Citation
Guardian ad Litem and judicial decision-making in cases of child abuse and neglect

Material Information

Title:
Guardian ad Litem and judicial decision-making in cases of child abuse and neglect
Creator:
McDanal, Cynthia Sutton
Publication Date:
Language:
English
Physical Description:
vii, 74 leaves : ; 29 cm.

Subjects

Subjects / Keywords:
Child abuse ( jstor )
Child neglect ( jstor )
Child psychology ( jstor )
Children ( jstor )
Demography ( jstor )
Judicial rulings ( jstor )
Logistic regression ( jstor )
Mathematical variables ( jstor )
Parents ( jstor )
Recommendations ( jstor )
Child Abuse ( mesh )
Decision Making ( mesh )
Department of Clinical and Health Psychology thesis Ph.D ( mesh )
Dissertations, Academic -- College of Health Related Professions -- Department of Clinical and Health Psychology -- UF ( mesh )
Jurisprudence ( mesh )
Legal Guardians ( mesh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1994.
Bibliography:
Includes bibliographical references (leaves 51-53).
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Cynthia Sutton McDanal.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright [name of dissertation author]. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
002390141 ( ALEPH )
ALZ5024 ( NOTIS )
51314415 ( OCLC )

Downloads

This item has the following downloads:

guardianadlitemj00mcda.pdf

guardianadlitemj00mcda_0054.txt

guardianadlitemj00mcda_0074.txt

guardianadlitemj00mcda_0043.txt

guardianadlitemj00mcda_0040.txt

guardianadlitemj00mcda_0036.txt

guardianadlitemj00mcda_0044.txt

guardianadlitemj00mcda_0005.txt

guardianadlitemj00mcda_0003.txt

guardianadlitemj00mcda_0057.txt

guardianadlitemj00mcda_0082.txt

guardianadlitemj00mcda_0068.txt

guardianadlitemj00mcda_0017.txt

guardianadlitemj00mcda_0037.txt

guardianadlitemj00mcda_0022.txt

guardianadlitemj00mcda_0033.txt

guardianadlitemj00mcda_0004.txt

guardianadlitemj00mcda_pdf.txt

guardianadlitemj00mcda_0028.txt

guardianadlitemj00mcda_0069.txt

guardianadlitemj00mcda_0006.txt

guardianadlitemj00mcda_0062.txt

guardianadlitemj00mcda_0010.txt

guardianadlitemj00mcda_0047.txt

guardianadlitemj00mcda_0059.txt

guardianadlitemj00mcda_0041.txt

guardianadlitemj00mcda_0002.txt

guardianadlitemj00mcda_0083.txt

guardianadlitemj00mcda_0077.txt

guardianadlitemj00mcda_0032.txt

guardianadlitemj00mcda_0081.txt

guardianadlitemj00mcda_0024.txt

guardianadlitemj00mcda_0075.txt

guardianadlitemj00mcda_0056.txt

guardianadlitemj00mcda_0048.txt

guardianadlitemj00mcda_0060.txt

guardianadlitemj00mcda_0084.txt

guardianadlitemj00mcda_0064.txt

guardianadlitemj00mcda_0052.txt

guardianadlitemj00mcda_0023.txt

guardianadlitemj00mcda_0029.txt

guardianadlitemj00mcda_0021.txt

guardianadlitemj00mcda_0000.txt

guardianadlitemj00mcda_0018.txt

guardianadlitemj00mcda_0019.txt

guardianadlitemj00mcda_0031.txt

guardianadlitemj00mcda_0035.txt

guardianadlitemj00mcda_0080.txt

guardianadlitemj00mcda_0053.txt

guardianadlitemj00mcda_0050.txt

guardianadlitemj00mcda_0008.txt

guardianadlitemj00mcda_0020.txt

guardianadlitemj00mcda_0061.txt

guardianadlitemj00mcda_0034.txt

guardianadlitemj00mcda_0026.txt

guardianadlitemj00mcda_0058.txt

guardianadlitemj00mcda_0001.txt

guardianadlitemj00mcda_0030.txt

guardianadlitemj00mcda_0071.txt

guardianadlitemj00mcda_0012.txt

guardianadlitemj00mcda_0011.txt

guardianadlitemj00mcda_0016.txt

AA00009050_00001_pdf.txt

guardianadlitemj00mcda_0039.txt

guardianadlitemj00mcda_0046.txt

guardianadlitemj00mcda_0072.txt

guardianadlitemj00mcda_0067.txt

guardianadlitemj00mcda_0055.txt

guardianadlitemj00mcda_0025.txt

guardianadlitemj00mcda_0049.txt

guardianadlitemj00mcda_0027.txt

guardianadlitemj00mcda_0042.txt

guardianadlitemj00mcda_0065.txt

guardianadlitemj00mcda_0014.txt

guardianadlitemj00mcda_0066.txt

guardianadlitemj00mcda_0079.txt

guardianadlitemj00mcda_0070.txt

guardianadlitemj00mcda_0007.txt

guardianadlitemj00mcda_0045.txt

guardianadlitemj00mcda_0013.txt

guardianadlitemj00mcda_0038.txt

guardianadlitemj00mcda_0015.txt

guardianadlitemj00mcda_0078.txt

AA00009050_00001.pdf

guardianadlitemj00mcda_0076.txt

guardianadlitemj00mcda_0073.txt

guardianadlitemj00mcda_0063.txt

guardianadlitemj00mcda_0051.txt

guardianadlitemj00mcda_0009.txt


Full Text
















GUARDIAN AD LITEM AND JUDICIAL DECISION-MAKING
IN CASES OF CHILD ABUSE AND NEGLECT







By

CYNTHIA SUTTON MCDANAL


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


1994




GUARDIAN AD LITEM AND JUDICIAL DECISION-MAKING
IN CASES OF CHILD ABUSE AND NEGLECT
By
CYNTHIA SUTTON MCDANAL
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1994


ACKNOWLEDGEMENTS
The author extends appreciation to her parents, Mr. and Mrs. M. R. Sutton, and
her grandparents, Mr. and Mrs. D. A. Ball, for their continued patience and
encouragement. Particular gratitude is extended to the authors husband, Mr. Charles
McDanal, Jr., for his unconditional guidance and support.
ii


TABLE OF CONTENTS
ACKNOWLEDGEMENTS ii
LIST OF TABLES v
ABSTRACT vi
INTRODUCTION 1
Overview 1
Definition of Abuse 3
Incidence and Prevalence 4
Role of Child Protection Agencies 6
The Guardian ad Litem Program 8
REVIEW OF LITERATURE 11
Theoretical Models of Decision-Making 11
Empirical Studies of Decision-Making 15
RESEARCH OBJECTIVES AND HYPOTHESES 20
METHOD 22
Subjects 22
Measure 22
Procedures 25
RESULTS 26
Comparison of Judges and Guardians Decisions 26
Rate of Removal by Abuse Type 28
Logistic Regression Equations 30
Comparison of Cases of Physical Abuse Only and Neglect Only 37
DISCUSSION 41
REFERENCES 51
iii


APPENDIX A GAL QUESTIONNAIRE FOR DEPENDENCY CASES . 54
APPENDIX B CORRELATIONS AMONG INDEPENDENT
VARIABLES 65
APPENDIX C RATE OF REMOVAL BY TYPE OF ABUSE 69
APPENDIX D DESCRIPTIVE STATISTICS FOR VARIABLES IN
CASES OF PHYSICAL ABUSE ALONE AND NEGLECT
ALONE 71
APPENDIX E PEARSON THREE-WEEK TEST-RETEST
CORRELATIONS 73
BIOGRAPHICAL SKETCH 74
iv


1
2
3
4
5
6
7
8
9
10
11
12
13
LIST OF TABLES
page
Comparison of Guardian Recommendations and Judicial
Decisions 27
Results of Logistic Regression Analysis Using Guardian
Recommendations to Predict Judicial Decisions 27
Variable Values In Cases Where Judges and Guardians Disagreed
Regarding Case Outcome 29
Results of Logistic Regression Analysis Using Demographics to
Predict Judicial Decisions 32
Results of Logistic Regression Analysis Using Demographic
Variables to Predict Guardian Recommendations 32
Results of Logistic Regression Analysis Using Legal and Case
Variables to Predict Judicial Decisions 34
Results of Logistic Regression Analysis Using Legal and Case
Variables to Predict Guardian Recommendations 34
Results of Logistic Regression Analysis Using Family
Psychological Variables to Predict Judicial Decisions 36
Results of Logistic Regression Analysis Using Family
Psychological Variables to Predict Guardian Recommendations .... 36
Results of Logistic Regression Analysis Using All Significant
Predictor Variables to Predict Judicial Decisions 38
Results of Logistic Regression Analysis Using All Significant
Predictor Variables to Predict Guardian Recommendations 38
Distribution of Interval Variables 42
Frequencies of Dichotomous Variables 42
v


Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
GUARDIAN AD LITEM AND JUDICIAL DECISION-MAKING
IN CASES OF CHILD ABUSE AND NEGLECT
By
Cynthia Sutton McDanal
April, 1994
Chairman: Jacquelin Goldman
Major Department: Clinical and Health Psychology
It has long been recognized that cases involving child abuse and neglect confront
the legal system with the difficult moral and practical dilemma of whether to remove
children from their parents care. While broad guidelines are provided under federal and
state laws, there remains a great deal of ambiguity regarding appropriate factors to
consider when making such decisions. Several researchers have put forth rational models
to guide decision-making and a few have examined samples in attempts to determine
which factors are most relevant to court personnel. However, these studies generally
have not sought to generate predictive models for such decision-making, and they tend
to rely on examination of case records rather than direct report from those involved in
the decision-making process. The current study sought to extend our current knowledge
of the determinants of case disposition in cases of child abuse and neglect by considering
the influence of potential factors as reported by the childs court-appointed representative,
vi


or Guardian ad Litem. Guardian recommendations for placement were considered as
well as case adjudication to determine whether significant differences exist in the
decision-making processes of the two groups. Judicial decisions were also compared to
Guardian recommendations to determine the extent to which judges and Guardians
concur.
Results indicated that while judges are highly likely to concur with
recommendations of the Guardians, they tend to place a somewhat different emphasis on
the variables they use to determine whether to remove a child from his home. Judges
were found to rely most heavily on the presence of allegations of sexual abuse and on
the degree of parental warmth and support for the child. Guardians, however, were
found to rely most heavily on the presence of previous reports of abuse against the child.
Neither judges nor Guardians appeared to place heavy weight on much of the information
regarding family psychological functioning. Implications for this finding are discussed.
vii


INTRODUCTION
Overview
Recent estimates of the prevalence of child sexual abuse, physical abuse, and
neglect indicate that almost two million children per year become victims of some form
of maltreatment (National Center on Child Abuse and Neglect, 1988). Numerous studies
have documented the host of psychological and behavioral problems that often result from
such abuse (see Lamphear, 1985; Brown & Finkelhor, 1986; Houck & King, 1988, for
reviews). Such research clearly indicates that child abuse is a serious problem of wide
scope in our society.
The considerable task of assessing and intervening in cases of child abuse and
neglect has fallen to child protection agencies and the judicial system. To facilitate this
process, specific guidelines have been set, and Guardian ad Litem programs have been
established. Working within the guidelines, the Guardian ad Litem serves as a court-
appointed legal representative of the child. The Guardian gathers information concerning
the family and makes a recommendation to the court regarding appropriate placement of
the child and treatment for the family. It is then the responsibility of the judge to make
a final decision in the best interests of the child, the family, and society.
Such a process requires a great deal of subjective judgment, especially on the part
of the Guardian. Several researchers have put forth theoretical models to guide decision
making in such situations (e.g., Wasserman & Rosenfeld, 1988; Faller, 1988), and other
1


2
researchers have sought to describe the decision-making process by considering social,
family, and child characteristics which are associated with case disposition (e.g., Katz,
Hampton, Newberger, Bowles, & Snyder, 1986; Finkelhor, 1983).
Although the approaches applied in these latter studies are more direct in that they
analyze actual placement decisions, the studies themselves are largely descriptive and
have limitations in methodology or scope. The present study seeks to extend our current
knowledge of the decision-making process involved in cases of child abuse and neglect
by identifying factors which influence the Guardians recommendation and the final
adjudication of the case. A large number of potential determinants will be considered.
Guardians and judges will be evaluated separately in order to determine whether
differences exist in the decision-making processes of the two groups. The results of such
analyses can then be compared to theoretical models of decision-making in order to
evaluate the appropriateness and effectiveness of the process.
Historical Background
The problem of child abuse and neglect is not a new phenomenon. On the
contrary, children have been mistreated by adults since the beginning of civilization
(Radbill, 1968). Practices such as harsh discipline, child labor, incest, ritual sacrifice,
and infanticide attest to the extent to which child abuse is deeply rooted in cultural and
religious history (Eskin, 1980). The historical view of children as "property" of their
parents and the generally accepted concept that family matters were private contributed
to such a situation.


3
It was not until the late nineteenth century that society began to shift its view of
children as property to include the concept of childrens rights. The landmark case of
"Mary Ellen" in 1870 marked the first time that child abuse was recognized by a U.S.
court. The court intervened by removing Mary Ellen from her home and by prosecuting
her custodian. This case demonstrated the need for a child protection agency, and
shortly thereafter, the New York Society for the Prevention of Cruelty to Children was
chartered by the New York legislature (Landau, 1982). This agency served as the
prototype for many that followed.
Definition of Abuse
While the above case brought about a recognition that parental freedom is not
limitless and that children have a legal right to be protected from cruelty and neglect,
there has been no clear consensus of exactly what constitutes child abuse. Wasserman
and Rosenfeld (1986) note that early definitions of child abuse generally included only
the most severe cases, while recent definitions have been expanded to include any form
of corporal punishment. Emotional abuse has recently been recognized in the literature,
but it has also proved difficult to define precisely. State statutes and mandatory reporting
laws defining child abuse typically refer to harm or threatened harm to a childs physical
or mental well-being, but they do not provide specific guidelines.
Sloan (1983) has put forth a definition of abuse which is operationally useful. He
divides child abuse and neglect into four categories: physical abuse, neglect, sexual
abuse, and emotional maltreatment. Physical abuse of children refers to any non
accidental injury caused by the childs caretaker. Neglect involves inattention to the


4
basic needs of the child, including food, shelter, clothing, medical care, and supervision.
Sexual abuse includes any interaction between a child and an adult in which the child is
being used for the sexual stimulation of the adult. Finally, emotional maltreatment
includes such behaviors as blaming, belittling, or rejecting a child; unequal treatment of
siblings; or lack of concern for the childs welfare. Such definitions provide more
specific criteria for determining the occurrence of abuse or neglect, and are frequently
used by child protection workers.
Incidence and Prevalence
The exact scope of child maltreatment in this country is subject to wide debate.
One problem in determining the incidence of child abuse is the lack of a standard
definition regarding exactly what constitutes abuse and neglect. A second problem is that
different research designs yield different estimates of child abuse. Daro (1988) notes that
interviews with a random sample of households or individuals consistently project higher
incidence and prevalence rates than indicated by formal reporting data.
The National Center on Child Abuse and Neglect (1988) has responded to such
difficulties in estimation of the incidence of child abuse by recognizing five "levels" of
abused and neglected children. Level 1 includes all abused and neglected children known
to state and local Child Protective Services (CPS) agencies. At Level 2 are those
children who are not known to CPS, but who are known to other investigative agencies,
such as police, courts, or health departments. Level 3 includes children who are not
known to any Level 1 or 2 agency, but who are known to professionals in other
community institutions, such as hospitals, schools, daycare centers, and mental health


5
clinics. The abused and neglected children at Level 4 are known only to individuals
outside of Levels 1, 2, and 3, such as neighbors or other family members. Finally,
Level 5 includes those children who have not been recognized by anyone as abused or
neglected. Although this description of all five levels is valuable in that it formally
recognizes that unreported cases of abuse do exist, the Centers incidence study includes
cases only at Levels 1 through 3.
The results of this study indicated that 10.7 children per 1,000 were physically
abused in the United States during 1988, or a total of 675,000 children. Rates of neglect
were even higher 15.9 children per 1,000, or a total of over one million children
nationwide. The rate of all forms of child sexual abuse was estimated to be 2.5 per
1,000, or a total of 159,000 children.
As mentioned above, random sample interview studies generally yield higher
estimates of the prevalence of child abuse, probably because they tap unreported abuse
as well as abuse known to public agencies and community professionals. For example,
Strauss and Gelles (1986) conducted a national study on the incidence and prevalence of
family violence by telephone interviewing a probability sample of 6,002 households.
Child abuse data were based on the 1,428 households in which there was a child between
the ages of 3 and 17 years, and in which two caretakers were present. The results of this
study indicated that 19 children per 1,000 were subject to at least one incident of "very
severe violence" during 1985.
Random samples of the prevalence of sexual abuse also yield significantly higher
estimates than the Centers study. Russell (1983) surveyed a probability sample of 930


6
women residents of San Francisco about any experience of sexual abuse they might have
had. Her findings indicated that 38% of the women had experienced at least one incident
of sexual abuse before the age of 18 years. Finkelhor (1984) reported that 6% of all
males and 14% of all females in his random sample had experienced sexual abuse before
the age of 16 by a person at least five years older.
Regardless of the specific definition or research methodology utilized, virtually
all estimates of the incidence and prevalence of child abuse and neglect indicate that it
is an increasingly prevalent problem in our society.
Role of Child Protection Agencies
Since the case of Mary Ellen, it has been the responsibility of the court and child
protection agencies to assess and intervene in cases of child abuse and neglect.
Considering the amount of abuse and neglect that has been documented, the lack of a
standard definition of abuse, and the precarious balance between individual rights
protected under the fourteenth amendment and the need to protect children, this is an
imposing responsibility.
In order to carry out the tasks of assessment and intervention in a fair and
effective manner, specific guidelines for child protective services have been put forth.
Although the specific details of such guidelines differ by state, the general principles of
the process remain the same. Landau (1982) describes six major aspects of the child
protective service process:
The referral process. All 50 states now have mandatory reporting laws designed
to encourage the reporting of suspected abuse and neglect. Most reporting systems also


7
include provisions which encourage investigation, intervention, and treatment services.
Many states have also set up central abuse registries.
Definitions. As mentioned above, each state has a slightly different definition of
what constitutes child abuse and neglect, and all state definitions are rather broad.
However, differences are minimal, and the use of a broad definition provides the
practitioner with leeway in decision-making. Most states distinguish between abuse and
neglect, and therefore define them separately.
Critical elements of a child protection system. Landau notes that there are a
number of criteria for a child protective agency to be an effective system. Practitioners
must have accurate knowledge of the incidence of child abuse and neglect. The system
must have a strong and well-publicized mandatory reporting law and an effective central
abuse registry. A specially trained Child Protection Unit must be available at all times.
The court system must be capable of dealing fairly and effectively with families, and
treatment and rehabilitative programs should be available for those families. Finally,
interdisciplinary communication and a system of coordination should be used to
implement the network of services for families.
Assessment of cases. A specially trained child protection worker must
systematically gather information regarding the child, the family, and the circumstances
of their lives. The worker must then use this information to determine whether abuse
has occurred; if so, to determine why it occurred; and to identify appropriate
interventions.


8
Initiation of the legal process. It is the responsibility of the child protection
worker to decide whether to initiate court proceedings. Guidelines for the consideration
of court action include situations in which (a) the child is determined to be in imminent
danger, (b) attempts at treatment have failed, and/or (c) the family refuses to cooperate
with the investigation. Caseworkers must also consider issues regarding the nature of
evidence and the possible negative consequences of bringing a case into court.
Emergency removal. All states have statutes which allow for the emergency
removal of abused or neglected children from the home, even prior to the filing of a
petition. A court order may or may not be necessary. All states have specific criteria
regarding the conditions necessary to justify emergency removal, and most have specific
requirements describing what must be done following the removal.
Caseworker-attorney interchange. Landau (1982) also stresses the importance of
the establishment of rapport between the child protection worker and the agency attorney
as they work to prepare cases. Clear guidelines regarding decision-making are
beneficial.
The Guardian ad Litem Program
Almost all states have implemented Guardian ad Litem or similar programs to
carry out the above functions, and those that have not require that legal counsel be
appointed to represent the child in abuse and neglect hearings (Sloan, 1983). A Guardian
ad Litem is a court-appointed legal representative who participates on behalf of children,
who are legally unable to initiate or function in litigation without adult assistance. The
function of Guardians is to identify the childs best interests in the litigation, and they


9
are legally obligated to do everything within their power to represent those interests
(Whitcomb, 1988).
Wiig (1982) notes that although appointment of a Guardian ad Litem is a
condition of the federal Child Abuse Prevention and Treatment Act (PL93-247), the Act
does not define the functions of a Guardian, distinguish the Guardians role from that of
the childs social worker and the childs attorney, or clearly define the relationship of the
Guardian to the court. She conducted a limited study of six Guardians in an attempt to
clarify these roles.
Wiigs findings led her to conclude that the following ten activities might be
defined as the functions of the Guardian because they were distinguished from the actions
of the childs lawyer and social worker in the cases she examined: (1) to function as a
mediator in an attempt to achieve consensus among parties prior to court proceedings;
(2) to coordinate between juvenile court and other court departments; (3) to provide
continuity across the entire legal process; (4) to persuade the social worker to consider
alternative placements, (5) to minimize trauma to the child and to provide continuity of
relationship with the child: (6) to insure court attention to cases by placing them on the
calendar; (7) to make recommendations in the best interests of the child; (8) to identify
special needs of the child and to insure that the child receives attention to those needs;
(9) to prevent continuances not in the best interests of the child or to request an attorney
to do so; and (10) to request independent counsel for the child (pp. 93-94). Wiig further
notes that while the attorney or social worker might in some cases fulfill some of these


10
roles, the functions numbered 3, 5, and 8 are distinct and may be considered strictly
functions of the Guardian ad Litem.
While it is generally accepted that the Guardian fulfills these functions and
ultimately makes a recommendation to the court regarding disposition of the case, the
process by which the Guardian arrives at the recommendation is less clear. Several
researchers have noted the problems inherent in such a process, and have put forth
theoretical models of decision-making to guide practitioners. These models are discussed
in detail below.


REVIEW OF LITERATURE
Theoretical Models of Decision-Making
Wasserman and Rosenfeld (1986) have noted several difficulties facing judges who
must decide when to intervene in cases of possible abuse or neglect, and these difficulties
also apply to Guardians as they struggle to make the most appropriate recommendation
based on the information they have. Such problems include the aforementioned lack of
clarity in the definition of abuse, difficulties in proving abuse, limited alternatives
available other than foster care, problems in the foster care system, and the large number
of children in abusive and neglectful situations.
Wasserman and Rosenfeld (1986) also note the positive and negative aspects of
separation that must be considered when deciding whether to remove a child from his
home. Separation provides for the safety of the child, sets limits on parental behavior,
and lets the child know that society will intervene to protect him if his parents do not.
It may also provide an opportunity for the child to live in a safer, warmer environment.
Separation also frees the parents from the burdens of childrearing, thereby giving them
the opportunity to make constructive changes in their situation.
Potential negative consequences of separation must also be carefully considered.
Even though they may be abused, most children have an affectional bond with their
parents and see that relationship as a source of identity and well-being. Thus, even when
faced with abuse, many children express a desire to remain with their families. Also,
11


12
feelings of inadequacy and failure in both parents and children may follow separation.
The authors state that children may even perceive themselves as responsible for the
separation.
In order to make appropriate decisions given such a difficult situation, Wasserman
and Rosenfeld (1986) recommend that practitioners consider the nature and quality of the
parent-child bond. They state, "A judge who is provided with careful evaluation of the
parent-child relationship is in a better position to decide wisely" (p. 524). In addition
to interviewing all individuals involved in the case, the authors recommend that the
caseworker visit the familys home. The caseworker should note issues such as how well
the parent knows the child, how able and willing the parent is to care for the child, how
reasonable the parents beliefs about discipline are, and how much nurturance the parent
provides the child. Once the caseworker has obtained this information, he or she can
make a reasonable evaluation of the relationship between parent and child. According
to Wasserman and Rosenfeld, this evaluation can then be used to make the most
appropriate decision regarding disposition of the case.
While the characteristics of the family relationships likely have legal and clinical
utility for decision-making, an approach considering only this factor seems overly
simplistic. Based on clinical experience with over 300 cases of child sexual abuse, Faller
(1988) has proposed a model of decision-making in cases of intrafamilial child sexual
abuse which is more comprehensive than the approach described above. Like
Wasserman and Rosenfeld (1986), she believes that the parent-child relationship is an
important consideration, especially the relationship between mother and child. However,


13
Faller also believes that other parental characteristics are crucial to decisions about
intervention.
These characteristics differ between mothers and fathers. According to Fallers
theory, important maternal factors are (1) the level of the mothers dependency,
especially on the father-perpetrator; (2) the degree of love and nurturance she conveys
to the victim; and (3) the degree to which she responded protectively when she
discovered the sexual abuse. Important paternal characteristics are (1) his overall level
of functioning; (2) his superego functioning; and (3) the severity of the sexual abuse.
In order to use such information to choose intervention strategies, Faller has
identified four types of cases based on different combinations of the above characteristics.
Type I cases are ones in which both parents have many strengths. In such cases Faller
recommends family therapy, without court involvement. In Type II cases, the mother
has many positive characteristics, but the father exhibits many areas of problematic
behavior. For these cases, Faller suggests maintaining the mother and child intact and
referring them for treatment but advises that the father-perpetrator be excluded from the
family. Legal intervention may be necessary to remove the father from the home.
In Type III cases, mothers are dependent, hostile toward their children, and
unprotective. Although the father has sexually abused the child, he functions much better
than the mother and he often has a closer relationship with the child than does the
mother. According to Faller, this is the most difficult type of case for which to
determine appropriate intervention. She recommends therapy for all family members for
three to six months, at which time careful re-evaluation should be conducted. During


14
this time period, it may be optimal to place the child with a relative, which allows for
less disruption in the childs life while still protecting him or her from further sexual
abuse.
In Type IV cases, the mother is similar to the Type III mother, and the father
exhibits poor overall functioning and shows no guilt over the (usually) extensive abuse.
Faller recommends that parental rights be terminated in such cases, and that the children
receive treatment for at least one year. Criminal prosecution may also be warranted for
the perpetrator.
Although Fallers approach to decision-making is comprehensive and specific, it
does have several limitations. First, her approach assumes the availability of alternative
placements for the child (i.e., placement with a relative or for adoption), which may not
always be readily available. Second, some possible combinations of parental
characteristics do not match any of the four case types described. Decision-making for
these families is therefore more ambiguous. Third, this theoretical model has been
proposed for cases of father-child sexual abuse only. It is not clear whether such a
model might be adapted to cases of physical abuse or neglect, or to cases of sexual abuse
by a non-father perpetrator. Finally, though her model is based on clinical experience,
it has yet to be empirically tested.
Several researchers have conducted analogue studies with child protection workers
or have analyzed actual case disposition data in an attempt to empirically describe aspects
of the decision making-process in cases of child physical abuse or neglect. Such cases
are reviewed below.


15
Empirical Studies of Decision-Making
Meddin (1984) conducted an analogue study of criteria used by child protection
workers to determine placement decisions in cases of abuse and/or neglect. She surveyed
81 child welfare workers using a questionnaire which described six situations containing
two instances each of physical abuse, sexual abuse, and neglect. Respondents were
provided with information on eight variables: age, gender, and race of the child; the
childs experience of previous placements; the severity of the current incident; risk to the
child of further abuse and/or neglect; and the cooperation and functioning of the primary
caretaker. For each situation, respondents indicated whether they would place the child
in foster care; they then described what variables influenced their decisions.
Results of Meddins study indicated that the primary variable used to determine
placement of the child was risk of further abuse and/or neglect. Such a finding seems
of dubious value, given that caseworkers are rarely, if ever, provided with that
information directly. More often, assessment of risk to the child is a part of the
caseworkers decision-making process.
Meddin did, however, note that four other variables clustered around risk to the
child, and she suggested that these variables might serve as operational definitions or
indicators of risk. The four variables were severity of the current incident, functioning
and cooperation of the primary caretaker, and age of the child.
While Meddins study is useful in that it indicates several variables which seem
to be important in caseworkers placement decisions, it does have several limitations.
In actual case situations, child protection workers typically have a great deal more


16
information than the eight variables Meddin provided. It is impossible to evaluate the
importance of this additional information using the analogue research design employed
here. Additionally, there is the possibility that the caseworkers actual decisions differ
from what they report in a hypothetical situation.
Several empirical studies of actual case dispositions have been conducted in an
attempt to overcome such limitations. Runyan, Gould, Trost, and Loda (1981) examined
the records of the North Carolina Registry of Child Abuse and Neglect to determine
which social, family, and child characteristics were most influential in the decision to
place a child in foster care. According to Runyan et al., these records contained all
theoretically relevant factors as well as demographic information.
Results of the study indicated that several child and family characteristics put
children at increased risk for placement. These variables included parents who perceive
severe punishment as acceptable, parents with substance abuse problems, severe child
injuries, child abandonment, and several other specific types of abuse. Age of the child
or parent, the presence of several maltreated children in one family, and several other
specific types of abuse were found to be not significant. Race, income, and education
were also nonsignificant, suggesting that there is no systematic bias in placement
decisions. Two non-family variables were also found to be associated with a higher
chance of placementcases referred by courts or police and geographic area.
Runyan et al. also employed a logistic regression model to predict placement in
foster care given all the above variables. They then compared this to actual placement.
Results of this procedure did not strongly associate the above characteristics with removal


17
from the home. In fact, the variance explained by the model was only 0.168. Runyan
et al. interpreted this result to mean that assignment to foster care approximates a random
process across a large population.
An alternative explanation for such a finding may be that child protection workers
use different criteria for making placement decisions for different types of abuse cases
(i.e., sexual abuse, physical abuse, or neglect). Since these data were collapsed across
type of abuse, such differences may have canceled each other out.
Katz, Hampton, Newberger, Bowles, and Snyder (1986) examined factors that
influence the decision to remove children from their parents care in cases of abuse
(physical or sexual) versus neglect (nonphysical injury). These researchers examined the
records of 185 suspected abused or neglected children at Childrens Hospital, Boston, in
an attempt to reveal how demographic characteristics, family history, family stress, the
nature of the injury, and aspects of the medical encounter influence the outcome of the
case.
Interestingly, the results of the study indicated that children with nonphysical
injuries were more likely to be removed from their parents care than were those with
physical injuries. The authors suggested several possible explanations for this finding,
including the possibility that neglect may be perceived by clinicians as evidence of a
chronic family problem rather than a single mishap.
Katz et al. also conducted Chi-square analyses to test the influence of other
independent variables on placement, although these analyses were collapsed across cases
of abuse and neglect. Their findings indicated that the following factors predicted


18
removal from the home: Medicaid-eligible families, previous reports of suspected
maltreatment, and suspicion of the mother being involved in the maltreatment. Minority
families, families in which the child was under age six, and families that were
experiencing more than one stressor were more likely to have their children returned.
Surprisingly, severity of condition was not significantly associated with outcome.
The results of this study yielded several interesting and even counterintuitive
findings. Again, however, because different categories of abuse were not considered
separately, the possibility of overgeneralization exists.
Finkelhor (1983) analyzed data on 6,096 cases of sexual abuse which were
officially reported in the U.S. in 1978 in an attempt to evaluate how the cases were
handled. His specific aim was to determine whether children were being unnecessarily
removed from their homes. Such an approach seems useful in that his consideration of
sexual abuse cases alone minimizes the risk of overgeneralization or of having significant
effects washed out by differences among groups.
Finkelhor compared the overall placement rate of sexual abuse cases to the
placement rate of physical abuse cases in the National Clearinghouse sample. His
findings indicated that foster care placement occurs significantly more often in cases of
sexual abuse, which led him to hypothesize that sexual abuse results in more outrage than
does physical abuse, regardless of the degree of danger to the child.
Finkelhors main finding of the study was that foster care does not seem to be
used indiscriminantly in cases of sexual abuse. For example, the most important
predictor of foster care placement was whether or not the victim initially reported the


19
abuse. A preference to be placed outside the home was more likely to be met than a
preference to stay. Also, foster care is more likely to be used in the case of an older
child than a younger child. This seems reasonable given that older children are typically
better able to adjust to living without their families.
Other factors predisposing to placement in foster care included abuse committed
by parents (especially if both parents were involved), very large families, the presence
of other types of abuse or neglect, and the presence of several other family stressors,
such as spouse abuse, substance abuse, family discord, and mental health problems.
Such findings led Finkelhor to conclude that "the characteristics of the children who are
placed in foster care do suggest some rational relationship between the disposition and
the facts of the case" (p. 199).
Finkelhors data also suggest that foster care is not used indiscriminantly in
"vulnerable" families. For example, he found that foster care placement is not more
likely to occur with poor families. Further, black children were not more likely to be
removed from their homes than were white children Thus, according to Finkelhors
findings, there does not seem to be any systematic bias in the decision-making process.
The strengths of Finkelhors study include its large N, its consideration of a large
number of potentially relevant factors, and its statistical analysis of only one type of
abuse. This methodology does not, however, indicate whether any of these factors
cluster together to form a meaningful aggregate of predictive characteristics. Also,
similar information regarding cases of child physical abuse and of neglect is needed.


RESEARCH OBJECTIVES AND HYPOTHESES
The current study seeks to extend our knowledge of the determinants of case
disposition by considering the influence of potential factors in cases of physical abuse and
in cases of neglect. Guardian ad Litem recommendations for placement will be
considered as well as final judicial decisions, in order to determine whether significant
differences exist in the decision-making processes of the two groups. Judicial decisions
will also be compared to Guardian recommendations to determine the extent to which
judges rely on those recommendations.
Based on review of the available literature, it is hypothesized that several groups
of variables will predict Guardians and judges decisions to place the child outside the
home in cases of child physical abuse and neglect. Among demographic variables, it
is hypothesized that increased age of the child will predict placement outside of the
home, but that family race and income will not be related to case disposition.
A second category of variables hypothesized to be related to case disposition
includes those which are descriptive of the abuse situation. Specifically, it is proposed
that abuse perpetrated by parents, very severe abuse, the presence of other types of
abuse, and previous reports of abuse will predict placement of the child outside the
home.
A third category of variables includes those that describe the psychological state
of the family. It is specifically hypothesized that poorer parent psychological
20


21
functioning, high degree of parental noncompliance with the investigation, the presence
of spouse abuse and substance abuse, high degree of family discord, and low degree of
parental warmth and support for the child will all predict placement outside the home.


METHOD
Subjects
Subjects for this study included 94 families involved in child physical abuse or
neglect cases in the state of Florida. All families were selected from adjudicated cases
in 12 judicial circuits. Only one abused or neglected child from each family was
considered in this study in order to avoid misleading duplication of family and case
variables. Child age ranged from 1 year to 17 years (mean =4 years, 9 months). Family
income ranged from $0 to $65,000 per year (mean=$8,661). Fifty-nine percent of the
families were white, 34% were black, 4% were Hispanic, and 3% were other races.
Measure
Data for this study were collected using a questionnaire designed by the team of
investigators to sample demographic, legal, and psychological factors relevant to the
family (see Appendix A). Demographic and legal information was of public record and
included the age, sex, and race of each family member; who the alleged perpetrator was;
family income; occupation and education level of each parent; whether or not the abuse
was substantiated; past history of abuse; whether criminal charges were brought against
the perpetrator; whether mental health evaluation was recommended and/or completed
for each family member; which placement options were recommended by the Guardian;
and which placement options were adjudicated. Placement options included the
following: (1) perpetrator removed from the home, (2) parents retain custody with
22


23
supervision, (3) temporary custody awarded to a relative, (4) temporary foster placement,
(5) child placement in a treatment facility, (6) placement for adoption, (7) child placed
in a detention facility, and (8) not dependent and returned home. A category designated
"other" was also provided in case final adjudication included a placement not described
above. For the purposes of this study, these options were collapsed to "returned home"
(options 1, 2, or 8) or "removed from the home" (options 3 through 7).
Psychological information for each case was provided by the Guardian, who made
judgments based on his or her interviews with the family. Judgments were made on
visual analog scales on the following issues: (1) who the non-abusing caretaker supports;
(2) severity of the abuse/neglect; (3) childs distress over the current family situation; (4)
conflict between the parents; (5) parents willingness to report family problems in
general; (6) amount the family favors family activities; (7) amount the family members
clarify and/or speak for other family members; (8) degree to which the physical needs
of the child were met (and who they were met by); and (9) degree to which the emotional
needs of the child were met (and whom they were met by).
Guardians also rated the following issues for each parent: (1) conflict between
parent and child; (2) warm and positive feelings for child; (3) involvement in childs
daily life; (4) reliance on the child; (5) psychological issues (loneliness, depression,
anxiety, irritability); (6) satisfaction with social support; and (7) response to the Guardian
in general. Guardians also indicated on visual analog scales their confidence in each of
their judgments.


24
Additional information was requested regarding whether the parents were abused
or neglected as children, whether the parents are accused of substance abuse and/or
spouse abuse, whether the parents have lost jobs or moved within the past year, whether
the parent or child suffers a mental or physical handicap, and whether the child exhibits
any behavioral problems.
Reliability. Guardians judgments on the psychological issues were compared to
those of licensed clinical psychologists and law students (Goldman et al., in preparation).
A videotape of a Guardian interviewing expert Guardians who were acting as family
members involved in a child abuse scenario was viewed by 22 Guardians, 17 clinical
psychologists, and 15 law students. The script for the mock interview was written by
the investigators in conjunction with two Guardian supervisors so that it would be
representative of an authentic family situation. After viewing the videotape, all subjects
completed the portion of the research questionnaire which involved psychological issues.
No significant differences were found between the Guardians, psychologists, and
lawyers judgments on any of the psychological issues, indicating that no specific bias
could be attributed to the subjects training or their area of expertise.
Test-retest reliability was also calculated for each of the 29 items rated. The 22
Guardians viewed the mock interview twice, at three week intervals. Of the variables
used in this study, Pearson correlations were significant for three variables at the p< .05
level (level of conflict, maternal and paternal warmth for child), for one variable at the
p<.01 level (maternal noncompliance), and for one variable at the p = .001 level
(severity of abuse/neglect). Pearson correlations were found to be non-significant for


25
paternal noncompliance and parental psychological functioning (depression, anxiety,
irritability, and loneliness). It should be noted, however, that the videotape viewed by
the Guardians contained relatively little information regarding the variables found to have
non-significant correlations, and it is therefore hypothesized that the low correlations
were due more to Guardians lack of confidence in their ratings than to instability in their
judgments over time.
Procedures
Data for each family were collected by the Guardian ad Litem (GAL) assigned
to that case. Each judicial circuit and each Guardian participated in this study on a
volunteer basis. Guardians are assigned cases on an individual basis by the GAL director
of each circuit. Each Guardian is appointed to a family when the case first becomes
active, and that Guardian represents the child until the case is adjudicated. For the
purpose of this study, Guardians collected data on each family when the case reached the
stage of adjudication. Thus, the cases used in this study represent those that were not
settled out of court and therefore may represent the most severe cases.


