents
proficient in child management and self control skills. It
was felt that training in self-management would facilitate
the use of the learned behavioral principles in crisis
situations. Meichenbaum (1982) has reviewed laboratory
studies suggesting that failure to respond appropriately in
stressful situations is not due to lack of skills. The
self-management component allows foster parents to reduce
negative self-referent ideation and emotional arousal in
crisis situations and thus reduces the probability of
responding with stereotypical behaviors. A similar
procedure has been used successfully in training abusive

parents in self-control (Wolfe et al., 1981).

82

The self-management component combined cognitive
restructuring (Ellis, 1970), positive self instruction
(Meichenbaum, 1977) and progressive muscle relaxation. The
cognitive restructuring component was designed to help
foster parents become aware of the cognitions that
negatively affect their appraisal of foster children's
behaviors and involved discussions of beliefs, values and
attitudes that underlie perceptions of problematic
behaviors. Parents were taught to realize that such
cognitions affected their ability to parent foster children
through their affects on their own behavior. Viewing
problem behavior as a result of learning was encouraged, as
was the foster parent's ability to teach new behaviors.
Parents were taught that since all behaviors are learned,
foster children can be taught adaptive behaviors that are
incompatible with deviant behavior. The training also
sought to help foster parents develop the belief that
deviant behavior is not catastrophic, but a problem to be
solved.

In keeping with Meichenbaum's (1977) proposition that
individuals can instruct themselves how to behave
differently in stressful situations, foster parents were
encouraged to develop positive self-referent ideation about
their ability to help a foster child eliminate an
undesirable behavior or to teach him a new one. Progressive

muscle relaxation was taught so that parents could reduce

83

their level of emotional arousal generated by anticipatory
thinking. Self efficacy theory posits that emotional
arousal constitutes one source of self perceptions of
efficacy. The greater a person's ability to anticipate or
engage in stressful situations in a calm and controlled
manner, the more confident he is likely to feel about his
ability to handle such situations (Bandura, 1982).

The main purpose for the second curriculum was to
provide a point of comparison for the parenting and self
management curriculum. Hampson 8 Tavormina (1980a) found
behaviorally-oriented group foster parent teaching to be
superior to reflective group foster parent training in
enhancing family functioning and eliminating deviant
behaviors. Another objective was to train foster parents in
child management skills whose effectiveness is well
established (Karoly & Rosenthal, 1977; Ross, 1981). As
Fairweather & Tornatzky (1977) argued, innovations should be
evaluated against such established programs rather than
compared to no treatment control group.

Both curricula were developed by this researcher and
were reviewed by experts and specialists in behavior
management and the foster home field. Based on comments and
criticisms resulting from such reviews, changes were made to

produce the final versions.

84

The Setting

 

The experiment was carried out in a foster care agency
in the Flint, Michigan area, that provided specialized
foster care services. The agency is responsible for finding
foster homes for children in need, and offers a variety of
rehabilitative services to foster children. The agency is
staffed by professional and paraprofessional mental health
workers. The professional staff includes psychologists,
consulting psychiatrists, therapists and caseworkers.
Approximately 180 foster homes were affiliated with and
supervised by the agency. The agency is also engaged in
recruiting, training and licensing of foster parents and
homes.

Agency foster parents had indicated a strong desire for
child management training in addition to the supervision
they received from the agency. They had undergone training
to orient them to the foster care field, the philosophy of
the agency, and state regulations and requirements regarding
foster care. The foster parents received financial
remuneration for caring for foster children.

The children served by the agency presented a variety
of behavioral and emotional problems that are representative
of the severe difficulties that foster children experience.
These difficulties ranged from none to severe emotional
disturbance, including delinquency. Data pertaining to the

distribution of deviant behaviors between the two groups is

85

presented in Chapter Four. All of the children who were
clients of the agency received therapy as part of their
placement, provided by agency staff who have Masters'
degrees in social work or counseling. Case management is
done by B.A. level caseworkers, who generally oversee the
placement. They are in frequent contact with foster parents
and foster children. The goal of the agency is to return
foster children to their natural parents as soon as possible
or when this is not possible, to make arrangements for
permanent placement.

The foster parents affiliated with the agency
represented a cross section of the population. Those foster
parents who reported for training tended to be from urban
and rural areas. Blacks and whites constituted the two main
ethnic groups. Foster parents varied in their amount of
experience, from several years as foster parents to none.
Both, two-parent and one-parent homes were represented. The
failure of the subjects to respond to demographic items in
the Foster Parent Profile Questionnaire prevents a more
detailed description of the characteristics of the subject

population.

Trainers

Four agency employees with substantial experience in
behavior management volunteered to become co-trainers. Two

had been involved in running foster parents support group

86

meetings. Three co-trainers had master's degrees in clinical
psychology, social work and special education and functioned
as therapists. One trainer had a B.A. in psychology and was
involved in recruiting and licensing foster homes.

Training for the co-trainers consisted of four two-hour
sessions during which co-trainers reviewed the principles
and techniques of behaviorally-based parenting and self-
management skills, and familiarized themselves with the
training manuals. Assessment of trainers consisted of
written responses to three stimulus situations requiring
behavioral intervention. Their responses were scored by the
researcher and all four volunteers were found qualified to
be co-trainers. Co-trainers' responses were evaluated with
respect to (l) mastery of content as expressed through
appropriateness of suggested techniques and (2) procedure
suggested, that is, the steps to be followed. A score of 90
on a scale of 100 was considered necessary to qualify.

Co-trainers assigned themselves to foster parent
training classes based on time availability. They were
rotated so that each co-trainer participated in all classes.

Initially, it was planned that the researcher would not
be involved in training and that trained volunteers were to
teach all the classes. Lack of time on the part of
volunteer co-trainers and their insufficient number did not
permit the implementation of the original plan. Thus, the

researcher was also involved as a co-trainer. This

87

procedure proved helpful as one co-trainer dropped out after
the fourth week of training and other co—trainers were
sometimes unable to lead sessions due to emergency
situations that required their presence.

Co-trainers met once a week for an hour to discuss
training needs and review previous sessions. These meetings
addressed such concerns as (l) keeping subjects on task
without seeming uncaring about non-training concerns, (2)
sporadic attendance, (3) pulling parents out of training,
and (4) non-completion of homework assignments. Decisions
taken after discussing these concerns were applied in all

the following week's training sessions.

Training Implementation

The first 20 minutes of the first training session in
all groups were used to (1) provide an overview of training
and of the research aspect of the project, (2) explain how
subjects were assigned to classes, and (3) seek voluntary
participation in the evaluation of training. Another 15-20
minutes were used for pre-testing. Subjects who did not
attend the first session of training completed this
procedure individually.

Training in both models started in the same week. The
decision as to which training model would start first was
decided by tossing a coin. Training sessions were held on

Monday and Friday, once in the morning and once in the

88

afternoon. A control and an experimental class met on each
training day at different times, such that one condition had
a morning class on one training day and an afternoon class
on the other day. Subjects were trained in classes whose
attendance ranged from two to seventeen participants.
Training classes lasted two hours, except for the first
and last class meeting. Half an hour was added to the time
periods for these sessions so subjects could complete
assessment instructions. Training in both models ended the
same week. Classes within the same teaching condition were
combined for the last day of training because agency
employees were given an unanticipated free ”stress day“ that
fell on the last Monday of training. Subjects were
contacted by telephone by agency personnel and asked to

attend the next class in their respective groups.

Implementation Difficulties

Training in both conditions was beset by several
difficulties, most of which persisted throughout the
training period. These problems consisted of (l) sporadic
attendance, (2) interruptions, (3) frequent recital of non-
related training concerns, (4) non-completion of homework
assignments.

Sporadic attendance interferred with the continuity of
training. Concepts and principles that had been already

discussed often had to be repeated for those who had missed

89

earlier sessions. This situation may have hindered the
mastery of concepts and principles that were introduced
toward the latter part of training.

On several occasions, agency personnel came to the room
where training was being held to ask to meet with a foster
parent. These interruptions persisted throughout training
but seemed to decline after this problem was discussed with
agency personnel.

Frequent recital of non-related concerns constituted
the most difficult problem in completing the training as
outlined. Subjects in both conditions, but more of those in
the experimental morning class, displayed a preference for
complaining about the agency's ”pro foster child approach".
The participants felt that behavior problems displayed by
their foster children were partly maintained by the
inadequate support they received‘ from caseworkers and
therapists. y They seemed to feel that these professionals
undermined their parenting roles and diminished the
effectiveness of their disciplinary measures. The subjects
conveyed the impression that the entire foster family system
gave too much power to the children and not enough to the
foster parents.

Some parents complained that they were underpaid or
questioned the monies they were getting. They felt that
they should be receiving the rates paid for severely

disturbed children, instead of the regular rates they were

9O

getting. Many parents wanted to know the criteria that were
used in determining rates.

Others complained that caseworkers behaved in an
arbitrary fashion that they found insensitive to their
concerns. They indicated that caseworkers sometimes
scheduled appointments for their foster children without
adequate prior notice. Some expressed irritation at the
attention given to the foster child during home visits,
feeling excluded by the caseworker.

This attention appeared to have some unwanted
consequences for training. A number of parents indicated
that the ”gripes of other parents' decreased their interest
in coming for training. Other parents approached the
researcher individually to express feelings of exasperation
with the “constant venting and griping.‘ They indicated
that they enjoyed the classes and learned when the co-
trainers were able to teach. These parents were encouraged
to express these thoughts and feelings during the training
sessions.

The apparent need of the foster parents to use training
as an outlet to express their feelings of dissatisfaction
with the agency placed co-trainers in a delicate position.
On one hand, there was the pressure to keep parents focused
on the task at hand, while on the other there was concern to
empathize with and avoid alienating the .participants. It

was thought that through these complaints, the subjects were

91

voicing their frustrations with a difficult job that is
seldom rewarding. Moreover, it was felt that co-trainers
needed to be careful not to exacerbate the apparent feelings
of powerlessness that seemed to underlie the complaints.
Based on this understanding, it was decided that the
following procedure would be employed. When subjects began
to voice non-training related concerns, co-trainers (1)
acknowledged the subject's concerns, (2) inquired as to how
many parents had the same concerns, (3) asked subjects to
try to relate the expressed concerns to the training topic
being discussed or those to be covered, (4) suggested that a
specific amount of time be set aside to discuss these
concerns after the session's instructional objectives had
been met. If subjects persisted in discussing these issues,
co-trainers and participants negotiated a specificed amount
of time to discuss the concerns. While this procedure did
not entirely eliminate the problem, it contributed to an

increase in the time used for training.

Training_Related Concerns

 

The participants appeared to have experience a moderate
degree of difficulty in accepting the usefulness of the
concept of reinforcement. Many subjects felt that
reinforcement was tantamount to bribing a child, and
preferred punishing techniques aimed at reducing deviant

behaviors. Some difficulties were also noted in discussing

92

contingency contracting and conflict resolution. There
appeared to be a marked difference in the responses of
participants in the two training conditions. Subjects in
one of the experimental classes tended to be vehement in
their opposition to negotiating contracts with their foster
children. These subjects also resisted doing self
management exercises that required them to explore their
beliefs, attitudes, and feelings toward deviant behavior.
They tended to express ultimatums, demanding that a foster
child either, ”do it my way or leave my house." Only a
small minority, about three parents, appeared to invest
efforts in these exercises. Unfortunately, this class had
the largest attendance of the two experimental classes.
Subjects in the other experimental class appeared more
cooperative, but much less interactive. These subjects
tended to join the co-trainers rather than oppose them.
Another training related concern involved homework
assignments. By the third session, it became obvious that
only a small number of subjects were doing the homework
assignments. This situation was pointed out to the subjects
with the explanation that doing the assignments would
provide the co-trainers with the opportunity to give them
feedback on how they were applying the principles. As this
situation did not improve, part of the time allocated for
discussion of homework assignments was added to the time

allocated for review. Data pertaining to the number of

93

assignments completed between the two groups will be
presented in Chapter Four. A

Apart from the difficulties noted above, training
followed the outline in the trainers' manual. While
participants were not required to do any out-of-class
reading, they were given access to and encouraged to read

Living with children (Patterson, 1976) and Families

 

(Patterson, 1971).

Instrumentation

 

Three main outcome measures were used. These were:
The Family Environment Scale (Moos & Moos, 1981), the Foster
Parent Parenting Efficacy Scale and the Foster-Placement

Information Sheet.

 

The FES is designed to assess perceptions of family
interactional patterns. It is composed of ten subscales,
each consisting of nine true/false statements. Only three
of these subscales were used in this study, the Cohesion,
Conflict and Control subscales. Consistent with Karoly and
Rosenthal (1977), the higher the score on the Cohension and
Control subscales, and the lower the score on the Conflict
subscale, the more positive the family climate was assumed
to be.

The scale is relatively easy to administer. Scale
items are short and written in simple language. Test-

retest reliability estimates obtained over a four month

94

period are .72 for Cohesion, .66 for Conflict and .78 for
Control (Moos & Moos, 1981). Internal consistency is
estimated at .78 for Cohesion, .75 for Conflict and .67 for
Control (Moos & Moos, 1981).

The FES has been used with a variety of populations and
ethnic groups with little loss in its ability to
discriminate between distressed and non-distressed families
(Moos & Moos, 1981). Moreover, the subscales appear to be
sensitive to intervention-related changes in perceptions of
family climate (Karoly & Rosenthal, 1977: Moos & Moos,
1981).

Subjects in this study appeared to respond with caution
to the scale's items. A large number of subjects expressed
concerns about having their family misperceived by agency
personnel. They appear to distinguish two families, one
with foster children and one without. It appeared as though
some subjects responded to the test with perceptions of
their family that discounted the presence of foster
children. Subjects inquired as to whether the test was
assessing their family or the foster child. When told to
respond according to the ways things are done in their
families, including everyone in the home, parents
volunteered the explanation that many of the items are true
of their family only because of the presence of the foster
children. The instructions that they respond to the test

items based on the way people in their home behave toward

95

and interact with one another appeared to leave them
perplexed.

The Foster Parent Parenting Efficacy Scale (FPPES) is a
lO-item test that was designed to assess foster parents'
feeling of confidence in their ability to function as foster
parents. The items were adapted from the Parenting Sense of
Competence Scale developed by Gibaud-Wallston and Wandersman
(1978) by rewording ten of the original scale's 16 items in
order to make them appropriate for foster parents. An
internal consistency reliability estimate of .71 was
computed for the scale from the pretest scores of 37
subjects who completed it.

The Foster Placement Information Sheet is a list of six
items that pertain to two types of placement-related events.
Of these, three items refer to types of placement breakdowns
and three to types of contacts between foster parents and
agency personnel.

Scoring consisted of recording the number of occurrence
of each placement-related event for each foster child in the
subject's home. Since length of placement varied for foster
children, each type of contact was divided by the number of
days spent in the home. The result was then summed for the
foster children to obtain a score for each parent for each
contact item. Similarly, the number of each type of
placement breakdown was divided by the number of foster

children in the home and then summed to obtain a score.

96

This' procedure is based on the assumption that the greater
the number of foster children in a home, the greater the
possibility of placement breakdowns. The higher the scores
for each type of placement-related event, the more
problematic the placement. The procedure for handling
presenting behavior consisted of writing the presenting
problem (or lack of problem) for each foster child in the
subject's home.

The FPIS was not administered to subjects, but completed
by reviewing agency records. As with the other measures,
assessments were done on three occasions. Pre-training
assessment covered a period of two months, posttest
assessment covered the two months of training and follow-up
assessment reviewed information pertaining to the two months
following the end of training.

The reliability of the items was established through
inter-recorder agreement. Initially it was planned that two
persons who were not affiliated with this study would serve
as recorders. Since only one person volunteered to carry
out that task, the researcher also served in the capacity of
recorder. While it was possible for this researcher to have
non-agency personnel function in that capacity, agency
personnel felt that this role would violate the
confidentiality of the information contained in the

subject's record.

97

Prior to completing the Foster Placement Information
Sheet, a volunteer agency employee was trained in how to
identify and record the desired information up to 90 percent
proficiency. Unfortunately, this employee terminated her
employment with the agency soon after training ended. As
there were no other volunteers, the researcher recorded the
information. The following procedure was used to establish
the reliability of the recorded information.

A random sample of 10 subjects was selected and
assigned to a trained recorder. This person received the
same training and met the same requirements as the previous
volunteer. After the recorder had completed the Foster
Placement Information Sheet on the subjects, his scores
were compared with those the researcher had given the same
set of subjects. The number of agreements was divided by
the number of agreements plus disagreements for each item.

Using this procedure, the following percentage of
agreement was obtained. Of the items related to contacts
with agency staff (Therapists & Caseworkers): home contacts
had a proportion of agreement of .80, office contacts also
had a proportion of agreement of .80 and telephone contacts
had the lowest proportion of agreement of .60. The items
pertaining to placement breakdowns had a lower proportion of
agreement than those referring to contacts. Unlike contact
items, type of placement breakdown was not always clearly

stated, and required some interpretation on the part of the

98

recorder. Request for removal had a proportion of agreement
of .50. A proportion of agreement of .60 was obtained for
unrequested removal, while number of children who ran away
had a proportion of agreement of .80. The relatively higher
level of agreement regarding this last item, as compared to
the two previous ones, relates to the greater documentation
regarding that behavior in agency records. A special report
is completed for each child who is on 'run away" status.

The low level of agreement regarding requests for
removal and unrequested removals was related to conflicting
reports between therapists and caseworkers and conflicting
statements within the same report regarding the party
who initiated the procedure for removing a child from his
foster home. One criticism that could be leveled against
the procedure used to establish reliability of the item on
the FPIS pertains to the relatively small number of subjects
selected. It was not possible to use more subjects because
of the volunteer's time constraints.

In addition to the outcome measures mentioned above,
four other assessment devices were used. These scales
provided information on the demograhic characteristics of
the sample, amount of information mastered, training
climate, and perceived usefulness of the principles and
techniques taught in training.

Information pertaining to demographic characteristics

was obtained by the Foster Parent ProfileQuestionnaire.

99

The questionnaire is a 22 item scale adapted from a scale of
the same name developed by Brown (1980). Questionnaire
items cover such areas as: experience, age, education,
housing, neighborhood, type of children fostered, feelings
toward the agency, frequency of contact with agency staff
and turnover in child care workers. Participants in this
study appeared to respond negatively to this questionnaire.
Many participants complained that the questions were too
personal. Some indicated that they found it confusing and
unclear, and left many items unanswered.

Amount of information mastered was assessed through the
administration of four multiple-choice quizzes. Each quiz
covered the information presented in the preceding session
and was written by the researcher and the co-trainers.
Quizzes were administered at the onset of the training
sessions. Subjects generally did not like the quizzes, and
they were made optional after the third one was given.
Subjecs did not volunteer to take the tests after they were
made optional. Data pertaining to the performance of the
two groups on the quizzes is presented in Chapter Four.

Training climate was measured by the Training
Information Sheet. This scale contains 7 items pertaining
to time of training, attendance, tardiness, group
participation, completion of assignments, amount of material
covered and length of session. The information provided by

this scale was used to maintain training homogeneity. This

100

scale was completed by the co—trainers at the end of each
session. Data pertaining to attendance and completion of
assignments is presented in Chapter Four.

Frequency of use and perceived usefulness of the
parenting skills taught were measured by the Utilization and
Usefulness Scale. This scale was made up of two Likert-type
items each consisting of a different subscale. One item
assessed frequency of use, while the other measured

perceived usefulness.

Data Collection

The procedure that was followed in collecting data for
each assessment period is described below under separate

heading for each period.

Pretesting

 

The Family Environment Scale, the Foster Parent
Parenting Efficagy Scale and the Foster Parent Profile

Questionnaire were administered to the subjects by the co-

 

trainers at the first meeting prior to the onset of
training. Subjects who were absent from that session were
tested during the first session they attended, prior to
participating in training.

The co-trainers collected the completed scales and

returned them to the researcher. Since many subjects

101

experienced difficulties with the Foster Parent Profile

 

Questionnaire, they were allowed to take it home with the

 

understanding that it was to be returned' at the next
training session. Subjects were encouraged to complete the
other scales at the training site.

While co-trainers were aware that two training
conditions were being manipulated, they were unaware of
which condition constituted the experimental and control
group. The trainers distributed the tests to the
participants, read test instructions and answered subjects'

questions.

Posttesting

 

Because of an unexpected holiday that fell on one of
the last training days, the classes within each training
condition were combined into one, forming only two classes.
Following the same procedure used previously, the co-
trainers administered the Family Environment Scale, the

Foster Parent Parenting Efficacy Scale and the Treatment

 

Information Scale 'to all subjects who attended the two

 

sessions. Successful subjects who did not attend the last
sessions were administered the scales at home or at the
agency by co-trainers within two weeks after training ended.
Four subjects declined to complete the scales, two from the

experimental group and two from the control group.

102

Follow-Up

 

Co-trainers and agency employees who had weekly
contacts with the participants were approached by the
researcher and asked to serve as examiners. Subjects were
contacted by telephone by the researcher to inform them that
they would be asked to complete the scales at their next
scheduled meeting with agency personel and to thank them for
their cooperation in the previous training sessions.
Subjects completed the scales individually at the agency or
at home. Of the 23 subjects who completed the scales at
posttest, 13 responded at follow-up. The examiners
indicated that other subjects declined to complete the
scales for various reasons stated earlier. Data for this
period were collected within a two-week period, two months
after training ended.

It is suspected that examiners' behavior may have
affected subjects' reluctance to complete scaled measures at
follow-up. Three examiners indicated they had forgotten to
administer the scales during their contact with the
subjects and it appeared as though administration of the
scales did not have the same priority it had had during the
first two testing periods. It was later learned that the
agency was experiencing difficulties related to staff
turnover and foster parent dissatisfaction with agency

personnel.

103

Analysis

Demographic variables were analyzed using the chi-
square technique and t-tests for differences between means.
A repeated-measure analysis of variance was performed on
outcome measures. A statistical significance level of .05

was chosen for the analysis of outcome measures.

CHAPTER FOUR

ANALYSIS OF DATA

Because of attrition, a three-tier analytic strategy
was employed. The results of the analysis are presented in
the following manner: results pertaining to subjects who
completed all assessment measures will be discussed first,
followed by results for subjects with complete assessment
measures at pretest and posttest only. Results relating to
the differential effects of training on more cooperative and
less cooperative subjects will also be presented.

Thirteen subjects, nine blacks and four whites,
completed assessment measures at follow-up. The ages of
these participants ranged from 32 to 59 years, with a mean
age of 46 years. Experience ranged from zero to six years,
with a mode of one year, and an overall mean of 1.13 years.
Fifty percent did not complete high school, while 10 percent
reported having had some college. Fifty-nine percent of the
subjects were single female heads of households.