RESULTS
Comparison of Judges and Guardians Decisions
The cross-tabular display provided in Table 1 shows the frequency with which
judges agreed with the recommendation of the Guardians ad Litem. Results indicate that
the child was removed from the home by the judge in 78 cases and returned home in 16
cases. Guardians recommended removal from the home in 69 cases and recommended
returning the child to the home in 25 cases. Table 1 also shows that the judge agreed
with the Guardians recommendation (to remove the child from the home or to return the
child home) in 79 of the 94 cases. There were only 12 cases in which the judgebut
not the Guardianfelt that the child should be removed, and only 3 cases in which the
Guardianbut not the judgefelt that the child should be removed. The Kappa statistic
(K) was calculated to determine the degree of concurrence observed between judges and
Guardians, and the obtained value of .53 indicates good overall agreement (beyond
chance) between the two groups. It can therefore be concluded that while Guardians
appear to be slightly more conservative in their recommendations to remove children
from their homes, there was considerable agreement between the two groups regarding
placement of the child.
To further assess the relationship between Guardian recommendations and judicial
decisions, a logistic regression analysis was conducted using Guardian recommendation
as a predictor variable for judicial decision (See Table 2). Results indicated that the
26


27
Table 1. Comparison of Guardian Recommendations and Judicial
Decisions
Judicial Decision
Guardian Recommendation
Removed
Not Removed
Total
Removed
66 cases
12 cases
78 cases
(70%)
(13%)
(83%)
Not Removed
3 cases
13 cases
16 cases
(3%)
(14%)
(17%)
Total
69 cases
25 cases
94 cases
(73%)
(27%)
(100%)
K = .53
Table 2. Results of Logistic Regression Analysis Using Guardian
Recommendations to Predict Judicial Decisions
Predicting Judicial Decisions
Variable
DF
Parameter
Estimate
Standard
Error
Chi-
square
Probability of >
Chi-square
Intercept
1
0.08
0.40
0.04
0.84
Guardian
1
-3.17
0.71
19.77
0.00*
Recommendation
p = .0001
Hosmer-Lemeshow goodness-of-fit statistic = 6.24 with 8 DF, p = .62


28
Guardians recommendation for removal is significantly associated (p = .0001) with a
higher probability of judicial decision to remove the child from the home. Predicted
probabilities based on the parameter estimate obtained for the Guardian recommendation
variable indicate that the probability of removal is 96% in cases where the Guardian
recommends it and 48% in cases where the Guardian recommends against it. It should
be noted again, however, that these data are based on only 25 cases in which the
guardian recommended returning the child to the home. The Hosmer and Lemeshow
goodness-of-fit statistic for the model (6.24 with 8 DF, p = .62) indicates that the
observed responses do not significantly differ from responses which would be predicted
based on this model (Hosmer & Lemeshow, 1989).
The cases in which the judge disagreed with the Guardians recommendation for
placement of the child were examined further to determine whether any trends among the
predictor variables could be identified. Examination of the values of the predictor
variables in those cases, however, did not reveal any differences which might account
for the difference of opinion between the judges and the Guardians (see Table 3).
Rate of Removal by Abuse Type
It was also of interest to compare the rate of removal depending on the type of
abuse or neglect allegations present. The overall distribution of abuse types in this
sample (whether removed or not removed) indicated that 80% of the cases involved
neglect, 52% involved physical abuse, and 28% involved sexual abuse. Among the cases
in which removal occurred, 83% involved allegations of neglect, 50% involved
allegations of physical abuse, and 24% involved allegations of sexual abuse (see


29
Table 3.
Variable Values In Cases Where Judges and Guardians Disagreed
Regarding Case Outcome
J
G
C
F
R
P
S
M
P
s
P
W
C
N
S
S
u
U
H
A
A
A
E
T
R
E
S
A
0
O
U
P
D
A
1
M
C
R
V
Y
E
X
Y
R
N
N
B
A
G
R
A
I
E
P
E
P
V
A
F
M
F
C
A
B
E
D
G
N
E
R
E
R
B
U
L
O
B
U
I
E
C
R
I
S
E
U
N
M
U
s
A
P
T
P
S
C
P
S
E
N
Y
E
T
E
R
N
.
.
W
Y
79
N
N
N
.
37
5
42
N
N
R
N
2
0
W
Y
93
Y
N
Y
51
27
44
54
Y
N
R
N
.
12
w
Y
84
Y
N
Y
65
85
20
37
Y
N
R
N
3
0
N
Y
97
N
N
N
69
68
2
27
N
N
R
N
3
0
N
Y
91
N
Y
N
96
13
65
96
Y
N
R
N
5
0
N
Y
64
N
Y
N
25
47
16
5
Y
N
R
N
1
8
W
Y
81
Y
N
N
59
14
54
64
N
N
N
R
6
8
w
Y
89
Y
Y
Y
46
50
2
47
N
N
R
N
2
0
N
Y
86
N
N
N
.
68
46
48
Y
N
N
R
1
N
Y
82
N
N
N
.
99
0
0
N
N
R
N
2
W
Y
69
Y
N
N
63
47
45
21
N
Y
R
N
3
N
Y
94
N
N
N
88
21
.
18
Y
N
R
N
17
W
Y
85
Y
N
N
52
57
94
15
N
N
R
N
16
w
N
95
N
N
Y
.
19
90
14
N
N
N
tt it
R
4
20
N
Y
rrr
60
Ti vF
N
N
rrirm
N
54
81
78
33
Y
N
JUDGE=
GUARDIAN=
CH1AGE=
FAMINC =
RACE=
PARPERP=
SEVERITY =
MTYPES =
PREVREP =
SEXABUSE=
PSYFUNCT =
WARM =
CONFL =
NONCOMP =
SUBABUSE=
SPABUSE=
Adjudication (R = Child removed from home; N=Not removed)
Recommendation (R=Recommended removal; N=Did not)
Child age (in years)
Annual family income (in thousands of dollars)
Family race (W=White; N=Non-White)
Was the alleged perpetrator a parent?
Severity of abuse (99 = Highest possible severity level)
Were multiple types of abuse alleged?
Were there previous reports of abuse?
Was sexual abuse alleged?
Parental psychological functioning (99=Poorest possible level)
Parental warmth/support for child (99=Highest possible level)
Family conflict (99=Highest possible level)
Parental noncompliance (99=Highest possible level)
Was parental substance abuse alleged?
Was spouse abuse alleged?


30
Appendix C). Results also indicated that children were removed from their homes in
87% of the cases involving allegations of neglect, in 80% of the cases involving
allegations of physical abuse, and in 73% of the cases involving allegations of sexual
abuse. While these results are complicated by the fact that multiple types of abuse
occurred in 45% of the cases, they nonetheless indicate removal was most likely when
allegations of neglect were present and least likely when allegations of sexual abuse were
present.
Logistic Regression Equations
The primary purpose of this study was to determine which demographic variables,
abusive situation characteristics, and family psychological characteristics influence
Guardians and judges decisions to remove the child from the home in cases of child
abuse and/or neglect. Because the outcome variables (the judges placement of the child
or the Guardians recommendation for placement) and several predictor variables are
dichotomous, logistic regression was used. In order to test for the possibility of
misleading multicollinearity among the independent variables, correlation matrices were
obtained for each model. The correlations among the independent variables ranged from
.01 to .50, indicating that the results of the analyses were not compromised by significant
multicollinearity.
Two series (one for judges decisions and one for Guardians recommendations)
of three logistic regression models were runone model for each category of variables
(demographic variables, abusive situation characteristics, and family psychological
characteristics). The results of each of these models will be discussed separately.


31
Finally, two logistic regression equations including only the significant predictor
variables from the above models were testedone using judges decisions as a dependent
variable and including all independent variables found to significantly predict those
decisions and a second using Guardians decisions as a dependent variable and including
all independent variables found to significantly predict their decisions.
Demographic Variables
Because information regarding family income was provided in only 54 of the 94
cases, it was not possible to test the impact of this variable. Results of the logistic
regression equation including child age and family race as predictors indicated that
neither variable significantly impacted the decision to place the child outside the home
(see Table 4). The analysis using the Guardians recommendation as the dependent
variable yielded similar results, as neither child age nor family race emerged as a
significant predictor of a Guardians decision to recommend placement outside of the
home (see Table 5).
Abusive Situation Characteristics
Among variables related to aspects of the abusive or neglectful situation, it was
hypothesized that abuse perpetrated by parents, very severe abuse, the presence of
multiple types of abuse, and previous reports of abuse would predict placement outside
the home. It was not possible to test the influence of the "parent perpetrator" variable,
however, as 96% of the cases in this sample involved parent perpetrators. When the
remaining variables were included in the logistic regression equation using the judges


32
Table 4. Results of Logistic Regression Analysis Using Demographics to Predict
Judicial Decisions
Predicting Judicial Decisions
Demographic Variables
DF
Parameter
Estimate
Standard
Error
Chi-
Square
Probability of >
Chi-Square
Intercept
1
-1.88
0.55
11.90
0.00
Child Age
1
0.00
0.06
0.00
0.96
Family Race
1
0.38
0.59
0.41
0.52
Table 5. Results of Logistic Regression Analysis Using Demographic Variables to
Predict Guardian Recommendations
Predicting Guardian Recommendations
Demographic Variables
DF
Parameter
Estimate
Standard
Error
Chi-
Square
Probability of >
Chi-Square
Intercept
1
-1.50
0.47
10.36
0.00
Child Age
1
0.03
0.06
0.36
0.55
Family Race
1
0.34
0.51
0.44
0.51


33
decision as a dependent variable, none emerged as significantly related to removal of the
child from the parents home (see Table 6). In the analysis using the Guardians
recommendation as a dependent variable, only the presence of previous reports of abuse
(p<.05) emerged as significantly related to recommendation for placement outside the
home (see Table 7). In this model, the negative parameter estimate indicates that the
absence of previous reports of abuse or neglect (which was treated first in the model) is
associated with a smaller probability of removing the child from the home. Guardians
were not more likely to recommend removal in cases in which the abuse was very severe
or in which multiple types of abuse occurred.
Family Psychological Characteristics
Several variables describing the psychological state of the family were
hypothesized to affect judges decision-making in cases of abuse or neglect. Specifically,
it was proposed that poorer parent psychological functioning, high degree of parental
noncompliance with the investigation, the presence of spouse abuse and substance abuse,
high degree of family discord, and low degree of parental warmth and support for the
child would predict placement of the child outside of the home. However, only the
degree of the parents warmth and support for the child (p < .01) emerged as significantly
related to judges decision-making when the variables were simultaneously entered into
the prediction equation (see Table 8). The positive parameter estimate indicates that
lower levels of parental warmth and support (which were treated first in the model) were
associated with a larger probability of having the child removed from the home.


34
Table 6. Results of Logistic Regression Analysis Using Legal and Case Variables to
Predict Judicial Decisions
Predicting Judicial Decisions
Legal and Case Variables
DF
Parameter
Estimate
Standard
Error
Chi-
Square
Probability of >
Chi-Square
Intercept
1
-0.47
0.89
0.27
0.60
Severity of Abuse
1
-0.15
0.01
1.46
0.23
Multiple Abuse Types
1
0.19
0.59
0.10
0.75
Previous Reports
1
-0.25
0.57
0.19
0.66
Table 7. Results of Logistic Regression Analysis Using Legal and Case Variables to
Predict Guardian Recommendations
Predicting Guardian Recommendations
Legal and Case Variables
DF
Parameter
Estimate
Standard
Error
Chi-
Square
Probability of >
Chi-Square
Intercept
1
-0.88
0.89
0.97
0.32
Severity of Abuse
1
-0.00
0.01
0.09
0.76
Multiple Abuse Types
1
0.15
0.51
0.09
0.77
Previous Reports
1
-1.08
0.51
4.46
0.03*
* p < .05


35
The results of the analysis using the Guardians recommendation as an
independent variable yielded similar results (see Table 9). Like judges, Guardians were
more likely to recommend removal of children from homes in which parents showed
little warmth and support for them (p< .05). Interestingly, however, none of the other
variables regarding family psychological functioning significantly impacted the
Guardians decision-making. It therefore appears that neither judges nor Guardians rely
heavily on information regarding the psychological functioning of the family when
making decisions regarding the placement of the abused child.
Models Including All Significant Predictors
In order to determine the best overall predictive model of judicial decision
making, a logistic regression equation was run which included only the predictor variable
found to significantly impact judicial decisions from the above modelslevel of parental
warmth and support. Again, the positive parameter estimate obtained for the predictor
variable in this model indicates that lower degrees of warmth and support were more
associated with removal from the home. Calculation of predicted probabilities based on
the obtained parameter estimates indicated that the probability of removal was 98% in
cases where parental warmth and support was rated 0 (the lowest possible level) and 50%
in cases where parental warmth and support was rated 99 (the highest possible level).
The Hosmer-Lemeshow goodness-of-fit statistic (7.90 with 8 DF, p = .44) indicates that
the observed responses do not differ significantly from those which would occur based
on the predicted probabilities from the model (see Table 10).


36
Table 8. Results of Logistic Regression Analysis Using Family Psychological
Variables to Predict Judicial Decisions
Family Psychological
Variables
Predicting Judicial Decisions
DF
Parameter
Estimate
Standard
Error
Chi-
Square
Probability of >
Chi-Square
Intercept
1
-4.79
1.76
7.41
0.01
Psychological Functioning
1
0.01
0.02
0.11
0.74
Parental Warmth and Support
for Child
1
0.04
0.02
8.14
0.00*
Family Conflict
1
0.00
0.01
0.00
0.96
Parental Noncompliance
1
0.02
0.01
2.32
0.13
Substance Abuse
1
-0.17
0.71
0.06
0.81
Spouse Abuse
1
-0.78
0.87
0.81
0.37
* p < .01
Table 9. Results of Logistic Regression Analysis Using Family Psychological
Variables to Predict Guardian Recommendations
Family Psychological
Variables
Predicting Guardian Recommendations
DF
Parameter
Estimate
Standard
Error
Chi-
Square
Probability of >
Chi-Square
Intercept
1
-3.18
1.39
5.21
0.02
Psychological Functioning
1
0.02
0.02
1.18
0.28
Parental Warmth and Support
for Child
1
0.03
0.01
4.93
0.03*
Family Conflict
1
-0.00
0.01
0.06
0.80
Parental Noncompliance
1
0.01
0.01
0.99
0.32
Substance Abuse
1
-0.42
0.60
0.49
0.49
Spouse Abuse
1
-0.80
0.72
1.23
0.27
p < .05


37
A similar analysis was conducted using the independent variables found to
significantly impact the decisions of the Guardians. When the variables of previous
reports of abuse and parental warmth and support for the child were entered into the
equation, only the presence of previous reports of abuse (p = .05) emerged as strongly
related to Guardians decisions. Since inclusion of the variable quantifying parental
warmth and support for the child did not significantly add to the predictive power of the
equation, the analysis was conducted again using only the presence of previous reports
as a predictor variable. The negative parameter estimate indicates that the presence of
previous reports of abuse was associated with a higher probability of making a
recommendation for removal. The Hosmer-Lemeshow goodness-of-fit statistic (9.00 with
8 DF, p = .34) indicates that the observed responses do not significantly differ from those
which would occur given the predicted probabilities based on this model (see Table 11).
Comparison of Cases of Physical Abuse Only and Neglect Only
An original aim of this study was to determine whether judicial and Guardian
decision-making processes differed between cases of physical abuse alone and neglect
alone. Because the sample included only 11 cases solely involving physical abuse and
36 cases solely involving neglect, it was not possible to analyze the data using logistic
regression equationswhich require a larger sample size to be interpretively useful.
Analysis using the Shapiro-Wilk statistic (W) indicated that all interval independent
variables were normally distributed in the samples. The means of these variables were
compared using t-tests in an attempt to determine whether potentially important
differences exist among demographic, legal, or family psychological characteristics


38
Table 10. Results of Logistic Regression Analysis Using All Significant Predictor
Variables to Predict Judicial Decisions
Predicting Judicial Decisions
All Significant Predictors
DF
Parameter
Estimate
Standard
Error
Chi-
Square
Probability of >
Chi-Square
Intercept
1
-3.73
0.86
18.87
0.00
Parental Warmth and
Support for Child
1
0.04
0.01
9.33
0.00*
*p < .01
Hosmer-Lemeshow goodness-of-fit statistic=7.9 with 8 DF, p = .44
Table 11. Results of Logistic Regression Analysis Using All Significant Predictor
Variables to Predict Guardian Recommendations
Predicting Guardian Recommendations
All Significant Predictors
DF
Parameter
Estimate
Standard
Error
Chi-
Square
Probability of >
Chi-Square
Intercept
1
-0.57
0.30
3.50
0.06
Previous Reports of Abuse
1
-1.02
0.49
4.25
0.04*
* p < .05
Hosmer-Lemeshow goodness-of-fit statistic=9.00 with 8 DF, p = .34


39
between the two groups (see Appendix D). In addition, the frequencies of the
dichotomous independent and dependent variables in each group were reviewed, although
the number of cases was insufficient to support Chi-square analysis.
Results of comparisons between cases of physical abuse alone and neglect alone
using t-tests indicated that the level of severity of neglect tended to be more extreme than
the level of severity of abuse (t=-2.55, p = .02) and that there tended to be less conflict
between parents of neglected children than between parents of abused children (t=2.09,
p = .04). No significant differences between the two groups were found in child age,
family income, level of parental psychological functioning, degree of parental warmth
and support for the child, or parental noncompliance with the investigation.
Visual review of the frequencies of dichotomous variables in the two groups
revealed the following: (1) Judges were somewhat more likely to remove neglected
children from their homes (89% of neglected children as opposed to 73% of abused
children), (2) Guardians were less likely to recommend removal of neglected children
than abused children (75% of neglected children as opposed to 82% of abused children),
(3) Among neglect cases, there tended to be fewer cases involving previous reports of
abuse or neglect against the child (47% as opposed to 57% in cases of physical abuse),
(4) In cases of neglect alone, there were fewer cases involving spouse abuse (14% as
opposed to 27% in families of abused children), and (5) Reports of substance abuse were
more frequent among parents of neglected children (72% as opposed to 55% in cases of
physical abuse). Family race and the number of non-parent alleged perpetrators did not
differ significantly between the two groups.


40
Because these findings are based on very small samples, caution must be used in
their interpretation. The number and magnitude of the differences found, however,
indicate the possibility of substantive differences between the characteristics of cases of
physical abuse only and neglect only. It thus logically follows that important differences
may exist in the process of determining appropriate adjudication for the two different
groups.


DISCUSSION
The results of this study provide descriptive information about cases involving
allegations of child abuse and/or neglect cases which are served by the GAL program in
the state of Florida. Such descriptive information about the cases included in the current
study is provided to establish a framework for interpreting and understanding the results
which were obtained. The distribution of demographic, abusive situation, and family
psychological characteristics of the sample are presented in Tables 12 and 13. As noted
above, 59% of the families in the current study were white and 41% were African
American, Hispanic, and other races (34%, 4%, and 3%, respectively). Of the 56 cases
in which information on family income was provided, the mean income was $8,6660 per
year, suggesting that the majority of the families live in poverty.
All interval variables related to the abusive situation and to the familys
psychological functioning in the study were rated on visual analog scales with values
ranging from 1 to 99. The average level of severity of abuse was 76 (with 99 being the
most severe), indicating that these cases generally represent the most severe incidents of
abuse and neglect. At least one of the parents was the alleged perpetrator in 96% of the
cases, and there existed previous reports of abuse in 50% of the cases. Multiple types
of abuse were reported in 45 % of the cases, and sexual abuse allegations were present
(either alone or in combination with abuse and/or neglect) in 28% of the cases.
41


42
Table 12. Distribution of Interval Variables
Variable
N
Mean
S.D.
Median
Child Age
90
4.77
4.46
3.00
Family Income
56
8.66
12.61
5.50
Severity of Abuse
94
76.05
21.85
84.00
Parental Psychological
Functioning
86
56.04
20.98
54.25
Parental Warmth and
Support for Child
94
47.33
29.42
48.50
Family Conflict
86
46.53
31.15
47.00
Parental Noncompliance
94
37.12
26.20
33.25
Table 13. Frequencies of Dichotomous Variables
Variable
Family Race
Parent Perpetrator
Multiple Types of Abuse
Previous Reports of Abuse
Sexual Abuse
Spouse Abuse
Substance Abuse
Frequency Percent
White
55
58
Non-White
39
42
Alleged
90
96
Not Alleged
4
4
Alleged
42
45
Not Alleged
52
55
Yes
47
50
No
47
50
Alleged
26
28
Not Alleged
68
72
Alleged
20
21
Not Alleged
74
79
Alleged
61
65
Not Alleged
33
35


43
There were allegations of spouse abuse (in addition to child abuse) in 21 % of the
cases, and allegations of substance abuse were made in 65% of the cases. The average
rating for overall psychological functioning of the family was 56 (with 99 being the
poorest possible level of psychological functioning), and the average rating for parental
warmth and support for the child was 47 (with 99 being the highest possible level of
warmth and support). The average conflict level reported for the families was 47 (with
99 being the highest possible level of conflict), and the average degree of parental
noncompliance with the investigation was 37 (with 99 being the highest possible level of
noncompliance).
One of the major aims of this study was to determine the frequency with which
judges agree with Guardians ad Litem regarding case outcome. Cross-tabular display of
Guardian recommendation and actual case outcome indicated that the two groups agreed
in 79 of the 94 cases, and the positive Kappa statistic indicated that this level of
agreement is beyond what would be expected by chance alone. In addition, results of
a logistic regression analysis indicated that Guardian recommendation for removal is
significantly associated with a higher probability of judicial decision to remove the child
from the home. The results indicate, in fact, that the probability of a judicial decision
for removal is 96% when the Guardian has recommended that the child be removed.
The main objective of the current study was to determine which demographic,
legal, and family psychological variables are most predictive of decisions to remove (or
recommend removal of) children from their homes in cases involving child abuse and
neglect. It is important at this point to note again that all cases included in this study had


44
reached the point of adjudication and therefore likely represent the most severe cases to
enter into the judicial system. Consideration of this population is felt to be important,
however, as it is these cases in which judges are most likely to render decisions.
Interestingly, very few of the variables assessed were found to be significantly associated
with judicial and Guardian decision-making.
In addition, while judges were quite likely to agree with the recommendations of
Guardians, their decisions were significantly associated with somewhat different
information from the Guardians. Review of the cases in which judges and Guardians
disagreed did not reveal any trends which would account for such differences. It is
therefore possible that this discrepancy is due either to chance or to the influence of
another factor (or factors) which was not accounted for in this study.
Overall, information regarding the degree of parental warmth and support for the
child was most predictive of final case outcome. The results of the separate analyses
using Guardians recommendations as the dependent variable indicated that the presence
of previous reports of abuse and on the degree of parental warmth and support for the
child are significantly associated with those recommendations. The results of a logistic
regression equation using only these predictor variables, however, indicated that only the
presence of previous reports of abuse or neglect against the child is significantly related
to Guardians recommendations for removal. Thus, at least in this sample, recurrent
abuse against a child is the single most predictive factor of Guardians recommendation
that he or she be removed from the home.


45
The results of previous studies have indicated that judges are more likely to
remove older children from the home than they are to remove younger children (e.g.,
Katz et al., 1986). In this sample, however, child age had no discernible effect on either
judges or Guardians decisions. Previous studies have also found that the race of the
family is not significantly associated with decisions of judges. The results of the current
study lend support to that finding, suggesting that racial discrimination on the part of
court personnel is not present in the process of adjudicating cases of child abuse and
neglect.
Several legal variables hypothesized to be related to judicial and Guardian ad
Litem decision-making were not found to significantly predict the outcome of the cases
included in this study. Neither judges nor Guardians decisions were significantly
associated with the fact that the abuse was very severe or on the fact that multiple types
of abuse were alleged. Interestingly, while the presence of previous reports of abuse was
found to be the most significant predictor of Guardians decisions to recommend removal
of children from their homes, this information was not significantly related to judicial
decision-making. Among the variables relating to the psychological functioning of the
family, only the level of the parents warmth and support for the child was found to be
significantly associated with decisions regarding case outcome. Information regarding
the psychological functioning of the parents, the parents compliance with the
investigation, the degree of family discord, or the presence of allegations of spouse or
substance abuse among the parents was not related to judges or Guardians decisions.


46
One source of concern which arises from the results of the current study is the
lack of association between judges and Guardians decisions and the severity level of the
alleged abuse. Florida Statute section 39.464 stipulates that grounds for termination of
parental rights include "egregious abuse," which is defined as "conduct that endangers
the life, health, or safety of the child." Yet visual analysis of the data indicated that a
number of children were returned home in cases where the severity of the abuse was
judged to be very high. If one assumes that children who are very severely abused are
at risk for greater injury in the future (if not at greater risk for more frequent future
abuse), this finding reflects a potentially serious oversight on the part of court personnel
involved in dependency case decision-making. In addition, there was little association
between judges decisions and the presence or absence of previous reports of abuse or
neglect against the childdespite the fact that Florida Statute section 39.464 stipulates
that "continuing abuse or neglect" is grounds for termination of parental rights.
Another area of concern is the lack of association between judicial and Guardian
decisions and information regarding the psychological and interpersonal functioning of
the family. The results obtained here differ from those obtained by Meddin (1984) in
her study of child protective direct service workers and supervisors. Subjects in that
study were found to treat the functioning and cooperation of the parents as important
indicators of future risk to the child. Despite considerable evidence that psychological
disturbance of parents (particularly depression), aberrant parenting, and marital discord
frequently lead to abuse and neglect of children and subsequently to poor adjustment of


47
those children (e.g., Lynch and Roberts, 1982; Martin, 1979; Green, 1976), factors such
as these were not found to predict decisions to remove children from their homes.
Review of Florida Statute section 39.467 provides a likely explanation for the
above finding. Judges are required to consider "the love, affection, and other emotional
ties existing between the child and the childs present parent or siblings"which they
appear to do, given the significant association between the degree of parental warmth and
support for the child and final adjudication. However, while judges are instructed to
consider "the capacity of the parent or parents to care for the child to the extent that the
childs health and well-being will not be endangered upon the childs return home," there
is no specific requirement to consider the psychological functioning of the parents.
Again, given our current knowledge about the detrimental effects of poor parent
psychological functioning on the well-being of children, failure to consider these factors
when determining whether to remove children from their homes would represent a gross
oversight.
Results of comparison of the variables among groups of children who were
physically abused (but not neglected or sexually abused) and those who were neglected
(but not physically or sexually abused) also indicated several potentially important
differences in both case type and outcomemost notably that while Guardians are
somewhat less likely to recommend removal of children in cases of neglect alone, judges
are somewhat more likely to adjudicate removal. Such a finding raises questions
regarding whether the decision-making processes of judges and Guardians differ when
allegations include different types or combinations of abuse or neglect. As a result, it


48
is recommended that future research consider groups of physically abused, sexually
abuse, and neglected children separately rather than risk "washing out" of potential
differences among them.
The current study has identified a number of variables important to judges and
Guardians in determining custody of abused and/or neglected children, as well as a
number of issues which may warrant more attention on the parts of these two groups.
There are, however, several methodological limitations present in this study which
deserve consideration. First, there are limitations which concern the generalizability of
the results obtained here. Despite considerable effort, the researchers were unable to
collect data on every case referred to the GAL program during the period of study.
Some of the judicial circuits in the state of Florida were more willing to cooperate than
were others, and even the most cooperative were able to provide data only from
Guardians willing to complete our forms on a volunteer basis. Our sample is therefore
not random and may not accurately represent the population of GAL referrals in the state
of Florida. Additionally, the current results are based on a relatively small sample, and
some of our findings may therefore be sample specific. Caution must thus be used when
generalizing the results obtained in this study to other populations. In addition, the
sample used in this study did not include a wide distribution of family income levels or
child ages. Because the sample included mostly very low income families with young
children, generalization of these findings to populations with different demographic
characteristics is not advised, as judges and Guardians may well weigh information
differently in cases involving wealthy families or those with older children.


49
While federal laws regarding cases involving allegations of child abuse or neglect
provide some degree of uniformity in determining case disposition, it may be that some
states have legal guidelines and political climates which differ from the state of Florida.
It is therefore recommended that the results obtained here not be generalized to other
geographic regions prior to further study in those areas. Another aspect of the current
sample which deserves consideration is that the cases involved both initial dependency
hearings and termination of parental rights hearings. Thus, some of the children were
removed from their homes on a temporary basis, while others were permanently
removed. Because court personnel may use somewhat different criteria when deciding
whether parental rights should be severed, it is recommended that additional research be
conducted on these two groups independently of each other.
Future research is clearly needed to assess the generalizability of the current
findings to other populations. It is also recommended that a similar analysis be
conducted using a larger sample of a comparable population in order to replicate these
findings and to determine whether the process of Guardian ad Litem and judicial
decision-making changes over time.
Further research is also needed to clarify the results obtained in the current study.
One rather puzzling finding of this study was that while judges tend to agree with the
recommendations of Guardians, they appear to place a somewhat different emphasis on
the variables they use to determine the final adjudication on the cases they consider.
This finding suggests that while judges rely on case information provided to them by the
Guardians, they may not rely heavily on the Guardians recommendations. Future


50
research using Guardian recommendation as an independent variable to predict judicial
decisions is thus warranted to further clarify this situation. In addition, it is
recommended that future studies be conducted which consider abused children and
neglected children as separate groups in order to determine whether important differences
exist in the decision-making of judges and Guardians between the two groups.
Another area requiring further research is the lack of consideration given to
psychological aspects of the family in making decisions regarding removing children
from their homes. Potential explanations for this finding include the possibility that
judges and Guardians are unaware of the importance of such factors on the short- and
long-term well-being of the children involved, that they are not adequately educated
about how to assess the information, and/or that they do not feel that the consideration
of such information is warranted under the legal guidelines for determining case
disposition. Future research aimed at elucidating these or other reasons is therefore
suggested.


REFERENCES
Alter, C.F. (1985). Decision-making factors in cases of child neglect. Child Welfare.
64, 99-111.
Brown, A., & Finkelhor, D. (1986). Impact of child sexual abuse: A review of the
research. Psychological Bulletin. 99, 66-77.
Daro, D. (1988). Confronting child abuse: Research for effective program design.
New York: The Free Press.
Eckenrode, J., Powers, J., Doris, L, Munsch, J., & Bolger, N. (1988). Substantiation
of child abuse and neglect reports. Journal of Consulting and Clinical
Psychology. 56, 9-16.
Eskin, M. (1980). Child abuse and neglect: A literature review and selected biblio
graphy. Washington, DC: U.S. Department of Justice.
Faller, K.C. (1988). Decision-making in cases of intrafamilial child sexual abuse.
American Journal of Orthopsychiatry. 58, 121-128.
Finkelhor, D. (1983). Removing the childprosecuting the offender in cases of sex
ual abuse: Evidence from the National Reporting System for Child Abuse and
Neglect. Child Abuse and Neglect. 7, 195-205.
Finkelhor, D. (1984). Child sexual abuse: New theory and research. New York:
Free Press.
Goldman, J., Albanese, I., Graves, L., McDanal, C.S., Sorenson, E., & Ward, M. (in
preparation). Reliability of Guardian ad Litem ratings on psychological issues in
cases of child abuse and neglect.
Green, A. (1976). A psychodynamic approach to the study and treatment of child-
abusing parents. Journal of Child Psychiatry. 15. 414-429.
Houck, G.M., & King, M.C. (1989). Child maltreatment: Family characteristics and
developmental consequences. Issues in Mental Health Nursing. 10, 193-208.
51


52
Hosmer, D.W, & Lemeshow, S. (1989). Applied Logistic Regression. New York:
John Wiley & Sons.
Katz, M.H., Hampton, R.L., Newberger, E.H., Bowles, R.T., & Snyder, J.C. (1986).
Returning children home: Clinical decision-making in cases of child abuse and
neglect. American Journal of Orthopsychiatry. 56, 253-262.
Knudsen, D.D. (1988). Child protective services: Discretion, decisions, dilemmas.
Springfield, IL: Charles C. Thomas.
Lamphear, V.S. (1985). The impact of maltreatment on childrens psychosocial
adjustment: A review of the research. Child Abuse and Neglect. 9, 251-263.
Landau, H.R. (1982). The abused and neglected child. Practising Law Institute.
Lynch, M.A., & Roberts, J. (1982). Consequences of Child Abuse. London, England:
Academic Press.
Martin, H.P. (1979). Child abuse and child development. Child Abuse and Neglect.
3, 415-421.
Meddin, B.J. (1984). Criteria for placement decisions in protective services. Child
Welfare. 63, 367-373.
National Center on Child Abuse and Neglect (1988). Study findings: Study of na
tional incidence and prevalence of child abuse and neglect. Washington, DC:
U.S. Department of Health and Human Services.
Radbill, S.X. (1968). A history of child abuse and infanticide. In B.E. Heifer & C.H.
Kempe (eds.), The battered child. Chicago: University of Chicago Press.
Runyan, D.K., Gould, C.L., Trost, D.C., & Loda, F.A. (1981). Determinants of foster
care placement for the maltreated child. American Journal of Public Health. 71,
706-711.
Russell, D.E.H. (1983). The incidence and prevalence of intrafamilial and extra-
familial sexual abuse of female children. Child Abuse and Neglect. 7, 133-146.
Sloan, I.J. (1983). Child abuse: Governing law and legislation. New York: Oceana
Publications, Inc.
Strauss, M.A., & Gelles, R.J. (1986). Societal change and change in family violence
from 1975 to 1985 as revealed by two national surveys. Journal of Marriage and
the Family. 48, 465-479.


53
Wasserman, S., & Rosenfeld, A. (1986). Decision-making in child abuse and neglect.
Child Welfare. 65, 515-529.
Whitcomb, D. (1988). Guardians Ad Litem in criminal courts. Washington: U.S.
Department of Justice.
Wiig, J.K. (1982). Functions of the Guardian ad Litem in child abuse and neglect
proceedingsLos Angeles Juvenile Court. In H. Landau (ed.), The abused and
neglected child. Practising Law Institute.