Since age and experience have been found to be
positively related to placement stability (Boyd & Remy,
1978), subjects in the two training conditions
were compared with respect to those variables. The

104

105

comparison with respect to age produced a t-value of 0.0 (g;
= 7, p = .1), which was not statistically significant at
the .05 level.

Similarly, no significant differences between the two
training groups were found with respect to experience. A 3-
value of -l.68 with 9 degrees of freedom was obtained. This
value is statistically significant at the .13 level.

Because the relatively high level of attrition raised
'concerns over the initial equivalence of the two training
groups, a t-test for significant differences between means
was conducted on pretest data. Table 4.1 summarizes the
results of those mean comparisons.

As can be seen from this table, there were no
significant statistical differences between the two groups
on pretest measures.

Since it has been reported that behavior problems are
positively associated with placement breakdowns and
maladaptive interactional patterns in families in general,
data pertaining to the behavior difficulties of foster
children were analyzed (Tables 4.2 to 4.4). As can be
observed from Table 4.2, participants in the two training
conditions did not differ significantly with respect to the
types of behavior they confronted at Time 1. Table 4.3
shows that the two groups significantly differed with
respect to the number of children, displaying stealing and

lying behaviors at posttest, while Table 4.4 reveals that at

106

Table 4.1

Summary of E-Test Results On Pretest Data

 

 

Scale d§_ EeValue p

Cohension ll -.54 .60
Conflict ll -.41 .69
Control 11 .10 .92
Efficacy ll -.84 .42
Reg. for removal 11 -.92 .38
Unreq. removals ll 0 1
Run aways ll 0 l
Home contacts 11 -.13 .90
Office Contacts 11 -l.53 .15

Phone Contacts 11 -1.28 .23

 

107

Table 4.2

Results of Fisher's Exact Test for Presenting
Problems at Time 1

 

 

Presenting Problem 2

Minor or no behavior problem .76
Noncompliance .76
Bed wetting --
Stealing/Lying .07
Truancy .70
InapprOpriate sexual behavior .88
Antisocial .39
Hostile aggressiveness .91
Emotional disturbance .37

Other .63

 

108

Table 4.3

Results of Chi Square and Fisher's Exact
Test for Presenting Problems at Time 2

 

Presenting Problem df X(chi) FET ‘2

 

Minor or no behavior

problem .86
Noncompliance l .31 .58
Bed wetting 1 .96 .33
Stealing/Lying l 5.70 .02*
Truancy - l 1.05 .30
Inappropriate sexual

behavior 1 2.01 .16
Antisocial l 2.01 .16
Hostile aggressive 1 .31 .58
Emotional disturbance l .002 .96
Other 1 2.01 .16

 

*Significant at .05 level

Chi Square Results for Presenting
Problems at Time 3

109

Table 4.4

 

 

Presenting Problem §£_ I§(Chi)

Minor or no Behavior

Problem 1 .01 .91

Noncompliance l 4.33 .04*
Bed wetting 1 2.30 .13

Stealing/Lying l 3.15 .08

Truancy 1 5.64 .02*
Inappropriate Sexual

behavior

Antisocial l .01 .94

Hostile Aggressive l .01 .91

Emotional disturbances l .002 .96

Other 1 2.30 .13

 

*Significant at .05 level

110

Times 3, subjects in the two training conditions differed
significantly in regard to children who displayed
noncompliant and truant behavior.

Because foster children moved into and out of the
participants' homes during the course of the study, an
analysis of the data pertaining to the number of children
cared for was performed. This was done in order to assess
the initial equivalence of the two groups with respect to
that variable and the maintenance of that equivalence over

time (Tables 4.5 and 4.6).

Table 4.5

Repeated Measures Analysis 0; Variance
of Foster Children

 

 

Source df gs g 2
Training 1 .22 .074 .79
Subjects 11 2.98

Time 2 1.87 2.78 .08
Training X Time 2 .71 1.05 .37
Subjects X Time 22 .67

a

 

Number of subjects: Experimental = 6, Control a 7

111

Table 4.6

Mean Number of Foster Children
for Training Conditions 3

 

 

 

Training Time Time Time Mean of
Conditions 1 2 3 Means
Experimental 1.33 2.33 2.50 2.06
Control 1.71 2.00 2.00 1.90
Mean of

Means 1.54 2.15 2.23 1.98

 

The results of the analysis suggest that the two
training conditions did not differ significantly with
respect to the number of foster children for whom the
participants cared [3 (1,11): .07, p a .79]. While those in
the experimental condition accepted more foster children
across the time periods, no training X time interaction
effect was obtained [£(2,22)=2.78, 22.37]. Those in both
training conditions experienced an increase in foster
children from Time 1 to Time 2. Experimental subjects
increased the number of children they cared for consistently
across the time periods observed. A 75 percent increase was
noted between Time 1 and Time 2 for those in the
experimental training condition, compared to a 17 percent
increase in number of children for the control group for the

same time period. From Time 2 to Time 3, a 7 percent

112

increase was noted for the experimental group, compared to
none for the control group.

In addition to the analysis of the data pertaining to
the number of children fostered, data related to attendance,
amount of material mastered and number of completed
assignments were also analyzed in an effort to assess
intergroup homogeneity. No significant differences between
the two groups was found with respect to attendance (E=‘-1r
dfall, 28.92). Experimental subjects had a mean attendance
of 5.5 sessions, compared to 5.6 for those in the control
group.

Analysis of the data related to amount of material
learned, as measured by classroom tests, revealed that
experimental subjects obtained a mean score of 1.83,
compared to 3.14 for control subjects. These results were
not significant (Ea-2.11, gfsll, p=.06). Similarly, the two
groups did not differ significantly with respect to the
number of assignments completed (t=.84, g§=ll,p=.42). The
experimental subjects obtained a mean score of 2.16,
compared to 3.43 for control subjects.

Data related to the utilization and usefulness of the
skills taught were also analyzed. Analysis of the
utilization data suggest that there was no significant
difference between the two groups (tal.45,d£-10,p-.18).
Experimental subjects obtained a mean score of 3.17,

compared to 2.5 for control subjects. There was also no

113

significant difference between the two groups with respect
to perceived usefulness of the techniques taught (t=-.35,
g;=1o, p=.73). A mean score of 2.5 was obtained by
experimental subjects, compared to 2.67 for control

subjects.

Hypothesis Testing

The research hypotheses were grouped by the research
questions from which they were generated. For example,
hypotheses 1.1 to 1.3 related to question '41, while
hypothesis 2.1 pertained to question #2 and so on.

Question #1: Does training in parenting and

self-management skills, as compared to training

in parenting skills only, lead to improved family

environment as perceived by foster parents?

Hypothesis 1.1: Foster parents trained in

parenting and self-management skills will score

significantly higher on the Cohesion scale of the

Family Environment Scale than those trained in

parenting skills only.

This hypothesis was not accepted. No significant
difference between the two groups was found at the .05 level
(F(l,1l)=.89, {p=.37]. In addition, there were no
significant effects for time (F(2,22)=.09,p=.92) and no
significant training by time interaction (F(2,22)=1.90,

22.17). Table 4.7 summarizes the results of the analysis,

while Table 4.8 presents the means.

1

Table 4.7

14

Repeated Measures Analysis of Variange
of Scores on the Cohesion Subscale

 

 

Source d_f_ _M_S F 2
Training 1 7.38 .887 .336
Subjects 11 8.33

Time 2 .102 .089 .915
Training X Time 2 2.19 1.897 .174
Subjects X Time 22 1.15

 

a
Number of subjects

Table 4.8

Experimental 2 6, Control = 7

Means of Scores on Cohesion Subscale

 

 

 

Training Time Time Time Mean of
Conditions 1 2 3 Means
Experimental 7.16 7.16 6.33 6.88
Control 7.71 7.42 8.14 7.76
Mean of

Means 7.46 7.30 7.30 7.35

 

115

As can be seen from Table 4.8, these results were
contrary to expectation. Those in the experimental training
condition experienced a decrease in perceived cohesion among
family members between Times 2 and 3, while those in the
control condition experienced an increase. Control subjects
experienced a decrease in score from Time 1 to 2, while
experimental participants underwent no perceived changes.
Those in both training conditions experienced a decrease in
perceived cohesion relative to their Time 1 scores.

Hypothesis 1.2: Foster parents trained in

parenting and self-management skills will score

significantly lower on the Conflict Scale of the

Family Environment Scale than those trained in

parenting skills only.

This hypothesis was not accepted. The analysis
revealed no significant effect for training condition
(§(l,ll)=.18, p=.68) nor for time (F(2,22)=.05, ‘p=.96) at
the .05 level. Training condition by time interaction was
also not significant at the .05 level (F(2,22)=.015, p=.99).
Table 4.9 summarizes the results of the analysis, while
Table 4.10 presents the means for training conditions and
time periods.

Table 4.10 revealed that the results were in the
predicted direction. The experimental group had a lower
mean score for conflict than the control gorup at Times 2
and 3. Contrary to the control group, however, it

experienced an increase in conflict from Time 1 to Time 2.

116

Table 4.9

Repeated Measures Analysis of Variance
of Scores on the Conflict Subscale of the PBS

 

 

 

Score df MS F p
.Training 1 .71 .176 .683
Subjects 11 4.01

Time 2 .08 .045 .956
Training X Time 2 .03 .015 .985
Subjects X Time 22 1.72

a

Number of subjects: Experimentals6, Control=7

Table 4.10

Means of Scores for Conflict Subscale

 

 

 

Training Time Time Time Mean of
Conditions 1 2 3 Means
Experimental 1.66 1.83 1.83 1.77
Control 2.00 2.00 2.14 2.04
Mean of

Means 1.84 1.92 2.00 1.91

 

117

Those in the experimental group perceived less stress in
their family environments across the time periods than did
those in the control group.

Hypothesis 1.3: Foster parents trained in

parenting and self-management skills will score

significantly higher on the Control Scale of the

Family Environment Scale than those trained in

parenting skills only.

This hypothesis was not accepted. The analysis
revealed no significant effect for training condition
(F(1,11)=.92, pé.36) nor for time (F(2,22)=.43, 42:.66) at
the .05 level of significance. Training X time interaction
was also not significant at the .05 level (§(2,22)=.79,
28.75). Table 4.11 summarizes the results of the analysis,
while Table 4.12 presents the means for training conditions
and time periods.

Table 4.12 shows that the results were not in the
predicted directions. Control subjects perceived their
family environments as more controlled than experimental
subjects. Both groups experienced a decrease from Time 1 to
Time 2. Experimental subjects, however, experienced a
greater decrease in perceived control than control group
subjects. While both groups experienced an increase from

Time 2 to Time 3, control group subjects obtained a greater

increase.

Repeated Measures Analysis of Variance
of Scores on the Control Subscale of the PBS

118

Table 4.11

a

 

 

 

 

 

 

Source g MS 3 _p
Training 1 1.59 .920 .358
Subjects 11 1.73
Time 2 1.00 .429 .657
Training X Time 2 .67 .286 .754
Subjects X Time 22 2.33
a
Number of subjects: Experimental = 6, Control a 7
Table 4.12

Means of Scores for Control Subscale
Training Time Time Time Mean of
Conditions 1 2 3 Means
Experimental 5.50 4.83 5.16 5.16
Control 5.42 5.28 6.00 5.57
Mean of
Means 5.46 5.07 5.61 5.37

 

119

Question #2: Does training in parenting and
self-management skills, as compared to training
in parenting skills only, lead to an increased
sense of parenting efficacy in foster parents?
Hypothesis 2.1: Foster parents trained in
parenting and self-management skills will score
significantly higher on the Foster Parent
Parenting Efficacy Scale than those trained in
parenting skills only.

This hypothesis was not accepted. The analysis
revealed no significant effects for training condition
(F(l,ll)=.10, p=.76) nor for time (§(2,22)=.10, p=.91) at
the .05 level. Training by time interaction, also, was not
significant at the .05 level (§(2,22)=2.25, 28.13). Table
4.13 summarizes the results of the analysis, while Table
4.14 presents the means for training conditions and time
periods.

Table 4.14 shows that the results were in the predicted
direction. An increase in mean score was observed across
the time periods for the experimental condition. Subjects
in the control group showed a mean increase in self-efficacy
from Time 1 to 2, but a decrease from Time 2 to Time 3. The
difference in mean score between the two training
conditions, however, was greatest at Time 1.

Question #3: Does training in parenting and

self-management skills, as compared to training

in parenting skills only, lead to fewer placement

breakdowns?

Hypothesis 3.1: Participants trained in

parenting and self-management skills will express

significantly fewer requests for removal than
those trained in parenting skills only.

120

Table 4.13

Results for Repeated Measures Analysis of

Variance of Scores on the FPPES‘a

 

 

 

 

 

 

Source df MS F p
Training 1 4.52 .099 .759
Subjects 11 45.54

Time 2 .72 .099 .906
Training X Time 2 16.34 2.25 .129
Subjects X Time 22 7.27

a

Number of subjects: Experimental = 6, Control a 7

Table 4.14
Means of Efficacy Scores

Training Time Time Time Mean of
Conditions 1 2 3 Means
Experimental 27.83 28.50 30.33 28.88
Control 30.00 30.28 28.42 29.57
Mean of
Means 29.00 29.46 29.30 29.25

 

121

This hypothesis was not accepted because analysis
revealed no significant effects for training (§(1,11)=.76,
2:40) or for time (§(2,22)=l.47, £2.25) at the .05 level.
Training by time interactions were also not significant
(F(2,22)=1.27, p=.30). Table 4.15 summarizes the results of
the analysis, while Table 4.16 presents the means for

training conditions and time periods.

Table 4.15

Repeated Measures Analysis of Variance
on Number of Requests for Removala

 

 

 

source a; E P. a
Training 1 .06 .758 .403
Subjects ll .08

Time 2 .08 1.474 .251
Training X Time 2 .07 1.274 .300
Subjects X Time 22 .05

a

Number of subjects: Experimental - 6, Control a 7

122

Table 4.16

Mean Number of Requests for Removal

 

 

 

Training Time Time Time Mean of
Conditions 1 2 3 Means
Experimental 0 .05 .12 .06
Control .04 .29 .07 .13
Mean of

Means .02 .18 .09 .10

 

Table 4.16 shows that the results were partially in the
predicted direction. Both groups experienced an increase in
requests for removal at Time 2, but this increase favored
the control group. While the experimental group continued
to experience an increase, from Time 2 to Time 3, a decrease
was noted for the control group. The means for the time
periods revealed an increase in removal requests from Time 1

to Time 2, that was followed by a decrease from Time 2 to

Time 3.
Hypothesis 3.2: Foster parents trained in
parenting and self-management skills will
experience signficiantly fewer unrequested

removals than those trained in parenting skills
only.

123

This hypothesis was not accepted. The analysis
revealed no significant effect for training (§(l,1l)=.59,
£8.46) at the .05 level, but a significant effect for time
was observed (§(2,22)=3.48, p=.049). No significant
training by time interaction was observed. Table 4.17
summarizes the results of the analysis, Table 4.18 presents

the means for training conditions and time periods.

Table 4.17

Repeated Measures Analysis of Variance
for Number of Unrequested Removals

 

 

 

source d_f as E. 2
Training 1 .003 .589 .459
Subjects 11 .005

Time 2 .020 3.477 .049
Training X Time 2 .003 .589 .564
Subjects X Time 22 .005

a

Number of subjects: Experimental . 6, Control a 7

As can be seen from Table 4.18, the results were not in
the predicted direction. While both groups experience an
increase from Time 1 to Time 2, those in the experimental
training condition experienced more requests for removal

than the control group. Visual inspection of the means for

124

Table 4.18

Mean Number for Unrequested Removals

 

 

 

Training Time Time Time Mean of
Conditions 1 2 3 Means
Experimental 0 .10 0 .033
Control 0 .04 0 .014
Mean of

Means 0 .07 0 .023

 

the three time periods suggested that the increase in
unrequested removals from Time 1 to Time 2 was equal to the
decrease noted from Time 2 to Time 3. This would suggest

that both sets of contrasts are significant.

Hypothesis 3.3: Foster parents trained in

parenting and self-management skills will have

significantly fewer foster children running away

from their homes than those trained in parenting

skills only.

This hypothesis was not accepted. The analysis
revealed no significant effects for training (§(l,ll)=.67,
p=.43) nor for time (E(2,22)=2.15, 23.14) at the .05 level.

Training X time interaction was also not significant

(§(2,22)=.81, p=.46). Table 4.19 summarizes the results of

125

the analysis; Table 4.20 presents the means for training

conditions and time periods.

Table 4.19

Results of Repeated Measures Analyiis of
Variance for Runaway Children

 

 

 

Source g M_S_ F_ p_
Training 1 .05 .666 .432
Subjects 11 .08

Time 2 .06 2.146 .141
Training X Time 2 .023 .805 .460
Subjects X Time 22 .03

a

Number of subjects: Experimental a 6, Control a 7

Table 4.20

Mean Number for Runaway Children

 

 

 

Training Time Time Time Mean of
Conditions 1 2 3 Means
Experimental 0 .17 .17 .11
Control 0 .11 0 .04
Mean of

Means 0 .14 .08 .07

 

126

Table 4.20 shows that the results were not in the
predicted direction. While both groups experienced an
increase in runaways, the experimental group obtained a
higher mean score than the control group. Moreover, while
the experimental group appeared to maintain the same mean
score from Time 2 to Time 3, the control group experienced a

decrease in runaways.

Question #4: Does training in parenting and
self-management skills, as compared to training
in parenting skills only, lead to fewer contacts
with agency staff?

Hypothesis 4.1: Foster parents trained in
parenting and self-management skills will need

significantly fewer home visits than those
trained in parenting skills only.

This hypothesis was not accepted. The analysis revealed
no significant effect for training (F(1,11)=.43, £8.53) nor
for time (F(2,22)=.73, ‘p-.49) at the .05 level of
significance. Training by time interaction was also not
significant (F(2,22)=.31, 22.74). Table 4.21 summarizes the
results of the analysis, while Table 4.22 presents the means
for training conditions and time periods.

Table 4.22 revealed that the results were not in the
expected direction. The experimental group had more home
contacts than the control group. An increase was noted from
Time 1 to Time 2 for the experimental group, while the

control group mean score remained stable. Though both

127

Table 4.21

Repeated Measures of Variance for Number
of Home Contacts 3

 

 

Source ' 9;, gs g» 3
Training 1 .004 .427 .527
Subjects 11 .010

Time 2 .004 .733 .492
Training X Time 2 .002 .306 .740
Subjects X Time 22 .006

a

 

Number of subjects: Experimental a 6, Control a 7

Table 4.22

Means of Home Contacts

 

 

 

Training Time Time Time Mean of
Conditions 1 2 3 Means
Experimental .07 .12 .07 .08
Control .07 .07 .04 .06
Mean of

Means .07 .09 .05 .07

 

128

groups experienced a drop in contacts from Time 2 to Time 3,

the change favored the control group.

Hypothesis 4.2: Foster parents trained in
parenting and self-management skills will have
significantly fewer office contacts than those
trained in parenting skills only.

This hypothesis was not accepted. The analysis
revealed no significant effect for training (§(1,11)=1.33,
28.27) nor for time (F(2,22)=2.10, 2:.15) at the .05 level.
Training by time interaction was also not significant
(§(2,22)=.85, 42:.44). Table 4.23 summarizes the results of
the analysis, while Table 4.24 presents the means for
training conditions and time periods.

As can be seen from Table 4.24, the results were in the
anticipated direction. The experimental group experienced
an increase in contacts from Time 1 to Time 2 and a decrease
from Time 2 to Time 3. While the control group experienced
a continual decrease over the three time periods, mean
scores did not drop below those of the experimental
subjects. The initial differences between the two scores
suggested that the two groups may not have been equivalent

with respect to this variable.

Hypothesis 4.3: Foster parents trained in
parenting and self-management skills will have
significantly fewer problem-related telephone
contacts than those trained in parenting skills
only.

129

Table 4.23

Results of Repeated Measures of Vagiance for
Number of Office Contacts

 

 

 

Source g; MS E 2
Training 1 .07 1.326 .274
Subjects 11 .05

Time 2 .04 2.10 .146
Training X Time 2 .01 .846 .443
Subjects X Time 22 .02

a

Number of subjects: Experimental = 6, Control a 7

Table 4.24

Means of Office Contact

 

 

 

Training Time Time Time Mean of
Conditions 1 2 3 Means
Experimental .08 .12 .05 .08
Control .24 .17 .09 .17
Mean of

Means .17 .15 .07 .13

 

130

This hypothesis was not accepted. The analysis
revealed no significant effect for training (F(1,ll)=.003,
28.95) nor for time (F(2,22)=l.22, 2:.32) at the .05 level.
Training by time interaction was also not significant
(F(2,22)=.132, p=.29). Table 4.25 presents the means for
training conditions and time periods.

Table 4.26 showed that the results were not in the
predicted direction. The experimental group had a greater
number of telephone contacts from Time 1 to Time 2, as
compared to control subjects. While both groups experienced
a drop in telephone contacts, the control condition had a

higher decrease than the experimental group.

Table 4.25

Results of Repeated Measures of Analygis of
Variance for Telephone Contacts

 

 

 

Source 115. '15 E. .9.
Training 1 0 .003 .954
Subjects 11 .016

Time 2 .007 1.215 .316
Training X Time 2 .007 1.323 .287
Subjects X Time 22 .005

a

Number of subjects: Experimental 3 6, Control a 7

131

TABLE 4.26

Means of Telephone Contacts

 

 

 

Training Time Time Time Mean of
Conditions 1 2 3 Means
Experimental 0 .08 .07 .05
Control .06 .07 .03 .05
Mean of

Means .03 .08 .05 .05

 

Data Analysis for Subjects With
Completed Scaled Measures at Posttest

Twenty-three subjects had complete data on outcome
measures at posttest. Of these, 74 percent were black and
26 percent were white. Seventy-one percent of the sample
consisted of single female heads of households, thirty-one
percent were married.

The ages of these subjects ranged from 23 to 61 years,
with a mean age of 44 years. Experience ranged from 0 to 20
years, with a mean of 2.85 years, a median of 1 year and a
mode of 0 years. Thirty percent did not finish high school
and 25 percent reported some college experience. Twenty
percent finished high school. Two subjects or 11 percent of

those who responded indicated that they were college

132

graduates. The preceding descriptive results are based on
the responses of 19 subjects to these items.

T-tests for differences between means for age and
experience proved not to be statistically significant at
the .05 level. A t-value of -1.04 with 16 degrees of
freedom, significant at the .31 level, was obtained for age.
Experimental subjects had a mean age of 41.78 years, while
control subajects had a mean age of 47.11 years.

Similarly, no statistically significant differences
between the two training groups were noted with respect to
experience. A t-value of -.04 with 11 degrees of freedom
was obtained, which was significant at the .97 level.

In order to assess the extent to which the equivalence
of the two groups with respect to the outcome measures
might have been vitiated by attrition, t-tests for
differences between means were conducted on pretest data.
Table 4.27 summarizes the results of those mean comparisons.

As can be seen from Table 4.27, there were no
significant statistical differences between the two groups
on pretest measures.

Since it has been reported that generally behavior
problems are positively associated with placement breakdowns
and maladaptive interactional patterns in families, data
pertaining to the behavior difficulties of foster children
were analyzed. Tables 4.28 to 4.30 present the results of

those analyses.

133

Table 4.27

Summary of t-Test Results

on Pretest Data

 

 

Scale df tevalue p

Cohesion 21 -1.31 .20
Conflict 21 .54 .60
Control 21 1.21 .24
Efficacy 21 - .48 .63
Reg. for removal 21 .70 .49
Unreq. removals 21 .87 .39
Runaways 21 .02 .99
Home contacts 21 .59 .56
Office contacts 21 -1.07 ..30
Phone contacts 21 -1.16 .26

 

134

Table 4.28

Chi Square for Presenting Problems at Time 1

 

 

Presenting Problem d_f_ 1(chi) p_
Minor or no Behavioral

problems 1 .80 .37
Noncompliance l .02 .87
Bed wetting 1 .02 .88
Stealing/Lying l 2.84 .09
Truancy 1 .07 .79
Inappropriate Sexual

Behavior 1 .89 .34
Antisocial 1 2.20 .14
Hostile Aggressive 1 0.0 1.0
Emotional disturbance l .05 .82
Other 1 .02 .88

 

135

Table 4.29

Chi Square for Presenting Problems at Time 2

 

Presenting Problem if 1(01'11 ) P.

 

Minor or no Behavior
Problems

Noncomplaince
Bed wetting
Stealing/Lying
Truancy

Inappropriate Sexual
Behavior

Antisocial
Hostile Aggressive
Emotional disturbance

Other

.07

1.01
4.82
.36

3.92
2.39

1.01

.79

1.0

.32
.03*
.55

.05*
.12

1.0
1.0

.32

 

*Significant at .05 level

136

Table 4.30

Chi Square for Presenting Problems at Time 3

 

 

Presenting Problem .3; .j£(chi) .3
Minor or no behavior

problems 1 1.88 .17
Noncompliance 1 .07 .79
Bed wetting 1 1.12 .29
Stealing/Lying l 3.90 .05*
Truancy l 1.81 .18
Inappropriate sexual

behavior 1 1.28 .27
Antisocial 1 .35 .67
Emotional disturbance l 0 1.0
Other 1 1.23 .27

 

*Significant at .05 level

137

As can be see from the above tables, subjects in the
two conditions did not differ significantly in relation to
the types of difficulties presented by their foster children
at Time 1. While Table 4.30 shows that the two groups
differed significantly with respect to experience with
chidren who exhibited stealing and lying and inappropriate
sexual behaviors at Time 3, Table 4.31 reveals that they
differed in regard to children who displayed inappropriate
sexual behaviors at Time 3.

Because foster children moved into and out of the
participants' homes throughout the study, an analysis of the
data pertaining to the number of children in the home was
performed. This was done in order to assess the initial
equivalence of the two groups with respect to that variable
and to test for the maintenance of that equivalence over
time. Tables 4.31 and 4.32 summarize the results.

The results of the analysis suggest that those in the
two training conditions did not differ significantly with
respect to the number of foster children they cared for
(§(l,21)=.53, 23.48). Although those in the experimental
condition accepted more foster children across the time
periods than control subjects, no training by time
interaction was observed (F(2,42)=.20, p=.82). A
significant effect for time was observed (E(2,42)-3.81,

28.03) at the .03 level of significance.

138

Table 4.31

Repeated Measures Analysis of Variance of

Foster Children a

 

 

 

 

 

 

Source df_ .MS. .1; IL
Training 1 2.344 .528 .475
Subjects 21 4.438
Time 2 3.304 3.814 .030
Training X Time 2 .170 .20 .823
Subjects X Time 42 .866
a
Number of subjects: Experimental 3 6, Control a 7
Table 4.32
Means of Foster Children
Training Time Time Time Mean of
Conditions 1 2 3 Means
Experimental 1.69 2.38 2.54 2.21
Control 1.50 2.0 2.0 1.83
Mean of
Means 1.61 2.22 2.30 2.04

 

139

Sheffé post hoc comparisons revealed a statistically
significant difference between Time 1 and Time 2 means
(F(2,42) p .025), and between Time 1 and Time 3 means
(fi(2,42) p .01).

Those in both training conditions had an increase in
foster children from Time 1 to Time 2. While experimental
subjects continued to work with increasing number of
children, control subjects kept their ”number of foster
children stable from Time 2 to Time 3.

In addition to analysis of data on the number of
children fostered, data related to attendance, amount of
material mastered and number of completed assignments were
also analyzed to assess intergroup homogeneity. No
significnt difference was found between the two groups with
respect to attendance at the .05 level of significance
(t=.65, g§=21, £2.52). Experimental subjects had a mean
attendance of 5.92 sessions, compared to 5.6 for control
subjects.

Analysis related to amount of material learned, as
measured by classroom quizzes, revealed a mean score of 2.23
for experimental subjects, compared to 3.0 for control
subjects. These results were not significant at the .05
level (t=-l.43, g§=21, p=.l7). As with the case for
subjects who completed all outcome measures at follow-up,
control subjects scored higher on the classroom quizzes than

experimental subjects.

140

Similarly, control subjects completed an average of 3.5
assignments, compared to an average of 2.0 for experimental
subjects. These means were not found to be significant at

the .05 level (t=-l.49, d_f_=21, 2:.15).

Hypothesis Testing

Since the hypotheses to be discussed in the following
sections are the same ones stated earlier, they wilant be
restated.

Hypothesis 1.1 was not accepted. The analysis revealed
no significant effect for training (§(1,21)=l.23, £2.28) nor
for time (§(l,21)=.02, £2.90) at the .05 level. Training X
time interaction was also not significant (g(1,21)=.42,
p=.52). Table 4.33 summarizes the results of the analysis,
while Table 4.34 presents the means for training conditions
and time periods.

Table 4.35 shows that the control group perceived their
families as more cohesive than experimental subjects. The
latter, however, had a gain of .23 from Time 1 to Time 2,
compared to a drop of .20 for control subjects. The control
group perceived their families as more cohesive‘ prior to
training. A difference of .98 favored the control group.

Hypothesis 1.2 was not accepted. The analysis revealed
no significant effect for training (§(l,21)=.35, 2:.56) nor
for time (§(l,21)=.02, £2.90) at the .05 level. Training X

time interaction was also not significant (§(l,21)=.01,

141

Table 4.33

Repeated Measures of Analysis of Variance
of Scores on the Cohesion Subscale of the PBS

 

 

 

 

 

 

Source (if MS F_ a
Training 1 6.555 1.233 .279
Subjects 21 5.317
~Time 1 .021 .018 .896
Training X Time 1 .524 .424 .522
Subjects X Time 21 1.235

a

Number of subjects: Experimental . 13, Control a 10

Table 4.34
Mean of Cohesion Subscale Scores

Training Time Time Mean of
Conditions 1 2 Means
Experimental 6.92 7.15 7.02
Control 7.90 7.70 7.80

Mean of

Means 7.34 7.39 7.37

 

142

12.91). Table 4.35 summarizes the results of the analysis,
while Table 4.36 presents the means for training conditions
and time periods.

As can be seen from Table 4.36, the results were not in
the predicted direction. An increase in conflict was noted
for the experimental group, while no gain was observed for
control subjects. Experimental subjects, however, had a

higher pretest mean score than control subjects.

Table 4.35

Repeated Measures of Analysis of Varia ce
of Scores on the Conflict Subscale

 

 

source <2; 1152. z. p.
Training 1 2.023 .352 .560
Subjects 21 5.754

Time 1 .021 .015 .902
Training X Time 1 .016 .012 .914

Subjects X Time 21 1.402

 

143

Table 4.36

Means of Conflict Subcale Scores

 

 

 

Training Time Time Mean of
Conditions 1 2 Means
Experimental 2.38 2.46 2.42
Control 2.0 2.0 2.0
Mean of

Means 2.21 2.2 2.23

 

Hypothesis 1.3 was not accepted. The analysis revealed
no significant effect for training (§(1,21)2.03, £2.87) nor
for time (§(l,21)21.23, £2.28) at the .05 level of
significance. Training by time interaction was also not
significant (§(l,21)21.74, £2.20). Table 4.37 summarizes
the results of the analysis, while Table 4.38 presents the
means for training conditions and time periods.

Table 4.38 reveals that the results were not in the
anticipated direction. Experimental subjects experienced
decrease in perceived control of their family environments,
while the control group experienced a gain in control.

Hypothesis 2.1 was not accepted. The analysis revealed
no significant effects for training (§(1,21)2.69, £2.42) nor

for time (§(1,21)2.03, £2.88) at the .05 level of

144

Table 4.37

Results of Repeated Measures of Analysis ofa
Variance of Scores on the Control Subscale

 

 

 

 

 

 

Source df M_S_ g 3
Training 1 .080 .027 .870
Subjects 21 2.969

Time 1 3.673 1.232 .280
Training X Time 1 5.179 1.736 .202
Subjects X Time 21 2.983

a

Number of subjects: Experimental 2 13, Control 2 10

Table 4.38
Means of Control Subscale Scores

Treatment Time Time Mean of
Conditions 1 2 Means
Experimental 6.46 5.30 5.88
Control 5.70 5.90 5.80

Mean of

Means 6.13 5.56 5.85

 

145

significance. Training X time interaction was also not
significant (§(l,21)2.21, £2.65). Table 4.39 summarizes the
results of the analysis, while Table 4.40 presents the means

for training conditions and time periods.

Table 4.39

Repeated Measures of Analysis of Variance of
Scores on the FPPES

 

 

Source .df MS .2 P.
Training 1 18.766 .685 .417
Subjects 21 27.414

Time 1 .195 ' .025 .877
Training X Time 1 1.638 .207 .654
Subjects X Time 21 7.936

a

 

Number of subjects: Experimental 2 13, Control 2 10

Table 4.40 reveals that the results were not in the
predicted direction. The control group experienced an
increase in sense of efficacy or competence, while the
experimental group had a decrease in that area. Control
subjects started with higher level of efficacy scores than

experimental subjects and increased these after training.

146

Table 4.40

Means of Efficacy Subscale Scores

 

 

 

Training Time Time Mean of
Conditions 1 2 Means
Experimental 28.68 28.23 28.46
Control 29.60 29.90 29.75
Mean of

Means 29.08 28.95 29.02

 

Since archival data was available, the following
analyses included follow-up information on the
subjects. Hypothesis 3.1 was not accepted. The analysis
revealed no significant effects for training (§(l,21)2.14,
£2.72) or for time (§(2,42)2.96, .E'-39) at the .05 level.
Training X time interaction was also not significant
(F(2,42)2l.66, £2.20). Table 4.41 summarizes the results of
the analysis, while Table 4.42 presents the means for
training conditions and time periods.

As can be seen from Table 4.42, the results were mixed.
While both groups experienced increased requests for removal
from Time 1 to Time 2, more children were removed from the
control group than the experimental group. A decrease in

such requests was observed for control subjects from Time 2

147

Table 4.41

Repeated Measures of Analysis of Variance on
Number of Requests for Removala

 

 

 

 

 

 

Source JfiL E 2
Training .01 .135 .717
Subjects .076

Time .061 .961 .391
Training X Time .105 1.663 .202
Subjects X Time .063

a

Number of subjects: Experimental 2 13, Control 2 10

Table 4.42
Mean for Removal Requests

Training Time Time Time Mean of
Conditions 1 2 3 Means
Experimental .06 .09 .21 .12
Control .03 .20 .05 .09
Mean of
Means .05 .14 .14 .11

 

148

to Time 3, but an increase was noted for experimental
subjects.

Hypothesis 3.2 was not accepted. The analysis revealed
no significant effects for training (§(1,21)2.63, £2.44) or
for time (F(2,42)21.74, £2.19) at the .05 level. Training X
time interaction was also not significant (F(2,42)2.15,
£2.86). Table 4.43 summarizes the results of the analysis,
while Table 4.44 presents the means for training conditions

and time periods.

Table 4.43

Repeated Measures of Analysis of gariance for
Unrequested Removals

 

 

 

source .4; he a a
Training 1 .003 .633 .435
Subjects 21 .005

Time 2 .009 1.742 .188
Training X Time 2 .001 .151 .860
Subjects X Time 42 .005

a

Number of subjects: Experimental 2 13, Control 2 10

149

Table 4.44

Means of Unrequested Removals

 

 

 

Training Time Time Time Mean of
Conditions 1 2 3 Means
Experimental .023 i .046 0 .023
Control 0 .030 0 .010
Mean of

Means .013 .039 0 .012

 

Table 4.44 shows that the results were not in the
expected direction. The experimental group had had more
foster children removed from their homes at Time 2 than had
the control group. The latter, however, had a greater
increase in such removals from pretest measures than
experimental subjects. Both subjects experienced a sharp
drop in such removals from Time 2 to Time 3, with the
experimental group experiencing a higher decrease than
control subjects in that interval.

Hypothesis 3.3 was not accepted. The analysis revealed
no significant effects for training (F(l,21)2.55, £2.47) or
for time (F(2,42)2.96, £2.39) at the .05 level. Training X
time interaction was also not significant (F(2,42)2.29,

£2.75). Table 4.45 summarizes the results of the analysis,

150

while Table 4.46 presents the means for training conditions
and time periods.

As can be seen from Table 4.46, the results were not in
the desired direction. Experimental subjects had more
children who ran away from their home than control subjects
across the three intervals. While both groups had the same
number of runaway children at Time 1, the experimental group

had an increase of runaway incidents from Time 1 to Time 2.

Table 4.45

Repeated Measures of Analysis of Variance
for Runaway Children

 

 

 

Source d2, M5 .3 2
Training 1 .057 .548 .467
Subjects 21 .104

Time 2 .049 .959 .392
Training X Time 2 .015 .291 .749
Subjects X Time 42 .051

a

Number of subjects: Experimental 2 13, Control 2 10

151

Table 4.46

Means for Runaway Children

 

 

 

Training Time Time Time Mean of
Conditions 1 2 3 Means
Experimental .10 .177 .077 .118
Control .10 .08 0 .06
Mean of

Means .10 .135 .043 .09

 

Hypothesis 4.1 was not accepted. The analysis revealed
no significant effects for training (§(1,21)2l.40, £2.25) or
for time (g(2,42)21.74, £2.19) at the .05 level. Training X
time interaction was also not significant (§(2,42)2.15,
£2.86). Table 4.47 summarizes the results of the analysis,
while Table 4.48 presents the means for training ennditions
and time periods.

Table 4.48 shows that the results were not in the
anticipated direction. Experimental subjects consistently
had more home visits than control subjects across time. No
increase in home visits was noted for the control group
between Time 1 and Time 2. While both groups experienced a
decrease in home visits from Time 2 to Time 3, the control
group still had fewer home visits than the experimental

group.

152

Table 4.47

Repeated Measures of Analysis of Variance
of Home Contacts 3

 

 

 

source if £4.8_ z a
Training 1 .012 1.401 .250
Subjects 21 .009

Time 2 .009 1.742 .188
Training X Time 2 .001 .151 .860
Subjects X Time 42 .005

a

Number of subjects: Experimental 2 13, Control 2 10

Table 4.48

Means of Home Contacts

 

 

 

Training Time Time Time Mean of
Conditions 1 2 3 Means
Experimental .08 .10 .054 .077
Control .06 .06 .03 .05
Mean of

Means .07 .08 .043 .064

 

153

Hypothesis 4.2 was not accepted. The analysis revealed
no significant effects for training (§(1,21)2.12, £2.73) or
for time (§(2,42)21.55, £2.23) at the .05 level. Training X
time interaction was also not significant (§(2,42)2l.28,
£2.29). Table 4.49 summarizes the results of the analysis.
Table 4.50 presents the means for training conditions and
time periods.

As can be seen from Table 4.50, the results were mixed.
The experimental group had fewer office contacts at Time 2,
but more such contacts at Time 3 than the control group.
While both groups had a decrease in number of office
contacts between Time 2 and Time 3, the control group had

slightly fewer office visits than the experimental group.

Table 4.49

Repeated Measures of Analysis of Variance
of Office Contactsa

 

 

Source _d_f MS _F_‘ a
Training 1 .007 .12 .73
Subjects 21 .056

Time 2 .018 1.55 .23
Training X Time 2 .015 1.28 .29

Subjects X Time 42 .012

 

154

Table 4.50

Means of Office Contacts

 

 

 

Training Time Time Time Mean of
Conditions 1 2 3 Means
Experimental .11 .12 .10 .11
Control .18 .14 .07 .13
Mean of

Means .14 .13 .09 .12

 

Hypothesis 4.3 was not accepted. The analysis revealed
no significant effects for training (§(l,21)20, £2.99) or
for time (£(2,42)2.64, £2.53) at the .05 level. Training X
time interaction was also not significant (£(2,42)2l.36,
£2.27). Table 4.51 summarizes the results of the analysis,
while Table 4.52 presents the means for training conditions
and time periods.

Table 4.52 suggests that the results were not in the
hypothesized direction. Experimental subjects had more
phone contacts than control subjects at Time 2 and Time 3.
While the control group maintained telephone contacts at
their pretest level at Time 2, experimental subjects had an
increase in telephone contacts from pretest to posttest.

Moreover, while the control group had a decrease of

155

 

 

 

 

 

 

Table 4.51
Repeated Measures of Analysis og‘Variance of
Telephone Contacts
Source d§_ MS .3 ‘2
Training 1 0 0 .99
Subjects 21 .012
Time 2 .004 .64 .53
Training X Time 2 .008 1.36 .27
Subjects X Time 42 .006
a
Number of subjects: Experimental 2 13, Control 2 10
Table 4.52
Means of Telephone Contacts
Training Time Time Time Mean of
Conditions 1 2 3 Means
Experimental .023 .062 .077 .054
Control .06 .06 .04 .053
Mean of
Means .039 .061 .061 .053

 

156

telephone contacts from Time 2 to Time 3, the experimental

group had an increase in such contacts.

Supplementary Analysis

 

Since there were no significant effects for type of
training, groups in the two training conditions were
combined and those who attended four or more training
sessions were compared with subjects who attended three or
fewer training sessions (but at least one session). This
was done in an attempt to assess the effects, if any, of
duration of training on the two groups. In the following
discussion, those subjects who attended more than three
sessions will be referred to as more cooperative subjects,
while those who attended three or fewer sessions will be
referred to as less cooperative subjects.

Of the 52 subjects who attended at least one training
sessions, 34 or 65 percent were single female heads of
households. Seventy-one percent were black, 29 percent were
white. These subjects had an age range of 23 to 67 years,
with a mean age of 46.36 years. Experience ranged from zero
to 21 years, with a mean of 2.51 years, a median of 1.07 and
mode of 1 year of experience. Twenty-eight percent had some
high school: 25 percent completed high school and 9 percent
were college graduates.

The two groups that formed this sample did not differ

significantly with respect to age. A E-value of -l.33 with

157

42 degrees of freedom, significant at the .19 level of
significance was obtained, when the two group means for age
were compared. Similarly, no significant differences were
noted with respect to experience (32.21, df243, £2.84).
More cooperative subjects had a mean of 2.65 years of
experience, as compared to one of 2.36 for the less
cooperative group.

Since the majority of the less cooperative subjects did
not complete scaled measures at pretest, it was not possible
to assess the two groups' equivalence with respect to the
pretest measures. Only ten less cooperative subjects turned
in completed pretest measures. Because archival data were
available, however, it was possible to compare the two
groups according to placement breakdowns and types of
contacts. Table 4.53 summarizes the results of the t—test
for these variables.

As can be seen from Table 4.53, .a significant
difference, favoring less cooperative subjects, was noted in
regards to requests for removal. No significant differences
at the .05 level were obtained with respect to the other
variables.

Chi square analysis of data relating to presenting
behaviors of foster children revealed no statistical
differences at the .05 level. Table 4.54 to 4.56 summarize

the results of these analyses.

158

Table 4.53

Summary of E-Test Results of Pretest Variables

 

 

Scale .gg tyvalue '2

Removal Requests 49 2.0 .05*
Unrequested removals 49 -.53 .60
Runaways 49 .72 .47
Home contacts 49 -l.21 .23
Office contacts 49 1.87 .07
Phone contacts 49 -.85 .40

 

*Significant at .05 level

159

Table 4.54

Chi Square Results for Presenting Problems at Time 1

 

 

Presenting Problem d_f _X (chi) a
Minor or no behavior

problems 1 .13 .71
Noncompliance l .02 .88
Bed wetting 1 .82 .36
Stealing/Lying l .02 .90
Truancy 1 1
Inappropriate sexual

behavior 1 .12 .73
Antisocial l .21 .65
Hostile aggressive l .23 .63
Emotional disturbance 1 1
Other 1 .22 .64

 

160

 

 

Table 4.55

Chi Square Results for Presenting Problems at Time 2
Presenting Problem if X_( Chi) £
Minor or no behavior
problems 1 .17 .68
Noncomplaince 1 .14 .71
Bed wetting l .26 .61
Stealing/Lying l .21 .64
Truancy 1
Inappropriate sexual
behavior 1 .33 .57
Antisocial 1 .86 .35
Hostile aggressive 1 .06 .81
Emotional disturbance l .99
Other 1 .71 .40

 

161

 

 

Table 4.56
Chi Square Results for Presenting Problems at Time 3

Presenting Problem df. X (chi) 9,
Minor or no behavior

problems 1 1.06 .30
Noncompliance 1 2.63 .10
Bed wetting l 1.41 .24
Stealing/Lying 1 .55 .46
Truancy 1 .02 .89
Inappropriate sexual

behavior 1 .62 .43
Antisocial l .71 .40
Hostile aggressive l .07 .78
Emotional disturbance l .26 .61
Other 1 .83 .36

 

162

 

 

As can be seen from the above tables, the two groups
did not differ significantly regarding the presenting
problems of the children they fostered.

As was the case for the two previous samples, an
analysis of the data relating to the number of children
fostered across the time periods was also performed. Tables
4.57 and 4.58 summarize the results of that analysis.