APPENDIX A
GAL QUESTIONNAIRE FOR DEPENDENCY CASES
Thank you for your time and support of our efforts to collect the following
information on dependency cases. Our goal is to provide information which will aid
the GAL program in its role as the advocate for children in dependency cases.
The questionnaire is divided into three broad sections. This first section deals with
basic demographic and legal information of your case. This information is similar to
that found in your case records.
The second sections requests information on individual and family psychological
issues. These questions may be easily answered after your interviews with the childs
family. This section requires that you mark on a line to indicate your rating of the
intrafamilial relationships. Some of these questions require a single mark on the line,
such as for conflict between the two parents:
CONFLICT BETWEEN THE TWO PARENTS:
Very Little Very Much
While other questions require you to make two ratings on each line, one for the
mother-child relationship and one for the father-child relationship:
CONFLICT BETWEEN EACH PARENT AND THE CHILD(ren) who was
reportedly abused or neglected, with an M, F, SM, and an SF.
M F
Very Little Very Much
Finally, the third section surveys information from the disposition hearing on the case
in terms of your recommendations to the court and the courts final determination.
Please return this form when completed to your GAL programs secretary.
Thank you again for your help.
54


55
Dependency Case
DEMOGRAPHIC AND LEGAL INFORMATION
Who is filling out this form? Guardian Coordinator/Director Other
Date of this hearing? Judge Code:
Circuit: County:
Circle any immediate family members who have not been interviewed by the
GAL: (Circle all that apply)
Mother Father Child Other (specify)
Stepmother Stepfather Grandparent
Are the parents represented by counsel: Yes No
Report for all family members living with the child victim (and the alleged
perpetrator if he/she does not live with the child) the following information:
Race: C=caucasian, H=hispanic, B=black, 0=other
V/P: V=victim of abuse/neglect, P=alleged perpetrator
Relation: (to the victim(s) P=parent, SP = step parent, GP=grandparent, S = sibling,
BF=boy/girlfriend of the parent, FP = foster parent, AP=adoptive parent,
ST=stranger, 0=other
AGE SEX RACE V/P RELATION
1.
2.
3.
4.
5.
6.
For each parent report the following information:
Yearly Years of
Income Education Age Occupation
MOTHER
FATHER
STEPMOTHER
STEPFATHER


56
Has one or more of the parents lost their jobs within the last year? Yes No
Has the family moved within the last year? Yes No
Circle the one category into which the reported child victim falls:
Born Prematurely Physically Handicapped Mentally Retarded
Neither Unknown
Which parent(s) may have difficulty caring for their child(ren) due to a physical
handicap, serious illness, or injury?
Mother Stepmother Father Stepfather Neither
For the following questions, place a check mark in the box corresponding to any
alleged or substantiated problem for any parent(s):
ALLEGATIONS SUBSTANTIATED
M
0
T
H
E
R
S M
T 0
E T
P H
E
R
F
A
T
H
E
R
S F
T A
E T
P H
E
R
N
E
I
T
H
E
R
M
0
T
H
E
R
S M
T 0
E T
P H
E
R
F
A
T
H
E
R
S F
T A
E T
P H
E
R
N
E
I
T
H
E
R
Alcohol Abuse
Drug Abuse
Hitting the Child
Child Physical Abuse
Child Sexual Abuse
Child Emotional Abuse
-
Child Neglect
Spousal Emotional
Abuse
Spousal Physical Abuse
Previous Reports of
Child Abuse or Neglect


57
INTERPERSONAL AND FAMILY ISSUES: Please evaluate items 1-9 from your
interviews with the family members. Place a mark on the line to indicate your rating
of the family. Also, please circle the number of each question if you have included
information on that issue in your report to the court.
1. Who does the non-abusing caretaker SUPPORT:
i
Child Victim
i
Reported Abuser
CONFIDENCE in your rating.
i
i
i
None
i
Total
2.
SEVERITY of the abuse/neglect:
i
i
i
Very Slight
i
Very Severe
CONFIDENCE in your rating.
i
i
i
None
i
Total
3.
CHILDS DISTRESS over the current family situation:
i
i
i
Very Little
i
Very Much
CONFIDENCE in your rating.
i
i
i
None
i
Total
4.
CONFLICT BETWEEN THE TWO PARENTS:
i...
i
i
Very Little
i
Very Much
CONFIDENCE in your rating.
i
i
i
None
i
Total


58
5. PARENTS WILLINGNESS TO REPORT FAMILY PROBLEMS in
general:
i i
i i
None Completely
CONFIDENCE in your rating.
i i
i i
None Total
6. Amount the FAMILY FAVORS FAMILY ACTIVITIES:
i i
i i
Too Little About Right Too Much
CONFIDENCE in your rating.
i i
i i
None Total
7. Amount that the family members CLARIFY AND/OR SPEAK FOR other
family members.
i i
i i
Very Little Very Much
CONFIDENCE in your rating.
i i
i i
None Total
8. DEGREE TO WHICH THE PHYSICAL NEEDS OF THE CHILD(ren)
were met, prior to the recent report of abuse/neglect.
i i
i i
None Total
CONFIDENCE in your rating.
i i
i i
None Total
These needs were primarily met by (circle one):
Mother Father Both Other (specify)


59
9. DEGREE TO WHICH THE EMOTIONAL NEEDS OF THE CHILD(ren)
were met, prior to the recent report of abuse/neglect.
i i
i i
None Total
CONFIDENCE in your rating.
i i
i i
None Total
These needs were primarily met by (circle one):
Mother Father Both Other (specify)
For items 10-16, use the following letters that apply to this family to represent your
evaluation of the parents on these issues. Place a mark on the line to represent your
rating of each parent and label each mark with one of the following letters. Please
make sure that you make separate ratings on each of them, including your ratings of
the CONFIDENCE you have in your rating on each of them.
M=Mother, SM=Stepmother, F = Father, SF=Stepfather
10. CONFLICT BETWEEN EACH PARENT AND THE CHILD(ren) who
was reportedly abused or neglected, with an M, F, SM, and SF.
i i
i i
Very Little Very Much
For each parent, is there as much conflict with other children?
Yes No Yes No
Father: Mother:
Stepfather: Stepmother:
CONFIDENCE in each of your ratings with an M, F, SM, and SF.
i i
i i
None Total


60
11. WARM AND POSITIVE FEELINGS FOR THE CHILD(ren) who was
reportedly abused or neglected held by each parent with an M, F, SM,
and SF.
! I
Very Little Very Much
CONFIDENCE in each of your ratings with an M, F, SM, and SF.
¡ .
None Total
12. INVOLVEMENT of each parent with the child(ren)s daily lives (e.g., peer
relationships, childs feelings, minor illnesses) with an M, F, SM, and SF.
¡ I
No Involvement Excessive Involvement
CONFIDENCE in each of your ratings with an M, F, SM, and SF.
i i
i i
None Total
13. EACH PARENTS RELIANCE ON THE CHILDREN FOR
COMPANIONSHIP, COMFORT, AND PROBLEM-SOLVING with an M,
F, SM, and SF.
¡ I
Very Little Very Much
CONFIDENCE in each of your ratings with an M, F, SM, and SF.
i i
i i
None Total
14.
PSYCHOLOGICAL ISSUES for each parent with an M, F, SM, and SF.
Lonely
Highly
Slightly


61
Depressed
1
Slightly
,
Highly
Anxious
i
i
i
Slightly
i
Highly
Irritable
i
i
i
Slightly
i
Highly
CONFIDENCE in each of your ratings with an M, F, SM, and SF.
i i
i i
None Total
15. SATISFACTION WITH SOCIAL SUPPORT that each parent receives for
any problems from relatives, friends, coworkers, and community agencies
with an M, F, SM, and SF.
i i
i i
Inadequate Excellent
CONFIDENCE in each of your ratings with an M, F, SM, and SF.
i i
i i
None Total
16. EACH PARENTS RESPONSE TO YOU in general with an M, F, SM,
and SF.
i i
i i
Very Friendly Hostile/Rejecting
CONFIDENCE in each of your ratings with an M, F, SM, and SF.
i i
i i
None Total
17. If either parent was abused/neglected as a child, report the types of that
abuse (P=physical, S=sexual, E=emotional, N=Neglect, and N/A=none)
Mother Father Stepmother Stepfather


62
18. Check the following problems if any child was reported to exhibit the
problem:
Child verbally fights with parents
Child physically fights with parents
Child physically fights with siblings
Child verbally or physically fights with non-family members
Child is more promiscuous than would be expected for
his/her age
Child has been absent from school for more than 2 weeks
in the last year, and it has not been due to any
medical illness
Child has run away from home frequently in the past
Child steals from family members or others
Child is withdrawn and does not participate in normal daily
activities, such as schoolwork or family activities
Child refuses to do what his parents tell him to do,
such as going to school or doing household chores
PLEASE CHECK TO MAKE SURE THAT YOU HAVE CIRCLED THE
NUMBER OF THE QUESTIONS WHICH CONTAIN INFORMATION YOU
REPORTED TO THE COURT
Report the total number and dates of any past report of abuse or neglect and for
the present report. (If no abuse write in N/A; if the neglect was chronic, use C in
place of the number of occurrences.)
DATES TOTAL NUMBER
PAST
PRESENT


63
Report whether criminal charges are being brought against the alleged
perpetrator for abuse/neglect in this case and previously.
This case: YES NO Previously (how many):
DISPOSITION
Does the alleged perpetrator accept responsibility to the abuse/neglect?
YES NO
Indicate who requested a mental health evaluation for any family members, and
who completed it:
REQUESTED BY:(l=judge 2 = Guardian 3=Attorney for Parent 4=HRS 5=N/A)
COMPLETED BY:(1 =Clinical Psychologist 2=Psychiatrist 3=Social Worker
4=Other Psychologist 5= Other 6=Not Completed)
Requested By Completed By
For the Mother
For the Father
For the Stepmother
For the Stepfather
For the Child(ren)
Circle any family member(s) for whom the GAL has recommended therapy:
Mother Stepmother Father Stepfather Children None
Do the parents admit dependency? YES NO
Please circle type of case:
Initial Dependency Termination of Parental Rights


64
Check any of the following options it they were being considered, then check
which option was recommended by the Guardian, and which option was
adjudicated.
Considered Guardian Adjudicated
Perpetrator Removed From the
Home
Parents Retain Custody With
Supervision
Temporary Custody Awarded to
a Relative
Temporary Foster Placement
Child Placed in a Treatment
Facility
Placement for Adoption
Child Placed in a Detention
Facility
Other (specify)
Not Dependent and Returned
Home


APPENDIX B
CORRELATIONS AMONG INDEPENDENT VARIABLES
Table B-l. Estimated Correlations Among Demographic Variables Entered
in Logistic Regression Equation Predicting Judicial Decisions
Child Age Family Race
Child Age 1.00 0.16
Family Race 0.16 1.00
Table B-2. Estimated Correlations Among Demographic Variables Entered
in Logistic Regression Equation Predicting Guardian Recommendations
Child Age Family Race
Child Age 1.00 0.18
Family Race 0.18 1.00
65


66
Table B-3. Estimated Correlations Among Legal and Case
Variables Entered in Logistic Regression Equation Predicting
Judicial Decisions
Severity of
Abuse
Multiple Types
of Abuse
Prior Reports
of Abuse
Severity of Abuse
LOO
-0.25
-0.06
Multiple Types of
Abuse
-0.25
1.00
-0.21
Prior Reports of
Abuse
-0.06
-0.21
1.00
Table B-4. Estimated Correlations Among Legal and Case
Variables Entered in Logistic Regression Equation Predicting
Guardian Recommendations
Severity of
Abuse
Severity of Abuse
1.00
Multiple Types of
-0.26
Abuse
Prior Reports of
-0.12
Abuse
Multiple Types
Prior Reports
of Abuse
of Abuse
-0.26
-0.12
1.00
-0.18
-0.18
1.00


67
Table B-5. Estimated Correlations Among Family Psychological Variable Entered in
Logistic Regression Equation Predicting Judicial Decisions
Psychological
Functioning
Parental
Warmth
Family
Conflict
Parental
Non-
compliance
Substance
Abuse
Spouse
Abuse
Psychological
Functioning
1.00
0.09
0.09
-0.23
-0.20
-0.22
Parental
Warmth
0.09
1.00
0.33
0.43
-0.16
0.07
Family
Conflict
0.09
0.33
1.00
-0.01
-0.24
-0.07
Parental Non-
compliance
-0.23
0.43
-0.01
1.00
0.04
-0.01
Substance
Abuse
-0.20
-0.16
-0.24
0.04
1.00
-0.20
Spouse Abuse
-0.02
0.07
-0.07
-0.01
-0.20
1.00
Table B-6. Estimated Correlations Among Family Psychological Variable Entered in
Logistic Regression Equation Predicting Guardian Recommendations
Psychological
Functioning
Parental
Warmth
Family
Conflict
Parental
Non-
compliance
Substance
Abuse
Spouse
Abuse
Psychological
Functioning
1.00
0.17
0.03
-0.20
-0.23
-0.01
Parental
Warmth
0.17
1.00
0.29
0.39
-0.09
-0.01
Family
Conflict
0.03
0.29
1.00
-0.06
-0.18
-0.11
Parental Non-
compliance
-0.20
0.39
-0.06
1.00
0.14
-0.01
Substance
Abuse
-0.23
-0.09
-0.18
0.14
1.00
-0.17
Spouse Abuse
-0.01
-0.01
-0.11
-0.01
-0.17
1.00


68
Table B-7. Estimated Correlations Among Significant Predictor Variables
Entered in Logistic Regression Equation Predicting Guardian
Recommendations
Prior Reports
Parental Warmth
of Abuse
and Support
Prior Reports
of Abuse
1.00
0.05
Parental Warmth
and Support
0.05
1.00


APPENDIX C
RATE OF REMOVAL BY TYPE OF ABUSE
Table C-l. Cross-Tabular Analysis of Removal Rate by Presence of Neglect
Allegation
Judicial Decision
Frequency
Total Percent
Not Removed
Removed
Total
Row Percent
Column Percent
Not Neglected
6
13
19
6.38
13.83
20.21
31.58
68.42
37.50
16.67
Neglected
10
65
75
10.64
69.15
79.79
13.33
86.67
62.50
83.33
Total
16
78
94
17.02
82.98
100.00
69


70
Table C-2. Cross-Tabular Analysis of Removal Rate by Presence of Physical
Abuse
Allegation
Frequency
Total Percent
Row Percent
Column Percent
Not Removed
Judicial Decision
Removed
Total
Not Physically
6
39
45
* Abused
6.38
41.49
47.87
13.33
86.67
37.50
50.00
Physically Abused
10
39
49
10.64
41.49
52.13
20.41
79.59
62.50
50.00
Total
16
78
94
17.02
82.98
100.00
Table C-3. Cross-Tabular Analysis of Removal Rate by Presence of Sexual
Abuse
Allegation
Judicial Decision
Frequency
Total Percent
Not Removed
Removed
Total
Row Percent
Column Percent
Not Sexually
9
59
68
Abused
9.57
62.77
72.34
13.24
86.76
56.25
75.64
Sexually Abused
7
19
26
7.45
20.21
27.66
26.92
73.08
43.75
24.36
Total
16
78
94
17.02
82.98
100.00


APPENDIX D
DESCRIPTIVE STATISTICS FOR VARIABLES IN CASES OF
PHYSICAL ABUSE ALONE AND NEGLECT ALONE
Table D-l. Distribution of Interval Variables in Cases of Physical Abuse Only
and Neglect Only
Physical Abuse Only
Neglect Only
T-Test
Variable
N
Mean
S.D.
N
Mean
S.D.
Value
Child Age
11
4.36
3.59
35
3.60
3.34
0.63
Family Income
6
13.17
9.00
21
4.95
7.30
2.05
Severity of Abuse
11
53.45
24.27
36
74.14
21.04
-2.55*
Parental Psychological
Functioning
10
48.66
18.65
30
58.99
23.34
-1.42
Parental Warmth and
Support for Child
11
57.45
18.83
36
54.94
31.06
0.33
Family Conflict
11
54.18
17.06
29
37.54
32.64
2.09*
Parental Noncompliance
11
37.36
23.63
36
31.89
28.78
0.64
* p < .05
71


72
Table D-2. Frequencies of Dichotomous Variables in Cases of Physical Abuse
Only and Neglect Only
Physical Abuse Only Neglect Only
Variable Frequency Percent Frequency Percent
Adjudication
Removed
8
73
32
89
Not Removed
3
27
4
11
Guardian
Remove
9
82
27
75
Recommendation
Do Not Remove
2
18
9
25
Family Race
White
5
45
16
44
Non-White
6
55
20
56
Parent Perpetrator
Alleged
Not Alleged
10
1
91
9
35
1
97
3
Previous Reports of
Yes
3
27
17
47
Abuse
No
8
73
19
53
Spouse Abuse
Alleged
Not Alleged
3
8
27
73
5
31
14
86
Substance Abuse
Alleged
6
55
26
72
Not Alleged
5
45
10
28


73
APPENDIX E
PEARSON THREE-WEEK TEST-RETEST CORRELATIONS
Table E-l. Pearson Three-Week Test-Retest Correlations of Psychological
Variables
Variable
r
Parental Conflict
.52*
Maternal Warmth
*
OO
Paternal Warmth
.49*
Maternal Depression
.36
Paternal Depression
.18
Maternal Anxiety
.68***
Paternal Anxiety
.21
Maternal Irritability
.60**
Paternal Irritability
.09
* p< .05
** p< .005
*** p< .001


BIOGRAPHICAL SKETCH
Cynthia McDanal is a psychology associate in the state of Maryland. She
currently works under supervision in a private practice conducting psychotherapy for
children and performing neuropsychological evaluations to rule out learning
disabilities and attention disorders in children, adolescents, and adults. Upon
completion of the requirements for the degree of Doctor of Philosophy, Cynthia will
pursue certification as a licensed clinical psychologist in Maryland and the District of
Columbia. She graduated magna cum laude from Stetson University in 1987 and
earned a Master of Science degree in Clinical and Health Psychology from the
University of Florida in 1990.
74


I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and
quality, as a dissertation for the degree of Doctor of Philosophy.
Jacqueun Goldman, Chair
Professor of Clinical and Health Psychology
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and
quality, as a dissertation for the degree of Doctor of Philosophy.
es(/lohnson
fessor of Clinical and Health Psychology
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and
quality, as a dissertation for the degree of Doctor of Philosophy.
^
Michael Robinson
Associate Professor of Clinical and Health
Psychology
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and
quality, as a dissertation for the degree of Doctor of Philosophy.
t'Vy?'
ephen Boees
Stephen Boggs
Associate Professor of Clinical and Health
Psychology
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and
quality, as a dissertation for the* degree of Doctor of Philosophy.
StevenJWitiis
Professor of Law


This dissertation was submitted to the Graduate Faculty of the College of
Health Related Professions and to the Graduate School was accepted as partial
fulfillment of the requirements for the degree of Doctor of Philosophy.
April, 1994 AsL-P
Dean, College of Health Related Professions
Dean, Graduate School


Full Text

PAGE 1

*8$5',$1 $' /,7(0 $1' -8',&,$/ '(&,6,210$.,1* ,1 &$6(6 2) &+,/' $%86( $1' 1(*/(&7 %\ &<17+,$ 687721 0&'$1$/ $ ',66(57$7,21 35(6(17(' 72 7+( *5$'8$7( 6&+22/ 2) 7+( 81,9(56,7< 2) )/25,'$ ,1 3$57,$/ )8/),//0(17 2) 7+( 5(48,5(0(176 )25 7+( '(*5(( 2) '2&725 2) 3+,/2623+< 81,9(56,7< 2) )/25,'$

PAGE 2

$&.12:/('*(0(176 7KH DXWKRU H[WHQGV DSSUHFLDWLRQ WR KHU SDUHQWV 0U DQG 0UV 0 5 6XWWRQ DQG KHU JUDQGSDUHQWV 0U DQG 0UV $ %DOO IRU WKHLU FRQWLQXHG SDWLHQFH DQG HQFRXUDJHPHQW 3DUWLFXODU JUDWLWXGH LV H[WHQGHG WR WKH DXWKRUfV KXVEDQG 0U &KDUOHV 0F'DQDO -U IRU KLV XQFRQGLWLRQDO JXLGDQFH DQG VXSSRUW LL

PAGE 3

7$%/( 2) &217(176 $&.12:/('*(0(176 LL /,67 2) 7$%/(6 Y $%675$&7 YL ,1752'8&7,21 2YHUYLHZ 'HILQLWLRQ RI $EXVH ,QFLGHQFH DQG 3UHYDOHQFH 5ROH RI &KLOG 3URWHFWLRQ $JHQFLHV 7KH *XDUGLDQ DG /LWHP 3URJUDP 5(9,(: 2) /,7(5$785( 7KHRUHWLFDO 0RGHOV RI 'HFLVLRQ0DNLQJ (PSLULFDO 6WXGLHV RI 'HFLVLRQ0DNLQJ 5(6($5&+ 2%-(&7,9(6 $1' +<327+(6(6 0(7+2' 6XEMHFWV 0HDVXUH 3URFHGXUHV 5(68/76 &RPSDULVRQ RI -XGJHVf DQG *XDUGLDQVf 'HFLVLRQV 5DWH RI 5HPRYDO E\ $EXVH 7\SH /RJLVWLF 5HJUHVVLRQ (TXDWLRQV &RPSDULVRQ RI &DVHV RI 3K\VLFDO $EXVH 2QO\ DQG 1HJOHFW 2QO\ ',6&866,21 5()(5(1&(6 LLL

PAGE 4

$33(1',; $ *$/ 48(67,211$,5( )25 '(3(1'(1&< &$6(6 $33(1',; % &255(/$7,216 $021* ,1'(3(1'(17 9$5,$%/(6 $33(1',; & 5$7( 2) 5(029$/ %< 7<3( 2) $%86( $33(1',; '(6&5,37,9( 67$7,67,&6 )25 9$5,$%/(6 ,1 &$6(6 2) 3+<6,&$/ $%86( $/21( $1' 1(*/(&7 $/21( $33(1',; ( 3($5621 7+5((:((. 7(675(7(67 &255(/$7,216 %,2*5$3+,&$/ 6.(7&+ LY

PAGE 5

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

PAGE 6

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f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fV FRXUWDSSRLQWHG UHSUHVHQWDWLYH YL

PAGE 7

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

PAGE 8

,1752'8&7,21 2YHUYLHZ 5HFHQW HVWLPDWHV RI WKH SUHYDOHQFH RI FKLOG VH[XDO DEXVH SK\VLFDO DEXVH DQG QHJOHFW LQGLFDWH WKDW DOPRVW WZR PLOOLRQ FKLOGUHQ SHU \HDU EHFRPH YLFWLPV RI VRPH IRUP RI PDOWUHDWPHQW 1DWLRQDO &HQWHU RQ &KLOG $EXVH DQG 1HJOHFW f 1XPHURXV VWXGLHV KDYH GRFXPHQWHG WKH KRVW RI SV\FKRORJLFDO DQG EHKDYLRUDO SUREOHPV WKDW RIWHQ UHVXOW IURP VXFK DEXVH VHH /DPSKHDU %URZQ t )LQNHOKRU +RXFN t .LQJ IRU UHYLHZVf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n PDNLQJ LQ VXFK VLWXDWLRQV HJ :DVVHUPDQ t 5RVHQIHOG )DOOHU f DQG RWKHU

PAGE 9

UHVHDUFKHUV KDYH VRXJKW WR GHVFULEH WKH GHFLVLRQPDNLQJ SURFHVV E\ FRQVLGHULQJ VRFLDO IDPLO\ DQG FKLOG FKDUDFWHULVWLFV ZKLFK DUH DVVRFLDWHG ZLWK FDVH GLVSRVLWLRQ HJ .DW] +DPSWRQ 1HZEHUJHU %RZOHV t 6Q\GHU )LQNHOKRU f $OWKRXJK WKH DSSURDFKHV DSSOLHG LQ WKHVH ODWWHU VWXGLHV DUH PRUH GLUHFW LQ WKDW WKH\ DQDO\]H DFWXDO SODFHPHQW GHFLVLRQV WKH VWXGLHV WKHPVHOYHV DUH ODUJHO\ GHVFULSWLYH DQG KDYH OLPLWDWLRQV LQ PHWKRGRORJ\ RU VFRSH 7KH SUHVHQW VWXG\ VHHNV WR H[WHQG RXU FXUUHQW NQRZOHGJH RI WKH GHFLVLRQPDNLQJ SURFHVV LQYROYHG LQ FDVHV RI FKLOG DEXVH DQG QHJOHFW E\ LGHQWLI\LQJ IDFWRUV ZKLFK LQIOXHQFH WKH *XDUGLDQf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f 3UDFWLFHV VXFK DV KDUVK GLVFLSOLQH FKLOG ODERU LQFHVW ULWXDO VDFULILFH DQG LQIDQWLFLGH DWWHVW WR WKH H[WHQW WR ZKLFK FKLOG DEXVH LV GHHSO\ URRWHG LQ FXOWXUDO DQG UHOLJLRXV KLVWRU\ (VNLQ f 7KH KLVWRULFDO YLHZ RI FKLOGUHQ DV SURSHUW\ RI WKHLU SDUHQWV DQG WKH JHQHUDOO\ DFFHSWHG FRQFHSW WKDW IDPLO\ PDWWHUV ZHUH SULYDWH FRQWULEXWHG WR VXFK D VLWXDWLRQ

PAGE 10

,W ZDV QRW XQWLO WKH ODWH QLQHWHHQWK FHQWXU\ WKDW VRFLHW\ EHJDQ WR VKLIW LWV YLHZ RI FKLOGUHQ DV SURSHUW\ WR LQFOXGH WKH FRQFHSW RI FKLOGUHQfV ULJKWV 7KH ODQGPDUN FDVH RI 0DU\ (OOHQ LQ PDUNHG WKH ILUVW WLPH WKDW FKLOG DEXVH ZDV UHFRJQL]HG E\ D 86 FRXUW 7KH FRXUW LQWHUYHQHG E\ UHPRYLQJ 0DU\ (OOHQ IURP KHU KRPH DQG E\ SURVHFXWLQJ KHU FXVWRGLDQ 7KLV FDVH GHPRQVWUDWHG WKH QHHG IRU D FKLOG SURWHFWLRQ DJHQF\ DQG VKRUWO\ WKHUHDIWHU WKH 1HZ
PAGE 11

EDVLF QHHGV RI WKH FKLOG LQFOXGLQJ IRRG VKHOWHU FORWKLQJ PHGLFDO FDUH DQG VXSHUYLVLRQ 6H[XDO DEXVH LQFOXGHV DQ\ LQWHUDFWLRQ EHWZHHQ D FKLOG DQG DQ DGXOW LQ ZKLFK WKH FKLOG LV EHLQJ XVHG IRU WKH VH[XDO VWLPXODWLRQ RI WKH DGXOW )LQDOO\ HPRWLRQDO PDOWUHDWPHQW LQFOXGHV VXFK EHKDYLRUV DV EODPLQJ EHOLWWOLQJ RU UHMHFWLQJ D FKLOG XQHTXDO WUHDWPHQW RI VLEOLQJV RU ODFN RI FRQFHUQ IRU WKH FKLOGf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f QRWHV WKDW LQWHUYLHZV ZLWK D UDQGRP VDPSOH RI KRXVHKROGV RU LQGLYLGXDOV FRQVLVWHQWO\ SURMHFW KLJKHU LQFLGHQFH DQG SUHYDOHQFH UDWHV WKDQ LQGLFDWHG E\ IRUPDO UHSRUWLQJ GDWD 7KH 1DWLRQDO &HQWHU RQ &KLOG $EXVH DQG 1HJOHFW f KDV UHVSRQGHG WR VXFK GLIILFXOWLHV LQ HVWLPDWLRQ RI WKH LQFLGHQFH RI FKLOG DEXVH E\ UHFRJQL]LQJ ILYH OHYHOV RI DEXVHG DQG QHJOHFWHG FKLOGUHQ /HYHO LQFOXGHV DOO DEXVHG DQG QHJOHFWHG FKLOGUHQ NQRZQ WR VWDWH DQG ORFDO &KLOG 3URWHFWLYH 6HUYLFHV &36f DJHQFLHV $W /HYHO DUH WKRVH FKLOGUHQ ZKR DUH QRW NQRZQ WR &36 EXW ZKR DUH NQRZQ WR RWKHU LQYHVWLJDWLYH DJHQFLHV VXFK DV SROLFH FRXUWV RU KHDOWK GHSDUWPHQWV /HYHO LQFOXGHV FKLOGUHQ ZKR DUH QRW NQRZQ WR DQ\ /HYHO RU DJHQF\ EXW ZKR DUH NQRZQ WR SURIHVVLRQDOV LQ RWKHU FRPPXQLW\ LQVWLWXWLRQV VXFK DV KRVSLWDOV VFKRROV GD\FDUH FHQWHUV DQG PHQWDO KHDOWK

PAGE 12

FOLQLFV 7KH DEXVHG DQG QHJOHFWHG FKLOGUHQ DW /HYHO DUH NQRZQ RQO\ WR LQGLYLGXDOV RXWVLGH RI /HYHOV DQG VXFK DV QHLJKERUV RU RWKHU IDPLO\ PHPEHUV )LQDOO\ /HYHO LQFOXGHV WKRVH FKLOGUHQ ZKR KDYH QRW EHHQ UHFRJQL]HG E\ DQ\RQH DV DEXVHG RU QHJOHFWHG $OWKRXJK WKLV GHVFULSWLRQ RI DOO ILYH OHYHOV LV YDOXDEOH LQ WKDW LW IRUPDOO\ UHFRJQL]HV WKDW XQUHSRUWHG FDVHV RI DEXVH GR H[LVW WKH &HQWHUfV LQFLGHQFH VWXG\ LQFOXGHV FDVHV RQO\ DW /HYHOV WKURXJK 7KH UHVXOWV RI WKLV VWXG\ LQGLFDWHG WKDW FKLOGUHQ SHU ZHUH SK\VLFDOO\ DEXVHG LQ WKH 8QLWHG 6WDWHV GXULQJ RU D WRWDO RI FKLOGUHQ 5DWHV RI QHJOHFW ZHUH HYHQ KLJKHU f§ FKLOGUHQ SHU RU D WRWDO RI RYHU RQH PLOOLRQ FKLOGUHQ QDWLRQZLGH 7KH UDWH RI DOO IRUPV RI FKLOG VH[XDO DEXVH ZDV HVWLPDWHG WR EH SHU RU D WRWDO RI FKLOGUHQ $V PHQWLRQHG DERYH UDQGRP VDPSOH LQWHUYLHZ VWXGLHV JHQHUDOO\ \LHOG KLJKHU HVWLPDWHV RI WKH SUHYDOHQFH RI FKLOG DEXVH SUREDEO\ EHFDXVH WKH\ WDS XQUHSRUWHG DEXVH DV ZHOO DV DEXVH NQRZQ WR SXEOLF DJHQFLHV DQG FRPPXQLW\ SURIHVVLRQDOV )RU H[DPSOH 6WUDXVV DQG *HOOHV f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fV VWXG\ 5XVVHOO f VXUYH\HG D SUREDELOLW\ VDPSOH RI

PAGE 13

ZRPHQ UHVLGHQWV RI 6DQ )UDQFLVFR DERXW DQ\ H[SHULHQFH RI VH[XDO DEXVH WKH\ PLJKW KDYH KDG +HU ILQGLQJV LQGLFDWHG WKDW b RI WKH ZRPHQ KDG H[SHULHQFHG DW OHDVW RQH LQFLGHQW RI VH[XDO DEXVH EHIRUH WKH DJH RI \HDUV )LQNHOKRU f UHSRUWHG WKDW b RI DOO PDOHV DQG b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f GHVFULEHV VL[ PDMRU DVSHFWV RI WKH FKLOG SURWHFWLYH VHUYLFH SURFHVV 7KH UHIHUUDO SURFHVV $OO VWDWHV QRZ KDYH PDQGDWRU\ UHSRUWLQJ ODZV GHVLJQHG WR HQFRXUDJH WKH UHSRUWLQJ RI VXVSHFWHG DEXVH DQG QHJOHFW 0RVW UHSRUWLQJ V\VWHPV DOVR

PAGE 14

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

PAGE 15

,QLWLDWLRQ RI WKH OHJDO SURFHVV ,W LV WKH UHVSRQVLELOLW\ RI WKH FKLOG SURWHFWLRQ ZRUNHU WR GHFLGH ZKHWKHU WR LQLWLDWH FRXUW SURFHHGLQJV *XLGHOLQHV IRU WKH FRQVLGHUDWLRQ RI FRXUW DFWLRQ LQFOXGH VLWXDWLRQV LQ ZKLFK Df WKH FKLOG LV GHWHUPLQHG WR EH LQ LPPLQHQW GDQJHU Ef DWWHPSWV DW WUHDWPHQW KDYH IDLOHG DQGRU Ff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f DOVR VWUHVVHV WKH LPSRUWDQFH RI WKH HVWDEOLVKPHQW RI UDSSRUW EHWZHHQ WKH FKLOG SURWHFWLRQ ZRUNHU DQG WKH DJHQF\ DWWRUQH\ DV WKH\ ZRUN WR SUHSDUH FDVHV &OHDU JXLGHOLQHV UHJDUGLQJ GHFLVLRQPDNLQJ DUH EHQHILFLDO 7KH *XDUGLDQ DG /LWHP 3URJUDP $OPRVW DOO VWDWHV KDYH LPSOHPHQWHG *XDUGLDQ DG /LWHP RU VLPLODU SURJUDPV WR FDUU\ RXW WKH DERYH IXQFWLRQV DQG WKRVH WKDW KDYH QRW UHTXLUH WKDW OHJDO FRXQVHO EH DSSRLQWHG WR UHSUHVHQW WKH FKLOG LQ DEXVH DQG QHJOHFW KHDULQJV 6ORDQ f $ *XDUGLDQ DG /LWHP LV D FRXUWDSSRLQWHG OHJDO UHSUHVHQWDWLYH ZKR SDUWLFLSDWHV RQ EHKDOI RI FKLOGUHQ ZKR DUH OHJDOO\ XQDEOH WR LQLWLDWH RU IXQFWLRQ LQ OLWLJDWLRQ ZLWKRXW DGXOW DVVLVWDQFH 7KH IXQFWLRQ RI *XDUGLDQV LV WR LGHQWLI\ WKH FKLOGfV EHVW LQWHUHVWV LQ WKH OLWLJDWLRQ DQG WKH\

PAGE 16

DUH OHJDOO\ REOLJDWHG WR GR HYHU\WKLQJ ZLWKLQ WKHLU SRZHU WR UHSUHVHQW WKRVH LQWHUHVWV :KLWFRPE f :LLJ f QRWHV WKDW DOWKRXJK DSSRLQWPHQW RI D *XDUGLDQ DG /LWHP LV D FRQGLWLRQ RI WKH IHGHUDO &KLOG $EXVH 3UHYHQWLRQ DQG 7UHDWPHQW $FW 3/f WKH $FW GRHV QRW GHILQH WKH IXQFWLRQV RI D *XDUGLDQ GLVWLQJXLVK WKH *XDUGLDQfV UROH IURP WKDW RI WKH FKLOGfV VRFLDO ZRUNHU DQG WKH FKLOGfV DWWRUQH\ RU FOHDUO\ GHILQH WKH UHODWLRQVKLS RI WKH *XDUGLDQ WR WKH FRXUW 6KH FRQGXFWHG D OLPLWHG VWXG\ RI VL[ *XDUGLDQV LQ DQ DWWHPSW WR FODULI\ WKHVH UROHV :LLJfV ILQGLQJV OHG KHU WR FRQFOXGH WKDW WKH IROORZLQJ WHQ DFWLYLWLHV PLJKW EH GHILQHG DV WKH IXQFWLRQV RI WKH *XDUGLDQ EHFDXVH WKH\ ZHUH GLVWLQJXLVKHG IURP WKH DFWLRQV RI WKH FKLOGfV ODZ\HU DQG VRFLDO ZRUNHU LQ WKH FDVHV VKH H[DPLQHG f WR IXQFWLRQ DV D PHGLDWRU LQ DQ DWWHPSW WR DFKLHYH FRQVHQVXV DPRQJ SDUWLHV SULRU WR FRXUW SURFHHGLQJV f WR FRRUGLQDWH EHWZHHQ MXYHQLOH FRXUW DQG RWKHU FRXUW GHSDUWPHQWV f WR SURYLGH FRQWLQXLW\ DFURVV WKH HQWLUH OHJDO SURFHVV f WR SHUVXDGH WKH VRFLDO ZRUNHU WR FRQVLGHU DOWHUQDWLYH SODFHPHQWV f WR PLQLPL]H WUDXPD WR WKH FKLOG DQG WR SURYLGH FRQWLQXLW\ RI UHODWLRQVKLS ZLWK WKH FKLOG f WR LQVXUH FRXUW DWWHQWLRQ WR FDVHV E\ SODFLQJ WKHP RQ WKH FDOHQGDU f WR PDNH UHFRPPHQGDWLRQV LQ WKH EHVW LQWHUHVWV RI WKH FKLOG f WR LGHQWLI\ VSHFLDO QHHGV RI WKH FKLOG DQG WR LQVXUH WKDW WKH FKLOG UHFHLYHV DWWHQWLRQ WR WKRVH QHHGV f WR SUHYHQW FRQWLQXDQFHV QRW LQ WKH EHVW LQWHUHVWV RI WKH FKLOG RU WR UHTXHVW DQ DWWRUQH\ WR GR VR DQG f WR UHTXHVW LQGHSHQGHQW FRXQVHO IRU WKH FKLOG SS f :LLJ IXUWKHU QRWHV WKDW ZKLOH WKH DWWRUQH\ RU VRFLDO ZRUNHU PLJKW LQ VRPH FDVHV IXOILOO VRPH RI WKHVH

PAGE 17

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

PAGE 18

5(9,(: 2) /,7(5$785( 7KHRUHWLFDO 0RGHOV RI 'HFLVLRQ0DNLQJ :DVVHUPDQ DQG 5RVHQIHOG f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f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

PAGE 19

IHHOLQJV RI LQDGHTXDF\ DQG IDLOXUH LQ ERWK SDUHQWV DQG FKLOGUHQ PD\ IROORZ VHSDUDWLRQ 7KH DXWKRUV VWDWH WKDW FKLOGUHQ PD\ HYHQ SHUFHLYH WKHPVHOYHV DV UHVSRQVLEOH IRU WKH VHSDUDWLRQ ,Q RUGHU WR PDNH DSSURSULDWH GHFLVLRQV JLYHQ VXFK D GLIILFXOW VLWXDWLRQ :DVVHUPDQ DQG 5RVHQIHOG f UHFRPPHQG WKDW SUDFWLWLRQHUV FRQVLGHU WKH QDWXUH DQG TXDOLW\ RI WKH SDUHQWFKLOG ERQG 7KH\ VWDWH $ MXGJH ZKR LV SURYLGHG ZLWK FDUHIXO HYDOXDWLRQ RI WKH SDUHQWFKLOG UHODWLRQVKLS LV LQ D EHWWHU SRVLWLRQ WR GHFLGH ZLVHO\ S f ,Q DGGLWLRQ WR LQWHUYLHZLQJ DOO LQGLYLGXDOV LQYROYHG LQ WKH FDVH WKH DXWKRUV UHFRPPHQG WKDW WKH FDVHZRUNHU YLVLW WKH IDPLO\fV KRPH 7KH FDVHZRUNHU VKRXOG QRWH LVVXHV VXFK DV KRZ ZHOO WKH SDUHQW NQRZV WKH FKLOG KRZ DEOH DQG ZLOOLQJ WKH SDUHQW LV WR FDUH IRU WKH FKLOG KRZ UHDVRQDEOH WKH SDUHQWf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f KDV SURSRVHG D PRGHO RI GHFLVLRQPDNLQJ LQ FDVHV RI LQWUDIDPLOLDO FKLOG VH[XDO DEXVH ZKLFK LV PRUH FRPSUHKHQVLYH WKDQ WKH DSSURDFK GHVFULEHG DERYH /LNH :DVVHUPDQ DQG 5RVHQIHOG f VKH EHOLHYHV WKDW WKH SDUHQWFKLOG UHODWLRQVKLS LV DQ LPSRUWDQW FRQVLGHUDWLRQ HVSHFLDOO\ WKH UHODWLRQVKLS EHWZHHQ PRWKHU DQG FKLOG +RZHYHU

PAGE 20

)DOOHU DOVR EHOLHYHV WKDW RWKHU SDUHQWDO FKDUDFWHULVWLFV DUH FUXFLDO WR GHFLVLRQV DERXW LQWHUYHQWLRQ 7KHVH FKDUDFWHULVWLFV GLIIHU EHWZHHQ PRWKHUV DQG IDWKHUV $FFRUGLQJ WR )DOOHUfV WKHRU\ LPSRUWDQW PDWHUQDO IDFWRUV DUH f WKH OHYHO RI WKH PRWKHUfV GHSHQGHQF\ HVSHFLDOO\ RQ WKH IDWKHUSHUSHWUDWRU f WKH GHJUHH RI ORYH DQG QXUWXUDQFH VKH FRQYH\V WR WKH YLFWLP DQG f WKH GHJUHH WR ZKLFK VKH UHVSRQGHG SURWHFWLYHO\ ZKHQ VKH GLVFRYHUHG WKH VH[XDO DEXVH ,PSRUWDQW SDWHUQDO FKDUDFWHULVWLFV DUH f KLV RYHUDOO OHYHO RI IXQFWLRQLQJ f KLV VXSHUHJR IXQFWLRQLQJ DQG f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

PAGE 21

WKLV WLPH SHULRG LW PD\ EH RSWLPDO WR SODFH WKH FKLOG ZLWK D UHODWLYH ZKLFK DOORZV IRU OHVV GLVUXSWLRQ LQ WKH FKLOGfV OLIH ZKLOH VWLOO SURWHFWLQJ KLP RU KHU IURP IXUWKHU VH[XDO DEXVH ,Q 7\SH ,9 FDVHV WKH PRWKHU LV VLPLODU WR WKH 7\SH ,,, PRWKHU DQG WKH IDWKHU H[KLELWV SRRU RYHUDOO IXQFWLRQLQJ DQG VKRZV QR JXLOW RYHU WKH XVXDOO\f H[WHQVLYH DEXVH )DOOHU UHFRPPHQGV WKDW SDUHQWDO ULJKWV EH WHUPLQDWHG LQ VXFK FDVHV DQG WKDW WKH FKLOGUHQ UHFHLYH WUHDWPHQW IRU DW OHDVW RQH \HDU &ULPLQDO SURVHFXWLRQ PD\ DOVR EH ZDUUDQWHG IRU WKH SHUSHWUDWRU $OWKRXJK )DOOHUfV DSSURDFK WR GHFLVLRQPDNLQJ LV FRPSUHKHQVLYH DQG VSHFLILF LW GRHV KDYH VHYHUDO OLPLWDWLRQV )LUVW KHU DSSURDFK DVVXPHV WKH DYDLODELOLW\ RI DOWHUQDWLYH SODFHPHQWV IRU WKH FKLOG LH SODFHPHQW ZLWK D UHODWLYH RU IRU DGRSWLRQf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

PAGE 22

(PSLULFDO 6WXGLHV RI 'HFLVLRQ0DNLQJ 0HGGLQ f FRQGXFWHG DQ DQDORJXH VWXG\ RI FULWHULD XVHG E\ FKLOG SURWHFWLRQ ZRUNHUV WR GHWHUPLQH SODFHPHQW GHFLVLRQV LQ FDVHV RI DEXVH DQGRU QHJOHFW 6KH VXUYH\HG FKLOG ZHOIDUH ZRUNHUV XVLQJ D TXHVWLRQQDLUH ZKLFK GHVFULEHG VL[ VLWXDWLRQV FRQWDLQLQJ WZR LQVWDQFHV HDFK RI SK\VLFDO DEXVH VH[XDO DEXVH DQG QHJOHFW 5HVSRQGHQWV ZHUH SURYLGHG ZLWK LQIRUPDWLRQ RQ HLJKW YDULDEOHV DJH JHQGHU DQG UDFH RI WKH FKLOG WKH FKLOGfV H[SHULHQFH RI SUHYLRXV SODFHPHQWV WKH VHYHULW\ RI WKH FXUUHQW LQFLGHQW ULVN WR WKH FKLOG RI IXUWKHU DEXVH DQGRU QHJOHFW DQG WKH FRRSHUDWLRQ DQG IXQFWLRQLQJ RI WKH SULPDU\ FDUHWDNHU )RU HDFK VLWXDWLRQ UHVSRQGHQWV LQGLFDWHG ZKHWKHU WKH\ ZRXOG SODFH WKH FKLOG LQ IRVWHU FDUH WKH\ WKHQ GHVFULEHG ZKDW YDULDEOHV LQIOXHQFHG WKHLU GHFLVLRQV 5HVXOWV RI 0HGGLQfV VWXG\ LQGLFDWHG WKDW WKH SULPDU\ YDULDEOH XVHG WR GHWHUPLQH SODFHPHQW RI WKH FKLOG ZDV ULVN RI IXUWKHU DEXVH DQGRU QHJOHFW 6XFK D ILQGLQJ VHHPV RI GXELRXV YDOXH JLYHQ WKDW FDVHZRUNHUV DUH UDUHO\ LI HYHU SURYLGHG ZLWK WKDW LQIRUPDWLRQ GLUHFWO\ 0RUH RIWHQ DVVHVVPHQW RI ULVN WR WKH FKLOG LV D SDUW RI WKH FDVHZRUNHUfV GHFLVLRQPDNLQJ SURFHVV 0HGGLQ GLG KRZHYHU QRWH WKDW IRXU RWKHU YDULDEOHV FOXVWHUHG DURXQG ULVN WR WKH FKLOG DQG VKH VXJJHVWHG WKDW WKHVH YDULDEOHV PLJKW VHUYH DV RSHUDWLRQDO GHILQLWLRQV RU LQGLFDWRUV RI ULVN 7KH IRXU YDULDEOHV ZHUH VHYHULW\ RI WKH FXUUHQW LQFLGHQW IXQFWLRQLQJ DQG FRRSHUDWLRQ RI WKH SULPDU\ FDUHWDNHU DQG DJH RI WKH FKLOG :KLOH 0HGGLQfV VWXG\ LV XVHIXO LQ WKDW LW LQGLFDWHV VHYHUDO YDULDEOHV ZKLFK VHHP WR EH LPSRUWDQW LQ FDVHZRUNHUVf SODFHPHQW GHFLVLRQV LW GRHV KDYH VHYHUDO OLPLWDWLRQV ,Q DFWXDO FDVH VLWXDWLRQV FKLOG SURWHFWLRQ ZRUNHUV W\SLFDOO\ KDYH D JUHDW GHDO PRUH

PAGE 23

LQIRUPDWLRQ WKDQ WKH HLJKW YDULDEOHV 0HGGLQ SURYLGHG ,W LV LPSRVVLEOH WR HYDOXDWH WKH LPSRUWDQFH RI WKLV DGGLWLRQDO LQIRUPDWLRQ XVLQJ WKH DQDORJXH UHVHDUFK GHVLJQ HPSOR\HG KHUH $GGLWLRQDOO\ WKHUH LV WKH SRVVLELOLW\ WKDW WKH FDVHZRUNHUVf DFWXDO GHFLVLRQV GLIIHU IURP ZKDW WKH\ UHSRUW LQ D K\SRWKHWLFDO VLWXDWLRQ 6HYHUDO HPSLULFDO VWXGLHV RI DFWXDO FDVH GLVSRVLWLRQV KDYH EHHQ FRQGXFWHG LQ DQ DWWHPSW WR RYHUFRPH VXFK OLPLWDWLRQV 5XQ\DQ *RXOG 7URVW DQG /RGD f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f§FDVHV UHIHUUHG E\ FRXUWV RU SROLFH DQG JHRJUDSKLF DUHD 5XQ\DQ HW DO DOVR HPSOR\HG D ORJLVWLF UHJUHVVLRQ PRGHO WR SUHGLFW SODFHPHQW LQ IRVWHU FDUH JLYHQ DOO WKH DERYH YDULDEOHV 7KH\ WKHQ FRPSDUHG WKLV WR DFWXDO SODFHPHQW 5HVXOWV RI WKLV SURFHGXUH GLG QRW VWURQJO\ DVVRFLDWH WKH DERYH FKDUDFWHULVWLFV ZLWK UHPRYDO

PAGE 24

IURP WKH KRPH ,Q IDFW WKH YDULDQFH H[SODLQHG E\ WKH PRGHO ZDV RQO\ 5XQ\DQ HW DO LQWHUSUHWHG WKLV UHVXOW WR PHDQ WKDW DVVLJQPHQW WR IRVWHU FDUH DSSUR[LPDWHV D UDQGRP SURFHVV DFURVV D ODUJH SRSXODWLRQ $Q DOWHUQDWLYH H[SODQDWLRQ IRU VXFK D ILQGLQJ PD\ EH WKDW FKLOG SURWHFWLRQ ZRUNHUV XVH GLIIHUHQW FULWHULD IRU PDNLQJ SODFHPHQW GHFLVLRQV IRU GLIIHUHQW W\SHV RI DEXVH FDVHV LH VH[XDO DEXVH SK\VLFDO DEXVH RU QHJOHFWf 6LQFH WKHVH GDWD ZHUH FROODSVHG DFURVV W\SH RI DEXVH VXFK GLIIHUHQFHV PD\ KDYH FDQFHOHG HDFK RWKHU RXW .DW] +DPSWRQ 1HZEHUJHU %RZOHV DQG 6Q\GHU f H[DPLQHG IDFWRUV WKDW LQIOXHQFH WKH GHFLVLRQ WR UHPRYH FKLOGUHQ IURP WKHLU SDUHQWVf FDUH LQ FDVHV RI DEXVH SK\VLFDO RU VH[XDOf YHUVXV QHJOHFW QRQSK\VLFDO LQMXU\f 7KHVH UHVHDUFKHUV H[DPLQHG WKH UHFRUGV RI VXVSHFWHG DEXVHG RU QHJOHFWHG FKLOGUHQ DW &KLOGUHQfV +RVSLWDO %RVWRQ LQ DQ DWWHPSW WR UHYHDO KRZ GHPRJUDSKLF FKDUDFWHULVWLFV IDPLO\ KLVWRU\ IDPLO\ VWUHVV WKH QDWXUH RI WKH LQMXU\ DQG DVSHFWV RI WKH PHGLFDO HQFRXQWHU LQIOXHQFH WKH RXWFRPH RI WKH FDVH ,QWHUHVWLQJO\ WKH UHVXOWV RI WKH VWXG\ LQGLFDWHG WKDW FKLOGUHQ ZLWK QRQSK\VLFDO LQMXULHV ZHUH PRUH OLNHO\ WR EH UHPRYHG IURP WKHLU SDUHQWVf FDUH WKDQ ZHUH WKRVH ZLWK SK\VLFDO LQMXULHV 7KH DXWKRUV VXJJHVWHG VHYHUDO SRVVLEOH H[SODQDWLRQV IRU WKLV ILQGLQJ LQFOXGLQJ WKH SRVVLELOLW\ WKDW QHJOHFW PD\ EH SHUFHLYHG E\ FOLQLFLDQV DV HYLGHQFH RI D FKURQLF IDPLO\ SUREOHP UDWKHU WKDQ D VLQJOH PLVKDS .DW] HW DO DOVR FRQGXFWHG &KLVTXDUH DQDO\VHV WR WHVW WKH LQIOXHQFH RI RWKHU LQGHSHQGHQW YDULDEOHV RQ SODFHPHQW DOWKRXJK WKHVH DQDO\VHV ZHUH FROODSVHG DFURVV FDVHV RI DEXVH DQG QHJOHFW 7KHLU ILQGLQJV LQGLFDWHG WKDW WKH IROORZLQJ IDFWRUV SUHGLFWHG

PAGE 25

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f DQDO\]HG GDWD RQ FDVHV RI VH[XDO DEXVH ZKLFK ZHUH RIILFLDOO\ UHSRUWHG LQ WKH 86 LQ LQ DQ DWWHPSW WR HYDOXDWH KRZ WKH FDVHV ZHUH KDQGOHG +LV VSHFLILF DLP ZDV WR GHWHUPLQH ZKHWKHU FKLOGUHQ ZHUH EHLQJ XQQHFHVVDULO\ UHPRYHG IURP WKHLU KRPHV 6XFK DQ DSSURDFK VHHPV XVHIXO LQ WKDW KLV FRQVLGHUDWLRQ RI VH[XDO DEXVH FDVHV DORQH PLQLPL]HV WKH ULVN RI RYHUJHQHUDOL]DWLRQ RU RI KDYLQJ VLJQLILFDQW HIIHFWV ZDVKHG RXW E\ GLIIHUHQFHV DPRQJ JURXSV )LQNHOKRU FRPSDUHG WKH RYHUDOO SODFHPHQW UDWH RI VH[XDO DEXVH FDVHV WR WKH SODFHPHQW UDWH RI SK\VLFDO DEXVH FDVHV LQ WKH 1DWLRQDO &OHDULQJKRXVH VDPSOH +LV ILQGLQJV LQGLFDWHG WKDW IRVWHU FDUH SODFHPHQW RFFXUV VLJQLILFDQWO\ PRUH RIWHQ LQ FDVHV RI VH[XDO DEXVH ZKLFK OHG KLP WR K\SRWKHVL]H WKDW VH[XDO DEXVH UHVXOWV LQ PRUH RXWUDJH WKDQ GRHV SK\VLFDO DEXVH UHJDUGOHVV RI WKH GHJUHH RI GDQJHU WR WKH FKLOG )LQNHOKRUfV PDLQ ILQGLQJ RI WKH VWXG\ ZDV WKDW IRVWHU FDUH GRHV QRW VHHP WR EH XVHG LQGLVFULPLQDQWO\ LQ FDVHV RI VH[XDO DEXVH )RU H[DPSOH WKH PRVW LPSRUWDQW SUHGLFWRU RI IRVWHU FDUH SODFHPHQW ZDV ZKHWKHU RU QRW WKH YLFWLP LQLWLDOO\ UHSRUWHG WKH

PAGE 26

DEXVH $ SUHIHUHQFH WR EH SODFHG RXWVLGH WKH KRPH ZDV PRUH OLNHO\ WR EH PHW WKDQ D SUHIHUHQFH WR VWD\ $OVR IRVWHU FDUH LV PRUH OLNHO\ WR EH XVHG LQ WKH FDVH RI DQ ROGHU FKLOG WKDQ D \RXQJHU FKLOG 7KLV VHHPV UHDVRQDEOH JLYHQ WKDW ROGHU FKLOGUHQ DUH W\SLFDOO\ EHWWHU DEOH WR DGMXVW WR OLYLQJ ZLWKRXW WKHLU IDPLOLHV 2WKHU IDFWRUV SUHGLVSRVLQJ WR SODFHPHQW LQ IRVWHU FDUH LQFOXGHG DEXVH FRPPLWWHG E\ SDUHQWV HVSHFLDOO\ LI ERWK SDUHQWV ZHUH LQYROYHGf YHU\ ODUJH IDPLOLHV WKH SUHVHQFH RI RWKHU W\SHV RI DEXVH RU QHJOHFW DQG WKH SUHVHQFH RI VHYHUDO RWKHU IDPLO\ VWUHVVRUV VXFK DV VSRXVH DEXVH VXEVWDQFH DEXVH IDPLO\ GLVFRUG DQG PHQWDO KHDOWK SUREOHPV 6XFK ILQGLQJV OHG )LQNHOKRU WR FRQFOXGH WKDW WKH FKDUDFWHULVWLFV RI WKH FKLOGUHQ ZKR DUH SODFHG LQ IRVWHU FDUH GR VXJJHVW VRPH UDWLRQDO UHODWLRQVKLS EHWZHHQ WKH GLVSRVLWLRQ DQG WKH IDFWV RI WKH FDVH S f )LQNHOKRUfV GDWD DOVR VXJJHVW WKDW IRVWHU FDUH LV QRW XVHG LQGLVFULPLQDQWO\ LQ YXOQHUDEOH IDPLOLHV )RU H[DPSOH KH IRXQG WKDW IRVWHU FDUH SODFHPHQW LV QRW PRUH OLNHO\ WR RFFXU ZLWK SRRU IDPLOLHV )XUWKHU EODFN FKLOGUHQ ZHUH QRW PRUH OLNHO\ WR EH UHPRYHG IURP WKHLU KRPHV WKDQ ZHUH ZKLWH FKLOGUHQ 7KXV DFFRUGLQJ WR )LQNHOKRUfV ILQGLQJV WKHUH GRHV QRW VHHP WR EH DQ\ V\VWHPDWLF ELDV LQ WKH GHFLVLRQPDNLQJ SURFHVV 7KH VWUHQJWKV RI )LQNHOKRUfV VWXG\ LQFOXGH LWV ODUJH 1 LWV FRQVLGHUDWLRQ RI D ODUJH QXPEHU RI SRWHQWLDOO\ UHOHYDQW IDFWRUV DQG LWV VWDWLVWLFDO DQDO\VLV RI RQO\ RQH W\SH RI DEXVH 7KLV PHWKRGRORJ\ GRHV QRW KRZHYHU LQGLFDWH ZKHWKHU DQ\ RI WKHVH IDFWRUV FOXVWHU WRJHWKHU WR IRUP D PHDQLQJIXO DJJUHJDWH RI SUHGLFWLYH FKDUDFWHULVWLFV $OVR VLPLODU LQIRUPDWLRQ UHJDUGLQJ FDVHV RI FKLOG SK\VLFDO DEXVH DQG RI QHJOHFW LV QHHGHG

PAGE 27

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f DQG MXGJHVf GHFLVLRQV WR SODFH WKH FKLOG RXWVLGH WKH KRPH LQ FDVHV RI FKLOG SK\VLFDO DEXVH DQG QHJOHFW $PRQJ GHPRJUDSKLF YDULDEOHV LW LV K\SRWKHVL]HG WKDW LQFUHDVHG DJH RI WKH FKLOG ZLOO SUHGLFW SODFHPHQW RXWVLGH RI WKH KRPH EXW WKDW IDPLO\ UDFH DQG LQFRPH ZLOO QRW EH UHODWHG WR FDVH GLVSRVLWLRQ $ VHFRQG FDWHJRU\ RI YDULDEOHV K\SRWKHVL]HG WR EH UHODWHG WR FDVH GLVSRVLWLRQ LQFOXGHV WKRVH ZKLFK DUH GHVFULSWLYH RI WKH DEXVH VLWXDWLRQ 6SHFLILFDOO\ LW LV SURSRVHG WKDW DEXVH SHUSHWUDWHG E\ SDUHQWV YHU\ VHYHUH DEXVH WKH SUHVHQFH RI RWKHU W\SHV RI DEXVH DQG SUHYLRXV UHSRUWV RI DEXVH ZLOO SUHGLFW SODFHPHQW RI WKH FKLOG RXWVLGH WKH KRPH $ WKLUG FDWHJRU\ RI YDULDEOHV LQFOXGHV WKRVH WKDW GHVFULEH WKH SV\FKRORJLFDO VWDWH RI WKH IDPLO\ ,W LV VSHFLILFDOO\ K\SRWKHVL]HG WKDW SRRUHU SDUHQW SV\FKRORJLFDO

PAGE 28

IXQFWLRQLQJ KLJK GHJUHH RI SDUHQWDO QRQFRPSOLDQFH ZLWK WKH LQYHVWLJDWLRQ WKH SUHVHQFH RI VSRXVH DEXVH DQG VXEVWDQFH DEXVH KLJK GHJUHH RI IDPLO\ GLVFRUG DQG ORZ GHJUHH RI SDUHQWDO ZDUPWK DQG VXSSRUW IRU WKH FKLOG ZLOO DOO SUHGLFW SODFHPHQW RXWVLGH WKH KRPH

PAGE 29

0(7+2' 6XEMHFWV 6XEMHFWV IRU WKLV VWXG\ LQFOXGHG IDPLOLHV LQYROYHG LQ FKLOG SK\VLFDO DEXVH RU QHJOHFW FDVHV LQ WKH VWDWH RI )ORULGD $OO IDPLOLHV ZHUH VHOHFWHG IURP DGMXGLFDWHG FDVHV LQ MXGLFLDO FLUFXLWV 2QO\ RQH DEXVHG RU QHJOHFWHG FKLOG IURP HDFK IDPLO\ ZDV FRQVLGHUHG LQ WKLV VWXG\ LQ RUGHU WR DYRLG PLVOHDGLQJ GXSOLFDWLRQ RI IDPLO\ DQG FDVH YDULDEOHV &KLOG DJH UDQJHG IURP \HDU WR \HDUV PHDQ \HDUV PRQWKVf )DPLO\ LQFRPH UDQJHG IURP WR SHU \HDU PHDQ f )LIW\QLQH SHUFHQW RI WKH IDPLOLHV ZHUH ZKLWH b ZHUH EODFN b ZHUH +LVSDQLF DQG b ZHUH RWKHU UDFHV 0HDVXUH 'DWD IRU WKLV VWXG\ ZHUH FROOHFWHG XVLQJ D TXHVWLRQQDLUH GHVLJQHG E\ WKH WHDP RI LQYHVWLJDWRUV WR VDPSOH GHPRJUDSKLF OHJDO DQG SV\FKRORJLFDO IDFWRUV UHOHYDQW WR WKH IDPLO\ VHH $SSHQGL[ $f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f SHUSHWUDWRU UHPRYHG IURP WKH KRPH f SDUHQWV UHWDLQ FXVWRG\ ZLWK

PAGE 30

VXSHUYLVLRQ f WHPSRUDU\ FXVWRG\ DZDUGHG WR D UHODWLYH f WHPSRUDU\ IRVWHU SODFHPHQW f FKLOG SODFHPHQW LQ D WUHDWPHQW IDFLOLW\ f SODFHPHQW IRU DGRSWLRQ f FKLOG SODFHG LQ D GHWHQWLRQ IDFLOLW\ DQG f QRW GHSHQGHQW DQG UHWXUQHG KRPH $ FDWHJRU\ GHVLJQDWHG RWKHU ZDV DOVR SURYLGHG LQ FDVH ILQDO DGMXGLFDWLRQ LQFOXGHG D SODFHPHQW QRW GHVFULEHG DERYH )RU WKH SXUSRVHV RI WKLV VWXG\ WKHVH RSWLRQV ZHUH FROODSVHG WR UHWXUQHG KRPH RSWLRQV RU f RU UHPRYHG IURP WKH KRPH RSWLRQV WKURXJK f 3V\FKRORJLFDO LQIRUPDWLRQ IRU HDFK FDVH ZDV SURYLGHG E\ WKH *XDUGLDQ ZKR PDGH MXGJPHQWV EDVHG RQ KLV RU KHU LQWHUYLHZV ZLWK WKH IDPLO\ -XGJPHQWV ZHUH PDGH RQ YLVXDO DQDORJ VFDOHV RQ WKH IROORZLQJ LVVXHV f ZKR WKH QRQDEXVLQJ FDUHWDNHU VXSSRUWV f VHYHULW\ RI WKH DEXVHQHJOHFW f FKLOGfV GLVWUHVV RYHU WKH FXUUHQW IDPLO\ VLWXDWLRQ f FRQIOLFW EHWZHHQ WKH SDUHQWV f SDUHQWVf ZLOOLQJQHVV WR UHSRUW IDPLO\ SUREOHPV LQ JHQHUDO f DPRXQW WKH IDPLO\ IDYRUV IDPLO\ DFWLYLWLHV f DPRXQW WKH IDPLO\ PHPEHUV FODULI\ DQGRU VSHDN IRU RWKHU IDPLO\ PHPEHUV f GHJUHH WR ZKLFK WKH SK\VLFDO QHHGV RI WKH FKLOG ZHUH PHW DQG ZKR WKH\ ZHUH PHW E\f DQG f GHJUHH WR ZKLFK WKH HPRWLRQDO QHHGV RI WKH FKLOG ZHUH PHW DQG ZKRP WKH\ ZHUH PHW E\f *XDUGLDQV DOVR UDWHG WKH IROORZLQJ LVVXHV IRU HDFK SDUHQW f FRQIOLFW EHWZHHQ SDUHQW DQG FKLOG f ZDUP DQG SRVLWLYH IHHOLQJV IRU FKLOG f LQYROYHPHQW LQ FKLOGfV GDLO\ OLIH f UHOLDQFH RQ WKH FKLOG f SV\FKRORJLFDO LVVXHV ORQHOLQHVV GHSUHVVLRQ DQ[LHW\ LUULWDELOLW\f f VDWLVIDFWLRQ ZLWK VRFLDO VXSSRUW DQG f UHVSRQVH WR WKH *XDUGLDQ LQ JHQHUDO *XDUGLDQV DOVR LQGLFDWHG RQ YLVXDO DQDORJ VFDOHV WKHLU FRQILGHQFH LQ HDFK RI WKHLU MXGJPHQWV

PAGE 31

$GGLWLRQDO LQIRUPDWLRQ ZDV UHTXHVWHG UHJDUGLQJ ZKHWKHU WKH SDUHQWV ZHUH DEXVHG RU QHJOHFWHG DV FKLOGUHQ ZKHWKHU WKH SDUHQWV DUH DFFXVHG RI VXEVWDQFH DEXVH DQGRU VSRXVH DEXVH ZKHWKHU WKH SDUHQWV KDYH ORVW MREV RU PRYHG ZLWKLQ WKH SDVW \HDU ZKHWKHU WKH SDUHQW RU FKLOG VXIIHUV D PHQWDO RU SK\VLFDO KDQGLFDS DQG ZKHWKHU WKH FKLOG H[KLELWV DQ\ EHKDYLRUDO SUREOHPV 5HOLDELOLW\ *XDUGLDQfV MXGJPHQWV RQ WKH SV\FKRORJLFDO LVVXHV ZHUH FRPSDUHG WR WKRVH RI OLFHQVHG FOLQLFDO SV\FKRORJLVWV DQG ODZ VWXGHQWV *ROGPDQ HW DO LQ SUHSDUDWLRQf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f SV\FKRORJLVWVf DQG ODZ\HUVf MXGJPHQWV RQ DQ\ RI WKH SV\FKRORJLFDO LVVXHV LQGLFDWLQJ WKDW QR VSHFLILF ELDV FRXOG EH DWWULEXWHG WR WKH VXEMHFWVf WUDLQLQJ RU WKHLU DUHD RI H[SHUWLVH 7HVWUHWHVW UHOLDELOLW\ ZDV DOVR FDOFXODWHG IRU HDFK RI WKH LWHPV UDWHG 7KH *XDUGLDQV YLHZHG WKH PRFN LQWHUYLHZ WZLFH DW WKUHH ZHHN LQWHUYDOV 2I WKH YDULDEOHV XVHG LQ WKLV VWXG\ 3HDUVRQ FRUUHODWLRQV ZHUH VLJQLILFDQW IRU WKUHH YDULDEOHV DW WKH S OHYHO OHYHO RI FRQIOLFW PDWHUQDO DQG SDWHUQDO ZDUPWK IRU FKLOGf IRU RQH YDULDEOH DW WKH S OHYHO PDWHUQDO QRQFRPSOLDQFHf DQG IRU RQH YDULDEOH DW WKH S OHYHO VHYHULW\ RI DEXVHQHJOHFWf 3HDUVRQ FRUUHODWLRQV ZHUH IRXQG WR EH QRQVLJQLILFDQW IRU

PAGE 32

SDWHUQDO QRQFRPSOLDQFH DQG SDUHQWDO SV\FKRORJLFDO IXQFWLRQLQJ GHSUHVVLRQ DQ[LHW\ LUULWDELOLW\ DQG ORQHOLQHVVf ,W VKRXOG EH QRWHG KRZHYHU WKDW WKH YLGHRWDSH YLHZHG E\ WKH *XDUGLDQV FRQWDLQHG UHODWLYHO\ OLWWOH LQIRUPDWLRQ UHJDUGLQJ WKH YDULDEOHV IRXQG WR KDYH QRQVLJQLILFDQW FRUUHODWLRQV DQG LW LV WKHUHIRUH K\SRWKHVL]HG WKDW WKH ORZ FRUUHODWLRQV ZHUH GXH PRUH WR *XDUGLDQVf ODFN RI FRQILGHQFH LQ WKHLU UDWLQJV WKDQ WR LQVWDELOLW\ LQ WKHLU MXGJPHQWV RYHU WLPH 3URFHGXUHV 'DWD IRU HDFK IDPLO\ ZHUH FROOHFWHG E\ WKH *XDUGLDQ DG /LWHP *$/f DVVLJQHG WR WKDW FDVH (DFK MXGLFLDO FLUFXLW DQG HDFK *XDUGLDQ SDUWLFLSDWHG LQ WKLV VWXG\ RQ D YROXQWHHU EDVLV *XDUGLDQV DUH DVVLJQHG FDVHV RQ DQ LQGLYLGXDO EDVLV E\ WKH *$/ GLUHFWRU RI HDFK FLUFXLW (DFK *XDUGLDQ LV DSSRLQWHG WR D IDPLO\ ZKHQ WKH FDVH ILUVW EHFRPHV DFWLYH DQG WKDW *XDUGLDQ UHSUHVHQWV WKH FKLOG XQWLO WKH FDVH LV DGMXGLFDWHG )RU WKH SXUSRVH RI WKLV VWXG\ *XDUGLDQV FROOHFWHG GDWD RQ HDFK IDPLO\ ZKHQ WKH FDVH UHDFKHG WKH VWDJH RI DGMXGLFDWLRQ 7KXV WKH FDVHV XVHG LQ WKLV VWXG\ UHSUHVHQW WKRVH WKDW ZHUH QRW VHWWOHG RXW RI FRXUW DQG WKHUHIRUH PD\ UHSUHVHQW WKH PRVW VHYHUH FDVHV

PAGE 33

5(68/76 &RPSDULVRQ RI -XGJHVf DQG *XDUGLDQVf 'HFLVLRQV 7KH FURVVWDEXODU GLVSOD\ SURYLGHG LQ 7DEOH VKRZV WKH IUHTXHQF\ ZLWK ZKLFK MXGJHV DJUHHG ZLWK WKH UHFRPPHQGDWLRQ RI WKH *XDUGLDQV DG /LWHP 5HVXOWV LQGLFDWH WKDW WKH FKLOG ZDV UHPRYHG IURP WKH KRPH E\ WKH MXGJH LQ FDVHV DQG UHWXUQHG KRPH LQ FDVHV *XDUGLDQV UHFRPPHQGHG UHPRYDO IURP WKH KRPH LQ FDVHV DQG UHFRPPHQGHG UHWXUQLQJ WKH FKLOG WR WKH KRPH LQ FDVHV 7DEOH DOVR VKRZV WKDW WKH MXGJH DJUHHG ZLWK WKH *XDUGLDQfV UHFRPPHQGDWLRQ WR UHPRYH WKH FKLOG IURP WKH KRPH RU WR UHWXUQ WKH FKLOG KRPHf LQ RI WKH FDVHV 7KHUH ZHUH RQO\ FDVHV LQ ZKLFK WKH MXGJHf§EXW QRW WKH *XDUGLDQf§IHOW WKDW WKH FKLOG VKRXOG EH UHPRYHG DQG RQO\ FDVHV LQ ZKLFK WKH *XDUGLDQf§EXW QRW WKH MXGJHf§IHOW WKDW WKH FKLOG VKRXOG EH UHPRYHG 7KH .DSSD VWDWLVWLF .f ZDV FDOFXODWHG WR GHWHUPLQH WKH GHJUHH RI FRQFXUUHQFH REVHUYHG EHWZHHQ MXGJHV DQG *XDUGLDQV DQG WKH REWDLQHG YDOXH RI LQGLFDWHV JRRG RYHUDOO DJUHHPHQW EH\RQG FKDQFHf EHWZHHQ WKH WZR JURXSV ,W FDQ WKHUHIRUH EH FRQFOXGHG WKDW ZKLOH *XDUGLDQV DSSHDU WR EH VOLJKWO\ PRUH FRQVHUYDWLYH LQ WKHLU UHFRPPHQGDWLRQV WR UHPRYH FKLOGUHQ IURP WKHLU KRPHV WKHUH ZDV FRQVLGHUDEOH DJUHHPHQW EHWZHHQ WKH WZR JURXSV UHJDUGLQJ SODFHPHQW RI WKH FKLOG 7R IXUWKHU DVVHVV WKH UHODWLRQVKLS EHWZHHQ *XDUGLDQ UHFRPPHQGDWLRQV DQG MXGLFLDO GHFLVLRQV D ORJLVWLF UHJUHVVLRQ DQDO\VLV ZDV FRQGXFWHG XVLQJ *XDUGLDQ UHFRPPHQGDWLRQ DV D SUHGLFWRU YDULDEOH IRU MXGLFLDO GHFLVLRQ 6HH 7DEOH f 5HVXOWV LQGLFDWHG WKDW WKH

PAGE 34

7DEOH &RPSDULVRQ RI *XDUGLDQ 5HFRPPHQGDWLRQV DQG -XGLFLDO 'HFLVLRQV -XGLFLDO 'HFLVLRQ *XDUGLDQ 5HFRPPHQGDWLRQ 5HPRYHG 1RW 5HPRYHG 7RWDO 5HPRYHG FDVHV FDVHV FDVHV bf bf bf 1RW 5HPRYHG FDVHV FDVHV FDVHV bf bf bf 7RWDO FDVHV FDVHV FDVHV bf bf bf 7DEOH 5HVXOWV RI /RJLVWLF 5HJUHVVLRQ $QDO\VLV 8VLQJ *XDUGLDQ 5HFRPPHQGDWLRQV WR 3UHGLFW -XGLFLDO 'HFLVLRQV 3UHGLFWLQJ -XGLFLDO 'HFLVLRQV 9DULDEOH ') 3DUDPHWHU (VWLPDWH 6WDQGDUG (UURU &KL VTXDUH 3UREDELOLW\ RI &KLVTXDUH ,QWHUFHSW *XDUGLDQ r 5HFRPPHQGDWLRQ S +RVPHU/HPHVKRZ JRRGQHVVRIILW VWDWLVWLF ZLWK ') S

PAGE 35

*XDUGLDQfV UHFRPPHQGDWLRQ IRU UHPRYDO LV VLJQLILFDQWO\ DVVRFLDWHG S f ZLWK D KLJKHU SUREDELOLW\ RI MXGLFLDO GHFLVLRQ WR UHPRYH WKH FKLOG IURP WKH KRPH 3UHGLFWHG SUREDELOLWLHV EDVHG RQ WKH SDUDPHWHU HVWLPDWH REWDLQHG IRU WKH *XDUGLDQ UHFRPPHQGDWLRQ YDULDEOH LQGLFDWH WKDW WKH SUREDELOLW\ RI UHPRYDO LV b LQ FDVHV ZKHUH WKH *XDUGLDQ UHFRPPHQGV LW DQG b LQ FDVHV ZKHUH WKH *XDUGLDQ UHFRPPHQGV DJDLQVW LW ,W VKRXOG EH QRWHG DJDLQ KRZHYHU WKDW WKHVH GDWD DUH EDVHG RQ RQO\ FDVHV LQ ZKLFK WKH JXDUGLDQ UHFRPPHQGHG UHWXUQLQJ WKH FKLOG WR WKH KRPH 7KH +RVPHU DQG /HPHVKRZ JRRGQHVVRIILW VWDWLVWLF IRU WKH PRGHO ZLWK ') S f LQGLFDWHV WKDW WKH REVHUYHG UHVSRQVHV GR QRW VLJQLILFDQWO\ GLIIHU IURP UHVSRQVHV ZKLFK ZRXOG EH SUHGLFWHG EDVHG RQ WKLV PRGHO +RVPHU t /HPHVKRZ f 7KH FDVHV LQ ZKLFK WKH MXGJH GLVDJUHHG ZLWK WKH *XDUGLDQfV UHFRPPHQGDWLRQ IRU SODFHPHQW RI WKH FKLOG ZHUH H[DPLQHG IXUWKHU WR GHWHUPLQH ZKHWKHU DQ\ WUHQGV DPRQJ WKH SUHGLFWRU YDULDEOHV FRXOG EH LGHQWLILHG ([DPLQDWLRQ RI WKH YDOXHV RI WKH SUHGLFWRU YDULDEOHV LQ WKRVH FDVHV KRZHYHU GLG QRW UHYHDO DQ\ GLIIHUHQFHV ZKLFK PLJKW DFFRXQW IRU WKH GLIIHUHQFH RI RSLQLRQ EHWZHHQ WKH MXGJHV DQG WKH *XDUGLDQV VHH 7DEOH f 5DWH RI 5HPRYDO E\ $EXVH 7\SH ,W ZDV DOVR RI LQWHUHVW WR FRPSDUH WKH UDWH RI UHPRYDO GHSHQGLQJ RQ WKH W\SH RI DEXVH RU QHJOHFW DOOHJDWLRQV SUHVHQW 7KH RYHUDOO GLVWULEXWLRQ RI DEXVH W\SHV LQ WKLV VDPSOH ZKHWKHU UHPRYHG RU QRW UHPRYHGf LQGLFDWHG WKDW b RI WKH FDVHV LQYROYHG QHJOHFW b LQYROYHG SK\VLFDO DEXVH DQG b LQYROYHG VH[XDO DEXVH $PRQJ WKH FDVHV LQ ZKLFK UHPRYDO RFFXUUHG b LQYROYHG DOOHJDWLRQV RI QHJOHFW b LQYROYHG DOOHJDWLRQV RI SK\VLFDO DEXVH DQG b LQYROYHG DOOHJDWLRQV RI VH[XDO DEXVH VHH

PAGE 36

7DEOH 9DULDEOH 9DOXHV ,Q &DVHV :KHUH -XGJHV DQG *XDUGLDQV 'LVDJUHHG 5HJDUGLQJ &DVH 2XWFRPH & ) 5 3 6 0 3 V 3 : & 1 6 6 X 8 + $ $ $ ( 7 5 ( 6 $ 2 8 3 $ 0 & 5 9 < ( ; < 5 1 1 % $ 5 $ ( 3 ( 3 9 $ ) 0 ) & $ % ( 1 ( 5 ( 5 % 8 / 2 % 8 ( & 5 6 ( 8 1 0 8 V $ 3 7 3 6 & 3 6 ( 1 < ( 7 ( 5 1 : < 1 1 1 1 1 5 1 : < < 1 < < 1 5 1 Z < < 1 < < 1 5 1 1 < 1 1 1 1 1 5 1 1 < 1 < 1 < 1 5 1 1 < 1 < 1 < 1 5 1 : < < 1 1 1 1 1 5 Z < < < < 1 1 5 1 1 < 1 1 1 < 1 1 5 1 < 1 1 1 1 1 5 1 : < < 1 1 1 < 5 1 1 < 1 1 1 < 1 5 1 : < < 1 1 1 1 5 1 Z 1 1 1 < 1 1 1 WW LW 5 1 < f§UUU 7L Y) 1 1 UULUPf§ 1 < 1 -8'*( *8$5',$1 &+$*( )$0,1& 5$&( 3$53(53 6(9(5,7< 07<3(6 35(95(3 6(;$%86( 36<)81&7 :$50 &21)/ 121&203 68%$%86( 63$%86( $GMXGLFDWLRQ 5 &KLOG UHPRYHG IURP KRPH 1 1RW UHPRYHGf 5HFRPPHQGDWLRQ 5 5HFRPPHQGHG UHPRYDO 1 'LG QRWf &KLOG DJH LQ \HDUVf $QQXDO IDPLO\ LQFRPH LQ WKRXVDQGV RI GROODUVf )DPLO\ UDFH : :KLWH 1 1RQ:KLWHf :DV WKH DOOHJHG SHUSHWUDWRU D SDUHQW" 6HYHULW\ RI DEXVH +LJKHVW SRVVLEOH VHYHULW\ OHYHOf :HUH PXOWLSOH W\SHV RI DEXVH DOOHJHG" :HUH WKHUH SUHYLRXV UHSRUWV RI DEXVH" :DV VH[XDO DEXVH DOOHJHG" 3DUHQWDO SV\FKRORJLFDO IXQFWLRQLQJ 3RRUHVW SRVVLEOH OHYHOf 3DUHQWDO ZDUPWKVXSSRUW IRU FKLOG +LJKHVW SRVVLEOH OHYHOf )DPLO\ FRQIOLFW +LJKHVW SRVVLEOH OHYHOf 3DUHQWDO QRQFRPSOLDQFH +LJKHVW SRVVLEOH OHYHOf :DV SDUHQWDO VXEVWDQFH DEXVH DOOHJHG" :DV VSRXVH DEXVH DOOHJHG"