Table 4.57
Repeated Measures of Analysis of Variance
of Number of Foster Childrena

Source _df .145. j: ,9
Training 1 3.11 .86 .36
Subjects 49 3.61

Time 2 3.65 5.23 .01
Training X Time 2 .11 .16 .85
Subjects X Time 98 .70

 

Table 4.57 shows that trainingtduration

did not differ

significantly with respect to the number of foster children

the two groups parented (§(l,49)2.86,

was

significant effect for time was

£2.01).

no significant training X time

£2.36). While there
interaction effect, a

observed (§(2,98)25.23,

163

Table 4.58

Means for Number of Foster Children

 

 

 

 

Training Time Time Time Mean of

Conditions 1 2 3 Means

More cooperative 1.70 2.22 2.22 2.04

Less cooperative 1.50 1.83 1.96 1.69

Mean of Means 1.61 2.04 2.10 1.92
Scheffé Post Hoc comparison revealed contrasts

significant beyond the .01 level. Differences between the

means for Time 1 and Time 2 were statistically significant

(F(2,98)26.73), £ .01), as were differences obtained in

comparing the means for Time 1 and Time 3 (F(2,98)29.11, £
.01).

In contrast to the other two samples, analyses of data
relating to attendance, amount of material mastered and
number of completed assignments, revealed that the more
cooperative and less cooperative groups differed
significantly with respect to those variables. A t-value of
13.48, g§=so, £2.00 was obtained for attendance. More
cooperative subjects attended an average of 5.66 sessions,
compared to an average of two sessions for the less

cooperative group.

164

Less cooperative subjects obtained a mean quiz score
of .44, while. the more cooperative group achieved an
average of 2.52. Comparison of the two means produced a 5-
value of 4.49 with 34 degrees of freedom. This value is
significant at the probability of .00.

Similarly, significant differences were found between
the two groups in number of assignments completed. A 5-
value of 2.60 with 35 degrees of freedom, significant at
the .01 level was obtained. More cooperative subjects
completed an average of 2.26 assignments, compared to a mean
of .20 assignments for the less cooperative subjects.

The analysis of data relating to requested removals
revealed no significant effect for training duration
(§(l,47)-2.88, £2.10) or for time (E(2,94)21.13, £2.33) at
the .05 level of significance. Training X time interaction
was also not significant (F(2,94)2.14, £2.87). Table 4.59
summarizes the results of the analysis, while Table 4.60
presents the means for training conditions and time periods.

Table 4.60 shows that the more cooperative group
initiated more requests for removal than the less
cooperative group. This pattern was consistent across time,
starting at pretesting. A slight decrease was noted for
more cooperative subjects while an increase was observed for
the less cooperative group from Time 2 to Time 3. While
both groups increased their requests for removal from Time 1
to Time 2, the less cooperative group experienced a greater

increase than the more cooperative group.

165

Table 4.59

Repeated Measures Analysis of Variance on
Number of Requests for Removal‘3

 

 

 

source if .242: 2 2
Training 1 .17 2.878 .096
Subjects 47 .06

Time 2 .064 1.129 .328
Training X Time 2 .008 .143 .867
Subjects X Time 94 .06

a

Number of subjects: More cooperative 2 27, Less
cooperative 2 24

Table 4.60

Mean Number of Requests for Removal

 

 

Training Time Time Time Mean of
Conditions 1 2 3 Means
More cooperative .09 .14 .13 .12
Less cooperative 0 .06 .09 .05

 

Mean of Means .04 .10 .11 .09

 

166

The analysis of data pertaining to unrequested removals
revealed no significant effects for training (§(l,47)2.20,
£2.66) or for time (F(2,94)2.66, £2.52) at the .05 level.
Training X time interaction was significant at the .05 level
(g(2,94)=3.59, £2.03). Table 4.61 summarizes the results of
the analysis, while Table 4.62 presents the means for

cooperation with training over the time periods.

Table 4.61

Repeated Measures Analysis of Variange
for Number of Unrequested Removals

 

 

 

source d_f _M_§ _F 2
Training 1 .005 .20 .657
Subjects 47 .03

Time 2 .01 .664 .517
Training X Time 2 .059 3.594 .031
Subjects X Time 94 .016

a

Number of subjects: More cooperative 2 27, Less
cooperative 2 24

167

Table 4.62

Mean Number of Unrequested Removals

 

 

 

Training Time Time Time Mean of
Conditions 1 2 3 Means
More cooperative .01 .07 0 .03
Less cooperative .02 .01 .08 .04
Mean of Means .014 .04 .04 .03

 

As can be observed from Table 4.62, the more
cooperative group experienced an increase in unrequested
removals from Time 1 to Time 2, while the less cooperative
subjects experienced a drop in such requests. From Time 2
to Time 3, the more cooperative subjects experienced a
significant decrease, while the less cooperative subjects
registered a significant increase in unrequested removals.
The results suggest that between Time 2 and Time 3, more
cooperative subjecs did not have as many children removed
from their homes as did less cooperative subjects.

The analysis of data related to runaways revealed no
significant effects for training (§(l,47)2.24, £2.62) or for
time (F(2,94)2.73, £2.49) at the .05 level. Training X time
interaction was also not significant (§(2,94)2.99, £2.38).
Table 4.63 summarizes the results of the analysis, while
Table 4.64 presents the means for cooperation with training

and time periods.

168

Table 4.63

Repeated Measures Analysisa of
Variance for Runaways

 

 

 

 

 

 

Source di. HE E 2
Training 1 .016 .245 .623
Subjects 47 .065
Time 2 .023 .726 .487
Training X Time 2 .032 .988 .376
Subjects X Time 94 .032
Table 4.64

Means for Runaways
Training Time Time Time Mean of
Conditions 1 2 3 Means
More cooperative .085 .122 .033 .08
Less cooperative .041 .066 .070 .059
Mean of Means .063 .094 .052 .070

 

169

Table 4.64 reveals an increase in runaways from Time 1
to Time 2 for subjects in both groups, with the less
cooperative subjects showing a steeper increase. While the
number of runaways then dropped for the more cooperative
subjects, the less cooperative group experienced another
slight increase.

Analysis of data relating to home contacts revealed no
significant effects for cooperation with training
(§(l,47)22.12, £2.15) or for time (F(2,94)22, £2.15) at
the .05 level of significance. Training X time interaction
was also not significant (F(2,94)21.56, £2.22). Table 4.65
summarizes the results of the analysis, while Table 4.66
presents the means for cooperation with training and time
periods.

As can be seen from Table 4.66, more cooperative
subjects had fewer home contacts than less cooperative
subjects. The latter group, however, experienced a greater
decrease in contacts from Time 1 to Time 2 than the former.
While both groups had fewer contacts with agency staff
between Time 2 and Time 3, contacts between more cooperative
subjects and agency personnel were lower than such contacts
for less cooperative subjects.

The analysis of data pertaining to office contacts
revealed no significant effects for cooperation with
training (§(1,47)2.47, £2.50) or for time (§(2,94)21.35,

£2.25) at the .05 level. Training X time interaction was

170

Table 4.65

Repeated Measures of Analysis of Variance for
Number of Home Contactsa

 

 

 

 

 

 

source as :18. _F p.
Training 1 .973 2.124 .152
Subjects 47 .458
Time 2 .863 1.936 .150
Training X Time 2 .693 1.555 .217
Subjects X Time 94 .446
a .
Number of subjects: more c00perative 2 27, less
cooperative 2 24
Table 4.66
Means of Home Contacts
Training Time Time Time Mean of
Conditions 1 2 3 Means
More
Cooperative .077 .075 .035 .063
Less
Cooperative .509 .095 .066 .223
Mean of

Means .293 .085 .051 .143

 

171

also not significant (g(2,94)21.91, £2.15). Table 4.67
summarizes the results of the analysis, while Table 4.68
presents the means for training cooperation and time
periods.

Table 4.68 shows that more cooperative subjects
experienced a decrease in office contacts while less
cooperative ones increased their contacts from Time 1 to
Time 2. Both groups showed a drop in number of office

contacts between Time 2 and Time 3.

Table 4.67

Repeated Measures of Analysis of Variance
for Number of Office Contactsa

 

 

 

Source ‘2; ‘MS 2 £
Training 1 .020 .469 .497
Subjects 47 .043

Time 2 .021 1.350 .254
Training X Time 2 .030 1.906 .154
Subjects X Time 94 .015

a

Number of subjects: More cooperative 2 27, Less
cooperative 2 24

172

Table 4.68

Means of Office Contacts

 

 

 

Training Time Time Time Mean of
Conditions 1 2 3 Means
More cooperative .150 .115 .080 .115
Less cooperative .070 .125 .081 .092
Mean of Means .110 .120 .080 .103

 

The analysis of data relating to telephone contacts
revealed no significant effects for cooperation with
training (§(l,47)2.18, £2.67) or for time (F(2,94)2.003,
£2.99) at the .05 level. Training X time interaction was
also not significant (F(2,94)2.72, £2.49). Table 4.69
summarizes the results of the analysis, while Table 4.70
presents the means for training cooperation and time
periods.

As can be seen from Table 4.70, the more cooperative
group had an increase in telephone contacts, while the less
cooperative group had a decrease in such contacts between
Time 1 and Time 2. The reverse was observed between Time 2
and Time 3, when more cooperative subjects had a decrease in
telephone contacts, and less cooperative subjects increased

their number of such contacts.

173

Table 4.69

Repeated Measures of Analysis of
Variance for Telephone Contactsa

 

 

 

source dz. 15 3 .9
Training 1 .003 .184 .670
Subjects 47 .019

Time 2 0 .003 .997
Training X Time 2 .006 .719 .490
Subjects X Time 94 .008

a

Number of subjects: More cooperative 2 27, Less
cooperative 2 24

TABLE 4.70

Mean Number of Telephone Contacts

 

 

Training Time Time Time Mean of
Conditions 1 2 3 Means
More cooperative .038 .059 .050 .049
Less cooperative .070 .047 .058 .058

 

Mean of Means .054 .053 .054 .054

 

174

In summary, subjects trained in parenting and self-
management skills showed no statistically significant
changes, when compared to those trained in parenting skills
only. Comparison of more cooperative and less cooperative
subjects revealed that the former had significantly fewer
children removed from their homes after training than did
the latter. No other significant differences were observed

between the two groups.

CHAPTER FIVE

SUMMARY, DISCUSSION AND IMPLICATIONS

Over the last hundred years, foster family care has
become the most widespread and preferred form of substitute
family care for children with special needs and others
requiring parental surrogates. Youngsters who do not have
or cannot remain in their own homes, who would have been
placed in institutions 50 years ago, are now placed with
substitute families, thus enabling them to remain in their
home communities. Children who are being kept out of
institutions or those removed from institutions are retained
in or returned to the community. Since the parents of
these youngsters are for a variety of reasons unable to meet
their needs, either temporarily or permanently, child
welfare professionals have relied on foster family care to
fill the parental void.

As indicated earlier, the increase of youngsters with
special needs into the foster family care system has been a
source of both challenge and strain for professionals in the
field. However, no one bears more of the brunt of the
strains and stresses created by such children than foster
parents. Traditional foster care for children without

175

176

parents has changed to include care for those with
ineffective or abusive parents, without concommitant changes
in the foster family care system to accommodate the new
clientele. Foster parents are expected to function as
parental surrogates to children who require therapeutic
handling in the form of parental care.

Realization of the new difficulties inherent in the
role of foster parents has brought forth a
reconceptualization in this role. Foster parents are
beginning to be perceived as therapeutic agents who are
essential for both the habilitation and rehabilition of
youths with special needs.

With this reconceptualization has come a recognition of
the need for specialized training for foster parents. Since
the 1970s, a great many programs have been developed to
provide this special training. The results of evaluative
studies of these training programs have been encouraging but
inconclusive. Reliance on paper and pencil measures and
limitations in the designs of the investigations prevent
definitive appraisal of their usefulness. Particularly,
these studies do not evaluate training programs with
respect to placement-related outcomes. Attitudinal measures
and measures of perceptions of change in foster children's
behavior tend to be the most common forms of outcome
measurement. Placement stability, foster parent turnover,

changes in family climate, and contact with agency child

177

welfare staff have not received sufficient attention as
outcome measures. Thus, there is a need for evaluation to
assess training programs with respect to placement-related
outcomes that have the potential to alleviate or increase
the burdens on the family foster care system.

Thus, it was the purpose of this investigation to
assess the differential effects of two l6-hour behaviorally
oriented foster parent training programs on perceptions of
family climate, parenting efficacy, placement stability and
dependence on agency staff. One curriculum emphasized child
behavior management and self-management skills, while the
other concerned only behavior management. Both types of
training curricula had been found useful with various parent
populations, but they had not yet been compared. The self-
management component is designed to train parents in
techniques of self-control that will allow them to reduce
emotional arousal and thus facilitate the use of behavioral
principles in crisis situations.

To evaluate the two curricula, a group X time research
design was used. There were two groups and three time
measurement points. Measurements were taken prior to
training, at completion of training and during a two-week
period 60 days after training ended.

Subjects for this research investigation were licensed
foster parents affiliated with a child placement agency, who

were required to attend training as one of the requirements

178

for maintaining their license. Assignment to training
groups was done on a random basis. Although 98 subjects
were assigned to the two training conditions, because
of attrition in various forms, results are reported for
three subsamples of the sample. These subsamples consisted
of (a) subjects who attended four or more training sessions
and completed all assessments at the three measurement
points and (b) subjects who attended four or more sessions,
but completed only pretest and posttest assessment measures.
Supplementary results were calculated for (c) subjects who
attended four or more training sessions vs. those who
attended three or fewer sessions.

While both blacks and whites were represented, the
majority of the subjects were black females, who are single
heads of households residing in an urban area. Due to the
subjects' reluctance to respond to items pertaining to
income, it was not possible to ascertain their economic
status. The majority of the subjects reported that they had
finished high school.

Discussion of Results

 

Four research questions were formulated for this study,
from which ten hypotheses were generated and experimentally
tested. The previous chapter presented the results of the
analyses pertaining to each hypothesis. In the review of

the results in the next section, subjects who completed

179

measures at all three measurement points are referred to as
"follow-up subjects,” while those who had completed pretest
and posttest measures only are called "posttest subjects."

Question #1: Does training in parenting and self

management skills as compared to training in

parenting skills only, lead to improved family
environment as perceived by foster parents?

The three hypotheses relating to this question were not
accepted. The two curricula did not significantly differ in
their effects on the foster parents' perceptions of family
climate.

The results for the follow-up subjects did differ,
however, from those of posttest subjects. While the results
consistently favored subjects trained in parenting only for
posttest sample, they were mixed for follow-up subjects.
Experimental subjects in the follow-up subsample perceived
their family environments as being less cohesive, with less
control and lower levels of conflict than did control group
subjects. Except for conflict, these results were not in
the direction predicted. No change in cohesion was observed
from pretest to posttest for experimental subjects, while a
drop was noted for control subjects over the same period.
Control subjects perceived their families as more cohesive
over the time periods, with an increase in perceived
cohesion noted after training ended.

Experimental subjects perceived less conflict in their

family environments than did control subjects at all three

180

measurement points. An increase in perceived conflict
was noted after training, but this leveled off in the two-
month period between completion and follow-up. Control
subjects, on the other hand, maintained their pretest scores
at posttest, but perceived greater conflict in the time
between completion of training and follow-up.

Both groups suffered a drop in perceived control from
their pretest scores, but the control group perceived more
control than did experimental subjects. Both groups
increased in perceived control after training, although the
increase favored control group subjects.

Experimental subjects in the posttest subsample
consistently scored lower than control subjects on all
measures of family environment after training ended.
Experimental subjects perceived their family environments as
less cohesive and more conflictual, with less control than
did control subjects.

The above results suggest that training in parenting
and self-management skills, as compared to training in
parenting only, does not lead to more positive perceptions

of family climate.

Question #2: Does training in parenting and self
management skills, as compared to training in
parenting skills only, lead to an increased sense
of parenting efficacy in foster parents?

181

The hypothesis generated by this question was rejected.
No significant changes in perceived parenting efficacy were
observed between the two groups. Experimental subjects in
the follow-up subsample consistently improved on their
pretest score, while the control group varied after pretest.
An improvement on pretest mean score was observed at the end
of training, while a decrease was noted two months following
training.

The results for the posttest subsample varied from
those of the follow-up subjects. Control group subjects
posted an increase in sense of parenting efficacy, while
experimental subjects experienced a decrease.

These findings suggest that training in parenting and
self-management skills, as compared to training in parenting
only, does not lead to significant changes in foster

parents' sense of parenting efficacy.

Question #3: Does training in parenting and
self-management skills, as compared to training
in parenting skills only, lead to fewer placement
breakdowns?

The three hypotheses that were generated by this
question were rejected. It was hypothesized that
experimental subjects would make fewer requests for removal
of foster children from their homes, have fewer children
removed by agency personnel and have fewer children running

away from their homes than would control group subjects.

182

The results were contrary to those expected for both
subsamples. Except for requests for removal, for which
results varied, control subjects did better in retaining
their foster children than did experimental subjects.
Training groups in both subsamples made increased request to
have foster children removed while in training. While the
experimental groups in both subsamples slightly increased
their requests, the control groups made more numerous
requests. In the time period that followed the end of
training, the opposite was true. Experimental groups made
relatively more removal requests than control subjects. A
significant time effect for unrequested removals was noted
for, the follow-up subjects. Both training groups in that
subsample experienced an increase from pretraining scores
that was followed by a decrease after completion of
training. It is unclear at this time as to what was
responsible for the significant time effect.

Since none of the observed differences between training
conditions were significant, it appears that training in
parenting and self-management skills, as compared to
training in parenting skills alone, does not lead to greater

placement stability.

Question #4: Does training in parenting and
self-management skills, as compared to training
in parenting skills only, lead to fewer contacts
with agency staff?

183

The hypotheses based on this question were not
confirmed. It was hypothesized that experimental subjects
would have fewer home visits, office contacts and problem-
related telephone contacts after training than the control
group. As has been the case with the previous results,
control subjects in both subsamples did better than the
experimental group, with the exception of number of office
contacts. Experimental subjects in the "follow-up
subsample" had fewer office contacts during and after
training than did the control subjects. The situation was
slightly different for the "posttest subsample.” The
control subjects had more office contacts during training
than they did afterwards, while the experimental group had
more contacts after than during training.

Based on these results, it would appear that training
in parenting and self-management skills, as opposed to
training in parenting alone, does not lead to decreased

dependence on agency staff.

Discussion

 

The lack of any significant training effect cannot be
explained merely by relative inefficacy of the experimental
training in producing the hypothesized changes. In spite
of efforts to control for extraneous variables and reduce
the probability for alternative explanations, the results of

this investigation defy explanation on the basis of the

184

relative strength of the two training conditions. Several
other factors relating to subject selection, intergroup
homogeneity, attrition, training implementation, and
dependent measures compete for the explanation of the
obtained results.

The relatively high level of attritioin may have
affected the equivalence of the training groups. The
analysis of the data relating to deviant behaviour presented
by foster children showed that the two groups of the
”follow-up” subsample differed significantly on one variable
at Time 2 and on two variables at Time 3. Results for the
'posttest'°subsample indicated that the two groups differed
on two variables at Time 2 and on one variable at Time 3.
While these results may have been due to chance because of
the number of statistical tests conducted, they cannot be
discounted entirely. In addition, there are reasons to
think that the two groups may not have been equivalent with
respect to interest in training. Control group subjects
scored higher on classroom quizzes, completed more homework
assignments and were perceived as more receptive by trainers
than subjects in the experimental groups. Although these
findings were not significant, they cannot be ignored in
evaluating the results, for classroom quizzes and homework
assignments afforded subjects the opportunity to receive
feedback on their application of the principles and

techniques that were taught. Interestingly, experimental

185

subjects, in the follow-up subsample, reported using the
skills they learned more frequently than the control group,
but with less satisfaction with their results. In short,
attrition may have interacted with the selection process, to
produce training groups that may have differed in their
receptiveness to the training.

In addition to selection, intergroup homogeneity also
needs to be considered in assessing the results of this
investigation. It was observed that parents in the
experimental groups tended to introduce nonrelevant topics
into training sessions more often than control subjects and
were less willing to participate in interactional exercises
in the classroom that entailed engaging in some degree of
introspection and self-examination. It is quite probable
that the experimental subjects' behaviors were attempts to
avoid participation in such exercises in a public forum
prior to establishing some sense of trust with the trainers
and other subjects.

Difficulties encountered in implementing training
suggest that the time used for training may have been
insufficient. Time difficulties stemmed from several
sources. The agency and foster parents appeared to function
in a state of almost continual crisis, resulting in frequent
interruption of training sessions when agency personnel

needed to speak to subjects. Subjects were pulled out of

186

training sessions, for a variety of reasons, which took time
away from training.

Absenteeism also depleted the time allocated for
training. Missed sessions frequently required repetition of
previously covered material, reducing the time available to
cover new concepts and practice new skills.

Another problem was some subjects' introduction of
nonrelevant topics. This angered other subjects and may
have contributed to poor attendance, .as well as depleting
training time. Nonrelated topics tended to concern
controversial agency policies or irritation and anger with
agency staff.

The loss of training time is suspected to have had more
negative effects on experimental than control subjects,
since the former group had to cover more material within the
same amount of time. It is possible that this situation
affected experimental subjects' ability to master and
integrate the information.

A fourth factor that contributed to the loss of
training time was related to the trainers' lack of knowledge
of the learning style of predominantly black, low S.E.S.,
urban foster parents. While all of the trainers were
sufficiently well versed in the concepts and techniques that
were taught, they lacked experience in how to adapt their
material to the needs and mode of learning of the majority

of the parents. This situation may ahve contributed to the

187

difficulties they encountered in getting parents to focus on
training-related issues.

A fifth factor that may have affected the results was
reading and writing ability. Performance on classroom
quizzes and written exercises suggested that many subjects
in both training conditions had difficulties with reading
and writing. The superior performance of control subjects
on classroom quizzes and homework assignments suggests that
these subjects may not have been as affected by inadequate
reading and writing ability as they experimental subjects.

Apart from implementation difficulties, the dependent
measures used may have contributed to the lack of any
statistically significant findings. Subjects were very
careful and cautious in completing the scaled measures.
Many expressed concerns about being misunderstood and
misperceived. Since the desirable responses could be easily
recognized, it is quite possible that social desirability
may have affected subjects' responses.

Another factor that may have affected the results was
staff turnover at the agency. After the end of training,
the agency's staff changes seemed to affect recorded
information. It was observed that recent reports in the
files were shorter and tended to have less pertinent
information than those prior to the end of training. In
light of this, the significant time effect observed for

unrequestd removals and the overall tendency of the results,

188

based on archival data, to improve after training ended may
be more of an artifact of insufficient information than the
effect of the combined training curricula over time.