PAGE 37

$SSHQGL[ &f 5HVXOWV DOVR LQGLFDWHG WKDW FKLOGUHQ ZHUH UHPRYHG IURP WKHLU KRPHV LQ b RI WKH FDVHV LQYROYLQJ DOOHJDWLRQV RI QHJOHFW LQ b RI WKH FDVHV LQYROYLQJ DOOHJDWLRQV RI SK\VLFDO DEXVH DQG LQ b RI WKH FDVHV LQYROYLQJ DOOHJDWLRQV RI VH[XDO DEXVH :KLOH WKHVH UHVXOWV DUH FRPSOLFDWHG E\ WKH IDFW WKDW PXOWLSOH W\SHV RI DEXVH RFFXUUHG LQ b RI WKH FDVHV WKH\ QRQHWKHOHVV LQGLFDWH UHPRYDO ZDV PRVW OLNHO\ ZKHQ DOOHJDWLRQV RI QHJOHFW ZHUH SUHVHQW DQG OHDVW OLNHO\ ZKHQ DOOHJDWLRQV RI VH[XDO DEXVH ZHUH SUHVHQW /RJLVWLF 5HJUHVVLRQ (TXDWLRQV 7KH SULPDU\ SXUSRVH RI WKLV VWXG\ ZDV WR GHWHUPLQH ZKLFK GHPRJUDSKLF YDULDEOHV DEXVLYH VLWXDWLRQ FKDUDFWHULVWLFV DQG IDPLO\ SV\FKRORJLFDO FKDUDFWHULVWLFV LQIOXHQFH *XDUGLDQVf DQG MXGJHVf GHFLVLRQV WR UHPRYH WKH FKLOG IURP WKH KRPH LQ FDVHV RI FKLOG DEXVH DQGRU QHJOHFW %HFDXVH WKH RXWFRPH YDULDEOHV WKH MXGJHfV SODFHPHQW RI WKH FKLOG RU WKH *XDUGLDQfV UHFRPPHQGDWLRQ IRU SODFHPHQWf DQG VHYHUDO SUHGLFWRU YDULDEOHV DUH GLFKRWRPRXV ORJLVWLF UHJUHVVLRQ ZDV XVHG ,Q RUGHU WR WHVW IRU WKH SRVVLELOLW\ RI PLVOHDGLQJ PXOWLFROOLQHDULW\ DPRQJ WKH LQGHSHQGHQW YDULDEOHV FRUUHODWLRQ PDWULFHV ZHUH REWDLQHG IRU HDFK PRGHO 7KH FRUUHODWLRQV DPRQJ WKH LQGHSHQGHQW YDULDEOHV UDQJHG IURP WR LQGLFDWLQJ WKDW WKH UHVXOWV RI WKH DQDO\VHV ZHUH QRW FRPSURPLVHG E\ VLJQLILFDQW PXOWLFROOLQHDULW\ 7ZR VHULHV RQH IRU MXGJHVf GHFLVLRQV DQG RQH IRU *XDUGLDQVf UHFRPPHQGDWLRQVf RI WKUHH ORJLVWLF UHJUHVVLRQ PRGHOV ZHUH UXQf§RQH PRGHO IRU HDFK FDWHJRU\ RI YDULDEOHV GHPRJUDSKLF YDULDEOHV DEXVLYH VLWXDWLRQ FKDUDFWHULVWLFV DQG IDPLO\ SV\FKRORJLFDO FKDUDFWHULVWLFVf 7KH UHVXOWV RI HDFK RI WKHVH PRGHOV ZLOO EH GLVFXVVHG VHSDUDWHO\

PAGE 38

)LQDOO\ WZR ORJLVWLF UHJUHVVLRQ HTXDWLRQV LQFOXGLQJ RQO\ WKH VLJQLILFDQW SUHGLFWRU YDULDEOHV IURP WKH DERYH PRGHOV ZHUH WHVWHGf§RQH XVLQJ MXGJHVf GHFLVLRQV DV D GHSHQGHQW YDULDEOH DQG LQFOXGLQJ DOO LQGHSHQGHQW YDULDEOHV IRXQG WR VLJQLILFDQWO\ SUHGLFW WKRVH GHFLVLRQV DQG D VHFRQG XVLQJ *XDUGLDQVf GHFLVLRQV DV D GHSHQGHQW YDULDEOH DQG LQFOXGLQJ DOO LQGHSHQGHQW YDULDEOHV IRXQG WR VLJQLILFDQWO\ SUHGLFW WKHLU GHFLVLRQV 'HPRJUDSKLF 9DULDEOHV %HFDXVH LQIRUPDWLRQ UHJDUGLQJ IDPLO\ LQFRPH ZDV SURYLGHG LQ RQO\ RI WKH FDVHV LW ZDV QRW SRVVLEOH WR WHVW WKH LPSDFW RI WKLV YDULDEOH 5HVXOWV RI WKH ORJLVWLF UHJUHVVLRQ HTXDWLRQ LQFOXGLQJ FKLOG DJH DQG IDPLO\ UDFH DV SUHGLFWRUV LQGLFDWHG WKDW QHLWKHU YDULDEOH VLJQLILFDQWO\ LPSDFWHG WKH GHFLVLRQ WR SODFH WKH FKLOG RXWVLGH WKH KRPH VHH 7DEOH f 7KH DQDO\VLV XVLQJ WKH *XDUGLDQfV UHFRPPHQGDWLRQ DV WKH GHSHQGHQW YDULDEOH \LHOGHG VLPLODU UHVXOWV DV QHLWKHU FKLOG DJH QRU IDPLO\ UDFH HPHUJHG DV D VLJQLILFDQW SUHGLFWRU RI D *XDUGLDQfV GHFLVLRQ WR UHFRPPHQG SODFHPHQW RXWVLGH RI WKH KRPH VHH 7DEOH f $EXVLYH 6LWXDWLRQ &KDUDFWHULVWLFV $PRQJ YDULDEOHV UHODWHG WR DVSHFWV RI WKH DEXVLYH RU QHJOHFWIXO VLWXDWLRQ LW ZDV K\SRWKHVL]HG WKDW DEXVH SHUSHWUDWHG E\ SDUHQWV YHU\ VHYHUH DEXVH WKH SUHVHQFH RI PXOWLSOH W\SHV RI DEXVH DQG SUHYLRXV UHSRUWV RI DEXVH ZRXOG SUHGLFW SODFHPHQW RXWVLGH WKH KRPH ,W ZDV QRW SRVVLEOH WR WHVW WKH LQIOXHQFH RI WKH SDUHQW SHUSHWUDWRU YDULDEOH KRZHYHU DV b RI WKH FDVHV LQ WKLV VDPSOH LQYROYHG SDUHQW SHUSHWUDWRUV :KHQ WKH UHPDLQLQJ YDULDEOHV ZHUH LQFOXGHG LQ WKH ORJLVWLF UHJUHVVLRQ HTXDWLRQ XVLQJ WKH MXGJHfV

PAGE 39

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

PAGE 40

GHFLVLRQ DV D GHSHQGHQW YDULDEOH QRQH HPHUJHG DV VLJQLILFDQWO\ UHODWHG WR UHPRYDO RI WKH FKLOG IURP WKH SDUHQWVf KRPH VHH 7DEOH f ,Q WKH DQDO\VLV XVLQJ WKH *XDUGLDQfV UHFRPPHQGDWLRQ DV D GHSHQGHQW YDULDEOH RQO\ WKH SUHVHQFH RI SUHYLRXV UHSRUWV RI DEXVH Sf HPHUJHG DV VLJQLILFDQWO\ UHODWHG WR UHFRPPHQGDWLRQ IRU SODFHPHQW RXWVLGH WKH KRPH VHH 7DEOH f ,Q WKLV PRGHO WKH QHJDWLYH SDUDPHWHU HVWLPDWH LQGLFDWHV WKDW WKH DEVHQFH RI SUHYLRXV UHSRUWV RI DEXVH RU QHJOHFW ZKLFK ZDV WUHDWHG ILUVW LQ WKH PRGHOf LV DVVRFLDWHG ZLWK D VPDOOHU SUREDELOLW\ RI UHPRYLQJ WKH FKLOG IURP WKH KRPH *XDUGLDQV ZHUH QRW PRUH OLNHO\ WR UHFRPPHQG UHPRYDO LQ FDVHV LQ ZKLFK WKH DEXVH ZDV YHU\ VHYHUH RU LQ ZKLFK PXOWLSOH W\SHV RI DEXVH RFFXUUHG )DPLO\ 3V\FKRORJLFDO &KDUDFWHULVWLFV 6HYHUDO YDULDEOHV GHVFULELQJ WKH SV\FKRORJLFDO VWDWH RI WKH IDPLO\ ZHUH K\SRWKHVL]HG WR DIIHFW MXGJHVf GHFLVLRQPDNLQJ LQ FDVHV RI DEXVH RU QHJOHFW 6SHFLILFDOO\ LW ZDV SURSRVHG WKDW SRRUHU SDUHQW SV\FKRORJLFDO IXQFWLRQLQJ KLJK GHJUHH RI SDUHQWDO QRQFRPSOLDQFH ZLWK WKH LQYHVWLJDWLRQ WKH SUHVHQFH RI VSRXVH DEXVH DQG VXEVWDQFH DEXVH KLJK GHJUHH RI IDPLO\ GLVFRUG DQG ORZ GHJUHH RI SDUHQWDO ZDUPWK DQG VXSSRUW IRU WKH FKLOG ZRXOG SUHGLFW SODFHPHQW RI WKH FKLOG RXWVLGH RI WKH KRPH +RZHYHU RQO\ WKH GHJUHH RI WKH SDUHQWVf ZDUPWK DQG VXSSRUW IRU WKH FKLOG S f HPHUJHG DV VLJQLILFDQWO\ UHODWHG WR MXGJHVf GHFLVLRQPDNLQJ ZKHQ WKH YDULDEOHV ZHUH VLPXOWDQHRXVO\ HQWHUHG LQWR WKH SUHGLFWLRQ HTXDWLRQ VHH 7DEOH f 7KH SRVLWLYH SDUDPHWHU HVWLPDWH LQGLFDWHV WKDW ORZHU OHYHOV RI SDUHQWDO ZDUPWK DQG VXSSRUW ZKLFK ZHUH WUHDWHG ILUVW LQ WKH PRGHOf ZHUH DVVRFLDWHG ZLWK D ODUJHU SUREDELOLW\ RI KDYLQJ WKH FKLOG UHPRYHG IURP WKH KRPH

PAGE 41

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r r S

PAGE 42

7KH UHVXOWV RI WKH DQDO\VLV XVLQJ WKH *XDUGLDQfV UHFRPPHQGDWLRQ DV DQ LQGHSHQGHQW YDULDEOH \LHOGHG VLPLODU UHVXOWV VHH 7DEOH f /LNH MXGJHV *XDUGLDQV ZHUH PRUH OLNHO\ WR UHFRPPHQG UHPRYDO RI FKLOGUHQ IURP KRPHV LQ ZKLFK SDUHQWV VKRZHG OLWWOH ZDUPWK DQG VXSSRUW IRU WKHP S f ,QWHUHVWLQJO\ KRZHYHU QRQH RI WKH RWKHU YDULDEOHV UHJDUGLQJ IDPLO\ SV\FKRORJLFDO IXQFWLRQLQJ VLJQLILFDQWO\ LPSDFWHG WKH *XDUGLDQVf GHFLVLRQPDNLQJ ,W WKHUHIRUH DSSHDUV WKDW QHLWKHU MXGJHV QRU *XDUGLDQV UHO\ KHDYLO\ RQ LQIRUPDWLRQ UHJDUGLQJ WKH SV\FKRORJLFDO IXQFWLRQLQJ RI WKH IDPLO\ ZKHQ PDNLQJ GHFLVLRQV UHJDUGLQJ WKH SODFHPHQW RI WKH DEXVHG FKLOG 0RGHOV ,QFOXGLQJ $OO 6LJQLILFDQW 3UHGLFWRUV ,Q RUGHU WR GHWHUPLQH WKH EHVW RYHUDOO SUHGLFWLYH PRGHO RI MXGLFLDO GHFLVLRQn PDNLQJ D ORJLVWLF UHJUHVVLRQ HTXDWLRQ ZDV UXQ ZKLFK LQFOXGHG RQO\ WKH SUHGLFWRU YDULDEOH IRXQG WR VLJQLILFDQWO\ LPSDFW MXGLFLDO GHFLVLRQV IURP WKH DERYH PRGHOVf§OHYHO RI SDUHQWDO ZDUPWK DQG VXSSRUW $JDLQ WKH SRVLWLYH SDUDPHWHU HVWLPDWH REWDLQHG IRU WKH SUHGLFWRU YDULDEOH LQ WKLV PRGHO LQGLFDWHV WKDW ORZHU GHJUHHV RI ZDUPWK DQG VXSSRUW ZHUH PRUH DVVRFLDWHG ZLWK UHPRYDO IURP WKH KRPH &DOFXODWLRQ RI SUHGLFWHG SUREDELOLWLHV EDVHG RQ WKH REWDLQHG SDUDPHWHU HVWLPDWHV LQGLFDWHG WKDW WKH SUREDELOLW\ RI UHPRYDO ZDV b LQ FDVHV ZKHUH SDUHQWDO ZDUPWK DQG VXSSRUW ZDV UDWHG WKH ORZHVW SRVVLEOH OHYHOf DQG b LQ FDVHV ZKHUH SDUHQWDO ZDUPWK DQG VXSSRUW ZDV UDWHG WKH KLJKHVW SRVVLEOH OHYHOf 7KH +RVPHU/HPHVKRZ JRRGQHVVRIILW VWDWLVWLF ZLWK ') S f LQGLFDWHV WKDW WKH REVHUYHG UHVSRQVHV GR QRW GLIIHU VLJQLILFDQWO\ IURP WKRVH ZKLFK ZRXOG RFFXU EDVHG RQ WKH SUHGLFWHG SUREDELOLWLHV IURP WKH PRGHO VHH 7DEOH f

PAGE 43

7DEOH 5HVXOWV RI /RJLVWLF 5HJUHVVLRQ $QDO\VLV 8VLQJ )DPLO\ 3V\FKRORJLFDO 9DULDEOHV WR 3UHGLFW -XGLFLDO 'HFLVLRQV )DPLO\ 3V\FKRORJLFDO 9DULDEOHV 3UHGLFWLQJ -XGLFLDO 'HFLVLRQV ') 3DUDPHWHU (VWLPDWH 6WDQGDUG (UURU &KL 6TXDUH 3UREDELOLW\ RI &KL6TXDUH ,QWHUFHSW 3V\FKRORJLFDO )XQFWLRQLQJ 3DUHQWDO :DUPWK DQG 6XSSRUW IRU &KLOG r )DPLO\ &RQIOLFW 3DUHQWDO 1RQFRPSOLDQFH 6XEVWDQFH $EXVH 6SRXVH $EXVH r S 7DEOH 5HVXOWV RI /RJLVWLF 5HJUHVVLRQ $QDO\VLV 8VLQJ )DPLO\ 3V\FKRORJLFDO 9DULDEOHV WR 3UHGLFW *XDUGLDQ 5HFRPPHQGDWLRQV )DPLO\ 3V\FKRORJLFDO 9DULDEOHV 3UHGLFWLQJ *XDUGLDQ 5HFRPPHQGDWLRQV ') 3DUDPHWHU (VWLPDWH 6WDQGDUG (UURU &KL 6TXDUH 3UREDELOLW\ RI &KL6TXDUH ,QWHUFHSW 3V\FKRORJLFDO )XQFWLRQLQJ 3DUHQWDO :DUPWK DQG 6XSSRUW IRU &KLOG r )DPLO\ &RQIOLFW 3DUHQWDO 1RQFRPSOLDQFH 6XEVWDQFH $EXVH 6SRXVH $EXVH S

PAGE 44

$ VLPLODU DQDO\VLV ZDV FRQGXFWHG XVLQJ WKH LQGHSHQGHQW YDULDEOHV IRXQG WR VLJQLILFDQWO\ LPSDFW WKH GHFLVLRQV RI WKH *XDUGLDQV :KHQ WKH YDULDEOHV RI SUHYLRXV UHSRUWV RI DEXVH DQG SDUHQWDO ZDUPWK DQG VXSSRUW IRU WKH FKLOG ZHUH HQWHUHG LQWR WKH HTXDWLRQ RQO\ WKH SUHVHQFH RI SUHYLRXV UHSRUWV RI DEXVH S f HPHUJHG DV VWURQJO\ UHODWHG WR *XDUGLDQfV GHFLVLRQV 6LQFH LQFOXVLRQ RI WKH YDULDEOH TXDQWLI\LQJ SDUHQWDO ZDUPWK DQG VXSSRUW IRU WKH FKLOG GLG QRW VLJQLILFDQWO\ DGG WR WKH SUHGLFWLYH SRZHU RI WKH HTXDWLRQ WKH DQDO\VLV ZDV FRQGXFWHG DJDLQ XVLQJ RQO\ WKH SUHVHQFH RI SUHYLRXV UHSRUWV DV D SUHGLFWRU YDULDEOH 7KH QHJDWLYH SDUDPHWHU HVWLPDWH LQGLFDWHV WKDW WKH SUHVHQFH RI SUHYLRXV UHSRUWV RI DEXVH ZDV DVVRFLDWHG ZLWK D KLJKHU SUREDELOLW\ RI PDNLQJ D UHFRPPHQGDWLRQ IRU UHPRYDO 7KH +RVPHU/HPHVKRZ JRRGQHVVRIILW VWDWLVWLF ZLWK ') S f LQGLFDWHV WKDW WKH REVHUYHG UHVSRQVHV GR QRW VLJQLILFDQWO\ GLIIHU IURP WKRVH ZKLFK ZRXOG RFFXU JLYHQ WKH SUHGLFWHG SUREDELOLWLHV EDVHG RQ WKLV PRGHO VHH 7DEOH f &RPSDULVRQ RI &DVHV RI 3K\VLFDO $EXVH 2QO\ DQG 1HJOHFW 2QO\ $Q RULJLQDO DLP RI WKLV VWXG\ ZDV WR GHWHUPLQH ZKHWKHU MXGLFLDO DQG *XDUGLDQ GHFLVLRQPDNLQJ SURFHVVHV GLIIHUHG EHWZHHQ FDVHV RI SK\VLFDO DEXVH DORQH DQG QHJOHFW DORQH %HFDXVH WKH VDPSOH LQFOXGHG RQO\ FDVHV VROHO\ LQYROYLQJ SK\VLFDO DEXVH DQG FDVHV VROHO\ LQYROYLQJ QHJOHFW LW ZDV QRW SRVVLEOH WR DQDO\]H WKH GDWD XVLQJ ORJLVWLF UHJUHVVLRQ HTXDWLRQVf§ZKLFK UHTXLUH D ODUJHU VDPSOH VL]H WR EH LQWHUSUHWLYHO\ XVHIXO $QDO\VLV XVLQJ WKH 6KDSLUR:LON VWDWLVWLF :f LQGLFDWHG WKDW DOO LQWHUYDO LQGHSHQGHQW YDULDEOHV ZHUH QRUPDOO\ GLVWULEXWHG LQ WKH VDPSOHV 7KH PHDQV RI WKHVH YDULDEOHV ZHUH FRPSDUHG XVLQJ WWHVWV LQ DQ DWWHPSW WR GHWHUPLQH ZKHWKHU SRWHQWLDOO\ LPSRUWDQW GLIIHUHQFHV H[LVW DPRQJ GHPRJUDSKLF OHJDO RU IDPLO\ SV\FKRORJLFDO FKDUDFWHULVWLFV

PAGE 45

7DEOH 5HVXOWV RI /RJLVWLF 5HJUHVVLRQ $QDO\VLV 8VLQJ $OO 6LJQLILFDQW 3UHGLFWRU 9DULDEOHV WR 3UHGLFW -XGLFLDO 'HFLVLRQV 3UHGLFWLQJ -XGLFLDO 'HFLVLRQV $OO 6LJQLILFDQW 3UHGLFWRUV ') 3DUDPHWHU (VWLPDWH 6WDQGDUG (UURU &KL 6TXDUH 3UREDELOLW\ RI &KL6TXDUH ,QWHUFHSW 3DUHQWDO :DUPWK DQG 6XSSRUW IRU &KLOG r rS +RVPHU/HPHVKRZ JRRGQHVVRIILW VWDWLVWLF ZLWK ') S 7DEOH 5HVXOWV RI /RJLVWLF 5HJUHVVLRQ $QDO\VLV 8VLQJ $OO 6LJQLILFDQW 3UHGLFWRU 9DULDEOHV WR 3UHGLFW *XDUGLDQ 5HFRPPHQGDWLRQV 3UHGLFWLQJ *XDUGLDQ 5HFRPPHQGDWLRQV $OO 6LJQLILFDQW 3UHGLFWRUV ') 3DUDPHWHU (VWLPDWH 6WDQGDUG (UURU &KL 6TXDUH 3UREDELOLW\ RI &KL6TXDUH ,QWHUFHSW 3UHYLRXV 5HSRUWV RI $EXVH r r S +RVPHU/HPHVKRZ JRRGQHVVRIILW VWDWLVWLF ZLWK ') S

PAGE 46

EHWZHHQ WKH WZR JURXSV VHH $SSHQGL[ 'f ,Q DGGLWLRQ WKH IUHTXHQFLHV RI WKH GLFKRWRPRXV LQGHSHQGHQW DQG GHSHQGHQW YDULDEOHV LQ HDFK JURXS ZHUH UHYLHZHG DOWKRXJK WKH QXPEHU RI FDVHV ZDV LQVXIILFLHQW WR VXSSRUW &KLVTXDUH DQDO\VLV 5HVXOWV RI FRPSDULVRQV EHWZHHQ FDVHV RI SK\VLFDO DEXVH DORQH DQG QHJOHFW DORQH XVLQJ WWHVWV LQGLFDWHG WKDW WKH OHYHO RI VHYHULW\ RI QHJOHFW WHQGHG WR EH PRUH H[WUHPH WKDQ WKH OHYHO RI VHYHULW\ RI DEXVH W S f DQG WKDW WKHUH WHQGHG WR EH OHVV FRQIOLFW EHWZHHQ SDUHQWV RI QHJOHFWHG FKLOGUHQ WKDQ EHWZHHQ SDUHQWV RI DEXVHG FKLOGUHQ W S f 1R VLJQLILFDQW GLIIHUHQFHV EHWZHHQ WKH WZR JURXSV ZHUH IRXQG LQ FKLOG DJH IDPLO\ LQFRPH OHYHO RI SDUHQWDO SV\FKRORJLFDO IXQFWLRQLQJ GHJUHH RI SDUHQWDO ZDUPWK DQG VXSSRUW IRU WKH FKLOG RU SDUHQWDO QRQFRPSOLDQFH ZLWK WKH LQYHVWLJDWLRQ 9LVXDO UHYLHZ RI WKH IUHTXHQFLHV RI GLFKRWRPRXV YDULDEOHV LQ WKH WZR JURXSV UHYHDOHG WKH IROORZLQJ f -XGJHV ZHUH VRPHZKDW PRUH OLNHO\ WR UHPRYH QHJOHFWHG FKLOGUHQ IURP WKHLU KRPHV b RI QHJOHFWHG FKLOGUHQ DV RSSRVHG WR b RI DEXVHG FKLOGUHQf f *XDUGLDQV ZHUH OHVV OLNHO\ WR UHFRPPHQG UHPRYDO RI QHJOHFWHG FKLOGUHQ WKDQ DEXVHG FKLOGUHQ b RI QHJOHFWHG FKLOGUHQ DV RSSRVHG WR b RI DEXVHG FKLOGUHQf f $PRQJ QHJOHFW FDVHV WKHUH WHQGHG WR EH IHZHU FDVHV LQYROYLQJ SUHYLRXV UHSRUWV RI DEXVH RU QHJOHFW DJDLQVW WKH FKLOG b DV RSSRVHG WR b LQ FDVHV RI SK\VLFDO DEXVHf f ,Q FDVHV RI QHJOHFW DORQH WKHUH ZHUH IHZHU FDVHV LQYROYLQJ VSRXVH DEXVH b DV RSSRVHG WR b LQ IDPLOLHV RI DEXVHG FKLOGUHQf DQG f 5HSRUWV RI VXEVWDQFH DEXVH ZHUH PRUH IUHTXHQW DPRQJ SDUHQWV RI QHJOHFWHG FKLOGUHQ b DV RSSRVHG WR b LQ FDVHV RI SK\VLFDO DEXVHf )DPLO\ UDFH DQG WKH QXPEHU RI QRQSDUHQW DOOHJHG SHUSHWUDWRUV GLG QRW GLIIHU VLJQLILFDQWO\ EHWZHHQ WKH WZR JURXSV

PAGE 47

%HFDXVH WKHVH ILQGLQJV DUH EDVHG RQ YHU\ VPDOO VDPSOHV FDXWLRQ PXVW EH XVHG LQ WKHLU LQWHUSUHWDWLRQ 7KH QXPEHU DQG PDJQLWXGH RI WKH GLIIHUHQFHV IRXQG KRZHYHU LQGLFDWH WKH SRVVLELOLW\ RI VXEVWDQWLYH GLIIHUHQFHV EHWZHHQ WKH FKDUDFWHULVWLFV RI FDVHV RI SK\VLFDO DEXVH RQO\ DQG QHJOHFW RQO\ ,W WKXV ORJLFDOO\ IROORZV WKDW LPSRUWDQW GLIIHUHQFHV PD\ H[LVW LQ WKH SURFHVV RI GHWHUPLQLQJ DSSURSULDWH DGMXGLFDWLRQ IRU WKH WZR GLIIHUHQW JURXSV

PAGE 48

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b RI WKH IDPLOLHV LQ WKH FXUUHQW VWXG\ ZHUH ZKLWH DQG b ZHUH $IULFDQ $PHULFDQ +LVSDQLF DQG RWKHU UDFHV b b DQG b UHVSHFWLYHO\f 2I WKH FDVHV LQ ZKLFK LQIRUPDWLRQ RQ IDPLO\ LQFRPH ZDV SURYLGHG WKH PHDQ LQFRPH ZDV SHU \HDU VXJJHVWLQJ WKDW WKH PDMRULW\ RI WKH IDPLOLHV OLYH LQ SRYHUW\ $OO LQWHUYDO YDULDEOHV UHODWHG WR WKH DEXVLYH VLWXDWLRQ DQG WR WKH IDPLO\fV SV\FKRORJLFDO IXQFWLRQLQJ LQ WKH VWXG\ ZHUH UDWHG RQ YLVXDO DQDORJ VFDOHV ZLWK YDOXHV UDQJLQJ IURP WR 7KH DYHUDJH OHYHO RI VHYHULW\ RI DEXVH ZDV ZLWK EHLQJ WKH PRVW VHYHUHf LQGLFDWLQJ WKDW WKHVH FDVHV JHQHUDOO\ UHSUHVHQW WKH PRVW VHYHUH LQFLGHQWV RI DEXVH DQG QHJOHFW $W OHDVW RQH RI WKH SDUHQWV ZDV WKH DOOHJHG SHUSHWUDWRU LQ b RI WKH FDVHV DQG WKHUH H[LVWHG SUHYLRXV UHSRUWV RI DEXVH LQ b RI WKH FDVHV 0XOWLSOH W\SHV RI DEXVH ZHUH UHSRUWHG LQ b RI WKH FDVHV DQG VH[XDO DEXVH DOOHJDWLRQV ZHUH SUHVHQW HLWKHU DORQH RU LQ FRPELQDWLRQ ZLWK DEXVH DQGRU QHJOHFWf LQ b RI WKH FDVHV

PAGE 49

7DEOH 'LVWULEXWLRQ RI ,QWHUYDO 9DULDEOHV 9DULDEOH 1 0HDQ 6' 0HGLDQ &KLOG $JH )DPLO\ ,QFRPH 6HYHULW\ RI $EXVH 3DUHQWDO 3V\FKRORJLFDO )XQFWLRQLQJ 3DUHQWDO :DUPWK DQG 6XSSRUW IRU &KLOG )DPLO\ &RQIOLFW 3DUHQWDO 1RQFRPSOLDQFH 7DEOH )UHTXHQFLHV RI 'LFKRWRPRXV 9DULDEOHV 9DULDEOH )DPLO\ 5DFH 3DUHQW 3HUSHWUDWRU 0XOWLSOH 7\SHV RI $EXVH 3UHYLRXV 5HSRUWV RI $EXVH 6H[XDO $EXVH 6SRXVH $EXVH 6XEVWDQFH $EXVH )UHTXHQF\ 3HUFHQW :KLWH 1RQ:KLWH $OOHJHG 1RW $OOHJHG $OOHJHG 1RW $OOHJHG
PAGE 50

7KHUH ZHUH DOOHJDWLRQV RI VSRXVH DEXVH LQ DGGLWLRQ WR FKLOG DEXVHf LQ b RI WKH FDVHV DQG DOOHJDWLRQV RI VXEVWDQFH DEXVH ZHUH PDGH LQ b RI WKH FDVHV 7KH DYHUDJH UDWLQJ IRU RYHUDOO SV\FKRORJLFDO IXQFWLRQLQJ RI WKH IDPLO\ ZDV ZLWK EHLQJ WKH SRRUHVW SRVVLEOH OHYHO RI SV\FKRORJLFDO IXQFWLRQLQJf DQG WKH DYHUDJH UDWLQJ IRU SDUHQWDO ZDUPWK DQG VXSSRUW IRU WKH FKLOG ZDV ZLWK EHLQJ WKH KLJKHVW SRVVLEOH OHYHO RI ZDUPWK DQG VXSSRUWf 7KH DYHUDJH FRQIOLFW OHYHO UHSRUWHG IRU WKH IDPLOLHV ZDV ZLWK EHLQJ WKH KLJKHVW SRVVLEOH OHYHO RI FRQIOLFWf DQG WKH DYHUDJH GHJUHH RI SDUHQWDO QRQFRPSOLDQFH ZLWK WKH LQYHVWLJDWLRQ ZDV ZLWK EHLQJ WKH KLJKHVW SRVVLEOH OHYHO RI QRQFRPSOLDQFHf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b ZKHQ WKH *XDUGLDQ KDV UHFRPPHQGHG WKDW WKH FKLOG EH UHPRYHG 7KH PDLQ REMHFWLYH RI WKH FXUUHQW VWXG\ ZDV WR GHWHUPLQH ZKLFK GHPRJUDSKLF OHJDO DQG IDPLO\ SV\FKRORJLFDO YDULDEOHV DUH PRVW SUHGLFWLYH RI GHFLVLRQV WR UHPRYH RU UHFRPPHQG UHPRYDO RIf FKLOGUHQ IURP WKHLU KRPHV LQ FDVHV LQYROYLQJ FKLOG DEXVH DQG QHJOHFW ,W LV LPSRUWDQW DW WKLV SRLQW WR QRWH DJDLQ WKDW DOO FDVHV LQFOXGHG LQ WKLV VWXG\ KDG

PAGE 51

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f ZKLFK ZDV QRW DFFRXQWHG IRU LQ WKLV VWXG\ 2YHUDOO LQIRUPDWLRQ UHJDUGLQJ WKH GHJUHH RI SDUHQWDO ZDUPWK DQG VXSSRUW IRU WKH FKLOG ZDV PRVW SUHGLFWLYH RI ILQDO FDVH RXWFRPH 7KH UHVXOWV RI WKH VHSDUDWH DQDO\VHV XVLQJ *XDUGLDQVf UHFRPPHQGDWLRQV DV WKH GHSHQGHQW YDULDEOH LQGLFDWHG WKDW WKH SUHVHQFH RI SUHYLRXV UHSRUWV RI DEXVH DQG RQ WKH GHJUHH RI SDUHQWDO ZDUPWK DQG VXSSRUW IRU WKH FKLOG DUH VLJQLILFDQWO\ DVVRFLDWHG ZLWK WKRVH UHFRPPHQGDWLRQV 7KH UHVXOWV RI D ORJLVWLF UHJUHVVLRQ HTXDWLRQ XVLQJ RQO\ WKHVH SUHGLFWRU YDULDEOHV KRZHYHU LQGLFDWHG WKDW RQO\ WKH SUHVHQFH RI SUHYLRXV UHSRUWV RI DEXVH RU QHJOHFW DJDLQVW WKH FKLOG LV VLJQLILFDQWO\ UHODWHG WR *XDUGLDQVf UHFRPPHQGDWLRQV IRU UHPRYDO 7KXV DW OHDVW LQ WKLV VDPSOH UHFXUUHQW DEXVH DJDLQVW D FKLOG LV WKH VLQJOH PRVW SUHGLFWLYH IDFWRU RI *XDUGLDQVf UHFRPPHQGDWLRQ WKDW KH RU VKH EH UHPRYHG IURP WKH KRPH

PAGE 52

7KH UHVXOWV RI SUHYLRXV VWXGLHV KDYH LQGLFDWHG WKDW MXGJHV DUH PRUH OLNHO\ WR UHPRYH ROGHU FKLOGUHQ IURP WKH KRPH WKDQ WKH\ DUH WR UHPRYH \RXQJHU FKLOGUHQ HJ .DW] HW DO f ,Q WKLV VDPSOH KRZHYHU FKLOG DJH KDG QR GLVFHUQLEOH HIIHFW RQ HLWKHU MXGJHVf RU *XDUGLDQVf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f QRU *XDUGLDQVf GHFLVLRQV ZHUH VLJQLILFDQWO\ DVVRFLDWHG ZLWK WKH IDFW WKDW WKH DEXVH ZDV YHU\ VHYHUH RU RQ WKH IDFW WKDW PXOWLSOH W\SHV RI DEXVH ZHUH DOOHJHG ,QWHUHVWLQJO\ ZKLOH WKH SUHVHQFH RI SUHYLRXV UHSRUWV RI DEXVH ZDV IRXQG WR EH WKH PRVW VLJQLILFDQW SUHGLFWRU RI *XDUGLDQfV GHFLVLRQV WR UHFRPPHQG UHPRYDO RI FKLOGUHQ IURP WKHLU KRPHV WKLV LQIRUPDWLRQ ZDV QRW VLJQLILFDQWO\ UHODWHG WR MXGLFLDO GHFLVLRQPDNLQJ $PRQJ WKH YDULDEOHV UHODWLQJ WR WKH SV\FKRORJLFDO IXQFWLRQLQJ RI WKH IDPLO\ RQO\ WKH OHYHO RI WKH SDUHQWVf ZDUPWK DQG VXSSRUW IRU WKH FKLOG ZDV IRXQG WR EH VLJQLILFDQWO\ DVVRFLDWHG ZLWK GHFLVLRQV UHJDUGLQJ FDVH RXWFRPH ,QIRUPDWLRQ UHJDUGLQJ WKH SV\FKRORJLFDO IXQFWLRQLQJ RI WKH SDUHQWV WKH SDUHQWVf FRPSOLDQFH ZLWK WKH LQYHVWLJDWLRQ WKH GHJUHH RI IDPLO\ GLVFRUG RU WKH SUHVHQFH RI DOOHJDWLRQV RI VSRXVH RU VXEVWDQFH DEXVH DPRQJ WKH SDUHQWV ZDV QRW UHODWHG WR MXGJHVf RU *XDUGLDQVf GHFLVLRQV

PAGE 53

2QH VRXUFH RI FRQFHUQ ZKLFK DULVHV IURP WKH UHVXOWV RI WKH FXUUHQW VWXG\ LV WKH ODFN RI DVVRFLDWLRQ EHWZHHQ MXGJHVf DQG *XDUGLDQVf GHFLVLRQV DQG WKH VHYHULW\ OHYHO RI WKH DOOHJHG DEXVH )ORULGD 6WDWXWH VHFWLRQ VWLSXODWHV WKDW JURXQGV IRU WHUPLQDWLRQ RI SDUHQWDO ULJKWV LQFOXGH HJUHJLRXV DEXVH ZKLFK LV GHILQHG DV FRQGXFW WKDW HQGDQJHUV WKH OLIH KHDOWK RU VDIHW\ RI WKH FKLOG
PAGE 54