In summary, it would appear that the answers to the
research questions posed by this investigation cannot be
answer conclusively. The many competing explanations that
have been discussed suggest the results of this study may
have been affected by factors other than the training

curricula.

Supplementary Analysis

 

Since no statistical significant changes were found for
the two training conditions, subjects in the two training
conditions were aggregated in one group and compared to
foster parents who had attended less than four sessions.

While these two groups did not differ significantly
with respect to the severity of behavior problems
confronting them, nor in the number of children under their
care, the more cooperative group had more requests for
removal at pretest, significant at the .05 level of
significance, than the less cooperative group. Another
result that was significant at the .07 level of
significance, in favor of the more cooperative group,
was number of office contacts. More cooperative subjects
had more office contacts than less cooperative subjects.

This would seem to suggest that foster parents who did not

189

drop out of training may have been experiencing more
difficulties with their children. This assertion is in
agreement with the observation that parents who tended to
participate in training were motivated to do so partly by
the difficulties presented by their foster children (Boyd &
Remy, 1978).

No statistically significant differences were noted in
requests for removal, unrequested removals, runaways, home
visits, office contacts and telephone contacts. A
significant group X time interaction was obtained for
unrequested removals (F(2,94)23.59, £2.03). . While more
cooperative subjects experienced a significant drop in
number of foster children removed from their homes, the less
cooperative group experienced a significant increase in such
incidents. Before training ended, however, more cooperative
subjects experienced a relatively higher number of such
incidents than less cooperative group. It is suspected the
group X time interaction reflects the rotation of difficult
children across the two groups. According to agency
personnel, approximately ten percent of their difficult
foster children move from home to home until no home is

suitable for them.

Implications for Future Research

 

The results of this investigation indicated that the

two training curricula did not statistically differ with

190

respect to their effects on family climate, sense of
parenting efficacy, placement stability and dependence on
agency personnel. These findings, however, are far from
being conclusive. Several factors appear to provide
competing explanations for the results. A replication of
this investigation is indicated.

In replicating this investigation, it is suggested that
arrangements be made to test all family members on the
measure assessing family climate. The lack of research
personnel prevented the application of that procedure in
this investigation. A measure of family functioning which
includes the scores of all family members is more likely to
reflect training-related changes than a measure of only one
person's perceptions.

Longer time periods are also recommended for future
replications. Archival data gathered over longer time
periods are less likely to be influenced by cyclical changes
in staff than data collected over relatively short periods
of time. A time period of six months would allow the data
to average and stabilize. Using a larger number of subjects
would also be helpful.

In view of the difficulties encountered with data
collection at follow-up, it is advisable to have examiners
who are part of the research team or to rely on paid

personnel. Employee volunteers tend to experience priority

191

conflicts, which are resolved to the detriment of research
objectives.

More important, however, the implementation of the
training programs has pointed out the need for training
curriculum to be adapted to the learning styles of
individuals from lower social economic backgrounds. The
increasing reliance on women from the lower classes to fill
the needs for foster parents suggests that more attention be
paid to the development of such programs.

Future programs need to include components that pertain
to the resolution of conflict between foster parents and
agency staff. In addition, it is suggested that future
training curricula also involve the participation of agency

staff.

Appendix A

192
APPENDIX A

FOSTER PARENT PROFILE QUESTIONNAIRE

Name
Address

 

 

 

Phone

 

Directions: Please fill out this form as completely as possible

1) Total Years as Foster Parents Total Number of Children Fostered

 

 

 

 

2) Husband's Present Age ___ Wife's Present Age ___
3) Husband's Occupation Wife's Occupation

4) Husband's Religion Wife's Religion

5) Husband's Race Wife's Race

 

 

6) Husband's Education (Check One): Wife's Education (Check One):

__Post Graduate Degree __Post Graduate Degree
__College Degree ;_College Degree
__Some College __Some College
__High School Graduate __High School Graduate
__Some High School ‘__Some High School

Completed 7th grade but less Completed 7th grade but less
__than 9th __than 9th

__Completed less than 7th grade __Completed less than 7th grade

7) Family Income (Check One): __Under $5,000
(Estimated where exact informa-
tion is not available) __3 5,001 to 8.7.000

__8 7,001 to 8 9,000
.__3 9,001 to 312,000
__512,001 to $15,000
__More than 315,000

8)

9)

10)

11)

12)

193

Please indicate the appropriate group for you. (Check one space
in each column)

a)
b)
e)
d)
6)

 

 

Other: please indicate

Husband Wife
Indian
Spanish-American _____
Black _____
White _____

 

Type of neighborhood you live in (Check one)

a)
b)
C)
d)
e)

Farm area

Small town (1,000 or less)
Large town (1,000 to 15,000)
Small city (15,000 to 50,000)

Large city (50,000 and above)

Type of housing (Check one)

a)
b)
e)

Single dwelling
Apartment

Other (please specify)

 

Is your home considered to be . . . . . . (Check one)

a)
b)
C)
d)
e)
How
a)
b)
C)
d)

Regular foster home
Foster family group home

Shelter or emergency home

Pre-adoption home
Other (please specify)

 

many children of your own (natural or/and adopted do you have?
Boys Ages

Girls Ages

Number of boys still living at home

Number of girls still living at home

l

\O

4

13) -30v many foster children do you have presently living at home?

14)

15)
16)

17)

a) Foste
b) .Foste

r boys

r girls

What are the a es of your foster children? What are their races?
What are their reli ions? Ans how long has each been living with
ars)§

you (months, ye

a) Boys:

b) Girls:

Altogether, how

Alto ether, how
you 1.6., past

a) Foster
b) Foster

Which of these
(Indicate one o

a) ____.Under

0) ____ Pre-sc
c) __ Ages 6
d) ____ Teenag

Reli ion Length of time witggygu

so
0)
O
(0

Ass

 

 

 

 

 

 

 

 

 

 

O

 

 

 

 

many years have you been foster parents?

 

many foster foster children have you had live with
and present)?

boys
girls

age ranges have you most enjoyed fostering?
nly.

two years old
hool over two years old
to 12

ers

18)

19)

20)

21)

22)

195

Please indicate which of the following types of children have been
placed in your home: Check any which apply)

a) Mentally retarded

 

b) Physically handicapped

c ) __ Delinquent

d) ______Emotionally disturbed

e) _____ Abused and neglected

f) No experience with these types of children

 

How rewarding have you found the experience of being a foster parent?
(Check one).

3) Generally, very rewarding

b) Generally, moderately rewarding
c) Generally, seldom rewarding

d) Generally, not rewarding at all

In the last year, how many different foster home case workers have
been assigned to your home?

Indicate Number

During the average month, about how many contacts do you have with
the foster home case worker assigned to your home? (Indicate
number of each type.)'

a) By phone

b) ' By visits in your home

0) By visits to the agency's office

d) Other

 

Please indicate your overall satisfaction with the service you
have been receiving from your agency. (Check one.)

a) _____ Generally, very satisfied

b) _____ Usually satisfied

c) _____ Usually dissatisfied

d) _____ Generally, very dissatisfied

Appendix B

196
APPENDIX B

MOOS FAMILY ENVIRONMENT SCALE

The following is a list of statements about families. You
are to decide which of these statements are true of your
family and which are false. If you think the statement is
True or mostly True of your family, circle T (True). If
you think the statement is False or mostly False of your
family, circle F (False). Give us your overall impression
of your family as it is now, not necessarily as it may have
been, but rather the way things are now. The people you
leve with may or may not be blood relatives, but for the
purpose of these questions, consider anyone you live with

to comprise your "family".

TRUE FALSE

1. Family members really help and

support one another. T F
2. We fight a lot in our family T F
3. Family members are rarely ordered

around T F
4. We often seems to be killing

time at home T F
5. Family members rarely become

openly angry T F
6. There are very few rules to

follow in our family T F
7. We put a lot of energy into what

we do at home T F

8. Family members sometimes get so
angry, they throw things T F

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

197

There is one family member who
makes most of the decisions

There is a feeling of togerther-
ness in our family

Family members hardly ever lose
their tempers

There are set ways of doing things
at home

We rarely volunteer when something
has to be done at home

Family members often criticize
each other

There is a strong emphasis on
following rules in our family

Family members really back each
other up

Family members sometimes hit
each other ‘

Everyone has an equal say in
family decisions

There is little group spirit in
our family

If there is a disagreement in
our family, we try hard to smooth
things over and keep the peace

We can do whatever we want to do
in our family

We really get along well with
each other

Family members often try to
one-up or out-do each other

Rules are pretty inflexible in
our household

F1:1

hi:1

25.

26.

27.

198

There is plenty of time and
attention for everyon in our
family

In our family, we believe you
don't ever get anywhere by
raising your voice

You can't get away with much in
our family

Appendix C

Listed below are a number of statements about the experiences of

foster parents.

199
APPENDIX C

FOSTER PARENT PARENTING EFFICACY SCALEW

feel.

3.

9.

10.

1. Strongly Disagree
2. Disagree

3. Agree

4. Strongly Agree

The problems of taking care of a foster child.are
easy to solve once you know your actions affect
the child.

It is difficult to be a foster parent because you
seldom know whether you are doing a good or a bad
job.

I believe I can adequately handle the behavioral
problems presented by foster children under my
care.

I feel I am successful most of the time when I
try to get my foster children to do or not to do
something.

Since I seldom snow what my foster children will
do, I usually feel tense and anxious.

Even though being a foster parent could be
rewarding, I feel frustrated in my efforts to
help my foster children.

There are times when I feel I cannot be helpful
to my foster children.

If I don't consult my caseworker or therapist, I
don't know what to do when my foster children
misbehave.

Since becoming a foster parent, I go to bed
feeling I have not accomplished a whole lot.

'Being a foster parent is harder than I expected

Circle the number that best describes how you

SE

SA

Appendix D

200

APPENDIX D

UTILIZATION, "AND usgrymsgs _SCALE

For each of the questions below, choose the answer that best
describes your experiences with the parenting techniques that

you learned in training.

1. How often have you used the parenting skills taught in training
in interacting with your foster children since training ended?.

A. All of the time

B. Most of the time
0. Some of the time
D. None of the time

2. How often have you found these techniques useful?

A. All of the time
B. Most of the time
C. Some of the time
D. None of the time

Appendix E

201
APPENDIX E
FOSTER PLACEMENT INFORMATION SHEET

Nameaof parent -

Name of examiner -

Directions: List the presenting problems experienced by
each foster child in the foster home at the
time of his placement.

Foster child 1 presenting problem(s)

 

 

Foster child 2 presenting problem(s)

 

 

Foster child 3 presenting problem(s)

 

 

Directions: Record the number of times each of the incidents
listed below occurred in the last 2 months.

1. # of requests for removal of foster child

 

2. # of unrequested removals

 

3. # of foster child(ren) who ran away __

4. # of home contacts

 

5. # of office contacts

 

6. # of telephone contacts

 

Appendix F

202

APPENDIX 3'

Test # 1

NAME

 

DATE

 

Choose the best answer from among the choices given for each
question.

1.

A 3 year old child has been crying and has been irritable
all morning. You notice that the child has quieted down
and is playing with a toy. What is the best thing for
you to do? . -

a.
b.
c.

d.

Ignore the child as he may start crying and fussing
again, if you say anything to him.

Tell him you will take the toy away if he starts
acting again.

Tell him you are pleased with his behavior since he
is playing quietly.

Tell him you are surprised that he is playing quietly
because he has been so bad that morning.

"Behaviors are maintained by their consequences" means:

a.
b.
C.

d.

Consequences are not related to behaviors.
Changing consequences will not change behaviors.
A parent may help a youth change his behavior by
applying positive or negative consequences.
Nothing to me.

Reinforcement is:

a.

b.

A reward that is given after a child or youth has
produced the desired behavior.

Any gifts or presents that a foster parent gives to
a child or youth.

The act of taking something away from a youth.

A reward given before the youth has produced the
desired behaviors.

203

Test # 2

NAME

 

DATE

 

Choose the best answer from among the choices given for each
question.

1.

You obsreve that each time you tell your 15 year old
foster youth it is his turn to do the dishes, both of
you become involved in an argument that usually leaves
you frustrated and angry and the dishes undone. What
is the best thing for you to do?

a. Tell the youth you will have him removed from your
home if he does not do the dishes as everyone else.

b. Tell the caseworker/therapist and let him handle
the problem.

0. Set a specific time period by which the dishes should
be done and give the youth the choice of earning a
positive or a negative consequence depending of his
behavior.

d. Do the dishes yourself and save yourself further
aggravation.

Which of the following is an example of the proper way
to use extinction?

a. You consistently ignore the foster youth whenever he
swears but pay attention to him when he does not swear

b. You ignore the youth completely as if he no longer
existed.

0. You ignore the swearing behavior but tell your
husband to punish the youth.

d. You tell the youth you will wash his mouth with soap.

Which type of consequences is most helpful in teaching a
youth a new behavior:

a. Negative consequences

b. Positive consequences

c. Ignoring

d. Yelling for not doing the right thing.

204

4. A teenage foster child has gotten into your favorite
perfume without permission and accidentally brokewthe
bottle. What is the best thing for you to do?

a. Lecture to her about respect of other people's
property for an hour.

b. Break a favorite belonging of hers to show her how
it feels to have her property ruined.

0. Ground her for one month with no special privileges.

d. Have her clean up the mess and replace the perfume
out of her allowance money.

205

Test # 3

NAME

 

DATE

 

Choose the best answer from among the choices given for each
question.

1.

When the child fulfills the behavioral contract, you
should always:_

a. Rip up the contract.

b. Write another contract.

0. Give the negotiated reward with praise.

d. Ask what reward the child wants at that time.

Jill is a 14 year old girl who her foster mother describes
as lazy and disobedient. In an effort to get Jill to do

her work, her foster.mother developed a contract-with her

in which she would get a sweater she has been wanting if

she can be good for two weeks. Although Jill really wants
the sweater and promised to be good, there was little change
in her behavior. Why didn't this contract work?

a. Jill did not really want the sweater.

b. Jill needed to be punished for each day that she was
not a good girl.

c. Jill needed to know what specific behavior her foster
mother wanted and needed to be rewarded on a daily
basis for each behavior that she changed to please
her foster mother.

d. Jill will always be lazy because her natural mother
was lazy and did not teach Jill how to obey society's
ru 980

If the child misbehaves while under the terms of the
contract, you should always:

a. Rip up the contract and forget about any negotiated
rewards.

b. Write a new contract since this one was not effective.

0. Add a new condition to the behavioral contract.

d. Deliver the consequences in a matter-0f fact tone,
with very little lecturing on scolding.

206

Contingency contracting is a situation in which:

a.

b.

The parent writes a contract to change a child's
behavior. -

A legal and binding contract which is signed by
two persons.

The parent and child decide together on the terms
of behavior change in the contract.

A contract is ripped up and is no longer binding.

Appendix G

207
APPENDIX G
TRAINER'S MANUAL
Session I

Materials Needed
1. Blackboard
2. Chalk

Goals:

1., discuss the need for specialized parenting training (10-
15 minutes).

2. show that behaviors are maitained by their consequences
(20 to 30 minutes).

3. make parents aware of the importance of modeling (5 to 10
minutes).

4. introduce and define the concept of reinforcement (20 to
30 minutes).

5. specify conditions optimizing the effectiveness of re-
inforcers (20 to 30 minutes).

6. explain homework assignment (5 minutes).

Sequence:
1. State the goals verbally and write them on the board. Do
not encourage prolonged discussion of goals at this time.
Tell participants each goal will be dealt with in this session.
Say the following:
Child rearing is not new to most of you. Many of you
have a great deal of experience in that area and are
very skilled parents. Each of you has discovered a way
to raise children and has become good at it. All of
you have foster parenting experiences. What similarities
-do you see between parenting natural and foster
children.
For those foster parents who have never had children, ask
them to describe their experience as foster parents in light

of what they expected when they decided to become foster

208

parents. Allow time for the parents to respond. Acknowledge
parents' contribution by a nod or a paraphrased statement.
List parents' statements on the board. Avoid judgemental
responses such as that's good or bad. Once the participants
have exhausted the subject, ask about differences they have
observed in the two groups of children. What are the simi-
larities and differences between the two parental roles.

List participants' ideas on the board. If the following
points are not brought up in the discussion, introduce them.

a. Many foster children who have been placed require
more time and effort.

b. Many have not had the opportunity to learn more
desirable behaviors.

c. Many foster parents have expectations that may be
too high for their foster children. '

d. Foster parents' authority is limited and parental
role is temporary.

Explain: Foster parents function under the supervision of
the State via the placement agency and as a result, may not
be able to use techniques that have worked with their own

children. i.e. corporral punishment.

2. After it is clear that foster children's needs are
different from those of natural children, state:

There are ways of raising children with problematic
behaviors that have been found helpful. Before we go
any further, we need to establish some common ground.
We need a common language and understanding. What I
mean is some basic understanding of how we acquire
behaviors. If we are to be effective in influencing
behaviors, it would be helpful to know something about
how we come to behave the way we do.

209

Read the following situations

a. Kevin, a 7 year old boy, is screaming his lungs out
in the local supermarket because his mother would
not give him a chocolate bar. Kevin's mother,
distraught over his behavior, gives him the choco-
late bar to stop him.

What is Kevin likely to do next time his mother refuses to
give him a chocolate bar?

b. Pamela, a 14 year old girl, wants her parents to
buy her a pair of designer jeans. Her parents
explained to her that they are unable to buy them
for her as the family budget was in a state of,
crisis. Pamela became furious and told her parents
she wouldn't speak to any of them until they have
bought the jeans. After a week of not talking to
her parents, Pamela's father, in spite of the
mother's objection bought Pamela the desired Jeans.

What would happen to Pamela's behavior toward her parents
when she wants something?

c. Kevin is screaming his lungs out because his father
would not give him a chocolate bar. His father
spanked him to stop him.

Same question as above. Ask clients to share similar ex-
periences with the group. Explain that just as we are able
to influence the behavior of our children, they are equally
able to influence our behavior toward them. Read the
following:

d. Marylou bought a toy for her foster son Bill. At
the sight of the toy, Bill jumps up and hugs Marylou.

How would that influence Marylou's gift giving behavior to
Bill? How might Marylou feel toward Bill?
e. Marjorie buys a toy for foster daughter. When she

gives her the toy, she acts casually, as if to say
so what, what do you want? a star?

210

What effects would the foster daughter's behavior have on
Marjorie's gift giving behaviors with respect to the foster
daughter? Ask parents to share similar experiences. Conclude
by stating that behaviors are maitained by their consequences.
If these are thought to be good or rewarding, behaviors

that precede them will increase. When consequences are seen
as negative or non rewarding, behaviors that precede them

will decrease.

3. The discussion dealt with how behaviors are maintained
and not how they are acquired. Ask group to think about how
they learned the things that they learned. If they don't
bring up modeling, state:

a great many of the behaviors we have in our repertoire
were acquired by observing other people.

Read the following example:
Foster parents scream at foster child for misbehaving.
Later in the week, foster child screams at foster
brother for playing with his toys.

Pause (15 sec.) - discuss example - ask for other similar

experiences. Then state:
Modeling allows us to acquire new behaviors quicker.
Parents should always model the desired behavior as part
of any behavioral intervention. This tells the child
what is expected and provides him with an opportunity
to acquire new skills.

Conclude by making the points that:
1. behaviors are acquired through modeling.
2. consequences associated with acquired behaviors

determine whether or not, and under what conditions
behaviors will be displayed.

211

3. the likelyhood of repeating an exhibited behavior.

4. Move to goal #4.

We have been talking about consequences. Now it is

time to be more specific about what consequences are.
Consequences are behaviors that are produced in response
to another behavior.

Pause (15 sec.)

If we let a child watch television an extra hour at
night for putting out the garbage at 7 o'clock each
evening, watching television an extra hour is a con-
sequence of his behavior, just as putting out the ~
garbage on the agreed time is a consequence of our
behavior of letting him watch television an extra hour.
The two behaviors are reciprocally determined.

 

EL

P

Parents' behavior affects child's; child's behavior
affects parents'.

Pause: discuss

A consequence that increases a behavior is called
positive reinforcement. It is the same as reward.

In effect, a positive reinforcement is a reward. It
has the effect of increasing the probability that the
behavior preceding the reward would be repeated.

There are several types of reinforcement

a. Social reinforcement: verbal praise, smiles, hugs,
touching.

Jim, your 14 year old son, has sheveled the snow
on the sidewalk in front of your house without
being asked. When you run into him later in the
day, you say to him "that was quite a job, Jim,
thanks”.

b. Extra privileges: staying up late, playing with
parents, using the family car.

0. Rewarding activities which the child enjoy, but
seldom has the opportunity to participate in:
movies, skating, etc. or using positive activities
to reward less probable ones.

212

d. Tokens, points, stars, when combined with back up
reinforcers. These may be any of the reinforcers.

Jane, your 16 year old, has cleaned her room
everyday, has done the dishes and the laundry as
promised. For each chore, she received a specified
number of points. At the end of the week, the
points are tallied and she is given an extra 35

in addition to her allowance, so she could buy a
present for her boyfriend.

e. Tangible rewards: snacks, toys, books, allowance, etc.

Pause. Ask for questions, thoughts, ideas, as ways of

assessing participants' understanding. Then state:

Read

Another type of reinforcement that also increase
behaviors preceding it is negative reinforcement.
Negative reinforcement occurs when parents remove a
child from a punishing situation or remove the punishing
consequences before they have been effective in reducing
the problem behavior..

the following:

Sue, a 16 year old, did not do her chores as she agreed
in the contract she negotiated with her parents. Her
privileges to stay out until 11 PM on friday and saturday
were cancelled in accordance with the contract. Sue,
who had promised her boyfriend to go the rock concert
with him, became despondent when her mother told her she
would have to be home by dinner time ( 6PM) on friday
and saturday to make up for the chores she didn't do

by working on some previously agreed task. Sue became
agitated and started screaming "it is not fair, you
can't do this to me, you are just being mean". She

went to her room, came out a few minutes later and
yelled to her foster mother: "if it's blood you want,
you ----- , I'll give you blood, I'll kill myself if

I don't go to the concert t0night". Sue's foster

mother became very concerned and after a few minutes

of silence said to Sue "well I guess nothing is worth
your life, go ahead, but you'll have to do twice as
much work tomorrow".

Pause (15 sec.)

By removing the consequences that were attached to f
failing to do her chores, her foster mother has nega-
tively reinforced Sue's use of threat and not doing

213

her chores. It can be expected that whenever the foster
mother attempts to get Sue to carry out her agreement
she will resort to threat.

Encourage participants to share similar experiences. Explain

that the experiences do not have to be as melodramatic.

This one was used to illustrate a point. Pause (30 sec.)

before going to nest section.

Why is reinforcement important?

8.

Because by reinforcing desirable behaviors, we
diminish the opportunity for undesirable behaviors
to occur. Since desirable and undesirable behaviors
are incompatible, increases in desirable behaviors
automatically leads to a decrease in undesirable
behaviors.

Many parents spend more time interacting with their
children when they exhibit undesirable behaviors
than when they engage in desirable ones. How many
of you have "caught your children being good".

Since parents' attention to undesirable behaviors
may increase their occurrences instead of decreasing
them.

By using reinforcement, parents acquire reinforcing
value for the child.

to use positive reinforcement.

Determine what is rewarding to the child. This can
be done by observing his behavior or by asking him.
To think something will be rewarding for him because
it is for you is to deny the individuality and the
feelings of the child. Moreover, it will not work.

A husband asked his wife what she wanted for her
birthday, she replied that she wanted a bouquet

of roses. The husband shocked, retorted "what can
you do with roses" to himself and decided to buy

a microwave oven. 0n the day of her birthday, he
happily presented her with the gift. To his dismay,
she became sad and started to cry saying "I thought
you cared for my feelings".

Tell the child exactly what he has to do to earn the
reward. In another chapter we will teach you how

214

to be specific in your behavioral requests.

0. Don't set goals too high. Always start slightly
below the child's ability. If a child swears at
least once every hour, positive reinforcement will
have no effect on his behavior if he is to be
rewarded for not swearing for an entire day. Start
by rewarding him for every hour that he does not
swear, then gradually increase the time.

d. Be consistent. Reward each desirable behavior.

e. Reinforce immediately unless a point system is
being used. With small children, always reinforce
immediately.

f. Give appropriate amount of reinforcement.

g. Reinforce'the desired behavior every time when
attempting to increase its occurrence.

h. Always combine verbal messages that emphasize the
child's accomplishment when delivering non verbal
reinforcers.

You are really working hard at not swearing; look
at all the points you have earned.

1. Once desired behavior becomes a habit, gradually
phase out reinforcement.

6. Homework
Before telling parents about the homework, ask them to state
how often on the average they reinforce and reprimand their
foster child in a day. Record the number of reinforcements
and reprimands for each parent. Introduce homework assign-
ment by saying:

there is usually some discrepancy between what we think

we do and what we actually do, there fore, in order

for us to know for sure, it is best that we observe

and record our own behavior.

Hand out the homework sheets (handout #1). Tell parents

they are to put an 0 each time they reinforce their child,

215

and an X each time they reprimand him. Ask parents with
more than one foster children to choose the one they find
most difficult to handle. Tell parents to bring the form
with them to the next session. Suggest that occurrences

of either behavior should be recorded right away.

216

Session II

 

Materials Needed
1. Blackboard
2. Chalk

Goals:

1. review homework and discuss parents' reactions (10
minutes).

2. establish guidelines for selecting a target behavior
(20 minutes).

3. train parents in pinpointing target behaviors (25
minutes).

4. train parents to observe and record target behaviors
(30 minutes).

5. train parents to graph occurrences of target behaviors
(25 minutes).

6. explain homework assignment (10 minutes).

Sequence
1. Review homework and discuss parents' reactions. The
important point here is what was learned from the experience.
Say the following:
In order to influence the behavior of our children, we
need to be aware of how we generally respond to them.
As parents, we function as models to our children. The
more we know about our behavior toward our children,
the more control we can have over the behaviors we
display toward them.
Remind parents that behaviors are maintained by their
consequences and as parents they exert control over a wide
range of reinforcers. Avoid lecturing, the previous thoughts

are suggestions around which to structure the discussion

generated from reviewing the homework assignment.

217

2. Some guidelines for choosing problem behaviors. State
Before we can help a child change his behavior, we need
to know which behavior we would like him to eliminate
and more important which one we would like him to learn.
In order to do that in a way that is fair to the child
and conducive to his growth and development, we need
some guidelines to help us decide which behaviors need
to be eliminated.

Ask for suggestions; pause (15 sec.); discuss suggestions.

Say and write on board: here are some guidelines
a. Behaviors that are destructive of the child and of

others. i.e. aggressive behaviors, fire setting,
stealing.

b. Behaviors that are important for the long range
general functioning of the child. i.e. education,
employment, social interaction.

Point out that foster parents need to refrain from inducing

foster children to change behaviors that are not in agree-

ment with the parents' religious convictions or their view

of what is appropriate behaviors when these in no way detract

from the social/psychological functioning of the child and

present no danger to the child and others. Examples of such
behaviors include: child's refusal to go to church, clothing
preference, hair style and the like. Encourage parents to
verbalize their views. Do not attempt to persuade. Play

the role of facilitator, once the point has been made. The

purpose in bringing that issue is to sensitize foster parents

to the rights of their foster children and establish some
limitations on their use of behavioral technology. Trainer

may want to state that the foster child's stay is not

permanent; he will be returning to his home. To attempt to

218

change everything about the child would probably result in
more problematic behaviors. Such attempt may cause the child
to experience a need to retain a sense of his identity by
defying his foster parents' efforts through increased display
of negative behaviors (i.e. noncompliance, running away).
Point out that this is very important to keep in mind when
dealing with all foster children, especially adolescents.

Be careful not to lecture. Bring the point up only if none
of the participants brought it up. Be reflective in your
responses as opposed to persuasive. Trainer may ask par-
ticipants how would they feel if someone were to try to change
everything about them? What kind of messages would they get?
Is it possible that they might resent such attempt? Is that
in some way similar to telling the child he is all bad and
hasn't got anything good to keep? Do not provide any answers
to these questions, let the parents react to them. Accept

responses 0

3. Choosing and defining a target behavior.

Selecting and defining target behavior is best done in
positive terms. Remind participants that many foster
children carry with them a history of negative experiences.

By defining target behavior in positive terms, the child
is more likely to earn rewards or reinforcements as
opposed to negative consequences. Although it is
probably the negative behaviors that will catch your
attention at first, with a little effort, you can train
yourself to thind of an incompatible behavior that can
replace the negative one.

Pause (15 sec.); for reaction to above statement. Do not

219

rush this point. State:
Always select a behavior that can be counted. If it
can be counted, it can be specific and thus can be
recorded reliably. This is very important because it
allows both parents and children to know exactly what
is expected and to determine objectively when and how
often the target behavior occurs.
Write on board 1
Target behaviors:
a. are best described in positive terms
b. must be specific so that they can be easily counted.
Exercises. .
Go over Burchard and Leitenberg's list of translated negative
behaviors (p. 114-115 of the Vermont Foster Parent Training
Program Trainers Manual).
Ask parents to make a list of problem behaviors. Practice
the following.
a. state each behavior in Specific terms.
b. identify a positive incompatible behavior.
Joey is hyper
Use own examples if available. If not, go with the following.
The word hyper could mean several things. When expressed in

specific terms, it could be interpreted to mean that Joey:

a. has moved from one place to another twenty times in
one hour with no clear purpose.

b. is involved in more than one thing at a time.
c. turns the television dial six times in 15 minutes.
The desired behaviors to replace each of the above might be:

a. move three or four times an hour with clear purpose
in mind.

220

b. involved in one thing at a time.

c. turns the television dial 4 times in 15 minutes.
Note that when hyper is broken down into specific behaviors,
observing and counting are easier to handle and can be more
reliably recorded by both parents. The translation in
positive incompatible behavior helps parents know what
behavior to teach and reinforce and tell the child what
behavior is appropriate or desired. Start with small amount

of desired behavior at first.

4. Recording target behavior

Observing and recording target behavior are important because

they allow parents to determine the effectiveness of their

intervention in an objective manner.
Since there is always some discrepancy between what we
think we do and what we actually do, the recording of
actual behavior provides us with feedback of the amount
of time we are on and off task. When applied to target
behavior of.a child, it provides feedback for the child
about his behavior and provides us information aoout
the actual occurrences of the target behavior.

Pause (15 sec.). Discuss.

Before intervening, parents need to get a count of the

occurrences of the target behavior. Getting such a count

before applying any corrective procedures is called a

baseline. Work on worksheets.

5. Graphing

In addition to counting occurrences, it is also helpful
to graph the counted occurrences. When occurrences are
graphed, they provide a quick and visual reflection of
the amount of progress being made.

221

Seeing this progress is likely to be reinforcing for the child
and the foster parents. Moreover, graphs allow parents to
evaluate success of intervention. Though graphs may appear
to be tedious, the gains far outweight the efforts put into
graphing the occurrences of the desired behaviors. Refer to
Burchard and Leitenberg data work sheets (1978, p 125-130).,
What to do when graphing occurrences of target behavior?
a. Define target behavior in specific and positive terms.
b. Choose an observation period.

1. observe throughout the day for behaviors that
occur infrequently. ' .

2. observe for a limited period for behavior with high
frequency of occurrence (i.e. bet. 3 & 6 PM).

0. Record the number of times the behavior was observed
to occur. -

d. Graph the data to show how often and when the
behavior occurs. (Burchard and Leitenberg, 1978,130)

\

6. Homework

3. Select a negative behavior you would like to help
your foster child change.

b. Define the behavior in specific terms.

0. Identify a positive incompatible behavior that you
would like your child to increase.

d. Define that behavior so it can be counted.
e. Set a period for observation.

f. Count number of times the desired behavior occurs
during the period of observation.

g. Record data in data sheets.

h. Graph occurrences of desired behavior.

222

Session III

Materials Needed

1. Blackboard

2. Chalk

3. Video tape (optional)
Goals

1. review homework and discuss parents' reactions (40
minutes).

2. acquaint parents with methods of decreasing negative
behaviors (60-70 minutes).

3. explain homework assignment (10 minutes).

Sequence

1. Homework review

Review the homework assignment. Ask parents to exchange
assignment with another non related person, to give each
other feedback. Acknowledge that this was a difficult
assignmnet. Indicate, however, that other foster parents
have been able to learn the procedure through practice.

Move about the room, give each person positive reinforcement
for what he did. More time may be spent on any part of the

previous session requiring it.

2. Decreasing negative behaviors

Two sessions ago, we talked about positive reinforcements.
As you remember, these are ways of increasing desirable
behaviors. Since parents are unlikely to prevent the
occurrence of all problem behaviors, there are times

when there will be a need to intervene, to stop the
continuation of undesirable behaviors.

Pause (15 sec.)

223

Sometimes it is better to let a behavior run its
natural course without intervening. That is letting
the child face the natural consequences of his behavior.
Read the following example.
Extinction
Joey, a 6 year old boy, wants to play with his foster
father's tool box. When his foster father refuses to
let him play with the box, Joey throws a temper tantrum.
Foster father leaves the room and let Joey have his
temper tantrum.
Explain that this procedure is known as extinction. It
consists of purposely witholding reinforcers. In order for
it to be effective in reducing problem behavior, parents
must systematically and consistently ignore the undesirable
.behavior. Do not argue nor discuss the behavior. Arguing
may be reinforcing. Point out that the problem will often
get worse before it gets better and both parents must ignore
the behavior. If one parent ignores and the other reinforces
the behavior, this method will not be effective. It is good
to tell the child that the undesirable behavior will be
ignored. Reinforcement of appropriate behavior will increase

the effectiveness of this procedure. Summarize the following:

a. Problem behavior will get worse before it is
discontinued.

b. Parents, other adults and children residing in the
home must ignore the undesirable behavior. If the
behavior is being reinforced, it will be more
difficult to eliminate.

c. Undesired behavior must be consistently ignored.

Discuss participants' reactions and clarify points that are

unclear. Encourage participants to identify and relate

224

experiences in using this method. Evaluate advantages and
disadvantages of this procedure. Make the following points
if they were not brought up.' Encourage the participants to
add to the list.

a. does not require physical contact with the child.
No coercion necessary.

b. may be difficult to get members of the home to
ignore the problem behavior consistently. ie
grandparents.

0. limited to behaviors that are not destructive of
self and others.

Pause (15 sec.)

At other times it is useful to apply some logical
consequences to negative behaviors.

Read the following example
Response Cost
17 year old Jake took the family car for a drive with
his friend without asking his foster parents' permission
as they have agreed after a similar incident 3 months
ago. His foster parents ground him for two consecutive
weekends. '
Explain that this procedure, which is very familiar to parents
of teenage children, is known as response cost. It consists
of penalizing a child for the occurrence of undesired
behavior by removing or withdrawing a specified reinforcer.
Such reinforeers might be privileges, like staying out late,
having friends over, going to the mocies, allowance and the
like. This procedure involves the loss of something prized

to the child. Which reinforcer is forfeited and the degree

of the loss must be agreed on in advance by foster parents

225

and child. Arbitraryness in the use of this technique may

lead to resentment. Summarize the steps

a.

b.

f.

Specify undesired behaviors clearly so that its
occurrence can be readily identified.

Negotiate with the adolescent as to the reinforcer
to be withdrawn and the amount of the loss. i.e.
a parent may fine a teenager a specified amount of
money from his allowance as opposed to the entire
allowance.

Consistently apply the procedure whenever the problem
behavior occurs.

Avoid nagging, scolding or lecturing when adolescent
is penalized.

Set a specific time period for the duration of the
agreement with a set date for evaluation and res
negotiation. Two weeks or a month, depending on
the frequency and severity of behavior.

Renegotiate choice of reinforcer, if the one chosen
loses its effectiveness.

Discuss participants' reactions and clarify points that are

unclear.

Point out that this topic will come up again when

discussing interpersonal problem solving and designing

behavioral intervention. Encourage participants to identify

and relate experiences in using this procedure. Evaluate

advantages and disadvantages as the participants see them.

Make or summarize the following points. Encourage parti-

cipants to add their own views.

the adolescent is involved in the process.
requires parents and adolescent interaction.

may influence adolescent perception of parents as
reasonable.

may increase adolescent's self esteem, giving him
some sense of control.

226

e. adolescent may resent having to give up privileges.

f. parents may feel negotiation is abdicating their
parental responsibility.

g. reinforeers may lose their effectiveness, causing
parents to appear capricious in their behavior.

h. it may be difficult to identify each instances of
the occurrence of undesired behavior.

Pause (30 sec.) then go to the next section
Restitutional Overcorrection
Kevin, an 8 year old boy, becomes angry at his foster
parents and throws his glass of milk on the kitchen
floor. His foster mother tells him that in order to
help him control his temper better, next time he will
have to mop the entire kitchen floor plus the dining
area before he could go play baseball with his friend.
Explain that this procedure is known as restitutional over-
correction. It involves having the child face natural
consequences of his negative behaviors, in addition,
completing some related added tasks. If this procedure is
to be employed, it is good to tell the child about it ahead
of time. Point out that parents should not allow themselves
to get embroiled in a discussion over the fairness of the
procedure.
a. Specify the undesired behavior clearly.

b. tell child of the consequence of negative behaviors.

0. apply restitutional overcorrection after the
behavior has occurred.

Discuss parents' reactions and elaborate on points that are
not clear. Encourage participants to identify their expe-
riences in using this procedure. Evaluate advantages and

disadvantages.

227

a. child may acquire new behaviors.
Pause (30 sec.) then go to the next section.
Positive Practice Overcorrection
Nicole, who is 9 years old, has a tendency of running
into the house and slamming the door behind her.
Despite repeated requests to close the door gently,
Nicole has persisted in that behavior. Upon advice
from a friend, Nicole's foster mother firmly told her
to go back outside and come in again the way she knows
she is supposed to. Nicole went out and came back
into the house at an appr0priate pace. She repeated
this behavior five times each time she came into the
running and slamming the door. Before long Nicole was
observed to slow her pace down before she goes inside
the house.
Explain that this procedure is known as positive practice
overcorrection. It consists of having the child practice a
behavior that is incompatible with the problem behavior
repeatedly. Parents need to model the desired behavior for
their children and be sure that they understand what it is
they have to do. Suggest that only one practice trial be
required of the child after the first onset of the negative
behavior. After the appropriate behavior has been modeled,
parents should tellrthe child that future incidences of the
negative behavior would require that he practices the
modeled behavior for a specified number of times. Number of
practice trials should be aversive enough to discourage the
occurrence of the problem behavior.

a. Tell the child that his behavior is inappropriate,
teach him the desired behavior.

b. Have the child repeat the desired behavior only
once, the first time the procedure is used.

228

c. Inform that future occurrences of the negative
behaviors will require that he practices the modeled
behavior a specified number of times.

d. Apply procedure consistently after each time the
unwanted behavior occurs.

e. Parents may increase the number of required trials,
if the previously mentioned number does not appear
effective in reducing the problem behavior. The
added trials will have to be practiced at the next
occurrence of the problem.

Discuss participants' reactions. Encourage them to share
their experiences in using thiS‘procedure. Evaluate advan-
tages and disadvantages as*parents see them. Summarize or
bring up the following points:

a. help child learn appropriate behavior.

b. has more enduring effects than punishment.

Pause (30 see.)
Time Out
Jennifer, a 7 year old girl, starts screaming and
yelling after being told that she had to eat breakfast
before she could watch television. Her foster mother
tells her to stop screaming on the count of 3 or she
will have to stand in the corner until she has been
quiet for a full three minutes. She did not stop. The
foster mother took her firmly by the hand and took her
to the corner. She remained there until she stops for
the full 3 minutes.
Ask how many of them have used this technique before and how
it has worked for them. Explain that time out consists of
removing the child from the presence of positive reinforcers.
Parents are not always aware of what these reinforcers are.
By removing the child from the immediate environment and

putting him in a dull, non stimulating one, in a calm but

229

firm manner, parents reduce the influence of reinforcers

inherent in the situation, even though they may not be

aware what these are. When no reinforcers have been iden-

tified, the procedure is known as social isolation.

Steps in using T.O.

a.

D.

f.

g.

Tell the child clearly and specifically what
behaviors will bring this consequence prior to its
actual use.

Role play the procedure with the child prior to
using it. Introduce it as a game in which going
to T.O. means losing, doing what is required is
winning.

Select angarea or a room that is dull, no toys, no
television, no games or any other distractions.

Use T.0. for 5 minutes or less of the desired
behavior. Do not remove the child from T.O. when
he is engaged in undesirable behavior.

Let the child out promptly after he has complied

and spent the required time. Do not mention the

problem behavior, scold or nag the child. If the
child pouts or sulks, it is ok. That may be his

way of expressing his anger. Act as casually as

possible and go about your usual routine.

Do not hug the child or socially reinforce him
right after coming out of T.0. You can plan some
tasks for him, in which he is likely to succeed and
when he does, reinforce him for successfully comple-
ting that task. In the above example, if after
coming from T.O. Jennifer eats her breakfast, you
can reinforce that behavior. For compliance
training, do have the child comply with original
command after leaving T.O.

Use T.O. consistently whenever the problem behavior
occurs or if you tell the child you are going to
use it.

Use T.O. immediatley after the occurrence of the
problem behavior.

230

i. Be calm and authoritative in administering T.O.
Take the child firmly by the hand to the designated
corner or room.

3. Do not argue or plead with the child. If the child
physically resists being taken, than another
technique should be employed. Very early in
negative behavior chain, give choice between T.O.
and some other consequences.

After going through these steps, ask the parents to role
play the parts of parent and child. Divide them into groups
of players and observers. Each participant should play both
roles (parent and observer). Parents should role play, pre,
during and after T.O. behaviors. If video recorder is
available, record role play and use for discussion. After
the role playing exercises, discuss the participants'
reactions. Point out the advantages and disadvantages if’
they do not surface. In any event, bring up or summarize

the following

a. More effective with children who have desired to
please parents.

b. May be difficult to find a dull place in the home.

c. Many of the foster children the parents work with,
may require physical coercion in order to comply.

d. May not work with parents who do not exert authority
over their children.

e. Useful only with younger children, usually 10 years
old and younger.

Some general considerations when deciding to use punishment.

a. Since punishment is experienced as unpleasant, its
use may lead to:

1. avoidance of parents

2. acquisition of other undesirable behaviors to
avoid parents

3. Homework

3.
4.

To
to
1.

2.

3.

4.

231

a. running away

b. lying

inhibition of desired behavior
parental dependence on punishment

minimize the occurrence of the above parents need
be

very clear and very specific about the behavior
they want changed and the one that is acceptable

use punishment very infrequently. Many foster
children have had prior contacts with aversive
adults.

use punishment in conjuction with reinforcement
for desired behavior

be consistent as much as possible

Start using any one of these methods whenever you think they

are appropriate instead of your usual disciplinary techniques.

Observe.

Record number of times each technique has been used

and observe what happens. Remember to:

1.
2.

Pay attention to your child when he is good.

Begin to get into the habit of paying attention to
non problematic behaviors, although these may not
be extraordinary. What may seem simple for you,
may be difficult for the child.

232

Session IV*

 

Materials Needed

1. Blackboard

2. Chalk

3. ‘Video Camera (optional)
Goals:

1. review homework and discuss parents' reactions (30- 40
minutes).

2. familiarize participants with contingency contracting
(45 minutes).

3. rehearse the steps in contingency contracting (30
minutes).

4. homework (5 minutes).

Sequence
1. Homework review
Ask parents to share their experience in using the techniques
presented last week. Clarify misconceptions and summarize
the materials covered thus far. Suggest that the best way
to handle problem behaviors is to prevent their occurrences.
Explain the need to pay attention to positive behaviors and
to try to ignore negative ones. Emphasize the importance of
catching your foster child being good.
2. Contingency contracting
State:

It is important to get the child involved and to accept

some responsibility for changing his behaviors. This
can be done through contingency contracting.

233

Explain that

When we freely commit ourselves, we are more likely to
respect our commitment.

Define contingency contracting

A situation in which the parent and child decide
together on the terms for behavior change. This
includes the behavior to be changed and the con-
sequences for both desirable and undesirable behaviors.

Pause (15 sec.). State:

The following-procedures must be-followed in order for the
method to be effective. '

1.

This "contract" is drawn up and agreed upon by both the
child and parent(s) through active negotiation. Specific
positive behaviors are outlined with Specific positive
reinforeers available if the child engages in those
positive behaviors. Specific negative behaviors are

also specified along with specific negative consequences
if the child engages in those negative behaviors.

It is critical during these negotiations that the child
be allowed to provide genuine input into the process of
deciding what behaviors to change, how to change these
behaviors, and what should be the reinforeers and
punishers.

It is often useful to write up this "contract" with all
the behaviors and consequences clearly specified. All
parties should then sign this ”contract". Such a
procedure makes it more likely that everyone will feel
invested in fulfilling the terms of this negotiated
"contract".