WKRVH FKLOGUHQ HJ /\QFK DQG 5REHUWV 0DUWLQ *UHHQ f IDFWRUV VXFK DV WKHVH ZHUH QRW IRXQG WR SUHGLFW GHFLVLRQV WR UHPRYH FKLOGUHQ IURP WKHLU KRPHV 5HYLHZ RI )ORULGD 6WDWXWH VHFWLRQ SURYLGHV D OLNHO\ H[SODQDWLRQ IRU WKH DERYH ILQGLQJ -XGJHV DUH UHTXLUHG WR FRQVLGHU WKH ORYH DIIHFWLRQ DQG RWKHU HPRWLRQDO WLHV H[LVWLQJ EHWZHHQ WKH FKLOG DQG WKH FKLOGfV SUHVHQW SDUHQW RU VLEOLQJVf§ZKLFK WKH\ DSSHDU WR GR JLYHQ WKH VLJQLILFDQW DVVRFLDWLRQ EHWZHHQ WKH GHJUHH RI SDUHQWDO ZDUPWK DQG VXSSRUW IRU WKH FKLOG DQG ILQDO DGMXGLFDWLRQ +RZHYHU ZKLOH MXGJHV DUH LQVWUXFWHG WR FRQVLGHU WKH FDSDFLW\ RI WKH SDUHQW RU SDUHQWV WR FDUH IRU WKH FKLOG WR WKH H[WHQW WKDW WKH FKLOGfV KHDOWK DQG ZHOOEHLQJ ZLOO QRW EH HQGDQJHUHG XSRQ WKH FKLOGfV UHWXUQ KRPH WKHUH LV QR VSHFLILF UHTXLUHPHQW WR FRQVLGHU WKH SV\FKRORJLFDO IXQFWLRQLQJ RI WKH SDUHQWV $JDLQ JLYHQ RXU FXUUHQW NQRZOHGJH DERXW WKH GHWULPHQWDO HIIHFWV RI SRRU SDUHQW SV\FKRORJLFDO IXQFWLRQLQJ RQ WKH ZHOOEHLQJ RI FKLOGUHQ IDLOXUH WR FRQVLGHU WKHVH IDFWRUV ZKHQ GHWHUPLQLQJ ZKHWKHU WR UHPRYH FKLOGUHQ IURP WKHLU KRPHV ZRXOG UHSUHVHQW D JURVV RYHUVLJKW 5HVXOWV RI FRPSDULVRQ RI WKH YDULDEOHV DPRQJ JURXSV RI FKLOGUHQ ZKR ZHUH SK\VLFDOO\ DEXVHG EXW QRW QHJOHFWHG RU VH[XDOO\ DEXVHGf DQG WKRVH ZKR ZHUH QHJOHFWHG EXW QRW SK\VLFDOO\ RU VH[XDOO\ DEXVHGf DOVR LQGLFDWHG VHYHUDO SRWHQWLDOO\ LPSRUWDQW GLIIHUHQFHV LQ ERWK FDVH W\SH DQG RXWFRPHf§PRVW QRWDEO\ WKDW ZKLOH *XDUGLDQV DUH VRPHZKDW OHVV OLNHO\ WR UHFRPPHQG UHPRYDO RI FKLOGUHQ LQ FDVHV RI QHJOHFW DORQH MXGJHV DUH VRPHZKDW PRUH OLNHO\ WR DGMXGLFDWH UHPRYDO 6XFK D ILQGLQJ UDLVHV TXHVWLRQV UHJDUGLQJ ZKHWKHU WKH GHFLVLRQPDNLQJ SURFHVVHV RI MXGJHV DQG *XDUGLDQV GLIIHU ZKHQ DOOHJDWLRQV LQFOXGH GLIIHUHQW W\SHV RU FRPELQDWLRQV RI DEXVH RU QHJOHFW $V D UHVXOW LW

PAGE 55

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

PAGE 56

:KLOH IHGHUDO ODZV UHJDUGLQJ FDVHV LQYROYLQJ DOOHJDWLRQV RI FKLOG DEXVH RU QHJOHFW SURYLGH VRPH GHJUHH RI XQLIRUPLW\ LQ GHWHUPLQLQJ FDVH GLVSRVLWLRQ LW PD\ EH WKDW VRPH VWDWHV KDYH OHJDO JXLGHOLQHV DQG SROLWLFDO FOLPDWHV ZKLFK GLIIHU IURP WKH VWDWH RI )ORULGD ,W LV WKHUHIRUH UHFRPPHQGHG WKDW WKH UHVXOWV REWDLQHG KHUH QRW EH JHQHUDOL]HG WR RWKHU JHRJUDSKLF UHJLRQV SULRU WR IXUWKHU VWXG\ LQ WKRVH DUHDV $QRWKHU DVSHFW RI WKH FXUUHQW VDPSOH ZKLFK GHVHUYHV FRQVLGHUDWLRQ LV WKDW WKH FDVHV LQYROYHG ERWK LQLWLDO GHSHQGHQF\ KHDULQJV DQG WHUPLQDWLRQ RI SDUHQWDO ULJKWV KHDULQJV 7KXV VRPH RI WKH FKLOGUHQ ZHUH UHPRYHG IURP WKHLU KRPHV RQ D WHPSRUDU\ EDVLV ZKLOH RWKHUV ZHUH SHUPDQHQWO\ UHPRYHG %HFDXVH FRXUW SHUVRQQHO PD\ XVH VRPHZKDW GLIIHUHQW FULWHULD ZKHQ GHFLGLQJ ZKHWKHU SDUHQWDO ULJKWV VKRXOG EH VHYHUHG LW LV UHFRPPHQGHG WKDW DGGLWLRQDO UHVHDUFK EH FRQGXFWHG RQ WKHVH WZR JURXSV LQGHSHQGHQWO\ RI HDFK RWKHU )XWXUH UHVHDUFK LV FOHDUO\ QHHGHG WR DVVHVV WKH JHQHUDOL]DELOLW\ RI WKH FXUUHQW ILQGLQJV WR RWKHU SRSXODWLRQV ,W LV DOVR UHFRPPHQGHG WKDW D VLPLODU DQDO\VLV EH FRQGXFWHG XVLQJ D ODUJHU VDPSOH RI D FRPSDUDEOH SRSXODWLRQ LQ RUGHU WR UHSOLFDWH WKHVH ILQGLQJV DQG WR GHWHUPLQH ZKHWKHU WKH SURFHVV RI *XDUGLDQ DG /LWHP DQG MXGLFLDO GHFLVLRQPDNLQJ FKDQJHV RYHU WLPH )XUWKHU UHVHDUFK LV DOVR QHHGHG WR FODULI\ WKH UHVXOWV REWDLQHG LQ WKH FXUUHQW VWXG\ 2QH UDWKHU SX]]OLQJ ILQGLQJ RI WKLV VWXG\ ZDV WKDW ZKLOH MXGJHV WHQG WR DJUHH ZLWK WKH UHFRPPHQGDWLRQV RI *XDUGLDQV WKH\ DSSHDU WR SODFH D VRPHZKDW GLIIHUHQW HPSKDVLV RQ WKH YDULDEOHV WKH\ XVH WR GHWHUPLQH WKH ILQDO DGMXGLFDWLRQ RQ WKH FDVHV WKH\ FRQVLGHU 7KLV ILQGLQJ VXJJHVWV WKDW ZKLOH MXGJHV UHO\ RQ FDVH LQIRUPDWLRQ SURYLGHG WR WKHP E\ WKH *XDUGLDQV WKH\ PD\ QRW UHO\ KHDYLO\ RQ WKH *XDUGLDQVf UHFRPPHQGDWLRQV )XWXUH

PAGE 57

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

PAGE 58

5()(5(1&(6 $OWHU &) f 'HFLVLRQPDNLQJ IDFWRUV LQ FDVHV RI FKLOG QHJOHFW &KLOG :HOIDUH %URZQ $ t )LQNHOKRU f ,PSDFW RI FKLOG VH[XDO DEXVH $ UHYLHZ RI WKH UHVHDUFK 3V\FKRORJLFDO %XOOHWLQ 'DUR f &RQIURQWLQJ FKLOG DEXVH 5HVHDUFK IRU HIIHFWLYH SURJUDP GHVLJQ 1HZ
PAGE 59

+RVPHU ': t /HPHVKRZ 6 f $SSOLHG /RJLVWLF 5HJUHVVLRQ 1HZ
PAGE 60

:DVVHUPDQ 6 t 5RVHQIHOG $ f 'HFLVLRQPDNLQJ LQ FKLOG DEXVH DQG QHJOHFW &KLOG :HOIDUH :KLWFRPE f *XDUGLDQV $G /LWHP LQ FULPLQDO FRXUWV :DVKLQJWRQ 86 'HSDUWPHQW RI -XVWLFH :LLJ -. f )XQFWLRQV RI WKH *XDUGLDQ DG /LWHP LQ FKLOG DEXVH DQG QHJOHFW SURFHHGLQJVf§/RV $QJHOHV -XYHQLOH &RXUW ,Q + /DQGDX HGf 7KH DEXVHG DQG QHJOHFWHG FKLOG 3UDFWLVLQJ /DZ ,QVWLWXWH

PAGE 61

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fV IDPLO\ 7KLV VHFWLRQ UHTXLUHV WKDW \RX PDUN RQ D OLQH WR LQGLFDWH \RXU UDWLQJ RI WKH LQWUDIDPLOLDO UHODWLRQVKLSV 6RPH RI WKHVH TXHVWLRQV UHTXLUH D VLQJOH PDUN RQ WKH OLQH VXFK DV IRU FRQIOLFW EHWZHHQ WKH WZR SDUHQWV &21)/,&7 %(7:((1 7+( 7:2 3$5(176 9HU\ /LWWOH 9HU\ 0XFK :KLOH RWKHU TXHVWLRQV UHTXLUH \RX WR PDNH WZR UDWLQJV RQ HDFK OLQH RQH IRU WKH PRWKHUFKLOG UHODWLRQVKLS DQG RQH IRU WKH IDWKHUFKLOG UHODWLRQVKLS &21)/,&7 %(7:((1 ($&+ 3$5(17 $1' 7+( &+,/'UHQf ZKR ZDV UHSRUWHGO\ DEXVHG RU QHJOHFWHG ZLWK DQ 0 ) 60 DQG DQ 6) 0 ) 9HU\ /LWWOH 9HU\ 0XFK )LQDOO\ WKH WKLUG VHFWLRQ VXUYH\V LQIRUPDWLRQ IURP WKH GLVSRVLWLRQ KHDULQJ RQ WKH FDVH LQ WHUPV RI \RXU UHFRPPHQGDWLRQV WR WKH FRXUW DQG WKH FRXUWfV ILQDO GHWHUPLQDWLRQ 3OHDVH UHWXUQ WKLV IRUP ZKHQ FRPSOHWHG WR \RXU *$/ SURJUDPfV VHFUHWDU\ 7KDQN \RX DJDLQ IRU \RXU KHOS

PAGE 62

'HSHQGHQF\ &DVH '(02*5$3+,& $1' /(*$/ ,1)250$7,21 :KR LV ILOOLQJ RXW WKLV IRUP" *XDUGLDQ &RRUGLQDWRU'LUHFWRU 2WKHU 'DWH RI WKLV KHDULQJ" -XGJH &RGH &LUFXLW &RXQW\ &LUFOH DQ\ LPPHGLDWH IDPLO\ PHPEHUV ZKR KDYH QRW EHHQ LQWHUYLHZHG E\ WKH *$/ &LUFOH DOO WKDW DSSO\f 0RWKHU )DWKHU &KLOG 2WKHU VSHFLI\f 6WHSPRWKHU 6WHSIDWKHU *UDQGSDUHQW $UH WKH SDUHQWV UHSUHVHQWHG E\ FRXQVHO
PAGE 63

+DV RQH RU PRUH RI WKH SDUHQWV ORVW WKHLU MREV ZLWKLQ WKH ODVW \HDU"
PAGE 64

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f6 ',675(66 RYHU WKH FXUUHQW IDPLO\ VLWXDWLRQ L L L 9HU\ /LWWOH L 9HU\ 0XFK &21),'(1&( LQ \RXU UDWLQJ L L L 1RQH L 7RWDO &21)/,&7 %(7:((1 7+( 7:2 3$5(176 L L L 9HU\ /LWWOH L 9HU\ 0XFK &21),'(1&( LQ \RXU UDWLQJ L L L 1RQH L 7RWDO

PAGE 65

3$5(176f :,//,1*1(66 72 5(3257 )$0,/< 352%/(06 LQ JHQHUDO L L L L 1RQH &RPSOHWHO\ &21),'(1&( LQ \RXU UDWLQJ L L L L 1RQH 7RWDO $PRXQW WKH )$0,/< )$9256 )$0,/< $&7,9,7,(6 L L L L 7RR /LWWOH $ERXW 5LJKW 7RR 0XFK &21),'(1&( LQ \RXU UDWLQJ L L L L 1RQH 7RWDO $PRXQW WKDW WKH IDPLO\ PHPEHUV &/$5,)< $1'25 63($. )25 RWKHU IDPLO\ PHPEHUV L L L L 9HU\ /LWWOH 9HU\ 0XFK &21),'(1&( LQ \RXU UDWLQJ L L L L 1RQH 7RWDO '(*5(( 72 :+,&+ 7+( 3+<6,&$/ 1(('6 2) 7+( &+,/'UHQf ZHUH PHW SULRU WR WKH UHFHQW UHSRUW RI DEXVHQHJOHFW L L L L 1RQH 7RWDO &21),'(1&( LQ \RXU UDWLQJ L L L L 1RQH 7RWDO 7KHVH QHHGV ZHUH SULPDULO\ PHW E\ FLUFOH RQHf 0RWKHU )DWKHU %RWK 2WKHU VSHFLI\f

PAGE 66

'(*5(( 72 :+,&+ 7+( (027,21$/ 1(('6 2) 7+( &+,/'UHQf ZHUH PHW SULRU WR WKH UHFHQW UHSRUW RI DEXVHQHJOHFW L L L L 1RQH 7RWDO &21),'(1&( LQ \RXU UDWLQJ L L L L 1RQH 7RWDO 7KHVH QHHGV ZHUH SULPDULO\ PHW E\ FLUFOH RQHf 0RWKHU )DWKHU %RWK 2WKHU VSHFLI\f )RU LWHPV XVH WKH IROORZLQJ OHWWHUV WKDW DSSO\ WR WKLV IDPLO\ WR UHSUHVHQW \RXU HYDOXDWLRQ RI WKH SDUHQWV RQ WKHVH LVVXHV 3ODFH D PDUN RQ WKH OLQH WR UHSUHVHQW \RXU UDWLQJ RI HDFK SDUHQW DQG ODEHO HDFK PDUN ZLWK RQH RI WKH IROORZLQJ OHWWHUV 3OHDVH PDNH VXUH WKDW \RX PDNH VHSDUDWH UDWLQJV RQ HDFK RI WKHP LQFOXGLQJ \RXU UDWLQJV RI WKH &21),'(1&( \RX KDYH LQ \RXU UDWLQJ RQ HDFK RI WKHP 0 0RWKHU 60 6WHSPRWKHU ) )DWKHU 6) 6WHSIDWKHU &21)/,&7 %(7:((1 ($&+ 3$5(17 $1' 7+( &+,/'UHQf ZKR ZDV UHSRUWHGO\ DEXVHG RU QHJOHFWHG ZLWK DQ 0 ) 60 DQG 6) L L L L 9HU\ /LWWOH 9HU\ 0XFK )RU HDFK SDUHQW LV WKHUH DV PXFK FRQIOLFW ZLWK RWKHU FKLOGUHQ"
PAGE 67

:$50 $1' 326,7,9( )((/,1*6 )25 7+( &+,/'UHQf ZKR ZDV UHSRUWHGO\ DEXVHG RU QHJOHFWHG KHOG E\ HDFK SDUHQW ZLWK DQ 0 ) 60 DQG 6) 9HU\ /LWWOH 9HU\ 0XFK &21),'(1&( LQ HDFK RI \RXU UDWLQJV ZLWK DQ 0 ) 60 DQG 6) c 1RQH 7RWDO ,192/9(0(17 RI HDFK SDUHQW ZLWK WKH FKLOGUHQffV GDLO\ OLYHV HJ SHHU UHODWLRQVKLSV FKLOGfV IHHOLQJV PLQRU LOOQHVVHVf ZLWK DQ 0 ) 60 DQG 6) c 1R ,QYROYHPHQW ([FHVVLYH ,QYROYHPHQW &21),'(1&( LQ HDFK RI \RXU UDWLQJV ZLWK DQ 0 ) 60 DQG 6) L L L L 1RQH 7RWDO ($&+ 3$5(17f6 5(/,$1&( 21 7+( &+,/'5(1 )25 &203$1,216+,3 &20)257 $1' 352%/(062/9,1* ZLWK DQ 0 ) 60 DQG 6) c 9HU\ /LWWOH 9HU\ 0XFK &21),'(1&( LQ HDFK RI \RXU UDWLQJV ZLWK DQ 0 ) 60 DQG 6) L L L L 1RQH 7RWDO 36<&+2/2*,&$/ ,668(6 IRU HDFK SDUHQW ZLWK DQ 0 ) 60 DQG 6) /RQHO\ +LJKO\ 6OLJKWO\

PAGE 68

'HSUHVVHG 6OLJKWO\ +LJKO\ $Q[LRXV L L L 6OLJKWO\ L +LJKO\ ,UULWDEOH L L L 6OLJKWO\ L +LJKO\ &21),'(1&( LQ HDFK RI \RXU UDWLQJV ZLWK DQ 0 ) 60 DQG 6) L L L L 1RQH 7RWDO 6$7,6)$&7,21 :,7+ 62&,$/ 6833257 WKDW HDFK SDUHQW UHFHLYHV IRU DQ\ SUREOHPV IURP UHODWLYHV IULHQGV FRZRUNHUV DQG FRPPXQLW\ DJHQFLHV ZLWK DQ 0 ) 60 DQG 6) L L L L ,QDGHTXDWH ([FHOOHQW &21),'(1&( LQ HDFK RI \RXU UDWLQJV ZLWK DQ 0 ) 60 DQG 6) L L L L 1RQH 7RWDO ($&+ 3$5(17f6 5(63216( 72 <28 LQ JHQHUDO ZLWK DQ 0 ) 60 DQG 6) L L L L 9HU\ )ULHQGO\ +RVWLOH5HMHFWLQJ &21),'(1&( LQ HDFK RI \RXU UDWLQJV ZLWK DQ 0 ) 60 DQG 6) L L L L 1RQH 7RWDO ,I HLWKHU SDUHQW ZDV DEXVHGQHJOHFWHG DV D FKLOG UHSRUW WKH W\SHV RI WKDW DEXVH 3 SK\VLFDO 6 VH[XDO ( HPRWLRQDO 1 1HJOHFW DQG 1$ QRQHf 0RWKHU )DWKHU 6WHSPRWKHU 6WHSIDWKHU

PAGE 69

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f '$7(6 727$/ 180%(5 3$67 35(6(17

PAGE 70

5HSRUW ZKHWKHU FULPLQDO FKDUJHV DUH EHLQJ EURXJKW DJDLQVW WKH DOOHJHG SHUSHWUDWRU IRU DEXVHQHJOHFW LQ WKLV FDVH DQG SUHYLRXVO\ 7KLV FDVH <(6 12 3UHYLRXVO\ KRZ PDQ\f ',6326,7,21 'RHV WKH DOOHJHG SHUSHWUDWRU DFFHSW UHVSRQVLELOLW\ WR WKH DEXVHQHJOHFW" <(6 12 ,QGLFDWH ZKR UHTXHVWHG D PHQWDO KHDOWK HYDOXDWLRQ IRU DQ\ IDPLO\ PHPEHUV DQG ZKR FRPSOHWHG LW 5(48(67(' %
PAGE 71

&KHFN DQ\ RI WKH IROORZLQJ RSWLRQV LW WKH\ ZHUH EHLQJ FRQVLGHUHG WKHQ FKHFN ZKLFK RSWLRQ ZDV UHFRPPHQGHG E\ WKH *XDUGLDQ DQG ZKLFK RSWLRQ ZDV DGMXGLFDWHG &RQVLGHUHG *XDUGLDQ $GMXGLFDWHG 3HUSHWUDWRU 5HPRYHG )URP WKH +RPH 3DUHQWV 5HWDLQ &XVWRG\ :LWK 6XSHUYLVLRQ 7HPSRUDU\ &XVWRG\ $ZDUGHG WR D 5HODWLYH 7HPSRUDU\ )RVWHU 3ODFHPHQW &KLOG 3ODFHG LQ D 7UHDWPHQW )DFLOLW\ 3ODFHPHQW IRU $GRSWLRQ &KLOG 3ODFHG LQ D 'HWHQWLRQ )DFLOLW\ 2WKHU VSHFLI\f 1RW 'HSHQGHQW DQG 5HWXUQHG +RPH

PAGE 72

$33(1',; % &255(/$7,216 $021* ,1'(3(1'(17 9$5,$%/(6 7DEOH %O (VWLPDWHG &RUUHODWLRQV $PRQJ 'HPRJUDSKLF 9DULDEOHV (QWHUHG LQ /RJLVWLF 5HJUHVVLRQ (TXDWLRQ 3UHGLFWLQJ -XGLFLDO 'HFLVLRQV &KLOG $JH )DPLO\ 5DFH &KLOG $JH )DPLO\ 5DFH 7DEOH % (VWLPDWHG &RUUHODWLRQV $PRQJ 'HPRJUDSKLF 9DULDEOHV (QWHUHG LQ /RJLVWLF 5HJUHVVLRQ (TXDWLRQ 3UHGLFWLQJ *XDUGLDQ 5HFRPPHQGDWLRQV &KLOG $JH )DPLO\ 5DFH &KLOG $JH )DPLO\ 5DFH

PAGE 73

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

PAGE 74

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

PAGE 75

7DEOH % (VWLPDWHG &RUUHODWLRQV $PRQJ 6LJQLILFDQW 3UHGLFWRU 9DULDEOHV (QWHUHG LQ /RJLVWLF 5HJUHVVLRQ (TXDWLRQ 3UHGLFWLQJ *XDUGLDQ 5HFRPPHQGDWLRQV 3ULRU 5HSRUWV 3DUHQWDO :DUPWK RI $EXVH DQG 6XSSRUW 3ULRU 5HSRUWV RI $EXVH 3DUHQWDO :DUPWK DQG 6XSSRUW

PAGE 76

$33(1',; & 5$7( 2) 5(029$/ %< 7<3( 2) $%86( 7DEOH &O &URVV7DEXODU $QDO\VLV RI 5HPRYDO 5DWH E\ 3UHVHQFH RI 1HJOHFW $OOHJDWLRQ -XGLFLDO 'HFLVLRQ )UHTXHQF\ 7RWDO 3HUFHQW 1RW 5HPRYHG 5HPRYHG 7RWDO 5RZ 3HUFHQW &ROXPQ 3HUFHQW 1RW 1HJOHFWHG 1HJOHFWHG 7RWDO

PAGE 77

7DEOH & &URVV7DEXODU $QDO\VLV RI 5HPRYDO 5DWH E\ 3UHVHQFH RI 3K\VLFDO $EXVH $OOHJDWLRQ )UHTXHQF\ 7RWDO 3HUFHQW 5RZ 3HUFHQW &ROXPQ 3HUFHQW 1RW 5HPRYHG -XGLFLDO 'HFLVLRQ 5HPRYHG 7RWDO 1RW 3K\VLFDOO\ r $EXVHG 3K\VLFDOO\ $EXVHG 7RWDO 7DEOH & &URVV7DEXODU $QDO\VLV RI 5HPRYDO 5DWH E\ 3UHVHQFH RI 6H[XDO $EXVH $OOHJDWLRQ -XGLFLDO 'HFLVLRQ )UHTXHQF\ 7RWDO 3HUFHQW 1RW 5HPRYHG 5HPRYHG 7RWDO 5RZ 3HUFHQW &ROXPQ 3HUFHQW 1RW 6H[XDOO\ $EXVHG 6H[XDOO\ $EXVHG 7RWDO

PAGE 78

$33(1',; '(6&5,37,9( 67$7,67,&6 )25 9$5,$%/(6 ,1 &$6(6 2) 3+<6,&$/ $%86( $/21( $1' 1(*/(&7 $/21( 7DEOH 'O 'LVWULEXWLRQ RI ,QWHUYDO 9DULDEOHV LQ &DVHV RI 3K\VLFDO $EXVH 2QO\ DQG 1HJOHFW 2QO\ 3K\VLFDO $EXVH 2QO\ 1HJOHFW 2QO\ 77HVW 9DULDEOH 1 0HDQ 6' 1 0HDQ 6' 9DOXH &KLOG $JH )DPLO\ ,QFRPH 6HYHULW\ RI $EXVH r 3DUHQWDO 3V\FKRORJLFDO )XQFWLRQLQJ 3DUHQWDO :DUPWK DQG 6XSSRUW IRU &KLOG )DPLO\ &RQIOLFW r 3DUHQWDO 1RQFRPSOLDQFH r S

PAGE 79

7DEOH )UHTXHQFLHV RI 'LFKRWRPRXV 9DULDEOHV LQ &DVHV RI 3K\VLFDO $EXVH 2QO\ DQG 1HJOHFW 2QO\ 3K\VLFDO $EXVH 2QO\ 1HJOHFW 2QO\ 9DULDEOH )UHTXHQF\ 3HUFHQW )UHTXHQF\ 3HUFHQW $GMXGLFDWLRQ 5HPRYHG 1RW 5HPRYHG *XDUGLDQ 5HPRYH 5HFRPPHQGDWLRQ 'R 1RW 5HPRYH )DPLO\ 5DFH :KLWH 1RQ:KLWH 3DUHQW 3HUSHWUDWRU $OOHJHG 1RW $OOHJHG 3UHYLRXV 5HSRUWV RI
PAGE 80

$33(1',; ( 3($5621 7+5((:((. 7(675(7(67 &255(/$7,216 7DEOH (O 3HDUVRQ 7KUHH:HHN 7HVW5HWHVW &RUUHODWLRQV RI 3V\FKRORJLFDO 9DULDEOHV 9DULDEOH U 3DUHQWDO &RQIOLFW r 0DWHUQDO :DUPWK r 22 3DWHUQDO :DUPWK r 0DWHUQDO 'HSUHVVLRQ 3DWHUQDO 'HSUHVVLRQ 0DWHUQDO $Q[LHW\ rrr 3DWHUQDO $Q[LHW\ 0DWHUQDO ,UULWDELOLW\ rr 3DWHUQDO ,UULWDELOLW\ r S rr S rrr S

PAGE 81

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

PAGE 82

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n9\"n HSKHQ %RHHV 6WHSKHQ %RJJV $VVRFLDWH 3URIHVVRU RI &OLQLFDO DQG +HDOWK 3V\FKRORJ\ FHUWLI\ WKDW KDYH UHDG WKLV VWXG\ DQG WKDW LQ P\ RSLQLRQ LW FRQIRUPV WR DFFHSWDEOH VWDQGDUGV RI VFKRODUO\ SUHVHQWDWLRQ DQG LV IXOO\ DGHTXDWH LQ VFRSH DQG TXDOLW\ DV D GLVVHUWDWLRQ IRU WKHr GHJUHH RI 'RFWRU RI 3KLORVRSK\ 6WHYHQ-:LWLLV 3URIHVVRU RI /DZ

PAGE 83

7KLV GLVVHUWDWLRQ ZDV VXEPLWWHG WR WKH *UDGXDWH )DFXOW\ RI WKH &ROOHJH RI +HDOWK 5HODWHG 3URIHVVLRQV DQG WR WKH *UDGXDWH 6FKRRO ZDV DFFHSWHG DV SDUWLDO IXOILOOPHQW RI WKH UHTXLUHPHQWV IRU WKH GHJUHH RI 'RFWRU RI 3KLORVRSK\ $SULO $V/3 'HDQ &ROOHJH RI +HDOWK 5HODWHG 3URIHVVLRQV 'HDQ *UDGXDWH 6FKRRO

PAGE 84

81,9(56,7< 2) )/25,'$


UNIVERSITY OF FLORIDA
3 1262 08554 5324


GUARDIAN AD LITEM AND JUDICIAL DECISION-MAKING
IN CASES OF CHILD ABUSE AND NEGLECT
By
CYNTHIA SUTTON MCDANAL
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1994

ACKNOWLEDGEMENTS
The author extends appreciation to her parents, Mr. and Mrs. M. R. Sutton, and
her grandparents, Mr. and Mrs. D. A. Ball, for their continued patience and
encouragement. Particular gratitude is extended to the author’s husband, Mr. Charles
McDanal, Jr., for his unconditional guidance and support.
ii

TABLE OF CONTENTS
ACKNOWLEDGEMENTS ii
LIST OF TABLES v
ABSTRACT vi
INTRODUCTION 1
Overview 1
Definition of Abuse 3
Incidence and Prevalence 4
Role of Child Protection Agencies 6
The Guardian ad Litem Program 8
REVIEW OF LITERATURE 11
Theoretical Models of Decision-Making 11
Empirical Studies of Decision-Making 15
RESEARCH OBJECTIVES AND HYPOTHESES 20
METHOD 22
Subjects 22
Measure 22
Procedures 25
RESULTS 26
Comparison of Judges’ and Guardians’ Decisions 26
Rate of Removal by Abuse Type 28
Logistic Regression Equations 30
Comparison of Cases of Physical Abuse Only and Neglect Only 37
DISCUSSION 41
REFERENCES 51
iii

APPENDIX A GAL QUESTIONNAIRE FOR DEPENDENCY CASES . . 54
APPENDIX B CORRELATIONS AMONG INDEPENDENT
VARIABLES 65
APPENDIX C RATE OF REMOVAL BY TYPE OF ABUSE 69
APPENDIX D DESCRIPTIVE STATISTICS FOR VARIABLES IN
CASES OF PHYSICAL ABUSE ALONE AND NEGLECT
ALONE 71
APPENDIX E PEARSON THREE-WEEK TEST-RETEST
CORRELATIONS 73
BIOGRAPHICAL SKETCH 74
iv

1
2
3
4
5
6
7
8
9
10
11
12
13
LIST OF TABLES
page
Comparison of Guardian Recommendations and Judicial
Decisions 27
Results of Logistic Regression Analysis Using Guardian
Recommendations to Predict Judicial Decisions 27
Variable Values In Cases Where Judges and Guardians Disagreed
Regarding Case Outcome 29
Results of Logistic Regression Analysis Using Demographics to
Predict Judicial Decisions 32
Results of Logistic Regression Analysis Using Demographic
Variables to Predict Guardian Recommendations 32
Results of Logistic Regression Analysis Using Legal and Case
Variables to Predict Judicial Decisions 34
Results of Logistic Regression Analysis Using Legal and Case
Variables to Predict Guardian Recommendations 34
Results of Logistic Regression Analysis Using Family
Psychological Variables to Predict Judicial Decisions 36
Results of Logistic Regression Analysis Using Family
Psychological Variables to Predict Guardian Recommendations .... 36
Results of Logistic Regression Analysis Using All Significant
Predictor Variables to Predict Judicial Decisions 38
Results of Logistic Regression Analysis Using All Significant
Predictor Variables to Predict Guardian Recommendations 38
Distribution of Interval Variables 42
Frequencies of Dichotomous Variables 42
v

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
GUARDIAN AD LITEM AND JUDICIAL DECISION-MAKING
IN CASES OF CHILD ABUSE AND NEGLECT
By
Cynthia Sutton McDanal
April, 1994
Chairman: Jacquelin Goldman
Major Department: Clinical and Health Psychology
It has long been recognized that cases involving child abuse and neglect confront
the legal system with the difficult moral and practical dilemma of whether to remove
children from their parents’ care. While broad guidelines are provided under federal and
state laws, there remains a great deal of ambiguity regarding appropriate factors to
consider when making such decisions. Several researchers have put forth rational models
to guide decision-making and a few have examined samples in attempts to determine
which factors are most relevant to court personnel. However, these studies generally
have not sought to generate predictive models for such decision-making, and they tend
to rely on examination of case records rather than direct report from those involved in
the decision-making process. The current study sought to extend our current knowledge
of the determinants of case disposition in cases of child abuse and neglect by considering
the influence of potential factors as reported by the child’s court-appointed representative,
vi

or Guardian ad Litem. Guardian recommendations for placement were considered as
well as case adjudication to determine whether significant differences exist in the
decision-making processes of the two groups. Judicial decisions were also compared to
Guardian recommendations to determine the extent to which judges and Guardians
concur.
Results indicated that while judges are highly likely to concur with
recommendations of the Guardians, they tend to place a somewhat different emphasis on
the variables they use to determine whether to remove a child from his home. Judges
were found to rely most heavily on the presence of allegations of sexual abuse and on
the degree of parental warmth and support for the child. Guardians, however, were
found to rely most heavily on the presence of previous reports of abuse against the child.
Neither judges nor Guardians appeared to place heavy weight on much of the information
regarding family psychological functioning. Implications for this finding are discussed.
vii

INTRODUCTION
Overview
Recent estimates of the prevalence of child sexual abuse, physical abuse, and
neglect indicate that almost two million children per year become victims of some form
of maltreatment (National Center on Child Abuse and Neglect, 1988). Numerous studies
have documented the host of psychological and behavioral problems that often result from
such abuse (see Lamphear, 1985; Brown & Finkelhor, 1986; Houck & King, 1988, for
reviews). Such research clearly indicates that child abuse is a serious problem of wide
scope in our society.
The considerable task of assessing and intervening in cases of child abuse and
neglect has fallen to child protection agencies and the judicial system. To facilitate this
process, specific guidelines have been set, and Guardian ad Litem programs have been
established. Working within the guidelines, the Guardian ad Litem serves as a court-
appointed legal representative of the child. The Guardian gathers information concerning
the family and makes a recommendation to the court regarding appropriate placement of
the child and treatment for the family. It is then the responsibility of the judge to make
a final decision in the best interests of the child, the family, and society.
Such a process requires a great deal of subjective judgment, especially on the part
of the Guardian. Several researchers have put forth theoretical models to guide decision¬
making in such situations (e.g., Wasserman & Rosenfeld, 1988; Faller, 1988), and other
1

2
researchers have sought to describe the decision-making process by considering social,
family, and child characteristics which are associated with case disposition (e.g., Katz,
Hampton, Newberger, Bowles, & Snyder, 1986; Finkelhor, 1983).
Although the approaches applied in these latter studies are more direct in that they
analyze actual placement decisions, the studies themselves are largely descriptive and
have limitations in methodology or scope. The present study seeks to extend our current
knowledge of the decision-making process involved in cases of child abuse and neglect
by identifying factors which influence the Guardian’s recommendation and the final
adjudication of the case. A large number of potential determinants will be considered.
Guardians and judges will be evaluated separately in order to determine whether
differences exist in the decision-making processes of the two groups. The results of such
analyses can then be compared to theoretical models of decision-making in order to
evaluate the appropriateness and effectiveness of the process.
Historical Background
The problem of child abuse and neglect is not a new phenomenon. On the
contrary, children have been mistreated by adults since the beginning of civilization
(Radbill, 1968). Practices such as harsh discipline, child labor, incest, ritual sacrifice,
and infanticide attest to the extent to which child abuse is deeply rooted in cultural and
religious history (Eskin, 1980). The historical view of children as "property" of their
parents and the generally accepted concept that family matters were private contributed
to such a situation.

3
It was not until the late nineteenth century that society began to shift its view of
children as property to include the concept of children’s rights. The landmark case of
"Mary Ellen" in 1870 marked the first time that child abuse was recognized by a U.S.
court. The court intervened by removing Mary Ellen from her home and by prosecuting
her custodian. This case demonstrated the need for a child protection agency, and
shortly thereafter, the New York Society for the Prevention of Cruelty to Children was
chartered by the New York legislature (Landau, 1982). This agency served as the
prototype for many that followed.
Definition of Abuse
While the above case brought about a recognition that parental freedom is not
limitless and that children have a legal right to be protected from cruelty and neglect,
there has been no clear consensus of exactly what constitutes child abuse. Wasserman
and Rosenfeld (1986) note that early definitions of child abuse generally included only
the most severe cases, while recent definitions have been expanded to include any form
of corporal punishment. Emotional abuse has recently been recognized in the literature,
but it has also proved difficult to define precisely. State statutes and mandatory reporting
laws defining child abuse typically refer to harm or threatened harm to a child’s physical
or mental well-being, but they do not provide specific guidelines.
Sloan (1983) has put forth a definition of abuse which is operationally useful. He
divides child abuse and neglect into four categories: physical abuse, neglect, sexual
abuse, and emotional maltreatment. Physical abuse of children refers to any non¬
accidental injury caused by the child’s caretaker. Neglect involves inattention to the

4
basic needs of the child, including food, shelter, clothing, medical care, and supervision.
Sexual abuse includes any interaction between a child and an adult in which the child is
being used for the sexual stimulation of the adult. Finally, emotional maltreatment
includes such behaviors as blaming, belittling, or rejecting a child; unequal treatment of
siblings; or lack of concern for the child’s welfare. Such definitions provide more
specific criteria for determining the occurrence of abuse or neglect, and are frequently
used by child protection workers.
Incidence and Prevalence
The exact scope of child maltreatment in this country is subject to wide debate.
One problem in determining the incidence of child abuse is the lack of a standard
definition regarding exactly what constitutes abuse and neglect. A second problem is that
different research designs yield different estimates of child abuse. Daro (1988) notes that
interviews with a random sample of households or individuals consistently project higher
incidence and prevalence rates than indicated by formal reporting data.
The National Center on Child Abuse and Neglect (1988) has responded to such
difficulties in estimation of the incidence of child abuse by recognizing five "levels" of
abused and neglected children. Level 1 includes all abused and neglected children known
to state and local Child Protective Services (CPS) agencies. At Level 2 are those
children who are not known to CPS, but who are known to other investigative agencies,
such as police, courts, or health departments. Level 3 includes children who are not
known to any Level 1 or 2 agency, but who are known to professionals in other
community institutions, such as hospitals, schools, daycare centers, and mental health

5
clinics. The abused and neglected children at Level 4 are known only to individuals
outside of Levels 1, 2, and 3, such as neighbors or other family members. Finally,
Level 5 includes those children who have not been recognized by anyone as abused or
neglected. Although this description of all five levels is valuable in that it formally
recognizes that unreported cases of abuse do exist, the Center’s incidence study includes
cases only at Levels 1 through 3.
The results of this study indicated that 10.7 children per 1,000 were physically
abused in the United States during 1988, or a total of 675,000 children. Rates of neglect
were even higher — 15.9 children per 1,000, or a total of over one million children
nationwide. The rate of all forms of child sexual abuse was estimated to be 2.5 per
1,000, or a total of 159,000 children.
As mentioned above, random sample interview studies generally yield higher
estimates of the prevalence of child abuse, probably because they tap unreported abuse
as well as abuse known to public agencies and community professionals. For example,
Strauss and Gelles (1986) conducted a national study on the incidence and prevalence of
family violence by telephone interviewing a probability sample of 6,002 households.
Child abuse data were based on the 1,428 households in which there was a child between
the ages of 3 and 17 years, and in which two caretakers were present. The results of this
study indicated that 19 children per 1,000 were subject to at least one incident of "very
severe violence" during 1985.
Random samples of the prevalence of sexual abuse also yield significantly higher
estimates than the Center’s study. Russell (1983) surveyed a probability sample of 930

6
women residents of San Francisco about any experience of sexual abuse they might have
had. Her findings indicated that 38% of the women had experienced at least one incident
of sexual abuse before the age of 18 years. Finkelhor (1984) reported that 6% of all
males and 14% of all females in his random sample had experienced sexual abuse before
the age of 16 by a person at least five years older.
Regardless of the specific definition or research methodology utilized, virtually
all estimates of the incidence and prevalence of child abuse and neglect indicate that it
is an increasingly prevalent problem in our society.
Role of Child Protection Agencies
Since the case of Mary Ellen, it has been the responsibility of the court and child
protection agencies to assess and intervene in cases of child abuse and neglect.
Considering the amount of abuse and neglect that has been documented, the lack of a
standard definition of abuse, and the precarious balance between individual rights
protected under the fourteenth amendment and the need to protect children, this is an
imposing responsibility.
In order to carry out the tasks of assessment and intervention in a fair and
effective manner, specific guidelines for child protective services have been put forth.
Although the specific details of such guidelines differ by state, the general principles of
the process remain the same. Landau (1982) describes six major aspects of the child
protective service process:
The referral process. All 50 states now have mandatory reporting laws designed
to encourage the reporting of suspected abuse and neglect. Most reporting systems also

7
include provisions which encourage investigation, intervention, and treatment services.
Many states have also set up central abuse registries.
Definitions. As mentioned above, each state has a slightly different definition of
what constitutes child abuse and neglect, and all state definitions are rather broad.
However, differences are minimal, and the use of a broad definition provides the
practitioner with leeway in decision-making. Most states distinguish between abuse and
neglect, and therefore define them separately.
Critical elements of a child protection system. Landau notes that there are a
number of criteria for a child protective agency to be an effective system. Practitioners
must have accurate knowledge of the incidence of child abuse and neglect. The system
must have a strong and well-publicized mandatory reporting law and an effective central
abuse registry. A specially trained Child Protection Unit must be available at all times.
The court system must be capable of dealing fairly and effectively with families, and
treatment and rehabilitative programs should be available for those families. Finally,
interdisciplinary communication and a system of coordination should be used to
implement the network of services for families.
Assessment of cases. A specially trained child protection worker must
systematically gather information regarding the child, the family, and the circumstances
of their lives. The worker must then use this information to determine whether abuse
has occurred; if so, to determine why it occurred; and to identify appropriate
interventions.