When doing behavioral contracting, it is important to
begin with a contract the child is likely to complete.
Then, one can gradually increase the performance
standards. Collection of baseline information is
critical for determining where to start (i.e. begin the
program just a little above the child's current level
of functioning).

It is important to incorporate chartin into this
project; both to determine where the program should
begin, and to monitor progress.

234

When your child fulfills the positive behaviors specified
in the contract, it is important to follow through with
the negotiated rewards combined with social attention

and praise.

When your child misbehaves under the terms of the
contract, the consequences should be delivered in a
matter-of-fact tone, with little lecturing or scolding.

Before going into the next section, elicit parents' responses.

Clarify and explain as the need may be.

Steps in contingency contracting:

1.

Sit down with your child(ren), and state as specifically
as possible the behaviors which are concerning you.

State that you feel that if that behavior were to change,
you believe things would be more enjoyable in the home.

Elicit your child's view of the problem behavior. It
may be that your child simply does not know that his or
her behavior is problematic for you. In such a case,
your providing information about the behavior may be
sufficient to change it. However, in many cases this
will not be enough, and you will have to go through the
rest of the contracting steps.

After you have solicited your child's view of the
problem behavior, state that you have an idea that can
help change that behavior so that things are more enjoy-
able in the home, and that this idea is called
"contracting".

Identify as specifically as possible, the appropriate
behavior you want to substitute for the problem behavior
(e.g. "instead of leaving your dishes in the living
room, I would like you to place them in the sink in the
kitchen".). Then ask the child if that behavior is
possible for him to perform. Sometimes a child may have
a goo reason why a more appropriate behavior is not
possible for him to perform. This is a good time to
find out their reasoning.

State that the idea behind "contracting" is that you and
your child will to ether come up with a plan regarding
what should happen if your child does engage in the more
appr0priate behavior, as well as what should happen if
your child engages in the problem behavior. Then begin
negotiating the terms of the contract. You should try

235

to negotiate both daily (smaller) rewards and weekly
(larger) rewards. Often rewards will be in the form of
a special activity at the end of the week if a certain
number of daily rewards are earned during that week.
Daily punishments for misbehavior should be relatively
mild (e.g. no desert that evening); with 39 use of
weekly punishments. The idea is that your child should
be more influenced by trying to earn positive rewards,
then by avoiding punishments.

To determine an appropriate weekly goal (e.g. 4 evenings
of brushing teeth = going to the movies on friday),
charting can be used for one week to determine how often
the behavior naturally occurs. If you already have a
good idea of how often the appropriate behavior occurs,
you can skip this step. However, be sure to proceed in
small steps. Do not require perfect performance the
first week.

Gradually, you can increase the demands of the contract.
However,.this should be done through further negotiation.
Such increases in demands should occur when it is clear
your child is easily fulfilling the contract.

An example: A family decides that Johnny's fighting
with his brother is a problem. Consequently, the parents
sit down with Johnny and his brother and negotiate that
for every day that Johnny and his brother do not fight
with each other, they will earn a star on a star chart.
Based upon a baseline recording week, it was determined
that if at the and of the week, they have earned five
stars, then they will be able to purchase a special
treat (this treat could be a favorite dessert, a family
outing, a small toy, etc.). However, every day that
they d2 fight with each other will result in their not
being able to watch TV for the rest of the evening.

The important aspects of this example are:

1. The parents may decide the sibling fighting is a
problem, but the specifics of how to go about
changing this behavior are negotiated.

2. Charting remains an important part of the behavior
change plan. For example, it was only through
collecting one week of baseline that the family
knew to require 5 stars in order to get the weekly
reward. In other families more or less stars may
be appropriate to require for the weekly reward,
depending upon how frequently the children fight

236

with each other (ask how the parents might use such
a baseline to decide this!).

3. Consequences for both misbehavior and appropriate
behavior are clearIy specified.

3. Rehearse

Ask parents to divide into small groups and role play the
procedure. Discuss.

4. Homework

Ask parents to work out a contract to change the target
behavior selected earlier. They are to bring the contract

to class next week.

*this session is adapted from the Child Behavior Project,
Department of Psychology, Michigan State University.

237

Session V (Control)

Materials Needed

1.

Tape recorder

2. Video camera (optional)

3. Cassette tape

Goals:

1. review homework and discuss parents' reactions (30
minutes)

2. discuss rationale for inter ersonal conflict resolution
skills training (15 minutes).

3. acquaint foster parents with the conditions that
facilitate resolution of interpersonal conflict (30
minutes).

4. practice conflict resolution procedures (40 minutes).

5. explain homework assignment (5 minutes).

Session V (Experimental)

Materials Needed

1.

Tape recorder

2. Video camera (optional)

3. Cassette tape

Goals:

1. review homework and discuss parents' reactions (10
minutes).

2. discuss rationale for inter ersonal conflict resolution
skills training ( 5 minutes).

3. acquaint foster parents with the conditions that
facilitate resolution of interpersonal conflict (10
minutes).

4. practice conflict resolution procedures (30 minutes).

238

5. acquaint foster parents with self management techniques
(60 minutes).

6. explain homework assignment (5 minutes).

Sequence
1. Homework review
Go over last week's homework. Discuss parents' reactions
to the task. Encourage those who weren't successful to try
again without forcing the child to talk. Indicate that
today's main topic deals with conflict in interpersonal
relations. Such confilcts are usually very difficult to
resolve, especially when the behaviors of either party do
not violate some specific social norms or contrary to laws.
Ask parents to share examples of interpersonal conflicts
they have experienced either with their spouse, their
children or agency representative.
We mentioned the word conflict earlier in our discussion.
Though we all know when we are having a conflict, many
of us don't usually think of what a conflict really
entails. When we can break things down to their simplest
level, it helps us to get a better perspective of what
it is we are actually concerned with. To help us do
that, we are going to define what an interpersonal
conflict is and based on that definition, we will talk
about ways to reduce conflict in our relationship.

Using Patterson's definition:

Conflict is a situation in which one person demands
immediate behavior change and the other refuses.

What does it take to have a conflict?
a. at least two people
b. a demand

c. non compliance

239

A third aspect of conflict that is not included in the,
definition, is that over time the demands and refusals are
expressed in negative and painful statements. For instance:
George wants to talk to his son Bill; Bill does not want to
talk to his father, George. There is a conflict, but perhaps
no angry feelings. As George however, insists on talking

and Bill insists on not talking, one or both may become
irritated with the other and start making coercive statements
to each other.

George: Who the h--- you think you are? (angry)

Bill: Not who I think I am. Who the h--- you think you
are? (sarcastically)

Give a few of the examples below.
Some conflicts parents experience:

1. Sandra, 12, wants her parents to give her permission to
start dating.
Her parents, concerned that she might get hurt emotionally,
refuse to give her permission.

2. Father wants son Ken, 17, to remove the purple dye out
of his hair. Ken refuses, saying he has a right to do
what he wants with his hair.

3. Carol, 16, wants her mother to let her choose her own
friends.
Mother is worried that she is going out with the wrong
crowd.

4. Eric, 18, wants his parents to allow him to smoke pot
in his bedroom.
Parents sternly refuse.

5. Foster mother wants natural mother to stop visiting his
son while he is at her house.
Agency representative says this is not possible.

240

Foster parents believe in spanking and other corporal
punishment.
Agency representative does not condone such practices.

Sherry, 16, wants to drink beer, saying her girlfriend's
parents let her do it. Why can't her foster parents?
Her parents refuse. They believe drinking is a sin.

Jan, 14, invites boys over to her house when her parents
are not home.

Jan's foster parents want her to have boys over only
when one of them is home.

Ask parents to share some of their conflicts with the group.

Bring up the following points after the discussion.

2.

a. conflicts are a normal part of everyday life.

b. conflicts do not create bad feelings.

c. it is what we say when we are in conflict situations,
that create the bad feelings which then perpetuate
the conflict.

d. what we say in a conflict situation may contribute

to good feeling or diminish the intensity of bad
feelings, thereby helps resolve conflict.

Rationale

Why should we learn to handle conflict? This section is best

handled as a group discussion. Ask parents to state the

advantages of learning conflict resolution skills. Bring up

the points listed below, if no one in the group mentions them.

If they were brought up, state as a summary at the conclusion

of the discussion.

a. requests for changes in behavior are an integral
part of living together as a family.

b. escalation of conflict can be prevented.

c. accumulation of grievances may affect the way we
feel toward each other.

g.

241

there aren't always clear rules about what is right
or wrong or who is right and who is wrong. Many of
the conflicts we have, come from different sets of
values about the problem behavior.

the older the person, the less control we have over
reinforeers and punishers. Older persons have access
to a wide range of reinforcers and punishers, com-
peting for influence over their behavior. For
example, peer acceptance for adolescent may be more
reinforcing than parents' controlled reinforcers.

punishment is more effective when it is seen as a
choice and not as an imposition. When it is per-
ceived as an imposition, it is likely to lead to
resentment and failure to accept responsibility for
problem behaviors by adolescents.

allowing adolescents' real input into conflict
resolution fosters increased self responsibility.

Elaborate on the last point.

Adolescence is a time when children are trying to
assert their independence and define their identity.

Ask parents to share some experiences of their adolescent

years.

3. Resolving conflicts

Now

that we have agreed on what conflicts are, we focus

next on what to do when they occur, to prevent them
from escalating.

Negotiation.

1.

a.

Planning: allows us to schedule negotiation under
conditions that would facilitate resolution.

arrange a time and place for a discussion as soon
as the problem arises. The longer you wait the
more difficult it is to discuss the problem
without recriminations.

agree on a time when the persons involved are

likely to be relaxed. If any of the party is tense,
or experiencing other pressures, he is less likely
to be cooperative.

242

0. agree on a time and place when the persons involved
can be free from interruptions and distractions.

d. sessions should be brief 10 to 20 minutes.
e. discuss one problem per session.

f. agree on a specific time and place for further
sessions.

2. Procedures: these allow discussions to remain
focussed and reduce the likelyhood of coercive
behaviors and punishing statements. They only
serve to didetrack discussion of the problem.

a. victim is always ri ht. (victim is the person
who has a complaint

b. victim must follow the following rules
1. state what the other person did.
2. state how it made the victim feel.
3. do not guess the other person's intentions.
4. do not make attacking statements.

0. after victim has finished, listeners must para-
phrase in one or two sentences what the victim
has said.

d. change role - listener becomes the victim. This
is particularly important when the conflict in-
volves parents and children. If the adolescent
child does not get to bring out his side of the
problem, it is unlikely he will respect any
resolution of the conflict.

e. the listener paraphrases what the victim has said.

Read the examples below and ask parents to
1. identify the request implicit in each statement.

2. say how they would feel if they were approached in
that manner.

3. restate the request.

243

3. Laura turns toward her husband Bill, who has been
watching television since coming from work, and
says "you never do anything; you are always in
front of that stupid box; I am sick and tired of
doing everything in this house".

b. Michael looks around the living room of their
three bedroom house and says to his wife "what
kind of a wife are you supposed to be? Look at
this house, it gives me the creeps".

Summarize that in both examples, each speaker made coercive
punishing and accusatory statements. Such statements are
likely to lead to refusal to comply with the implicit request
and thus contribute to development of conflicts that could
have been avoided. An other example of discussion session
between Laura and Bill. Both are relaxed. The children are
in bed. The house is quiet. The television is off and they
had set this time to talk. Although they have many other
problems, they have chosen to talk about one thing tonight:
Bill's watching television after work. Ask for two volunteers
to roleplay the example below.
Bill: You say that I never do anything and that I am
always in front of that stupid box. When you say
these things to me, I feel hurt and uncared for.

Laura: You feel hurt and uncared for when I say you don't
do anything and are always in front of the set.

Bill: Yeap! (acknowledge that's what he said)

Laura: When I see you sitting in front of the T.V. and I
have to prepare dinner after working all day, I get
resentful and angry.

Bill: You get resentful and angry when you feel you have
to cook and see me watch television.

Participants may feel uncomfortable. Be persistent and

encouraging.

244

4. Practice
Divide parents into groups of observers and players and
practice the above steps. Ask parents to suggest conflict

situations.

5. Self management (Experimental)
Ask parents to reconvene as a large group. Help them
explore their reactions to the exercise. What was it like
for them? What stood out? What are-their impressions?
Summarize the discussion. Indicate that conflict tend to
trigger angry thoughts and feelings which may cause us to
behave in ways that we are likely to regret at a later time.
Pause (15 secs.)
Because of this we thought it would be useful to acquaint
you with a method that can help you stay in control of
your feelings
Pause (15 secs.). Allow parents time to respond. Some may
disagree, saying that we cannot help the way we feel.
Indicate that while many people feel that way, many others
have learned to control their feelings. Give the following
example:
Georgia, a 15 year old foster youth, verbally abuses
her foster mother in front of their visiting older
children in response to being asked to turn down her
portable cassette player. The foster parent turns
toward Georgia sisibly angry and tried to get the
casette player from her. The situation escalates into
a physical struggle between Georgia and the foster
parent's children.

After giving the example, ask the parents to put themselves

in the place of Georgia's foster mother. How would they

245

respond? List the responses on the board. After doing this
indicate that behaviors are usually preceded by feelings.
Ask what would they feel if they were Georgia's parents.
List the feelings on the board next to the behaviors mentioned
earlier. Indicate that our feelings are usually set off by
what we think of a situation or the way we see the situation.
Help them articulate how they would perceive the situation.
List the responses on the board next to the feelings mentioned
earlier. Point to the relationship between perceptions
(conclusions we reach), feelings and behavior. State:
The conclusions we reach about an event arouse feelings
which influence what we do. If we can change our
conclusions, our feelings will be different and so will
our behavior.
Pause (30 sec.). Allow the parents time to ponder about
what was said. Go back to the board and use the listed
responses to show that parents who do not perceive the
situation the same way, indicate they would behave differently
and vice-verse. If the responses were all in the same
direction, then give the following examples:
Foster parent 1:
"Georgia is probably feeling ignored and left out today.
She is angry with me because of this. Playing the radio
that way is probably one way to get my attention. I don't
like it when she behaves that way, I am going to have to
help her learn other ways to get my attention (or to express
anger toward me). My children are probably feeling the need
to protect me, right now there is no need to let things get
out of hand".

Foster parent directs her children away from the scene and

approaches Georgia calmly to discuss her behavior.

246

Foster parent 2

"No one disrespects me in my own house and get away with it.

She insulted me. She made me look bad in front of my

children. I'll be d--- if she doesn't do what I tell her to

do. This is my house. This child has no sense. She has

no consideration for other people's feelings...".

Foster parent menacingly tell Georgia to turn the volume

down. Georgia continue to refuse. Foster parent tries to

grab the radio.

After reading the second example, ask:
What conclusions have the first parent reached about
Georgia's behavior?' What conclusions have parent #2
reached about the same behavior?

After discussing the parents' responses, point out that if

parent #2 could learn to come to the same conclusions as

parent #1, her behavior would also be different. State:
Many of you are probably wondering how could this be
done. We respond so fast that we don't realize what
happened until after it occurred. To help prevent
this we need to look at the attitudes and beliefs we
hold toward foster children. These are important
because they generate self talk, which in turn
influence the way we feel. We may not realize this,
but we all talk to ourselves, we reason things out,
we question, we rationalize, we reassure and justify
our behaviors. We usually do that in the form of
thinking.

Go back to the listed responses given earlier and help the

group become aware of the beliefs underlying their perceptions.

Once this is done, emphasize the role of self talk in

maintaining our beliefs and influencing the way we think,

feel and behave. Conclude by saying that the meaning we

attribute to a youth's behavior determines how we will feel

about his behavior and what we will do in interacting with

the youth.

247

One way to change our beliefs is to change

what we say to ourselves.

5.
A.

Homework: paraphrase instruction for homework.

In order to help put into practice the procedures
outlined today, we would like you to do and tape
record the exercises below.

1.

First is to teach your child the procedures.
Select a quiet and relaxed time and place. After
dinner may be a good time. Select a conflict
situation and roleplay the procedures learned
today with your spouse or another person, with
your child (or children) observing. Then ask one
child to think of something he has not been
pleased with that he would like either you or your
spouse to change. Teach him to:

1.. state the problem in specific terms.

2. express feelings related to the problem.
(Reinforce his behavior socially by telling
him how hood he is doing, ruffling his hair,
winking at him and the like).

Practice the first two steps with different children or with

the same child using different conflicts before going to the

next step.

B.

A minimum of 3 trials is suggested for each child.

Begin to pay attention to the reasons you give for
your behaviors. Choose two incidents involving you
and one of your foster children. For each incident,
describe the following.

What happened during the incident or event?
What did you think about it?

What did you say to yourself?

What did you do?

Select one incident that you considered unpleasant and one

that you found pleasant.

248

Session VI (Control)
Materials Needed

1. Tape recorder
Goals:

1. review homework and discuss parents' reactions (40
minutes).

2. acquaint foster parents with interpersonal behavior
exchange (45 minutes).

3. practice behavior exchange (30 minutes).

4. discuss homework (5 minutes).

Session VI (Experimental)
Materials Needed

1. Blackboard

2. Tape recorder

Goals:

1. review homework and discuss parents' reactions (15
minutes).

2. distinguish between helpful and unhelpful self talk
(20 minutes).

3. acquaint foster parents with interpersonal behavior
exchange (10 minutes).

4. practice behavior exchange (30 minutes).
5. acquaint parents with relaxation techniques (40 minutes).

6. explain homework (5 minutes).

Sequence
1. Homework review
Ask parents to form small groups of two or three. Have them

take turn playing the recorded session and giving each other

249

feedback. Move around and spend about 5 minutes with each
group. Do not give feedback directly. Rather phrase comments
in the form of questions addressed to the entire group.

After each person has had a chance to play his tape, ask
parents to form a single group and discuss thoughts and
reactions to homework and exercises. Bring up the point

that the best way to break up ineffective conflict resolution
pattern is to practice new methods. Summarize the main

theme of the group discussion.

2. Self talk (experimental)

Ask parents to relate what they have learned about the
instructions that they give themselves. Facilitate parents'
sharing of their observation. After this is completed,
point out that self talk can either be helpful or unhelpful.
Read the following example.

Jake and his foster son, Brian, are trying to resolve
a conflict about curfew times. Brian behaves as if he
doesn't want to hear what Jake has to say, saying he
has nothing to say to Jake.

Unhelpful self talk

I knew it wouldn't work. This kid is more stubborn
than a mule. He is purposely ignoring me. He always
does that to me. He is doing this on purpose. I don't
care what he says, he is going to hear me out today.
Who the h--- does he think he is. Nobody ignores me

in my house, I knew this kid_wasnjt;going to respond.
Helpful self talk

I wonder why he is doing this now. This is new for me.
May be it is for him also. Perhaps this is not the
right time. I wish he would respond. I get angry when
he tries to ignore me. The therapist did say that Brian
is afraid of change. He did say that his behavior

250

sometimes trigger angry reaction in adults. I am going
to have to work at this. It is harder than I thought.
3. Behavior exchange
Since communication alone may not lead to changes in behavior,
it is helpful to get the parties to commit themselves to
such changes through an exchange of behaviors. Read and
discuss the following steps.

a. make a list of specific changes in the behavior of
the other person that you would like him to change.

b. make a list of the behavior of the other person
that you like.

0. exchange items on list. For each one that you get,
you give something in return.

d. establish consequences for the violation or the
agreement. Each person suggest some consequences
for his violating of the agreement. Consequences
should be negotiated when there is disagreement
regarding their practicality, appropriateness and
effectiveness. Parents may offer children a list
of consequences from which to choose. Allow child
to come up with at least one consequence of his
own choosing.

e. record all exchanges and consequences, so that
there will not be any future disagreement as to
who agreed to what. Set a specific date for
renegotiation.

f. apply consequences consistently.

4. Practice

Using a partner or another trainer, model the entire sequence.
Ask parents to form small groups and practice the above

steps on one of the conflicts they had mentioned previously.

Move about the room and spend 10 minutes in each group,

251

explaining and clarifying the task.

5. Relaxation

As you know conflicts sometimes engender unpleasant and angry
feelings, which make it difficult for us to remain calm and
in control. To help you reduce the effects of such feelings
on your behaviors, we would like to teach you a few tech-
niques to help you relax.

Pause (30 sec.). Ask if anyone had heard of relaxation
techniques or if anyone had had the opportunity to use any
such techniques. Do not rush to the exercises. Take time
to answer participants' questions. After this is done, do
the following: ask participants to seat as comfortably as
they can. Go over progressive muscle relaxation (Rim &

Masters, 1979).

6. Homework

A. Same as homework for the last session. Parents should
model and practice the exchange sequence with their
children. Ask them to tape record one session for
discussion next week.

B. Ask parents to practice the relaxation techniques
taught at least once a day and to record their
impressions.

252

Session VII (Control)
Materials Needed 3
1. Tape recorders

2. Blackboard

Goals:

1. review homework and discuss parents' reactions (45
minutes).

2. apply procedures taught earlier to problem behavior
(30 minutes).

3. develop a behavioral intervention program (45 minutes).

Session VII (Experimental)
Materials Needed

1. Tape recorders

2. Blackboard

Goals:

1. review homework and discuss parents' reactions (10
minutes).

2. apply procedures taught earlier to problem behavior
(60 minutes).

3. develop a behavioral intervention program (40 - 50
minutes). '

Sequence

1. Hovework review and small group feedback.

Ask parents to form small groups, listen to each other's

tape and give each other feedback. Spend 5 minutes in each

small group to model positive criticism. Ask each group to

select the best tape of their respective group. After each

group member has had the opportunity to receive reaction to

253

his tape, reconvene into a single group. Play the selected
tape and inquire as to what made them exemplary models.
Before going on to another section, make the point that
these procedures are not to be put away until there is some
major crisis. In order for parents to comfortable with them
and skilled in their use, they are to be employed as often
as possible. This will prevent the development of coercive

interactions and eliminate entrenched conflicts.

2. Applying learned procedures to problem behaviors.
In the previous sessions we discussed a number of
procedures that have been found helpful in elimination
problemabehaviors. Now in order to help us integrate
these procedures into skilled interventions, we are
going to apply them to selected behavior problems.
Ask parents to suggest problem behaviors they would like to
work on. Before focussing on specific behavior problems,
ask parents to state the procedures that should be followed
in every behavioral interventions. Write these on the
board. Once these have been reviewed, proceed to work on
specific behavior problems. Go through this section at
parents' pace. I
A. General procedures.
1. define undesirable behavior in specific terms.
2. pinpoint an incompatible, positive, desired behavior.
i.e. playing vs. fighting; compliance vs. non
compliance; absence of negative behavior vs.
presence of negative behavior.
3. note conditions surrounding occurrences of desired

behavior. i.e. child complies when foster father
is around, does not comply when around friends;

254

child swears after visit from natural parents or
therapy sessions. It is important to note the
conditions surrounding the occurrence of desired
and undesired behaviors, especially with foster
children. Some of these children's problem
behaviors may be cyclical, that is they have been
conditioned to certain events in the child's life.
Such events include placement into a new home,
visits from or to natural parents, therapy sessions,
change in therapist, new social worker, termination
and the introduction of another foster child
within the family unit. These events may be
experienced as stressful by the child, causing him
to respond in impulsive manner with whatever
behaviors have become conditioned with any of these
events. Knowledge of conditions preceeding problem
behaviors help parents
a. identify conditions associated with high and
low probability of the occurrences of desired
and undesired behavior.
b. determine what maintains these behaviors after
they have been exhibited.
0. structure task that will facilitate success and
not failure.

select an observation period. Depending on the
behavior, the observation may be spread over a day
or may be restricted to portion of a day. If the
behavior occurs frequently, observe for only a part
of a day. i.e. two hours. If behavior occurs
infrequently, it needs to be monitored for a longer
period of time. Observe behavior for 3 to 5 days.

record and graph occurrences of desired behaviors.
This will allow parents to determine the effectiveness
of intervention over time.

talk with child about problem behavior. Find out
if the child is aware of more socially acceptable
behaviors or is inhibited in his use of behaviors.
For children 12 and older, initiate conflict
resolution procedures. Regardless of age, give the
child a choice: he can choose to learn new ways of
behaving, be helped to use other behavior he has

in his repertoire that he is not using, or face
punishing consequences of his current behavior.

find our what can be used as a tangible reinforcer
to back up social reinforcement. Best ways to
identify reinforcers are to ask the child or to
observe his reactions to things. Some possible

10.

11.

12.

13.

14.

15.

256

non social reinforcers are money, special meal,
outings with parents, new clothes, cookies, etc..
Choice of reinforcers must depend on the prefe-
rence of the child and the practicality of that
preference.

teach the child the behavior he should use in the
future or tell him which one he should use if he
already has it in his behavioral repertoire. Be
specific and start by reinforcing small components
of the desired behavior.

tell the child about the procedure. Define your
role and define his. Inform when and why he will
be given a reward or be punished. Reward should be
given at rate slightly higher than the rate of the
desired behavior at first. Post record sheet in

a location where the child can see it. It is good
to ask the child where to post the record or graph
sheet.

gradually increase the time required to earn a
reward after the desired behavior has been learned.

always pair reward with verbal statements, empha-
sizing the child's control over his behavior. iie.
"look at all the points you earned, you must take
doing your work seriously", may be said to a child
who has done all his household chores.

deliver punishment immediately after problem
behavior occurs, when it is part of the program.
If it is not, ignore occurrences of undesired
behaviors.

fade out tangible reinforcers after the desired
behavior has been acquired, that is the child
behave in the desired way most of the time, usually
at a level tolerable to the parents. Inform the
child that sometimes he will get rewarded for
engaging in the behavior and at other times he

will not.

always pair delivery of tangible reinforcers with
social reinforcement.

if reinforcer or punishment is not effective,

review intervention program. Frequent reason for

program failure:

a. initial steps to get reinforcement may be too
large.

257

b. reinforcer might be too week.
0. mixing nagging, criticism and punishment with
reinforcement.
B. Rehearse relaxation techniques
1. deep breathing
2. progressive muscle relaxation
C. Brainstorming
You may use the examples below or preferably, those suggested
by parents. Read the problem and ask parents what methods
they would use in dealing with the behavior. Inform them
that the methods chosen may be different than the ones
presented. Make a list of the suggestions on the board and

ask parents to evaluate the consequences of each procedure.

a. Jane, a 13 year old, swears whenever she gets into an
argument.

Methods.

1. Response cost: withdrawing a reinforcer, i.e. no
TV-watching, no dessert, reduction in allowance
and the like.

2. Time out or social isolation: sending child to her
room.

3. Ignore.

4. Reinforce absence of swearing words: 1 pt. to be
converted into money for each 15 minutes he does
not swear.

5. Reinforce incompatible behavior: use of more
acceptable words to express her frustration (parents
must teach acceptable expressions).

6. Have her rephrase what she says each time a swear
word is used.

7. Lecture.

258

8. Wash mouth with soap.

9. Threaten to have her removed.
10. Slap her.

11. Yell and scream at her.

Discussion: review list and discuss the pros and cons of

each procedure. Evaluate their conSequences as the overall

foster parent/foster child relationship and possible effect
on the problem behavior. Solutions 1-6 are preferred over

the other 5.

b. Joey, a 15 year old, bullies the children of his foster
parents. He punches, kicks and pulls their hair when
they refuse to do what he says.

Methods.

1. Response cost. (If parents suggest this method,
ask them to state what privileges will be taken
away. Privileges should not be confused with
necessities.)

2. Social isolation.

3. Reinforce absence of any of the undesired behaviors.

4. Reinforce incompatible behaviors: verbal persuation,
moving away.

. Lecture.
. Threaten to have him removed.
. Beat with belt or paddle.

5

6

7

8. Ignore.

9. Yell or scream.
O

. Hold him and let the other children hit him in
return.

Discussion: same as before. Solutions 1-4 are preferred.

259

After the above examples have been discussed, move on to

next section.

3. Putting program together.
To help you put into practice the parenting and self-
management techniques you learned earlier, we are going
to practice putting a behavior program together.

Go over the example below with parents, asking them to

suggest how they would go about developing a behavior and

self management program.

1. Joey, 13 years old, bullies other children in his foster
home. '”

Develop the program with parents as a large group. Explain
that the purpose of this exercise is to help parents inte-
grate the techniques into a program.
2. Donna, 13 years old, consistently forgets to make her
bed in the morning, leaves her clothes sprawling all
over her room, and pospones doing the dishes until some-
one else does them.
ASk parents to do the same as for example #1.
After going through the examples, ask parents to form small
groups. Using the data they had gathered on the problem
behavior they had selected, they are to design a behavior
intervention program. Tell parents you will not be spending
any time with any of the groups. They should attempt to
answer each other's questions first before asking you to
intervene. Encourage parents to provide answer to each other's
questions. Each parent must design a behavior intervention plan.

Ask parents to implement their plan. Ask them to bring their

contract to class next week.

260

Session VIII
Materials Needed
1. Balckboard
2. Chalk

Goals:

1. review homework and discuss parents' reactions (60
minutes).

2. practice problem solving (50 minutes).

3. termination (10 minutes).

Sequence

1. Homework review

Ask parents to form small groups and go over each other's
program. Spend 5 minutes in each group. Try not to answer
questions directly, help the parents to find their own
answers. After 30 minutes of small group time, reconvene
into a large group and address the main problems that parents

encountered.

2. Problem solving*

Parents will divide into small groups and given the following
assignment. Each group is to identify a behavior for which
it will develop a behavioral plan. After this is done,
parents are to reconvene into a large group to evaluate their
plan. Encourage other parents to make constructive criticism.
When misconceptions come up, ask parents to provide another

answer.

261

3. Termination
Ask parents to reflect on the training, thank them for

doing a good job.

*Parents in the experimental group were instructed to
develop a self management plan'in'addition to a behavior
program.

Appendix H

262
APPENDIX H

_Informed Consent

I, agree to participate
in the Foster ParentiTraining Project, a clinical evaluation
program, examining the effectiveness of different foster parenting
carricula. I understand that I have been assigned to one of two
training curricula on a chance basis. I further understand there
is no guarantee that training will render me more effective as a
foster parent.

I understand that participation will involve:
1. periodic assessments of myself and members of my house-
hold 12 years and older a. before training b. after
training c. one month after training

2. ,8 weekly classes of 120 minutes each

3. the delivery of a certificate of achievement upon
successful completion of training

I understand that participation in this project is completely
voluntary and that neither myself nor members of my household are
under any obligations to participate. I am free to decline
entrance into the program, and I may withdraw my consent to
participate at any time during the program. I understand that

I may discontinue participation at any time without affecting

my status as a foster parent. All records will be kept

A confidential.

I have read this consent form, and all my questions have been
answered. I freely choose to participate. I understand that
I may withdraw at any time.

 

 

Signature. Date Witness ’DSte

263

REFERENCES

 

Abidin, R.R. Parenting skills. New York: Human Sciences
Press, 1976. -

Aldridge, M.J., Cautley, P.W., and Lichstein, D.P.
Guidelines for placement workers. Madison: Center for
SocIaI Services, University of’Wisconsin-Extension, 1974.

Ambinder, W. and Sargent D. Foster parent techniques of
management of pre-adolescent boys deviant behavior.
Child Welfare, 1965, 52, 90-94.

Arnold, J.E., Levine, A.G., and Patterson, G.R. Changes in
sibling behavior following family intervention. Journal
of Consulting and Clinical ngchology, 1975, 55, 683-688.

Babcock, C. Some psychodynamic factors in foster parent-
hood, Part I. Child welfare, 1965, 55, 485-493.

 

Bandura, A. Gauging the relationship between self efficacy
judgement and action. Psychological Review, 1977, 55,
191-215.

. Self-efficacy mechanism in human agency.
American Psychologists, 1982, 51, 122-147.

 

 

. Self-efficacy: Toward a unifying theory of
behavioral change. Psyghological Review, 1977, 55, 191-
215.

 

Bandura, A., Adams, N.E., and Beyer, J. Cognitive processes
mediating behavior change. Journal of Personality and
Social Psychology, 1977, 55, 1254139.

Barrett, M.C., and McKelvey, J. Stress and strains on the
child care worker: typologies for assessment. Child
Welfare, 1950, 55, 272-285.

Bauer, J., and Heinke, W. Treatment family care homes for
disturbed foster children. Child Welfare, 1976 55, 478-
490.

 

264

Berkowitz, N. Perceived family relationships in families
differing in adjustment level. Unpublished master's
thesis, University of Wisconsin, 1963.

Boyd, L.H., and Remy, L.L. Is foster parent training
worthwhile? Social Service Review, 1978, 275-296.

Brown, Jr., I. and Inouye, D.K. Learned helplessness
through modeling: The role of perceived similarity in
competence. Journal of Personality and Social

 

Brown, D.L. A comparative study of the effects of two
foster parent training methods on attitudes of parental
acceptance, sensitivity to children, and general foster
parent attitudes. Doctoral Dissertation, Michigan State
University, 1980.

Bryant, B. Special foster care: A history and rationale.
Journal of Clinical Child Psychology, 1981, 12, 8-20.

Burchard, J. and Leitenberg, 3. Vermont foster parent
training program: grainer's manugl. Vermont:
Department 0 Social and Rehabilitation Services, 1978.

Campbell, D.T. and Stanley, J.C. Experimental and quasi-

 

 

 

experimental designs for research. Chicago: Rana
McNa y Co ege P liabing Company, 1963.
Cautley, P.W. New foster parents. New York: Human

 

Services Press, I980}

Cautley, P.W. and Aldridge, M.J. Predicting success for
new foster parents. Social WOrk, 1975, 32,.48-53.

Child Welfare League of America. Newsletter, Summer-Pall,
1974.

Cobb, E.J., Leitenberg, 8., and Burchard, J.D. Poster
parents teaching foster parents: Communication and
conflict resolution skills training. Journal of
Community Psychology, 1982, 59, 240-249.

Crozier, J., and Roger, K. Social learning treatment of
child abuse. Journal of Behavioral Therapy and
ngchiatry, 1979, 12' 213-520.

 

Denicola, J. and Sandler, J. Training abusive parents in
child management and self-control skills. Behavior
Therapy, 1980, 55, 263-270.

265

Dinnage, R., and Pringle, M.L.K. Foster Home Care: Facts
and Fallacies. London: Longsman in association with the
material bureau for cooperation in child care, 1967.

Eisenberg, L. The sins of the fathers: Urban decay and
social pathology. American Journal of Orthopgychiatgy,

Ellis, A. Reason and emotion in psychotherapy. New York:
Lyle Stuart, I970.

Ellis, A. and Grieger, R. Handbook of rationale-emotive
therapy. New York: Springer, I977.

Fairweather, G.W. and Tornatzky, L. G. Egperimental methods
for social research. New York: Pergamon Press, 197

 

Fanshel, D. The role of foster parents in the future of
foster care. In H. D. Stone (ed. ) Foster care in
question: Anational assessment by twentygone experts.
New York: ChIIa WeIfare League of America,l970.

. Specializations within the foster parent
role: A research report. Child Welfare, 1961, fig, 17-
21.

 

Pain, 3., and Maluccio, A. Children leaving foster care:
Outcomes of permanency planning. Child Abuse & Neglect,
1984, 5, 425-431.

Freeman, H. Foster home care for mentally retarded
children: can it work? Child Welfare, 1978, 51, 113-
121.

Freeman, R.M., Sandler, J., Hernandez, M., and WOIfe, D.A.
Child abuse. In E.J. Mash & L.G. Terdal (eds.)
Behavioral Assessment of Childhood Disorders. New York:
The GuIIfora Press,l981.

Garrett, B. Foster care: America's lost children. Public

Gibaud-Wallston, J., and wandersman, L.P. Development and
utility of the parenting sense of competence scale.
Paper presented at the meeting of the American
Psychological Association, Toronto, Canada, August, 1978.

Gillham, A study of specialized foster care programs
operated by private agencies in Michigan and contracted
for by the Michigan Department of Social Services.
Michigan: Department of Social Services, 1983.

266

Goldstein, H. Group learning for foster parents. Children,
1967, 14, 180-184.

Guerney, L.F. Foster parent training project final report,
Part 11. Pennsylvania State University, 1976.

Haffey, N., and Levant, R. The differential effectiveness
of two models of skills training for working class
parents. Family Relations, 1984, 2;, 209-216.

Hampson, R.B., and Tavormina, J.B. Feedback from the
experts: A study of foster mothers. Social WOrk, 1980,
gg, 108-113.

 

Relative
effectiveness of behavioral and reflective group training
with foster mothers. Journal of Consulting and Clinical
Psychology, 1980a, 48(2), 294-29 .

 

Horejsi, C.R. Foster family care. Springfield, Illinois:
Thomas, 1979.

Horner, P.L. Family vulnerability to child-induced
stresses: Additional considerations for foster placement
of “difficult” children. Journal of Clinical Child
Psychology, 1981, 12, 42-46.

 

Jacobs, M. Foster Parent Training: An opportunity for
skills, enrichment and empowerment. Child welfare, 1980,
59, 615-624.

Kaplan, B.L., and Seitz, M. The Practical Guide to Foster
Family Care. Springfield, Illinois: Thomas,1 .

Karoly, P., and Rosenthal, M. Training parents in behavior
modification: effects on perceptions of family
interaction and deviant child behavior. Behavior Therapy
1977, Q: 406-410.

Kennedy, R. A foster parent looks at foster care. In H.D.
Stone (ed.) Foster Care in Question: A national
assessment by twenty-one experts. New York: Child
Welfare League of America, 1970.

Kline, D., and Overstreet, H.F. Foster Care of Children.
New York: Columbia University Press, 1972.

Langsam, M.Z. Children west. Cushing-Malloy, Inc., Ann
Arbor, Michigan, 1964.

267

Larson, 6., Allison, J., and Johnson, E. Alberta parent
counselors: A community treatment program for disturbed
youth. Child Welfare, 1978, 51, 47-52.

Lee, C. Self efficacy as a predictor of performance in
competitive gymnastics. Journal of Sport Psychology,
1982, 4, 405-409.

Levant, R.F., and Geer, M.F. A systematic skills approach
to the selection and training of foster parents as mental
health paraprofessionals, I: Project overview and
selection component. Journal of Community Psychology,
1981, 2, 224-230.

Levant, R.F., Slatery, S.C., and Slobodian, P.E. A
systematic skills approach to the selection and training
of foster parents as mental health paraprofessionals 11:
Training. Journal of Community Psychology, 1981, 9,
224-230.

Maas, H., and Engler, R.E. Children in need of parents.
New York: Comumbia University Press, 1959.

Mash, E.J., and Johnston, C. Parental-perceptions of child
behavioral problems, parenting, self-esteem, and mothers'
reported stress in younger and older hyperactive and
normal children. Journal of Consulting and Clinical
Psychology, 1983, 51, 86-99.

McCoy, J., and Donahue, J. Educating foster mothers through
the group process. Child welfare, 1961, 40, 29-31.

Meinchenbaum, D.H. Cognitive-behavior modification: an
integrative approach. New York: Plenum Press, 19 .

 

Meinchenbaum, D.H., Henshaw, D. and Himel, N. Coping with
stress a problem-solving process. In H.W. Krohne and L.
Laux (eds) Achievement; stress, and anxiety. washington,
Hemisphere Publishing Corporation, 1 .

Meichenbaum, D.H., and Turk, D. Stress, coping and disease:
a cognitive behavioral perspective. In R.W.J. Newfeld
(ed.) Ps chological Stress and Psychopathology. McGraw-
Hill Boo Company, New York, 1999.

Miller, J.G. Living systems: Basic concepts. Behavior
Science, 1965, 10, 193-253.

Minuchin, S., and Fishman, H.C. Famil therapy techniques,
Cambridge, Mass.: Harvard University Press, 1981.

268

Moos, R.B., and Moos, 3.3. Family environment scale manual.
Consulting Psychologists Press, Palo Alto, California,
1981.

Murray, L. A review of selected foster care-adoption
research from 1978 to mid-1982. Child Welfare, 1984,

 

Pardeck, J.T. The forgotten children: A study of the
stability and continuity of foster care. Washington,

 

D.C.: University Press of America, 1992.

Patterson, G.R. Families: Applications of social learning
to family life. Champaign, Illinois: Research Press,
1971.

 

 

 

 

 

 

. Living with children: New methods for
parents and teachers. Champaign, Illinois: Research
Press, 1976.

. Mothers: the unacknowledged victims.
Monographs of the Societ for Research in Child
Development, 1980, 45155, Serial No.*186.

 

Patterson, G.R., and Reid, J.B. Reciprocity and Coercion:
Two facets of social systems. In C. Neuringer and J.C.

 

Michael (eds.) Behavioral Modification in Clinical
Ps cholo . New* York: Appleton-Century-Crofts
(Prentice-Hall), 1970.

Patterson, G.R., Reid, J.B., Jones, R.R., and Conger, R.D.,
A social learning approach to family intgryention: Vol.
I Families with aggressive children. Eugene, Oregon:
Castalina, 1975.

Pearlin, P.I., and Schooler, C. The structure of coping.
Journal of Health and Social Behavior. 1978, 12, 2-21.

Penn, J.V. A model for training foster parents in behavior
modification techniques. Child welfare, 1978, El, 175-
180.

Reiter, C.F., and Kilmann, P.R. Mothers as family change
agents. Journal of Counseling Psychology. 1975, 2211!,
61-65.

Rimm, D.C. and Masters, J.C. Behavior Therapy: Techniques
and empirical findings. Academic Press, New York, 1979.

Ross, A.O. Child behavior therapy. New York: John Wiley &
Sons, 1981.

269

Ryan, P.R. Issues in_§ostering Ypsilanti: Eastern
Michigan University, 1979.

Sandler, J., VanDercar, C., and Mariann, M. Training child
abusers in the use of positive reinforcement practices.
Behavior Research and Therapy, 1978, 16, 169-175.

Schunk, D.H. Modeling and attributional effects on
children's achievement: A self-efficacy analysis.
Journal of Educational Psychology, 1981, 13, 93-105.

Scoresby, A. and Christensen, B. Differences in interaction
and environmental conditions of clinic and nonclinic
families: Implications for counselors.
gournal of Marriage and Family Counseling, 1976, 3, 63-

1.

Shyne, A., and Schroeder, A. National study of social
services to children and their families. washington,
D.C.: National Center for Child Advocacy. U.S.
Children's Bureau, Administration for Children, Youth and
Families, DHEW, 1978.

Skynner, A.C. Systems of family and marital psychotherapy.
Brunner Mazel: New York, 1976:

Slattery, S.C. The relationship between the foster parents'
level of interpersonal skills and the outcome of foster
placement. Unpublished doctoral dissertation, Boston
University, 1980.

Slobodian, P.E. An evaluation of the effectiveness of the
personal development approach to foster parents skills
training. Unpublished doctoral dissertation, Boston
University, 1980.

Soffen, J. The impact of group educational program for
foster parents. Child welfare, 1962, 41, 195-201.

Steele, B., and Pollock, D. A psychiatric study of parents
who abuse infants and small children. In R. Helfer and
C. Kempe (eds.) The Battered Child, University of
Chicago Press, Chicago, I968.

Stone, H.D., and Hunzeker, J.M. Education for Foster Family
Care: Models and methods for foster parents and social
workers. New York: Child welfare League of America.

Stone, N.M., and Stone, S.F. The prediction of successful
foster placement. The Journal of Contemporary Social
werk, 1983, 11-17.

270

Taylor, D.A., and Starr, P. Foster parenting: An
integrative review of the literature. Child welfare,
1967,‘41,371-383.

Thomas, C. The use of group methods with foster parents.
Children, 1961, 40, 33.

VanDerVeen, F. The parent's concept of the family unit and
child adjustment. Journal of Counseling Psychology,

Vasaly, S.M. Foster care in five states. Social Research
Group, Washington, D.C.: George Washington University,
1976.

Wahler, R.G., Winkel, G.H., Patterson, R.F., and Morrison,
D.C., Mothers as behavior therapists for their own
children. Behavioral Research and Therapy, 1965, 3, 113-
124.

Walter, H., and Gilmore, S.K. Placebo versus social learning
effects in parent training procedures designed to alter
the behaviors of aggressive boys. Behavior Research and
Therapy, 1973!.2! 361-377.

Wasserman, S. The abused parent of the abused child.
Children, l967,m 14, 175-179.

Weinberg, R., Gould, D., and Jackson, A. Expectations and
performance: Am empirical test of Bandura's self
efficacy theory. Journal of Sport Psychology, 1979, 1,
320-331.

Wilkes, J.R. The impact of fostering on the foster family.

 

Wiltz, N.A., Jr., and Patterson, G.R. An evaluation of
parent training procedures designed to alter
inappropriate aggressive behavior of boys. Behavior
Therapy, 1974, 5, 515-521.

Wolfe, D.A., and Sandler, J. Training abusive parents in
effective child management. Behavior Modifications,

Wolfe, D.A., and Sandler, J., and Kaufman, K. A competency
based parent training program for child abusers.
Journal of Consulting Psychology, 1981, 42, 633-640.

Wolins, M. Selectinggoster parents. New York: Columbia
University Press, 1963.