8
Initiation of the legal process. It is the responsibility of the child protection
worker to decide whether to initiate court proceedings. Guidelines for the consideration
of court action include situations in which (a) the child is determined to be in imminent
danger, (b) attempts at treatment have failed, and/or (c) the family refuses to cooperate
with the investigation. Caseworkers must also consider issues regarding the nature of
evidence and the possible negative consequences of bringing a case into court.
Emergency removal. All states have statutes which allow for the emergency
removal of abused or neglected children from the home, even prior to the filing of a
petition. A court order may or may not be necessary. All states have specific criteria
regarding the conditions necessary to justify emergency removal, and most have specific
requirements describing what must be done following the removal.
Caseworker-attorney interchange. Landau (1982) also stresses the importance of
the establishment of rapport between the child protection worker and the agency attorney
as they work to prepare cases. Clear guidelines regarding decision-making are
beneficial.
The Guardian ad Litem Program
Almost all states have implemented Guardian ad Litem or similar programs to
carry out the above functions, and those that have not require that legal counsel be
appointed to represent the child in abuse and neglect hearings (Sloan, 1983). A Guardian
ad Litem is a court-appointed legal representative who participates on behalf of children,
who are legally unable to initiate or function in litigation without adult assistance. The
function of Guardians is to identify the child’s best interests in the litigation, and they

9
are legally obligated to do everything within their power to represent those interests
(Whitcomb, 1988).
Wiig (1982) notes that although appointment of a Guardian ad Litem is a
condition of the federal Child Abuse Prevention and Treatment Act (PL93-247), the Act
does not define the functions of a Guardian, distinguish the Guardian’s role from that of
the child’s social worker and the child’s attorney, or clearly define the relationship of the
Guardian to the court. She conducted a limited study of six Guardians in an attempt to
clarify these roles.
Wiig’s findings led her to conclude that the following ten activities might be
defined as the functions of the Guardian because they were distinguished from the actions
of the child’s lawyer and social worker in the cases she examined: (1) to function as a
mediator in an attempt to achieve consensus among parties prior to court proceedings;
(2) to coordinate between juvenile court and other court departments; (3) to provide
continuity across the entire legal process; (4) to persuade the social worker to consider
alternative placements, (5) to minimize trauma to the child and to provide continuity of
relationship with the child: (6) to insure court attention to cases by placing them on the
calendar; (7) to make recommendations in the best interests of the child; (8) to identify
special needs of the child and to insure that the child receives attention to those needs;
(9) to prevent continuances not in the best interests of the child or to request an attorney
to do so; and (10) to request independent counsel for the child (pp. 93-94). Wiig further
notes that while the attorney or social worker might in some cases fulfill some of these

10
roles, the functions numbered 3, 5, and 8 are distinct and may be considered strictly
functions of the Guardian ad Litem.
While it is generally accepted that the Guardian fulfills these functions and
ultimately makes a recommendation to the court regarding disposition of the case, the
process by which the Guardian arrives at the recommendation is less clear. Several
researchers have noted the problems inherent in such a process, and have put forth
theoretical models of decision-making to guide practitioners. These models are discussed
in detail below.

REVIEW OF LITERATURE
Theoretical Models of Decision-Making
Wasserman and Rosenfeld (1986) have noted several difficulties facing judges who
must decide when to intervene in cases of possible abuse or neglect, and these difficulties
also apply to Guardians as they struggle to make the most appropriate recommendation
based on the information they have. Such problems include the aforementioned lack of
clarity in the definition of abuse, difficulties in proving abuse, limited alternatives
available other than foster care, problems in the foster care system, and the large number
of children in abusive and neglectful situations.
Wasserman and Rosenfeld (1986) also note the positive and negative aspects of
separation that must be considered when deciding whether to remove a child from his
home. Separation provides for the safety of the child, sets limits on parental behavior,
and lets the child know that society will intervene to protect him if his parents do not.
It may also provide an opportunity for the child to live in a safer, warmer environment.
Separation also frees the parents from the burdens of childrearing, thereby giving them
the opportunity to make constructive changes in their situation.
Potential negative consequences of separation must also be carefully considered.
Even though they may be abused, most children have an affectional bond with their
parents and see that relationship as a source of identity and well-being. Thus, even when
faced with abuse, many children express a desire to remain with their families. Also,
11

12
feelings of inadequacy and failure in both parents and children may follow separation.
The authors state that children may even perceive themselves as responsible for the
separation.
In order to make appropriate decisions given such a difficult situation, Wasserman
and Rosenfeld (1986) recommend that practitioners consider the nature and quality of the
parent-child bond. They state, "A judge who is provided with careful evaluation of the
parent-child relationship is in a better position to decide wisely" (p. 524). In addition
to interviewing all individuals involved in the case, the authors recommend that the
caseworker visit the family’s home. The caseworker should note issues such as how well
the parent knows the child, how able and willing the parent is to care for the child, how
reasonable the parent’s beliefs about discipline are, and how much nurturance the parent
provides the child. Once the caseworker has obtained this information, he or she can
make a reasonable evaluation of the relationship between parent and child. According
to Wasserman and Rosenfeld, this evaluation can then be used to make the most
appropriate decision regarding disposition of the case.
While the characteristics of the family relationships likely have legal and clinical
utility for decision-making, an approach considering only this factor seems overly
simplistic. Based on clinical experience with over 300 cases of child sexual abuse, Faller
(1988) has proposed a model of decision-making in cases of intrafamilial child sexual
abuse which is more comprehensive than the approach described above. Like
Wasserman and Rosenfeld (1986), she believes that the parent-child relationship is an
important consideration, especially the relationship between mother and child. However,

13
Faller also believes that other parental characteristics are crucial to decisions about
intervention.
These characteristics differ between mothers and fathers. According to Faller’s
theory, important maternal factors are (1) the level of the mother’s dependency,
especially on the father-perpetrator; (2) the degree of love and nurturance she conveys
to the victim; and (3) the degree to which she responded protectively when she
discovered the sexual abuse. Important paternal characteristics are (1) his overall level
of functioning; (2) his superego functioning; and (3) the severity of the sexual abuse.
In order to use such information to choose intervention strategies, Faller has
identified four types of cases based on different combinations of the above characteristics.
Type I cases are ones in which both parents have many strengths. In such cases Faller
recommends family therapy, without court involvement. In Type II cases, the mother
has many positive characteristics, but the father exhibits many areas of problematic
behavior. For these cases, Faller suggests maintaining the mother and child intact and
referring them for treatment but advises that the father-perpetrator be excluded from the
family. Legal intervention may be necessary to remove the father from the home.
In Type III cases, mothers are dependent, hostile toward their children, and
unprotective. Although the father has sexually abused the child, he functions much better
than the mother and he often has a closer relationship with the child than does the
mother. According to Faller, this is the most difficult type of case for which to
determine appropriate intervention. She recommends therapy for all family members for
three to six months, at which time careful re-evaluation should be conducted. During

14
this time period, it may be optimal to place the child with a relative, which allows for
less disruption in the child’s life while still protecting him or her from further sexual
abuse.
In Type IV cases, the mother is similar to the Type III mother, and the father
exhibits poor overall functioning and shows no guilt over the (usually) extensive abuse.
Faller recommends that parental rights be terminated in such cases, and that the children
receive treatment for at least one year. Criminal prosecution may also be warranted for
the perpetrator.
Although Faller’s approach to decision-making is comprehensive and specific, it
does have several limitations. First, her approach assumes the availability of alternative
placements for the child (i.e., placement with a relative or for adoption), which may not
always be readily available. Second, some possible combinations of parental
characteristics do not match any of the four case types described. Decision-making for
these families is therefore more ambiguous. Third, this theoretical model has been
proposed for cases of father-child sexual abuse only. It is not clear whether such a
model might be adapted to cases of physical abuse or neglect, or to cases of sexual abuse
by a non-father perpetrator. Finally, though her model is based on clinical experience,
it has yet to be empirically tested.
Several researchers have conducted analogue studies with child protection workers
or have analyzed actual case disposition data in an attempt to empirically describe aspects
of the decision making-process in cases of child physical abuse or neglect. Such cases
are reviewed below.

15
Empirical Studies of Decision-Making
Meddin (1984) conducted an analogue study of criteria used by child protection
workers to determine placement decisions in cases of abuse and/or neglect. She surveyed
81 child welfare workers using a questionnaire which described six situations containing
two instances each of physical abuse, sexual abuse, and neglect. Respondents were
provided with information on eight variables: age, gender, and race of the child; the
child’s experience of previous placements; the severity of the current incident; risk to the
child of further abuse and/or neglect; and the cooperation and functioning of the primary
caretaker. For each situation, respondents indicated whether they would place the child
in foster care; they then described what variables influenced their decisions.
Results of Meddin’s study indicated that the primary variable used to determine
placement of the child was risk of further abuse and/or neglect. Such a finding seems
of dubious value, given that caseworkers are rarely, if ever, provided with that
information directly. More often, assessment of risk to the child is a part of the
caseworker’s decision-making process.
Meddin did, however, note that four other variables clustered around risk to the
child, and she suggested that these variables might serve as operational definitions or
indicators of risk. The four variables were severity of the current incident, functioning
and cooperation of the primary caretaker, and age of the child.
While Meddin’s study is useful in that it indicates several variables which seem
to be important in caseworkers’ placement decisions, it does have several limitations.
In actual case situations, child protection workers typically have a great deal more

16
information than the eight variables Meddin provided. It is impossible to evaluate the
importance of this additional information using the analogue research design employed
here. Additionally, there is the possibility that the caseworkers’ actual decisions differ
from what they report in a hypothetical situation.
Several empirical studies of actual case dispositions have been conducted in an
attempt to overcome such limitations. Runyan, Gould, Trost, and Loda (1981) examined
the records of the North Carolina Registry of Child Abuse and Neglect to determine
which social, family, and child characteristics were most influential in the decision to
place a child in foster care. According to Runyan et al., these records contained all
theoretically relevant factors as well as demographic information.
Results of the study indicated that several child and family characteristics put
children at increased risk for placement. These variables included parents who perceive
severe punishment as acceptable, parents with substance abuse problems, severe child
injuries, child abandonment, and several other specific types of abuse. Age of the child
or parent, the presence of several maltreated children in one family, and several other
specific types of abuse were found to be not significant. Race, income, and education
were also nonsignificant, suggesting that there is no systematic bias in placement
decisions. Two non-family variables were also found to be associated with a higher
chance of placement—cases referred by courts or police and geographic area.
Runyan et al. also employed a logistic regression model to predict placement in
foster care given all the above variables. They then compared this to actual placement.
Results of this procedure did not strongly associate the above characteristics with removal

17
from the home. In fact, the variance explained by the model was only 0.168. Runyan
et al. interpreted this result to mean that assignment to foster care approximates a random
process across a large population.
An alternative explanation for such a finding may be that child protection workers
use different criteria for making placement decisions for different types of abuse cases
(i.e., sexual abuse, physical abuse, or neglect). Since these data were collapsed across
type of abuse, such differences may have canceled each other out.
Katz, Hampton, Newberger, Bowles, and Snyder (1986) examined factors that
influence the decision to remove children from their parents’ care in cases of abuse
(physical or sexual) versus neglect (nonphysical injury). These researchers examined the
records of 185 suspected abused or neglected children at Children’s Hospital, Boston, in
an attempt to reveal how demographic characteristics, family history, family stress, the
nature of the injury, and aspects of the medical encounter influence the outcome of the
case.
Interestingly, the results of the study indicated that children with nonphysical
injuries were more likely to be removed from their parents’ care than were those with
physical injuries. The authors suggested several possible explanations for this finding,
including the possibility that neglect may be perceived by clinicians as evidence of a
chronic family problem rather than a single mishap.
Katz et al. also conducted Chi-square analyses to test the influence of other
independent variables on placement, although these analyses were collapsed across cases
of abuse and neglect. Their findings indicated that the following factors predicted

18
removal from the home: Medicaid-eligible families, previous reports of suspected
maltreatment, and suspicion of the mother being involved in the maltreatment. Minority
families, families in which the child was under age six, and families that were
experiencing more than one stressor were more likely to have their children returned.
Surprisingly, severity of condition was not significantly associated with outcome.
The results of this study yielded several interesting and even counterintuitive
findings. Again, however, because different categories of abuse were not considered
separately, the possibility of overgeneralization exists.
Finkelhor (1983) analyzed data on 6,096 cases of sexual abuse which were
officially reported in the U.S. in 1978 in an attempt to evaluate how the cases were
handled. His specific aim was to determine whether children were being unnecessarily
removed from their homes. Such an approach seems useful in that his consideration of
sexual abuse cases alone minimizes the risk of overgeneralization or of having significant
effects washed out by differences among groups.
Finkelhor compared the overall placement rate of sexual abuse cases to the
placement rate of physical abuse cases in the National Clearinghouse sample. His
findings indicated that foster care placement occurs significantly more often in cases of
sexual abuse, which led him to hypothesize that sexual abuse results in more outrage than
does physical abuse, regardless of the degree of danger to the child.
Finkelhor’s main finding of the study was that foster care does not seem to be
used indiscriminantly in cases of sexual abuse. For example, the most important
predictor of foster care placement was whether or not the victim initially reported the

19
abuse. A preference to be placed outside the home was more likely to be met than a
preference to stay. Also, foster care is more likely to be used in the case of an older
child than a younger child. This seems reasonable given that older children are typically
better able to adjust to living without their families.
Other factors predisposing to placement in foster care included abuse committed
by parents (especially if both parents were involved), very large families, the presence
of other types of abuse or neglect, and the presence of several other family stressors,
such as spouse abuse, substance abuse, family discord, and mental health problems.
Such findings led Finkelhor to conclude that "the characteristics of the children who are
placed in foster care do suggest some rational relationship between the disposition and
the facts of the case" (p. 199).
Finkelhor’s data also suggest that foster care is not used indiscriminantly in
"vulnerable" families. For example, he found that foster care placement is not more
likely to occur with poor families. Further, black children were not more likely to be
removed from their homes than were white children Thus, according to Finkelhor’s
findings, there does not seem to be any systematic bias in the decision-making process.
The strengths of Finkelhor’s study include its large N, its consideration of a large
number of potentially relevant factors, and its statistical analysis of only one type of
abuse. This methodology does not, however, indicate whether any of these factors
cluster together to form a meaningful aggregate of predictive characteristics. Also,
similar information regarding cases of child physical abuse and of neglect is needed.

RESEARCH OBJECTIVES AND HYPOTHESES
The current study seeks to extend our knowledge of the determinants of case
disposition by considering the influence of potential factors in cases of physical abuse and
in cases of neglect. Guardian ad Litem recommendations for placement will be
considered as well as final judicial decisions, in order to determine whether significant
differences exist in the decision-making processes of the two groups. Judicial decisions
will also be compared to Guardian recommendations to determine the extent to which
judges rely on those recommendations.
Based on review of the available literature, it is hypothesized that several groups
of variables will predict Guardians’ and judges’ decisions to place the child outside the
home in cases of child physical abuse and neglect. Among demographic variables, it
is hypothesized that increased age of the child will predict placement outside of the
home, but that family race and income will not be related to case disposition.
A second category of variables hypothesized to be related to case disposition
includes those which are descriptive of the abuse situation. Specifically, it is proposed
that abuse perpetrated by parents, very severe abuse, the presence of other types of
abuse, and previous reports of abuse will predict placement of the child outside the
home.
A third category of variables includes those that describe the psychological state
of the family. It is specifically hypothesized that poorer parent psychological
20

21
functioning, high degree of parental noncompliance with the investigation, the presence
of spouse abuse and substance abuse, high degree of family discord, and low degree of
parental warmth and support for the child will all predict placement outside the home.

METHOD
Subjects
Subjects for this study included 94 families involved in child physical abuse or
neglect cases in the state of Florida. All families were selected from adjudicated cases
in 12 judicial circuits. Only one abused or neglected child from each family was
considered in this study in order to avoid misleading duplication of family and case
variables. Child age ranged from 1 year to 17 years (mean =4 years, 9 months). Family
income ranged from $0 to $65,000 per year (mean=$8,661). Fifty-nine percent of the
families were white, 34% were black, 4% were Hispanic, and 3% were other races.
Measure
Data for this study were collected using a questionnaire designed by the team of
investigators to sample demographic, legal, and psychological factors relevant to the
family (see Appendix A). Demographic and legal information was of public record and
included the age, sex, and race of each family member; who the alleged perpetrator was;
family income; occupation and education level of each parent; whether or not the abuse
was substantiated; past history of abuse; whether criminal charges were brought against
the perpetrator; whether mental health evaluation was recommended and/or completed
for each family member; which placement options were recommended by the Guardian;
and which placement options were adjudicated. Placement options included the
following: (1) perpetrator removed from the home, (2) parents retain custody with
22

23
supervision, (3) temporary custody awarded to a relative, (4) temporary foster placement,
(5) child placement in a treatment facility, (6) placement for adoption, (7) child placed
in a detention facility, and (8) not dependent and returned home. A category designated
"other" was also provided in case final adjudication included a placement not described
above. For the purposes of this study, these options were collapsed to "returned home"
(options 1, 2, or 8) or "removed from the home" (options 3 through 7).
Psychological information for each case was provided by the Guardian, who made
judgments based on his or her interviews with the family. Judgments were made on
visual analog scales on the following issues: (1) who the non-abusing caretaker supports;
(2) severity of the abuse/neglect; (3) child’s distress over the current family situation; (4)
conflict between the parents; (5) parents’ willingness to report family problems in
general; (6) amount the family favors family activities; (7) amount the family members
clarify and/or speak for other family members; (8) degree to which the physical needs
of the child were met (and who they were met by); and (9) degree to which the emotional
needs of the child were met (and whom they were met by).
Guardians also rated the following issues for each parent: (1) conflict between
parent and child; (2) warm and positive feelings for child; (3) involvement in child’s
daily life; (4) reliance on the child; (5) psychological issues (loneliness, depression,
anxiety, irritability); (6) satisfaction with social support; and (7) response to the Guardian
in general. Guardians also indicated on visual analog scales their confidence in each of
their judgments.

24
Additional information was requested regarding whether the parents were abused
or neglected as children, whether the parents are accused of substance abuse and/or
spouse abuse, whether the parents have lost jobs or moved within the past year, whether
the parent or child suffers a mental or physical handicap, and whether the child exhibits
any behavioral problems.
Reliability. Guardian’s judgments on the psychological issues were compared to
those of licensed clinical psychologists and law students (Goldman et al., in preparation).
A videotape of a Guardian interviewing expert Guardians who were acting as family
members involved in a child abuse scenario was viewed by 22 Guardians, 17 clinical
psychologists, and 15 law students. The script for the mock interview was written by
the investigators in conjunction with two Guardian supervisors so that it would be
representative of an authentic family situation. After viewing the videotape, all subjects
completed the portion of the research questionnaire which involved psychological issues.
No significant differences were found between the Guardians’, psychologists’, and
lawyers’ judgments on any of the psychological issues, indicating that no specific bias
could be attributed to the subjects’ training or their area of expertise.
Test-retest reliability was also calculated for each of the 29 items rated. The 22
Guardians viewed the mock interview twice, at three week intervals. Of the variables
used in this study, Pearson correlations were significant for three variables at the p< .05
level (level of conflict, maternal and paternal warmth for child), for one variable at the
p<.01 level (maternal noncompliance), and for one variable at the p = .001 level
(severity of abuse/neglect). Pearson correlations were found to be non-significant for

25
paternal noncompliance and parental psychological functioning (depression, anxiety,
irritability, and loneliness). It should be noted, however, that the videotape viewed by
the Guardians contained relatively little information regarding the variables found to have
non-significant correlations, and it is therefore hypothesized that the low correlations
were due more to Guardians’ lack of confidence in their ratings than to instability in their
judgments over time.
Procedures
Data for each family were collected by the Guardian ad Litem (GAL) assigned
to that case. Each judicial circuit and each Guardian participated in this study on a
volunteer basis. Guardians are assigned cases on an individual basis by the GAL director
of each circuit. Each Guardian is appointed to a family when the case first becomes
active, and that Guardian represents the child until the case is adjudicated. For the
purpose of this study, Guardians collected data on each family when the case reached the
stage of adjudication. Thus, the cases used in this study represent those that were not
settled out of court and therefore may represent the most severe cases.

RESULTS
Comparison of Judges’ and Guardians’ Decisions
The cross-tabular display provided in Table 1 shows the frequency with which
judges agreed with the recommendation of the Guardians ad Litem. Results indicate that
the child was removed from the home by the judge in 78 cases and returned home in 16
cases. Guardians recommended removal from the home in 69 cases and recommended
returning the child to the home in 25 cases. Table 1 also shows that the judge agreed
with the Guardian’s recommendation (to remove the child from the home or to return the
child home) in 79 of the 94 cases. There were only 12 cases in which the judge—but
not the Guardian—felt that the child should be removed, and only 3 cases in which the
Guardian—but not the judge—felt that the child should be removed. The Kappa statistic
(K) was calculated to determine the degree of concurrence observed between judges and
Guardians, and the obtained value of .53 indicates good overall agreement (beyond
chance) between the two groups. It can therefore be concluded that while Guardians
appear to be slightly more conservative in their recommendations to remove children
from their homes, there was considerable agreement between the two groups regarding
placement of the child.
To further assess the relationship between Guardian recommendations and judicial
decisions, a logistic regression analysis was conducted using Guardian recommendation
as a predictor variable for judicial decision (See Table 2). Results indicated that the
26

27
Table 1. Comparison of Guardian Recommendations and Judicial
Decisions
Judicial Decision
Guardian Recommendation
Removed
Not Removed
Total
Removed
66 cases
12 cases
78 cases
(70%)
(13%)
(83%)
Not Removed
3 cases
13 cases
16 cases
(3%)
(14%)
(17%)
Total
69 cases
25 cases
94 cases
(73%)
(27%)
(100%)
K = .53
Table 2. Results of Logistic Regression Analysis Using Guardian
Recommendations to Predict Judicial Decisions
Predicting Judicial Decisions
Variable
DF
Parameter
Estimate
Standard
Error
Chi-
square
Probability of >
Chi-square
Intercept
1
0.08
0.40
0.04
0.84
Guardian
1
-3.17
0.71
19.77
0.00*
Recommendation
p = .0001
Hosmer-Lemeshow goodness-of-fit statistic = 6.24 with 8 DF, p = .62

28
Guardian’s recommendation for removal is significantly associated (p = .0001) with a
higher probability of judicial decision to remove the child from the home. Predicted
probabilities based on the parameter estimate obtained for the Guardian recommendation
variable indicate that the probability of removal is 96% in cases where the Guardian
recommends it and 48% in cases where the Guardian recommends against it. It should
be noted again, however, that these data are based on only 25 cases in which the
guardian recommended returning the child to the home. The Hosmer and Lemeshow
goodness-of-fit statistic for the model (6.24 with 8 DF, p = .62) indicates that the
observed responses do not significantly differ from responses which would be predicted
based on this model (Hosmer & Lemeshow, 1989).
The cases in which the judge disagreed with the Guardian’s recommendation for
placement of the child were examined further to determine whether any trends among the
predictor variables could be identified. Examination of the values of the predictor
variables in those cases, however, did not reveal any differences which might account
for the difference of opinion between the judges and the Guardians (see Table 3).
Rate of Removal by Abuse Type
It was also of interest to compare the rate of removal depending on the type of
abuse or neglect allegations present. The overall distribution of abuse types in this
sample (whether removed or not removed) indicated that 80% of the cases involved
neglect, 52% involved physical abuse, and 28% involved sexual abuse. Among the cases
in which removal occurred, 83% involved allegations of neglect, 50% involved
allegations of physical abuse, and 24% involved allegations of sexual abuse (see

29
Table 3.
Variable Values In Cases Where Judges and Guardians Disagreed
Regarding Case Outcome
J
G
C
F
R
P
S
M
P
S
P
W
C
N
S
S
u
U
H
A
A
A
E
T
R
E
S
A
0
O
U
P
D
A
1
M
C
R
V
Y
E
X
Y
R
N
N
B
A
G
R
A
I
E
P
E
P
V
A
F
M
F
C
A
B
E
D
G
N
E
R
E
R
B
U
L
O
B
U
I
E
C
R
I
S
E
U
N
M
U
S
A
P
T
P
S
C
P
S
E
N
Y
E
T
E
R
N
.
.
W
Y
79
N
N
N
.
37
5
42
N
N
R
N
2
0
W
Y
93
Y
N
Y
51
27
44
54
Y
N
R
N
.
12
w
Y
84
Y
N
Y
65
85
20
37
Y
N
R
N
3
0
N
Y
97
N
N
N
69
68
2
27
N
N
R
N
3
0
N
Y
91
N
Y
N
96
13
65
96
Y
N
R
N
5
0
N
Y
64
N
Y
N
25
47
16
5
Y
N
R
N
1
8
W
Y
81
Y
N
N
59
14
54
64
N
N
N
R
6
8
w
Y
89
Y
Y
Y
46
50
2
47
N
N
R
N
2
0
N
Y
86
N
N
N
.
68
46
48
Y
N
N
R
1
N
Y
82
N
N
N
.
99
0
0
N
N
R
N
2
W
Y
69
Y
N
N
63
47
45
21
N
Y
R
N
3
N
Y
94
N
N
N
88
21
.
18
Y
N
R
N
17
W
Y
85
Y
N
N
52
57
94
15
N
N
R
N
16
w
N
95
N
N
Y
.
19
90
14
N
N
N
n rr
R
4
20
N
Y
—rrr
60
Ti \7
N
N
íTkTTI
N
54
81
78
33
Y
N
JUDGE=
GUARDIAN=
CH1AGE=
FAMINC =
RACE=
PARPERP=
SEVERITY =
MTYPES =
PREVREP =
SEXABUSE=
PSYFUNCT =
WARM =
CONFL =
NONCOMP =
SUBABUSE=
SPABUSE=
Adjudication (R = Child removed from home; N=Not removed)
Recommendation (R=Recommended removal; N=Did not)
Child age (in years)
Annual family income (in thousands of dollars)
Family race (W=White; N=Non-White)
Was the alleged perpetrator a parent?
Severity of abuse (99 = Highest possible severity level)
Were multiple types of abuse alleged?
Were there previous reports of abuse?
Was sexual abuse alleged?
Parental psychological functioning (99=Poorest possible level)
Parental warmth/support for child (99=Highest possible level)
Family conflict (99=Highest possible level)
Parental noncompliance (99=Highest possible level)
Was parental substance abuse alleged?
Was spouse abuse alleged?

30
Appendix C). Results also indicated that children were removed from their homes in
87% of the cases involving allegations of neglect, in 80% of the cases involving
allegations of physical abuse, and in 73% of the cases involving allegations of sexual
abuse. While these results are complicated by the fact that multiple types of abuse
occurred in 45% of the cases, they nonetheless indicate removal was most likely when
allegations of neglect were present and least likely when allegations of sexual abuse were
present.
Logistic Regression Equations
The primary purpose of this study was to determine which demographic variables,
abusive situation characteristics, and family psychological characteristics influence
Guardians’ and judges’ decisions to remove the child from the home in cases of child
abuse and/or neglect. Because the outcome variables (the judge’s placement of the child
or the Guardian’s recommendation for placement) and several predictor variables are
dichotomous, logistic regression was used. In order to test for the possibility of
misleading multicollinearity among the independent variables, correlation matrices were
obtained for each model. The correlations among the independent variables ranged from
.01 to .50, indicating that the results of the analyses were not compromised by significant
multicollinearity.
Two series (one for judges’ decisions and one for Guardians’ recommendations)
of three logistic regression models were run—one model for each category of variables
(demographic variables, abusive situation characteristics, and family psychological
characteristics). The results of each of these models will be discussed separately.

31
Finally, two logistic regression equations including only the significant predictor
variables from the above models were tested—one using judges’ decisions as a dependent
variable and including all independent variables found to significantly predict those
decisions and a second using Guardians’ decisions as a dependent variable and including
all independent variables found to significantly predict their decisions.
Demographic Variables
Because information regarding family income was provided in only 54 of the 94
cases, it was not possible to test the impact of this variable. Results of the logistic
regression equation including child age and family race as predictors indicated that
neither variable significantly impacted the decision to place the child outside the home
(see Table 4). The analysis using the Guardian’s recommendation as the dependent
variable yielded similar results, as neither child age nor family race emerged as a
significant predictor of a Guardian’s decision to recommend placement outside of the
home (see Table 5).
Abusive Situation Characteristics
Among variables related to aspects of the abusive or neglectful situation, it was
hypothesized that abuse perpetrated by parents, very severe abuse, the presence of
multiple types of abuse, and previous reports of abuse would predict placement outside
the home. It was not possible to test the influence of the "parent perpetrator" variable,
however, as 96% of the cases in this sample involved parent perpetrators. When the
remaining variables were included in the logistic regression equation using the judge’s

32
Table 4. Results of Logistic Regression Analysis Using Demographics to Predict
Judicial Decisions
Predicting Judicial Decisions
Demographic Variables
DF
Parameter
Estimate
Standard
Error
Chi-
Square
Probability of >
Chi-Square
Intercept
1
-1.88
0.55
11.90
0.00
Child Age
1
0.00
0.06
0.00
0.96
Family Race
1
0.38
0.59
0.41
0.52
Table 5. Results of Logistic Regression Analysis Using Demographic Variables to
Predict Guardian Recommendations
Predicting Guardian Recommendations
Demographic Variables
DF
Parameter
Estimate
Standard
Error
Chi-
Square
Probability of >
Chi-Square
Intercept
1
-1.50
0.47
10.36
0.00
Child Age
1
0.03
0.06
0.36
0.55
Family Race
1
0.34
0.51
0.44
0.51

33
decision as a dependent variable, none emerged as significantly related to removal of the
child from the parents’ home (see Table 6). In the analysis using the Guardian’s
recommendation as a dependent variable, only the presence of previous reports of abuse
(p<.05) emerged as significantly related to recommendation for placement outside the
home (see Table 7). In this model, the negative parameter estimate indicates that the
absence of previous reports of abuse or neglect (which was treated first in the model) is
associated with a smaller probability of removing the child from the home. Guardians
were not more likely to recommend removal in cases in which the abuse was very severe
or in which multiple types of abuse occurred.
Family Psychological Characteristics
Several variables describing the psychological state of the family were
hypothesized to affect judges’ decision-making in cases of abuse or neglect. Specifically,
it was proposed that poorer parent psychological functioning, high degree of parental
noncompliance with the investigation, the presence of spouse abuse and substance abuse,
high degree of family discord, and low degree of parental warmth and support for the
child would predict placement of the child outside of the home. However, only the
degree of the parents’ warmth and support for the child (p < .01) emerged as significantly
related to judges’ decision-making when the variables were simultaneously entered into
the prediction equation (see Table 8). The positive parameter estimate indicates that
lower levels of parental warmth and support (which were treated first in the model) were
associated with a larger probability of having the child removed from the home.

34
Table 6. Results of Logistic Regression Analysis Using Legal and Case Variables to
Predict Judicial Decisions
Predicting Judicial Decisions
Legal and Case Variables
DF
Parameter
Estimate
Standard
Error
Chi-
Square
Probability of >
Chi-Square
Intercept
1
-0.47
0.89
0.27
0.60
Severity of Abuse
1
-0.15
0.01
1.46
0.23
Multiple Abuse Types
1
0.19
0.59
0.10
0.75
Previous Reports
1
-0.25
0.57
0.19
0.66
Table 7. Results of Logistic Regression Analysis Using Legal and Case Variables to
Predict Guardian Recommendations
Predicting Guardian Recommendations
Legal and Case Variables
DF
Parameter
Estimate
Standard
Error
Chi-
Square
Probability of >
Chi-Square
Intercept
1
-0.88
0.89
0.97
0.32
Severity of Abuse
1
-0.00
0.01
0.09
0.76
Multiple Abuse Types
1
0.15
0.51
0.09
0.77
Previous Reports
1
-1.08
0.51
4.46
0.03*
* p < .05

35
The results of the analysis using the Guardian’s recommendation as an
independent variable yielded similar results (see Table 9). Like judges, Guardians were
more likely to recommend removal of children from homes in which parents showed
little warmth and support for them (p< .05). Interestingly, however, none of the other
variables regarding family psychological functioning significantly impacted the
Guardians’ decision-making. It therefore appears that neither judges nor Guardians rely
heavily on information regarding the psychological functioning of the family when
making decisions regarding the placement of the abused child.
Models Including All Significant Predictors
In order to determine the best overall predictive model of judicial decision¬
making, a logistic regression equation was run which included only the predictor variable
found to significantly impact judicial decisions from the above models—level of parental
warmth and support. Again, the positive parameter estimate obtained for the predictor
variable in this model indicates that lower degrees of warmth and support were more
associated with removal from the home. Calculation of predicted probabilities based on
the obtained parameter estimates indicated that the probability of removal was 98% in
cases where parental warmth and support was rated 0 (the lowest possible level) and 50%
in cases where parental warmth and support was rated 99 (the highest possible level).
The Hosmer-Lemeshow goodness-of-fit statistic (7.90 with 8 DF, p = .44) indicates that
the observed responses do not differ significantly from those which would occur based
on the predicted probabilities from the model (see Table 10).

36
Table 8. Results of Logistic Regression Analysis Using Family Psychological
Variables to Predict Judicial Decisions
Family Psychological
Variables
Predicting Judicial Decisions
DF
Parameter
Estimate
Standard
Error
Chi-
Square
Probability of >
Chi-Square
Intercept
1
-4.79
1.76
7.41
0.01
Psychological Functioning
1
0.01
0.02
0.11
0.74
Parental Warmth and Support
for Child
1
0.04
0.02
8.14
0.00*
Family Conflict
1
0.00
0.01
0.00
0.96
Parental Noncompliance
1
0.02
0.01
2.32
0.13
Substance Abuse
1
-0.17
0.71
0.06
0.81
Spouse Abuse
1
-0.78
0.87
0.81
0.37
* p < .01
Table 9. Results of Logistic Regression Analysis Using Family Psychological
Variables to Predict Guardian Recommendations
Family Psychological
Variables
Predicting Guardian Recommendations
DF
Parameter
Estimate
Standard
Error
Chi-
Square
Probability of >
Chi-Square
Intercept
1
-3.18
1.39
5.21
0.02
Psychological Functioning
1
0.02
0.02
1.18
0.28
Parental Warmth and Support
for Child
1
0.03
0.01
4.93
0.03*
Family Conflict
1
-0.00
0.01
0.06
0.80
Parental Noncompliance
1
0.01
0.01
0.99
0.32
Substance Abuse
1
-0.42
0.60
0.49
0.49
Spouse Abuse
1
-0.80
0.72
1.23
0.27
p < .05

37
A similar analysis was conducted using the independent variables found to
significantly impact the decisions of the Guardians. When the variables of previous
reports of abuse and parental warmth and support for the child were entered into the
equation, only the presence of previous reports of abuse (p = .05) emerged as strongly
related to Guardian’s decisions. Since inclusion of the variable quantifying parental
warmth and support for the child did not significantly add to the predictive power of the
equation, the analysis was conducted again using only the presence of previous reports
as a predictor variable. The negative parameter estimate indicates that the presence of
previous reports of abuse was associated with a higher probability of making a
recommendation for removal. The Hosmer-Lemeshow goodness-of-fit statistic (9.00 with
8 DF, p = .34) indicates that the observed responses do not significantly differ from those
which would occur given the predicted probabilities based on this model (see Table 11).
Comparison of Cases of Physical Abuse Only and Neglect Only
An original aim of this study was to determine whether judicial and Guardian
decision-making processes differed between cases of physical abuse alone and neglect
alone. Because the sample included only 11 cases solely involving physical abuse and
36 cases solely involving neglect, it was not possible to analyze the data using logistic
regression equations—which require a larger sample size to be interpretively useful.
Analysis using the Shapiro-Wilk statistic (W) indicated that all interval independent
variables were normally distributed in the samples. The means of these variables were
compared using t-tests in an attempt to determine whether potentially important
differences exist among demographic, legal, or family psychological characteristics

38
Table 10. Results of Logistic Regression Analysis Using All Significant Predictor
Variables to Predict Judicial Decisions
Predicting Judicial Decisions
All Significant Predictors
DF
Parameter
Estimate
Standard
Error
Chi-
Square
Probability of >
Chi-Square
Intercept
1
-3.73
0.86
18.87
0.00
Parental Warmth and
Support for Child
1
0.04
0.01
9.33
0.00*
*p < .01
Hosmer-Lemeshow goodness-of-fit statistic=7.9 with 8 DF, p = .44
Table 11. Results of Logistic Regression Analysis Using All Significant Predictor
Variables to Predict Guardian Recommendations
Predicting Guardian Recommendations
All Significant Predictors
DF
Parameter
Estimate
Standard
Error
Chi-
Square
Probability of >
Chi-Square
Intercept
1
-0.57
0.30
3.50
0.06
Previous Reports of Abuse
1
-1.02
0.49
4.25
0.04*
* p < .05
Hosmer-Lemeshow goodness-of-fit statistic=9.00 with 8 DF, p = .34

39
between the two groups (see Appendix D). In addition, the frequencies of the
dichotomous independent and dependent variables in each group were reviewed, although
the number of cases was insufficient to support Chi-square analysis.
Results of comparisons between cases of physical abuse alone and neglect alone
using t-tests indicated that the level of severity of neglect tended to be more extreme than
the level of severity of abuse (t=-2.55, p = .02) and that there tended to be less conflict
between parents of neglected children than between parents of abused children (t=2.09,
p = .04). No significant differences between the two groups were found in child age,
family income, level of parental psychological functioning, degree of parental warmth
and support for the child, or parental noncompliance with the investigation.
Visual review of the frequencies of dichotomous variables in the two groups
revealed the following: (1) Judges were somewhat more likely to remove neglected
children from their homes (89% of neglected children as opposed to 73% of abused
children), (2) Guardians were less likely to recommend removal of neglected children
than abused children (75% of neglected children as opposed to 82% of abused children),
(3) Among neglect cases, there tended to be fewer cases involving previous reports of
abuse or neglect against the child (47% as opposed to 57% in cases of physical abuse),
(4) In cases of neglect alone, there were fewer cases involving spouse abuse (14% as
opposed to 27% in families of abused children), and (5) Reports of substance abuse were
more frequent among parents of neglected children (72% as opposed to 55% in cases of
physical abuse). Family race and the number of non-parent alleged perpetrators did not
differ significantly between the two groups.

40
Because these findings are based on very small samples, caution must be used in
their interpretation. The number and magnitude of the differences found, however,
indicate the possibility of substantive differences between the characteristics of cases of
physical abuse only and neglect only. It thus logically follows that important differences
may exist in the process of determining appropriate adjudication for the two different
groups.

DISCUSSION
The results of this study provide descriptive information about cases involving
allegations of child abuse and/or neglect cases which are served by the GAL program in
the state of Florida. Such descriptive information about the cases included in the current
study is provided to establish a framework for interpreting and understanding the results
which were obtained. The distribution of demographic, abusive situation, and family
psychological characteristics of the sample are presented in Tables 12 and 13. As noted
above, 59% of the families in the current study were white and 41% were African
American, Hispanic, and other races (34%, 4%, and 3%, respectively). Of the 56 cases
in which information on family income was provided, the mean income was $8,6660 per
year, suggesting that the majority of the families live in poverty.
All interval variables related to the abusive situation and to the family’s
psychological functioning in the study were rated on visual analog scales with values
ranging from 1 to 99. The average level of severity of abuse was 76 (with 99 being the
most severe), indicating that these cases generally represent the most severe incidents of
abuse and neglect. At least one of the parents was the alleged perpetrator in 96% of the
cases, and there existed previous reports of abuse in 50% of the cases. Multiple types
of abuse were reported in 45 % of the cases, and sexual abuse allegations were present
(either alone or in combination with abuse and/or neglect) in 28% of the cases.
41

42
Table 12.
Distribution of Interval Variables
Variable
N
Mean
S.D.
Median
Child Age
90
4.77
4.46
3.00
Family Income
56
8.66
12.61
5.50
Severity of Abuse
94
76.05
21.85
84.00
Parental Psychological
Functioning
86
56.04
20.98
54.25
Parental Warmth and
Support for Child
94
47.33
29.42
48.50
Family Conflict
86
46.53
31.15
47.00
Parental Noncompliance
94
37.12
26.20
33.25
Table 13. Frequencies of Dichotomous Variables
Variable
Family Race
Parent Perpetrator
Multiple Types of Abuse
Previous Reports of Abuse
Sexual Abuse
Spouse Abuse
Substance Abuse
Frequency Percent
White
55
58
Non-White
39
42
Alleged
90
96
Not Alleged
4
4
Alleged
42
45
Not Alleged
52
55
Yes
47
50
No
47
50
Alleged
26
28
Not Alleged
68
72
Alleged
20
21
Not Alleged
74
79
Alleged
61
65
Not Alleged
33
35

43
There were allegations of spouse abuse (in addition to child abuse) in 21 % of the
cases, and allegations of substance abuse were made in 65% of the cases. The average
rating for overall psychological functioning of the family was 56 (with 99 being the
poorest possible level of psychological functioning), and the average rating for parental
warmth and support for the child was 47 (with 99 being the highest possible level of
warmth and support). The average conflict level reported for the families was 47 (with
99 being the highest possible level of conflict), and the average degree of parental
noncompliance with the investigation was 37 (with 99 being the highest possible level of
noncompliance).
One of the major aims of this study was to determine the frequency with which
judges agree with Guardians ad Litem regarding case outcome. Cross-tabular display of
Guardian recommendation and actual case outcome indicated that the two groups agreed
in 79 of the 94 cases, and the positive Kappa statistic indicated that this level of
agreement is beyond what would be expected by chance alone. In addition, results of
a logistic regression analysis indicated that Guardian recommendation for removal is
significantly associated with a higher probability of judicial decision to remove the child
from the home. The results indicate, in fact, that the probability of a judicial decision
for removal is 96% when the Guardian has recommended that the child be removed.
The main objective of the current study was to determine which demographic,
legal, and family psychological variables are most predictive of decisions to remove (or
recommend removal of) children from their homes in cases involving child abuse and
neglect. It is important at this point to note again that all cases included in this study had

44
reached the point of adjudication and therefore likely represent the most severe cases to
enter into the judicial system. Consideration of this population is felt to be important,
however, as it is these cases in which judges are most likely to render decisions.
Interestingly, very few of the variables assessed were found to be significantly associated
with judicial and Guardian decision-making.
In addition, while judges were quite likely to agree with the recommendations of
Guardians, their decisions were significantly associated with somewhat different
information from the Guardians. Review of the cases in which judges and Guardians
disagreed did not reveal any trends which would account for such differences. It is
therefore possible that this discrepancy is due either to chance or to the influence of
another factor (or factors) which was not accounted for in this study.
Overall, information regarding the degree of parental warmth and support for the
child was most predictive of final case outcome. The results of the separate analyses
using Guardians’ recommendations as the dependent variable indicated that the presence
of previous reports of abuse and on the degree of parental warmth and support for the
child are significantly associated with those recommendations. The results of a logistic
regression equation using only these predictor variables, however, indicated that only the
presence of previous reports of abuse or neglect against the child is significantly related
to Guardians’ recommendations for removal. Thus, at least in this sample, recurrent
abuse against a child is the single most predictive factor of Guardians’ recommendation
that he or she be removed from the home.

45
The results of previous studies have indicated that judges are more likely to
remove older children from the home than they are to remove younger children (e.g.,
Katz et al., 1986). In this sample, however, child age had no discernible effect on either
judges’ or Guardians’ decisions. Previous studies have also found that the race of the
family is not significantly associated with decisions of judges. The results of the current
study lend support to that finding, suggesting that racial discrimination on the part of
court personnel is not present in the process of adjudicating cases of child abuse and
neglect.
Several legal variables hypothesized to be related to judicial and Guardian ad
Litem decision-making were not found to significantly predict the outcome of the cases
included in this study. Neither judges’ nor Guardians’ decisions were significantly
associated with the fact that the abuse was very severe or on the fact that multiple types
of abuse were alleged. Interestingly, while the presence of previous reports of abuse was
found to be the most significant predictor of Guardian’s decisions to recommend removal
of children from their homes, this information was not significantly related to judicial
decision-making. Among the variables relating to the psychological functioning of the
family, only the level of the parents’ warmth and support for the child was found to be
significantly associated with decisions regarding case outcome. Information regarding
the psychological functioning of the parents, the parents’ compliance with the
investigation, the degree of family discord, or the presence of allegations of spouse or
substance abuse among the parents was not related to judges’ or Guardians’ decisions.

46
One source of concern which arises from the results of the current study is the
lack of association between judges’ and Guardians’ decisions and the severity level of the
alleged abuse. Florida Statute section 39.464 stipulates that grounds for termination of
parental rights include "egregious abuse," which is defined as "conduct that endangers
the life, health, or safety of the child." Yet visual analysis of the data indicated that a
number of children were returned home in cases where the severity of the abuse was
judged to be very high. If one assumes that children who are very severely abused are
at risk for greater injury in the future (if not at greater risk for more frequent future
abuse), this finding reflects a potentially serious oversight on the part of court personnel
involved in dependency case decision-making. In addition, there was little association
between judges’ decisions and the presence or absence of previous reports of abuse or
neglect against the child—despite the fact that Florida Statute section 39.464 stipulates
that "continuing abuse or neglect" is grounds for termination of parental rights.
Another area of concern is the lack of association between judicial and Guardian
decisions and information regarding the psychological and interpersonal functioning of
the family. The results obtained here differ from those obtained by Meddin (1984) in
her study of child protective direct service workers and supervisors. Subjects in that
study were found to treat the functioning and cooperation of the parents as important
indicators of future risk to the child. Despite considerable evidence that psychological
disturbance of parents (particularly depression), aberrant parenting, and marital discord
frequently lead to abuse and neglect of children and subsequently to poor adjustment of

47
those children (e.g., Lynch and Roberts, 1982; Martin, 1979; Green, 1976), factors such
as these were not found to predict decisions to remove children from their homes.
Review of Florida Statute section 39.467 provides a likely explanation for the
above finding. Judges are required to consider "the love, affection, and other emotional
ties existing between the child and the child’s present parent or siblings"—which they
appear to do, given the significant association between the degree of parental warmth and
support for the child and final adjudication. However, while judges are instructed to
consider "the capacity of the parent or parents to care for the child to the extent that the
child’s health and well-being will not be endangered upon the child’s return home," there
is no specific requirement to consider the psychological functioning of the parents.
Again, given our current knowledge about the detrimental effects of poor parent
psychological functioning on the well-being of children, failure to consider these factors
when determining whether to remove children from their homes would represent a gross
oversight.
Results of comparison of the variables among groups of children who were
physically abused (but not neglected or sexually abused) and those who were neglected
(but not physically or sexually abused) also indicated several potentially important
differences in both case type and outcome—most notably that while Guardians are
somewhat less likely to recommend removal of children in cases of neglect alone, judges
are somewhat more likely to adjudicate removal. Such a finding raises questions
regarding whether the decision-making processes of judges and Guardians differ when
allegations include different types or combinations of abuse or neglect. As a result, it

48
is recommended that future research consider groups of physically abused, sexually
abuse, and neglected children separately rather than risk "washing out" of potential
differences among them.
The current study has identified a number of variables important to judges and
Guardians in determining custody of abused and/or neglected children, as well as a
number of issues which may warrant more attention on the parts of these two groups.
There are, however, several methodological limitations present in this study which
deserve consideration. First, there are limitations which concern the generalizability of
the results obtained here. Despite considerable effort, the researchers were unable to
collect data on every case referred to the GAL program during the period of study.
Some of the judicial circuits in the state of Florida were more willing to cooperate than
were others, and even the most cooperative were able to provide data only from
Guardians willing to complete our forms on a volunteer basis. Our sample is therefore
not random and may not accurately represent the population of GAL referrals in the state
of Florida. Additionally, the current results are based on a relatively small sample, and
some of our findings may therefore be sample specific. Caution must thus be used when
generalizing the results obtained in this study to other populations. In addition, the
sample used in this study did not include a wide distribution of family income levels or
child ages. Because the sample included mostly very low income families with young
children, generalization of these findings to populations with different demographic
characteristics is not advised, as judges and Guardians may well weigh information
differently in cases involving wealthy families or those with older children.

49
While federal laws regarding cases involving allegations of child abuse or neglect
provide some degree of uniformity in determining case disposition, it may be that some
states have legal guidelines and political climates which differ from the state of Florida.
It is therefore recommended that the results obtained here not be generalized to other
geographic regions prior to further study in those areas. Another aspect of the current
sample which deserves consideration is that the cases involved both initial dependency
hearings and termination of parental rights hearings. Thus, some of the children were
removed from their homes on a temporary basis, while others were permanently
removed. Because court personnel may use somewhat different criteria when deciding
whether parental rights should be severed, it is recommended that additional research be
conducted on these two groups independently of each other.
Future research is clearly needed to assess the generalizability of the current
findings to other populations. It is also recommended that a similar analysis be
conducted using a larger sample of a comparable population in order to replicate these
findings and to determine whether the process of Guardian ad Litem and judicial
decision-making changes over time.
Further research is also needed to clarify the results obtained in the current study.
One rather puzzling finding of this study was that while judges tend to agree with the
recommendations of Guardians, they appear to place a somewhat different emphasis on
the variables they use to determine the final adjudication on the cases they consider.
This finding suggests that while judges rely on case information provided to them by the
Guardians, they may not rely heavily on the Guardians’ recommendations. Future

50
research using Guardian recommendation as an independent variable to predict judicial
decisions is thus warranted to further clarify this situation. In addition, it is
recommended that future studies be conducted which consider abused children and
neglected children as separate groups in order to determine whether important differences
exist in the decision-making of judges and Guardians between the two groups.
Another area requiring further research is the lack of consideration given to
psychological aspects of the family in making decisions regarding removing children
from their homes. Potential explanations for this finding include the possibility that
judges and Guardians are unaware of the importance of such factors on the short- and
long-term well-being of the children involved, that they are not adequately educated
about how to assess the information, and/or that they do not feel that the consideration
of such information is warranted under the legal guidelines for determining case
disposition. Future research aimed at elucidating these or other reasons is therefore
suggested.

REFERENCES
Alter, C.F. (1985). Decision-making factors in cases of child neglect. Child Welfare.
64, 99-111.
Brown, A., & Finkelhor, D. (1986). Impact of child sexual abuse: A review of the
research. Psychological Bulletin. 99, 66-77.
Daro, D. (1988). Confronting child abuse: Research for effective program design.
New York: The Free Press.
Eckenrode, J., Powers, J., Doris, L, Munsch, J., & Bolger, N. (1988). Substantiation
of child abuse and neglect reports. Journal of Consulting and Clinical
Psychology. 56, 9-16.
Eskin, M. (1980). Child abuse and neglect: A literature review and selected biblio¬
graphy. Washington, DC: U.S. Department of Justice.
Faller, K.C. (1988). Decision-making in cases of intrafamilial child sexual abuse.
American Journal of Orthopsychiatry. 58, 121-128.
Finkelhor, D. (1983). Removing the child—prosecuting the offender in cases of sex¬
ual abuse: Evidence from the National Reporting System for Child Abuse and
Neglect. Child Abuse and Neglect. 7, 195-205.
Finkelhor, D. (1984). Child sexual abuse: New theory and research. New York:
Free Press.
Goldman, J., Albanese, I., Graves, L., McDanal, C.S., Sorenson, E., & Ward, M. (in
preparation). Reliability of Guardian ad Litem ratings on psychological issues in
cases of child abuse and neglect.
Green, A. (1976). A psychodynamic approach to the study and treatment of child-
abusing parents. Journal of Child Psychiatry. 15. 414-429.
Houck, G.M., & King, M.C. (1989). Child maltreatment: Family characteristics and
developmental consequences. Issues in Mental Health Nursing. 10, 193-208.
51

52
Hosmer, D.W, & Lemeshow, S. (1989). Applied Logistic Regression. New York:
John Wiley & Sons.
Katz, M.H., Hampton, R.L., Newberger, E.H., Bowles, R.T., & Snyder, J.C. (1986).
Returning children home: Clinical decision-making in cases of child abuse and
neglect. American Journal of Orthopsychiatry. 56, 253-262.
Knudsen, D.D. (1988). Child protective services: Discretion, decisions, dilemmas.
Springfield, IL: Charles C. Thomas.
Lamphear, V.S. (1985). The impact of maltreatment on children’s psychosocial
adjustment: A review of the research. Child Abuse and Neglect. 9, 251-263.
Landau, H.R. (1982). The abused and neglected child. Practising Law Institute.
Lynch, M.A., & Roberts, J. (1982). Consequences of Child Abuse. London, England:
Academic Press.
Martin, H.P. (1979). Child abuse and child development. Child Abuse and Neglect.
3, 415-421.
Meddin, B.J. (1984). Criteria for placement decisions in protective services. Child
Welfare. 63, 367-373.
National Center on Child Abuse and Neglect (1988). Study findings: Study of na¬
tional incidence and prevalence of child abuse and neglect. Washington, DC:
U.S. Department of Health and Human Services.
Radbill, S.X. (1968). A history of child abuse and infanticide. In B.E. Heifer & C.H.
Kempe (eds.), The battered child. Chicago: University of Chicago Press.
Runyan, D.K., Gould, C.L., Trost, D.C., & Loda, F.A. (1981). Determinants of foster
care placement for the maltreated child. American Journal of Public Health. 71,
706-711.
Russell, D.E.H. (1983). The incidence and prevalence of intrafamilial and extra-
familial sexual abuse of female children. Child Abuse and Neglect. 7, 133-146.
Sloan, I.J. (1983). Child abuse: Governing law and legislation. New York: Oceana
Publications, Inc.
Strauss, M.A., & Gelles, R.J. (1986). Societal change and change in family violence
from 1975 to 1985 as revealed by two national surveys. Journal of Marriage and
the Family. 48, 465-479.

53
Wasserman, S., & Rosenfeld, A. (1986). Decision-making in child abuse and neglect.
Child Welfare. 65, 515-529.
Whitcomb, D. (1988). Guardians Ad Litem in criminal courts. Washington: U.S.
Department of Justice.
Wiig, J.K. (1982). Functions of the Guardian ad Litem in child abuse and neglect
proceedings—Los Angeles Juvenile Court. In H. Landau (ed.), The abused and
neglected child. Practising Law Institute.

APPENDIX A
GAL QUESTIONNAIRE FOR DEPENDENCY CASES
Thank you for your time and support of our efforts to collect the following
information on dependency cases. Our goal is to provide information which will aid
the GAL program in its role as the advocate for children in dependency cases.
The questionnaire is divided into three broad sections. This first section deals with
basic demographic and legal information of your case. This information is similar to
that found in your case records.
The second sections requests information on individual and family psychological
issues. These questions may be easily answered after your interviews with the child’s
family. This section requires that you mark on a line to indicate your rating of the
intrafamilial relationships. Some of these questions require a single mark on the line,
such as for conflict between the two parents:
CONFLICT BETWEEN THE TWO PARENTS:
Very Little Very Much
While other questions require you to make two ratings on each line, one for the
mother-child relationship and one for the father-child relationship:
CONFLICT BETWEEN EACH PARENT AND THE CHILD(ren) who was
reportedly abused or neglected, with an M, F, SM, and an SF.
M F
Very Little Very Much
Finally, the third section surveys information from the disposition hearing on the case
in terms of your recommendations to the court and the court’s final determination.
Please return this form when completed to your GAL program’s secretary.
Thank you again for your help.
54

55
Dependency Case
DEMOGRAPHIC AND LEGAL INFORMATION
Who is filling out this form? Guardian Coordinator/Director Other
Date of this hearing? Judge Code:
Circuit: County:
Circle any immediate family members who have not been interviewed by the
GAL: (Circle all that apply)
Mother Father Child Other (specify)
Stepmother Stepfather Grandparent
Are the parents represented by counsel: Yes No
Report for all family members living with the child victim (and the alleged
perpetrator if he/she does not live with the child) the following information:
Race: C=caucasian, H=hispanic, B=black, 0=other
V/P: V=victim of abuse/neglect, P=alleged perpetrator
Relation: (to the victim(s) P=parent, SP = step parent, GP=grandparent, S = sibling,
BF=boy/girlfriend of the parent, FP = foster parent, AP=adoptive parent,
ST=stranger, 0=other
AGE SEX RACE V/P RELATION
1.
2.
3.
4.
5.
6.
For each parent report the following information:
Yearly Years of
Income Education Age Occupation
MOTHER
FATHER
STEPMOTHER
STEPFATHER

56
Has one or more of the parents lost their jobs within the last year? Yes No
Has the family moved within the last year? Yes No
Circle the one category into which the reported child victim falls:
Born Prematurely Physically Handicapped Mentally Retarded
Neither Unknown
Which parent(s) may have difficulty caring for their child(ren) due to a physical
handicap, serious illness, or injury?
Mother Stepmother Father Stepfather Neither
For the following questions, place a check mark in the box corresponding to any
alleged or substantiated problem for any parent(s):
ALLEGATIONS SUBSTANTIATED
M
0
T
H
E
R
S M
T 0
E T
P H
E
R
F
A
T
H
E
R
S F
T A
E T
P H
E
R
N
E
I
T
H
E
R
M
0
T
H
E
R
S M
T 0
E T
P H
E
R
F
A
T
H
E
R
S F
T A
E T
P H
E
R
N
E
I
T
H
E
R
Alcohol Abuse
Drug Abuse
Hitting the Child
Child Physical Abuse
Child Sexual Abuse
Child Emotional Abuse
-
Child Neglect
Spousal Emotional
Abuse
Spousal Physical Abuse
Previous Reports of
Child Abuse or Neglect

57
INTERPERSONAL AND FAMILY ISSUES: Please evaluate items 1-9 from your
interviews with the family members. Place a mark on the line to indicate your rating
of the family. Also, please circle the number of each question if you have included
information on that issue in your report to the court.
1. Who does the non-abusing caretaker SUPPORT:
i
Child Victim
i
Reported Abuser
CONFIDENCE in your rating.
i
i
i
None
i
Total
2.
SEVERITY of the abuse/neglect:
i
i
i
Very Slight
i
Very Severe
CONFIDENCE in your rating.
i
i
i
None
i
Total
3.
CHILD’S DISTRESS over the current family situation:
i
i
i
Very Little
i
Very Much
CONFIDENCE in your rating.
i
i
i
None
i
Total
4.
CONFLICT BETWEEN THE TWO PARENTS:
i
i
i
Very Little
i
Very Much
CONFIDENCE in your rating.
i
i
i
None
i
Total

58
5. PARENTS’ WILLINGNESS TO REPORT FAMILY PROBLEMS in
general:
i i
i i
None Completely
CONFIDENCE in your rating.
i i
i i
None Total
6. Amount the FAMILY FAVORS FAMILY ACTIVITIES:
i i
i i
Too Little About Right Too Much
CONFIDENCE in your rating.
i i
i i
None Total
7. Amount that the family members CLARIFY AND/OR SPEAK FOR other
family members.
i i
i i
Very Little Very Much
CONFIDENCE in your rating.
None Total
8. DEGREE TO WHICH THE PHYSICAL NEEDS OF THE CHILD(ren)
were met, prior to the recent report of abuse/neglect.
i i
i i
None Total
CONFIDENCE in your rating.
i i
i i
None Total
These needs were primarily met by (circle one):
Mother Father Both Other (specify)

59
9. DEGREE TO WHICH THE EMOTIONAL NEEDS OF THE CHILD(ren)
were met, prior to the recent report of abuse/neglect.
i i
i i
None Total
CONFIDENCE in your rating.
i i
i i
None Total
These needs were primarily met by (circle one):
Mother Father Both Other (specify)
For items 10-16, use the following letters that apply to this family to represent your
evaluation of the parents on these issues. Place a mark on the line to represent your
rating of each parent and label each mark with one of the following letters. Please
make sure that you make separate ratings on each of them, including your ratings of
the CONFIDENCE you have in your rating on each of them.
M=Mother, SM=Stepmother, F = Father, SF=Stepfather
10. CONFLICT BETWEEN EACH PARENT AND THE CHILD(ren) who
was reportedly abused or neglected, with an M, F, SM, and SF.
i i
i i
Very Little Very Much
For each parent, is there as much conflict with other children?
Yes No Yes No
Father: Mother:
Stepfather: Stepmother:
CONFIDENCE in each of your ratings with an M, F, SM, and SF.
i i
i i
None Total

60
11. WARM AND POSITIVE FEELINGS FOR THE CHILD(ren) who was
reportedly abused or neglected held by each parent with an M, F, SM,
and SF.
I !
Very Little Very Much
CONFIDENCE in each of your ratings with an M, F, SM, and SF.
! ¡
None Total
12. INVOLVEMENT of each parent with the child(ren)’s daily lives (e.g., peer
relationships, child’s feelings, minor illnesses) with an M, F, SM, and SF.
¡ I
No Involvement Excessive Involvement
CONFIDENCE in each of your ratings with an M, F, SM, and SF.
i i
i i
None Total
13. EACH PARENT’S RELIANCE ON THE CHILDREN FOR
COMPANIONSHIP, COMFORT, AND PROBLEM-SOLVING with an M,
F, SM, and SF.
¡ I
Very Little Very Much
CONFIDENCE in each of your ratings with an M, F, SM, and SF.
i i
i i
None Total
14.
PSYCHOLOGICAL ISSUES for each parent with an M, F, SM, and SF.
Lonely
Highly
Slightly

61
Depressed
1
Slightly
,
Highly
Anxious
i
i
i
Slightly
i
Highly
Irritable
i
i
i
Slightly
i
Highly
CONFIDENCE in each of your ratings with an M, F, SM, and SF.
i i
i i
None Total
15. SATISFACTION WITH SOCIAL SUPPORT that each parent receives for
any problems from relatives, friends, coworkers, and community agencies
with an M, F, SM, and SF.
i i
i i
Inadequate Excellent
CONFIDENCE in each of your ratings with an M, F, SM, and SF.
i i
i i
None Total
16. EACH PARENT’S RESPONSE TO YOU in general with an M, F, SM,
and SF.
i i
i 1
Very Friendly Hostile/Rejecting
CONFIDENCE in each of your ratings with an M, F, SM, and SF.
i i
i i
None Total
17. If either parent was abused/neglected as a child, report the types of that
abuse (P=physical, S=sexual, E=emotional, N=Neglect, and N/A=none)
Mother Father Stepmother Stepfather

62
18. Check the following problems if any child was reported to exhibit the
problem:
Child verbally fights with parents
Child physically fights with parents
Child physically fights with siblings
Child verbally or physically fights with non-family members
Child is more promiscuous than would be expected for
his/her age
Child has been absent from school for more than 2 weeks
in the last year, and it has not been due to any
medical illness
Child has run away from home frequently in the past
Child steals from family members or others
Child is withdrawn and does not participate in normal daily
activities, such as schoolwork or family activities
Child refuses to do what his parents tell him to do,
such as going to school or doing household chores
PLEASE CHECK TO MAKE SURE THAT YOU HAVE CIRCLED THE
NUMBER OF THE QUESTIONS WHICH CONTAIN INFORMATION YOU
REPORTED TO THE COURT
Report the total number and dates of any past report of abuse or neglect and for
the present report. (If no abuse write in N/A; if the neglect was chronic, use C in
place of the number of occurrences.)
DATES TOTAL NUMBER
PAST
PRESENT

63
Report whether criminal charges are being brought against the alleged
perpetrator for abuse/neglect in this case and previously.
This case: YES NO Previously (how many):
DISPOSITION
Does the alleged perpetrator accept responsibility to the abuse/neglect?
YES NO
Indicate who requested a mental health evaluation for any family members, and
who completed it:
REQUESTED BY:(l=judge 2 = Guardian 3=Attorney for Parent 4=HRS 5=N/A)
COMPLETED BY:(1 =Clinical Psychologist 2=Psychiatrist 3=Social Worker
4=Other Psychologist 5= Other 6=Not Completed)
Requested By Completed By
For the Mother
For the Father
For the Stepmother
For the Stepfather
For the Child(ren)
Circle any family member(s) for whom the GAL has recommended therapy:
Mother Stepmother Father Stepfather Children None
Do the parents admit dependency? YES NO
Please circle type of case:
Initial Dependency Termination of Parental Rights

64
Check any of the following options it they were being considered, then check
which option was recommended by the Guardian, and which option was
adjudicated.
Considered Guardian Adjudicated
Perpetrator Removed From the
Home
Parents Retain Custody With
Supervision
Temporary Custody Awarded to
a Relative
Temporary Foster Placement
Child Placed in a Treatment
Facility
Placement for Adoption
Child Placed in a Detention
Facility
Other (specify)
Not Dependent and Returned
Home

APPENDIX B
CORRELATIONS AMONG INDEPENDENT VARIABLES
Table B-l. Estimated Correlations Among Demographic Variables Entered
in Logistic Regression Equation Predicting Judicial Decisions
Child Age Family Race
Child Age 1.00 0.16
Family Race 0.16 1.00
Table B-2. Estimated Correlations Among Demographic Variables Entered
in Logistic Regression Equation Predicting Guardian Recommendations
Child Age Family Race
Child Age 1.00 0.18
Family Race 0.18 1.00
65

66
Table B-3. Estimated Correlations Among Legal and Case
Variables Entered in Logistic Regression Equation Predicting
Judicial Decisions
Severity of
Abuse
Multiple Types
of Abuse
Prior Reports
of Abuse
Severity of Abuse
LOO
-0.25
-0.06
Multiple Types of
Abuse
-0.25
1.00
-0.21
Prior Reports of
Abuse
-0.06
-0.21
1.00
Table B-4. Estimated Correlations Among Legal and Case
Variables Entered in Logistic Regression Equation Predicting
Guardian Recommendations
Severity of
Abuse
Severity of Abuse
1.00
Multiple Types of
-0.26
Abuse
Prior Reports of
-0.12
Abuse
Multiple Types
Prior Reports
of Abuse
of Abuse
-0.26
-0.12
1.00
-0.18
-0.18
1.00

67
Table B-5. Estimated Correlations Among Family Psychological Variable Entered in
Logistic Regression Equation Predicting Judicial Decisions
Psychological
Functioning
Parental
Warmth
Family
Conflict
Parental
Non-
compliance
Substance
Abuse
Spouse
Abuse
Psychological
Functioning
1.00
0.09
0.09
-0.23
-0.20
-0.22
Parental
Warmth
0.09
1.00
0.33
0.43
-0.16
0.07
Family
Conflict
0.09
0.33
1.00
-0.01
-0.24
-0.07
Parental Non-
compliance
-0.23
0.43
-0.01
1.00
0.04
-0.01
Substance
Abuse
-0.20
-0.16
-0.24
0.04
1.00
-0.20
Spouse Abuse
-0.02
0.07
-0.07
-0.01
-0.20
1.00
Table B-6. Estimated Correlations Among Family Psychological Variable Entered in
Logistic Regression Equation Predicting Guardian Recommendations
Psychological
Functioning
Parental
Warmth
Family
Conflict
Parental
Non-
compliance
Substance
Abuse
Spouse
Abuse
Psychological
Functioning
1.00
0.17
0.03
-0.20
-0.23
-0.01
Parental
Warmth
0.17
1.00
0.29
0.39
-0.09
-0.01
Family
Conflict
0.03
0.29
1.00
-0.06
-0.18
-0.11
Parental Non-
compliance
-0.20
0.39
-0.06
1.00
0.14
-0.01
Substance
Abuse
-0.23
-0.09
-0.18
0.14
1.00
-0.17
Spouse Abuse
-0.01
-0.01
-0.11
-0.01
-0.17
1.00

68
Table B-7. Estimated Correlations Among Significant Predictor Variables
Entered in Logistic Regression Equation Predicting Guardian
Recommendations
Prior Reports
Parental Warmth
of Abuse
and Support
Prior Reports
of Abuse
1.00
0.05
Parental Warmth
and Support
0.05
1.00

APPENDIX C
RATE OF REMOVAL BY TYPE OF ABUSE
Table C-l. Cross-Tabular Analysis of Removal Rate by Presence of Neglect
Allegation
Judicial Decision
Frequency
Total Percent
Not Removed
Removed
Total
Row Percent
Column Percent
Not Neglected
6
13
19
6.38
13.83
20.21
31.58
68.42
37.50
16.67
Neglected
10
65
75
10.64
69.15
79.79
13.33
86.67
62.50
83.33
Total
16
78
94
17.02
82.98
100.00
69

70
Table C-2. Cross-Tabular Analysis of Removal Rate by Presence of Physical
Abuse
Allegation
Frequency
Total Percent
Row Percent
Column Percent
Not Removed
Judicial Decision
Removed
Total
Not Physically
6
39
45
* Abused
6.38
41.49
47.87
13.33
86.67
37.50
50.00
Physically Abused
10
39
49
10.64
41.49
52.13
20.41
79.59
62.50
50.00
Total
16
78
94
17.02
82.98
100.00
Table C-3. Cross-Tabular Analysis of Removal Rate by Presence of Sexual
Abuse
Allegation
Judicial Decision
Frequency
Total Percent
Not Removed
Removed
Total
Row Percent
Column Percent
Not Sexually
9
59
68
Abused
9.57
62.77
72.34
13.24
86.76
56.25
75.64
Sexually Abused
7
19
26
7.45
20.21
27.66
26.92
73.08
43.75
24.36
Total
16
78
94
17.02
82.98
100.00

APPENDIX D
DESCRIPTIVE STATISTICS FOR VARIABLES IN CASES OF
PHYSICAL ABUSE ALONE AND NEGLECT ALONE
Table D-l. Distribution of Interval Variables in Cases of Physical Abuse Only
and Neglect Only
Physical Abuse Only
Neglect Only
T-Test
Variable
N
Mean
S.D.
N
Mean
S.D.
Value
Child Age
11
4.36
3.59
35
3.60
3.34
0.63
Family Income
6
13.17
9.00
21
4.95
7.30
2.05
Severity of Abuse
11
53.45
24.27
36
74.14
21.04
-2.55*
Parental Psychological
Functioning
10
48.66
18.65
30
58.99
23.34
-1.42
Parental Warmth and
Support for Child
11
57.45
18.83
36
54.94
31.06
0.33
Family Conflict
11
54.18
17.06
29
37.54
32.64
2.09*
Parental Noncompliance
11
37.36
23.63
36
31.89
28.78
0.64
* p < .05
71

72
Table D-2. Frequencies of Dichotomous Variables in Cases of Physical Abuse
Only and Neglect Only
Physical Abuse Only Neglect Only
Variable Frequency Percent Frequency Percent
Adjudication
Removed
8
73
32
89
Not Removed
3
27
4
11
Guardian
Remove
9
82
27
75
Recommendation
Do Not Remove
2
18
9
25
Family Race
White
5
45
16
44
Non-White
6
55
20
56
Parent Perpetrator
Alleged
Not Alleged
10
1
91
9
35
1
97
3
Previous Reports of
Yes
3
27
17
47
Abuse
No
8
73
19
53
Spouse Abuse
Alleged
Not Alleged
3
8
27
73
5
31
14
86
Substance Abuse
Alleged
6
55
26
72
Not Alleged
5
45
10
28

73
APPENDIX E
PEARSON THREE-WEEK TEST-RETEST CORRELATIONS
Table E-l. Pearson Three-Week Test-Retest Correlations of Psychological
Variables
Variable
r
Parental Conflict
.52*
Maternal Warmth
*
OO
Paternal Warmth
.49*
Maternal Depression
.36
Paternal Depression
.18
Maternal Anxiety
.68***
Paternal Anxiety
.21
Maternal Irritability
.60**
Paternal Irritability
.09
* p< .05
** p< .005
*** p< .001

BIOGRAPHICAL SKETCH
Cynthia McDanal is a psychology associate in the state of Maryland. She
currently works under supervision in a private practice conducting psychotherapy for
children and performing neuropsychological evaluations to rule out learning
disabilities and attention disorders in children, adolescents, and adults. Upon
completion of the requirements for the degree of Doctor of Philosophy, Cynthia will
pursue certification as a licensed clinical psychologist in Maryland and the District of
Columbia. She graduated magna cum laude from Stetson University in 1987 and
earned a Master of Science degree in Clinical and Health Psychology from the
University of Florida in 1990.
74

I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and
quality, as a dissertation for the degree of Doctor of Philosophy.
â–  ^
Jacqueun Goldman, Chair
Professor of Clinical and Health Psychology
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and
quality, as a dissertation for the degree of Doctor of Philosophy.
% SJL
James^ohnson
Professor of Clinical and Health Psychology
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and
quality, as a dissertation for the degree of Doctor of Philosophy.
Michael Robinson
Associate Professor of Clinical and Health
Psychology
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and
quality, as a dissertation for the degree of Doctor of Philosophy.
ephen Boees
Stephen Boggs
Associate Professor of Clinical and Health
Psychology
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and
quality, as a dissertation for the* degree of Doctor of Philosophy.
Stevemwitiis
Professor of Law

This dissertation was submitted to the Graduate Faculty of the College of
Health Related Professions and to the Graduate School was accepted as partial
fulfillment of the requirements for the degree of Doctor of Philosophy.
April, 1994 AsL-P
Dean, College of Health Related Professions
Dean, Graduate School

UNIVERSITY OF FLORIDA
3 1262 08554 5324